Natural Resources Mgmt Plan for the UH Mgmt Areas on Mauna Kea (TPD0909014)

2.1.2          Surface Features and Soils

From 9,000 ft (2,743 m) upward Mauna Kea is a dry environment with much of its surface covered with rock that has been moderately altered by biogeochemical reactions. Geological changes occur here at  rates at either end of the spectrum: explosively fast and glacially slow. Due primarily to low rates of precipitation and a cool temperature regime, biogeochemical weathering of rocks is very slow and predominately mechanical in nature. This environmental setting is the primary reason why so much of the area does not contain soils, and why disturbances to surface features remain visible for long periods.

2.1.2.1         Ground Surface

Mauna Kea Science Reserve: The higher elevations of the MKSR have been classified as Very Stony Land or Cinder Land and are composed entirely of post-shield volcanic material.¹² A combination of coarse gravel to cobble-sized pieces of cinder and lava covers the ground surface of most of the summit area. Cinder is the dominant component of the cinder cones forming the summit and it is this debris that makes up the cones outer slopes (Porter 1972b; Wood 1980; Wolfe et al. 1997). Areas that were capped by lava flows at the summit plateau are relatively flat and dark grey to black in color, with a low albedo (surface reflectivity); ‘a‘ā flows deposited before glaciers covered the summit area later lost their original craggy surfaces when glaciers that slid over them. Exposed outcrops of moraine and till from these glacial icecaps are composed of poorly sorted cobbles, rocks, and boulders (Wolfe et al. 1997). Rills and small gullies incising the flanks of Pu‘u Poli‘ahu, Pu‘u Waiau, and other cones are indicative of a naturally altered layer that is less porous and more prone to erosion than cones that do not contain less porous layers of ash or other material (Wolfe et al. 1997).

Lava flow outcrops are scattered throughout the MKSR, poking out from layers of cinder, till, and a slowly increasing coating of finer particles as one descends the mountain. Many of these outcrop formations are the result of lava erupting under the icecaps of the glacial periods (see Section 2.1.1.4.2).

Hale Pōhaku and Mauna Kea Summit Access Road: The ground surface of the lower-elevation Hale Pōhaku facilities is covered with small particles that are several centimeters deep in some locations. The slopes of cinder cones in the vicinity of Hale Pōhaku are comprised of larger fragments than those of the summit and have been dusted with fine grained particles. The lowest lying areas are littered with cinder and small lava rocks.

Several trails at Hale Pōhaku and within the MKSR have been used frequently enough that they have  been etched into the ground surface. Only four trails are monitored by rangers: the Lake Waiau trail, the Mountain trail, the trail up Pu‘u Poli‘ahu, and the Summit Trail (see Section 3.1.3.3). Any new trails created by visitors are covered up by Ranger staff as soon as they are observed. Very little original  surface material remains on the more frequented trails, most having been crushed or kicked to the sides by hikers. During a site visit it was observed that trails are covered with a slippery layer of fine sediment, likely generated from the crushed cinder. In addition, hikers were seen walking in adjacent areas where footing is more stable, increasing the impact area of the trail.

12 See information on soils from the Natural Resources Conservation Service at: http://www.hi.nrcs.usda.gov/soils.html.

Other areas of disturbed surface features include areas adjacent to the Summit Access Road and the drainage networks around Hale Pōhaku. Along the road, culvert inlets are often filled with coarse and fine sediment, while at culvert outlets it appears that over time water and sediment have slowly cut into the mountain, forming larger and larger gullies. Similar impacts are evident along the edges of the parking areas, on adjacent land.

2.1.2.2         Soils

The process of soil formation, or pedogenesis, involves the interaction of five variables: parent rock material, time, climatic conditions, presence of vegetation, and topography (Brady and Weil 2000). While any of these five can be a limiting factor in soil formation, in the summit region of Mauna Kea, three of the five are limiting factors: the dry climate, a general lack of vegetation, and the topography. Constituents of pedogenically (naturally) derived soils such as ash can be valuable stratigraphic markers of volcanic activity (Porter 1973b). Because they form over long periods, soils may also be valuable environmental indicators and chemical databases of the mountain’s past. The following review focuses on the classification, locations and characteristics of the soils of Mauna Kea.

2.1.2.3         Soil Classification

The Soil Survey of the Island of Hawaii, State of Hawaii, published by the Department of Agriculture Natural Resources Conservation Service (NRCS), which houses the national soil survey, does not list any soils at the summit of Mauna Kea (Sato et al. 1973).¹³ However, formations that may be considered soils, that have soil-like properties, or both, have been found within the summit region (see Figure 2.1-13). These pockets of soil-like material have been classified as very young Andisols-Aridisols (lava) and Andisols-Entisols due primarily to the volcanic ash, cinders, and lava constituents available to create a soil (Juvik and Juvik 1998; Hotchkiss et al. 2000).

Deposits of volcanic lavas, ash, glacial till, and other materials have been weathered in-situ, making them soil-like. A few of these formations found at the summit include the potentially hydrothermally altered subsurface layers within Pu‘u Wēkiu and other cones (Ugolini 1974; Fan 1978), the lake sediments found at the bottom of Lake Waiau (Woodcock et al. 1966), and lower-elevation ash paleosols (soils formed on a past landscape) formed long ago, during long periods of volcanic rest (Porter 1997a; Sheldon 2006). Many of these have since been buried by more recent eruptions (Porter 1997a; Sheldon 2006). Contributions from aeolian sources are also likely, as are contributions from glaciers in the form of till (Ugolini 1974).

Due perhaps to events associated with lava-ice interaction and subsequent hydrothermal alteration, as well as glacial conditions, Ugolini (1974) suggests that pedogenesis has occurred within the upper 6 in (15 cm) of a soil profile found within Pu‘u Wēkiu. Below this depth, a clay-rich soil with secondary minerals is believed to have been altered in-situ, indicating that at some locations in the region processes related to soil formation are occurring within the subsurface (Ugolini 1974). Differences in texture and mineralogy observed at Pu‘u Waiau and other sites on the north side of the summit may indicate a more advanced stage of alteration due to the presence of clays and “X-ray amorphous colloids over crystalline material,” both of which represent alteration by hydrothermal processes (Ugolini 1974). The material of these and other cones is much more susceptible to weathering, its reduced permeability being responsible for the retention of water within Pu‘u Waiau and the persistence of Lake Waiau (Woodcock 1980; Wolfe et al. 1997). It has been suggested that the less porous substrate was formed by explosions caused by lava-ice interaction in an at least partially submerged glacial environment (Ugolini 1974; Porter 1979a; Woodcock 1980). A differing theory suggests that alteration of the substrate occurred not from lava-ice interaction, but through the effect of hot, sulfur-bearing gasses and hot water or steam percolating through the cones (Wolfe et al. 1997). Hydrothermal alteration of the substrate at these and other neighboring sites was also noted by Ugolini (1974) and Fan (1978). Paleosols are mentioned within the literature, and Porter (1997a) has conducted an extensive analysis of the late Pleistocene aeolian sediments. However, because soils that may have developed within the summit region have since been covered by flows, most exposed paleosols are not within the MKSR but at lower elevation construction sites and stream outcroppings.

Two buried soils (Humuula and Hookomo) separated by three distinct tephra layers were identified by Porter (1973b) just below Hale Pōhaku, around 9,186 ft (2,800 m). Other paleosols found by Porter, also at lower elevations, were considered “developed,” formed within dark-yellowish-brown loess and having soil horizons 8–12 in thick (20–30 cm) (Porter 1997a).

Below an elevation of 9,000 ft (2,743 m), three soil series have been identified by the NRCS: the Huikau Series (6,000 – 9,000 ft (1,829 – 2,743 m)), the Apakuie Series (5,500 – 8,000 ft (1,676 – 2,438 m)), and the Hanipoe Series (5,000 – 6,500 ft (1,524 – 1,981 m)) elevation (Sato et al. 1973; Wolfe and Morris 1996a). Two soils have been identified within these soil series; Andisols-Aridisols and Andisols-Entisols (Juvik and Juvik 1998). All three soil series are moderately to excessively well drained and consist of very fine to loamy sands on slopes that vary between gentle, at the lower elevations, to steeper slopes, at 9,000 ft (2,743 m) (Sato et al. 1973). As described within the NRCS soil survey, the lower Hanipoe soils have moderately rapid permeability with minimum erosion hazard and support a wide diversity of vegetation. Further upslope the Apakuie soils are more permeable and thus have a smaller erosion hazard. Huikau soils of the higher elevations are very permeable and due to stronger winds associated with these elevations have a high capacity for aeolian erosion.

2.1.2.4         Soil Surveys

The NRCS soil survey of the Island of Hawai‘i is undergoing a complete update, which is expected to be completed by 2011 (Jasper 2008). This survey is public, and information is available through the NRCS website.¹⁴ Other relevant work includes soil nutrient studies and site investigations that helped to identify ten habitat types based upon soil profiles found during work on Mauna Kea (Balakrishinan and Mueller- Dombois 1983). Additionally, the known limitations for soil development within the summit region were used to help create models that may lead to a better understanding of past ecosystem development and in modeling potential environmental conditions under a changing climate regime (Hotchkiss et al. 2000).

2.1.2.5         Threats to Surface Features and Soils

Threats to the existing soils and soil-like features of the UH Management Areas include the consistent displacement of particulates due to use of off-road areas by hikers and vehicles and any changes to the existing hydrogeology. While processes of erosion are natural occurrences, with increased human presence and associated increased use of the off-road areas comes a greater potential for the movement of materials through both natural and anthropogenic mechanisms. For example, as water is most likely one of the primary forces moving finer sediments downhill, any activity that alters existing hydrology will affect how much sediment is moved, how far it is moved, and the amount of impact the movement will have.

2.1.2.6         Soil Information Gaps

The following information gaps regarding the soils of the UH Management Areas have been identified through review of the literature:

1. Soil Components and Movement The NOAA Mauna Loa Observatory monitors air quality for a number of variables, one of which is the concentration of airborne particulates. Some of the aeolian debris found at the observatory, is dust originating in Asia and is seen every year (Barnes 2008). It is possible that some of this dust has been deposited in Lake Waiau and may change its chemistry and the composition of the sediment layer of its bottom. Other aeolian debris that gets suspended in the airshed and deposited across the region is likely fine materials generated from the exposed surfaces of Mauna Kea.

2.1.3          Hydrology

The following review focuses on the hydrology of Mauna Kea, including the various types of water inputs and their sources, surface and subsurface hydrologic pathways, water resources, and water quality.

2.1.3.1         Water Budget Analysis

A hydrologic cycle describes the movement of water on, above and below the earth’s surface. To understand Mauna Kea’s hydrologic cycle and effectively manage its components, it is necessary to know the spatial distribution of precipitation inputs. Spatial distribution is also needed to calculate a water budget analysis, which is a hydrologic assessment conducted to account for the inputs and losses and to identify flow paths and the fate of water in a given area. In general, a water budget considers inputs and losses. For Mauna Kea, inputs come in the form of precipitation and losses occur through infiltration, evapotranspiration, and sublimation.¹⁵

Primary water inputs to the hydrologic cycle of Mauna Kea are rainfall and snow, and to a lesser extent fog condensation (see Section 2.1.3).¹⁶ Anecdotal evidence and published literature agree that water input from rain and snow varies from year to year and that the range can be considerable. Snow’s contribution to the total precipitation of the upper slopes and summit area was found to be significant (Ehlmann et al. 2005).

 15 Infiltration is the process by which water penetrates into the ground surface. A portion of the water is tied up in the soil and may be extracted by plant roots, and a portion flows deeper until it encounters the groundwater. Evapotranspiration is the process by which water enters the atmosphere by evaporation and plant transpiration. The process by which water in its solid phase is transformed directly to vapor phase without first passing through the liquid phase.

16 On Mauna Kea, fog drip is associated with vegetated areas below 9,000 ft (2,743 m) and is not a contributing source of water for upper elevation watersheds(Arvidson 2002).

On Mauna Kea, above 9,000 ft (2,743 m), mean annual precipitation is low (see Section 2.1.4) and evaporation levels are high. Rates of pan evaporation¹⁷ have been quantified by at least two researchers during the past 30 years, with estimates of 70 inches (178 cm) per year (Ekern and Chang 1985). Although the pan evaporation estimate is not the actual evaporation, its high value means that meteorological conditions of the summit area are conducive to evaporation and that water loss is significant. Since precipitation inputs are low and evaporation is high, the amount of water available for infiltration is likely small, both relative to input and in absolute terms. Although the amount of precipitation that infiltrates into the ground is unknown, it is generally accepted, and is reported by the NRCS, that infiltration rates in the summit region are high, and that during heavy precipitation events, water reaching the ground surface quickly infiltrates (see Section 2.1.2). The redistribution of snow across the mountain by wind is known to occur, which affects water distribution across the landscape, increasing it in areas favorable for snow deposition and decreasing it in wind-blown areas. The scarcity of vegetation means that very little rainfall is intercepted by vegetation or evaporated off leaves or other plant surfaces. However, many areas of the landscape have broken rocky surfaces that protrude above the ground, increasing overall surface area. These surfaces may trap and hold water, exposing it to evaporation.

2.1.3.2         Watersheds

A watershed is defined as an area where runoff generated across the landscape discharges at a common outlet. Eight State of Hawai‘i delineated watersheds fall within the boundaries of the MKSR (see Figure 2.1-14). Three of the watersheds have only a few acres of their area in the MKSR. Most of the land within the MKSR falls within three watersheds: Pōhakuloa, along the southern flank; Wailuku, on the east side; and Kalopa, along the north. For all three of these watersheds, the water generated from the portion of their areas that falls within MKSR most probably constitutes only a small portion of their total water inputs, due largely to the low precipitation amounts on the upper slopes and summit area.

2.1.3.3         Surface Water

Rivers and Streams: According to the DLNR Commission on Water Resource Management (CWRM),  the State agency that defines stream flow status, none of the streams in MKSR watersheds are perennial (having continuous flow all year). The Wailuku River is the only river whose numerous gulches extend along the upper flanks of Mauna Kea, and where these coalesce, down slope near the 10,000 ft elevation (3,048 m), stream flow is considered to be perennial.¹⁸

Lake Waiau: The only persistent surface water body at the summit is Lake Waiau, a small, heart-shaped feature having an average surface area of approximately 1.5 ac (0.6 ha) (Arvidson 2002; Menviel n.d.). Located at the bottom of Pu‘u Waiau, the lake freezes almost entirely during colder times of the year and has never been known to dry up. Its depth varies between 1.6 ft (0.5 m) and 7.5 ft (2.3 m), at which point the water overflows and drains into Pōhakuloa Gulch from a low point along the lake’s west side (Woodcock 1980; Menviel n.d.). An extensive discussion of the hydrology and geology of Lake Waiau is presented in Section 2.1.1.4.3. There are no other perennial surface water bodies within the MKSR.

Seeps and Springs: Three low-volume spring-seeps Hopukani, Waihu, and Liloe Springs are known to emanate from Mauna Kea’s southwestern flank along Pōhakuloa Gulch within the Mauna Kea Forest Reserve, with other, smaller volume seeps found neighboring Waikahalulu Gulch (Arvidson 2002). While

17 Pan evaporation (potential evaporation rate) is a measure of evaporation from a pan of water that is continually supplied with water. It is representative for open water bodies. Actual evaporation varies and is some fraction of the total precipitation value.

18 Perennial/Significant Streams as defined by the CWRM, Hawaii Stream Assessment Project, 1993

the specific water sources of these springs are unknown, studies conducted by Woodcock (1980) using radioactive-isotope tritium indicated that the source water for Waihu Spring originated in relatively recent events and was probably not from relic permafrost or subsurface ice. Therefore precipitation and, possibly, Lake Waiau are believed to be the primary contributors to Waihu Spring, specifically, and perhaps other springs (Woodcock 1980). The hypothesis of Lake Waiau being the source for Waihu Spring is not supported by the conclusions of Ebel (2000) and Ehlmann et al. (2005) following their respective studies on the hydrology of the lake and the springs. Ehlmann concludes, on the basis of isotope ratios, that permafrost is not the source of water for Lake Waiau, while Ebel estimates that seepage from Lake Waiau contributes at most only 3 percent of the total flow volume of the springs. It may be that water is being transported along unknown, dense, subsurface lava flows, ash layers, or buried till, where it flows along the path of least resistance leading to a particular spring outlet. It is likely that the springs are fed by water infiltrating into the substrate from across the upslope watershed areas, percolating downward until it encounters a confining layer that directs flows towards the seeps and springs (see Section 2.1.1.4.3).

2.1.3.3         Groundwater

During the assembly of background information for this management plan, no studies were located that investigated and mapped groundwater flow paths or groundwater head levels specifically within the MKSR or Hale Pōhaku. Information regarding the substrate composition and their alignments was obtained from geologic studies that have identified and delineated most of the surface formations and some of the stratigraphy on the mountain, and from limited information gathered from soil borings drilled in support of construction for observatories and other infrastructure. Groundwater transportation rates in the summit region of Mauna Kea are unknown, and no flow paths have been identified. It is generally believed that groundwater flows along the direction of the ground surface slope, although the presence of variable subsurface features, such as dikes and sills, with low hydraulic conductivity, likely alter groundwater flow rates and flow paths. Groundwater flow paths are important, in part, to understanding the potential movement of leachate from underground waste water systems (see Section 3.1.1.2.6). Very limited information was found discussing the fate and transport of leachate from the summit region,¹⁹ and it is unknown how much of the total volume of leachate from these systems, if any, makes it to the mountain’s aquifers.

There have been studies using surrogates such as isotope signatures or surface-water mass balance estimates to infer flow paths and hydrologic connectivity of groundwater generated from: 1) the Astronomy Precinct to Lake Waiau and the springs along Pōhakalua Gulch, and 2) between the springs along Pōhakalua Gulch and Lake Waiau (Woodcock 1980; Ebel 2000; Johnson 2001; Ehlmann et al. 2005). In general, the studies found that rain and snowfall are the sources of water across and within the MKSR and for springs along Pōhakalua Gulch, and that Lake Waiau is not hydrologically connected to the water generated on lands outside its watershed basin, the basin delineated as the inner slopes of Pu‘u Waiau (see Section 2.1.3.3).

19 Nance conducted a limited investigation of groundwater transmission between Lake Waiau and existing and proposed septic systems located in the Astronomy Precinct (NASA 2005). He concluded that leachate from septic systems would not flow into or toward Lake Waiau.

Aquifers: The MKSR is located above five State of Hawai‘i delineated aquifer systems, while Hale Pōhaku is over one, the Waimea Aquifer (see Figure 2.1-14). The Waimea Aquifer system also lies under the land encompassed by the west half of the MKSR, including both NAR parcels. The southeast portion of the MKSR, approximately one-quarter of its surface area, lies over the Onomea Aquifer, which also lies within and beneath the Wailuku Watershed. The three other aquifers, Hakalau, Pa‘auilo and Honoka‘a, lie beneath the lands comprising the east and northeast areas of the MKSR. The Astronomy Precinct is located entirely above the Waimea Aquifer. It is possible, but unconfirmed, that water infiltrating into the substrate from the Astronomy Precinct flows out of the Waimea Aquifer boundary along preferential flow paths that route water to the other aquifer systems. As part of the 2005 Keck Outrigger EIS proceedings, authors calculated hypothetical impacts of nitrogen and phosphorus on the two Waiki‘i wells (Nos. 5239-01 and 02) located 13 mi (20 km) west of the summit. Using conservative assumptions, increases in nitrogen and phosphorus in the Waiki‘i well water were calculated to be approximately 0.4 and 1.6 percent, respectively, for the two chemicals (NASA 2005). Several of the observatories, including CFHT, Gemini, UH 2.2 m and UH 0.6 m are located along the easternmost boundary of the Waimea Aquifer, and it is possible that water discharged from their septic systems flows toward the Hakalau or Onomea Aquifer systems. However, as stated previously, the fate of the effluent is unknown.

2.1.3.4         Water Quality

Water quality parameters of Lake Waiau investigated by Massey (1978) and others in 2003 indicated a slightly alkaline water, with conductivity ranging between 109 and 121 µS/cm (at 25°C), and very low levels of dissolved constituents (NASA 2005). A turbid look and greenish tint to the lake water has been noted by observers for many years (Bryan 1939; Neal 1939; Wentworth and Powers 1941; Maciolek 1969; Group 70 1982; Arvidson 2002) and is attributed to algae mats growing on the bottom of the lake (Woodcock et al. 1966; Massey 1978; Dillon 1979). There are, however, accounts from visitors to the lake in which a green tint was not mentioned (Raine 1939). In 1977, a severe reduction in lake water levels with concomitant elevations in phytoplankton biomass was identified and classified as hypereutrophication (a significant increase in nutrients, including nitrogen and phosphorus) (Laws and Woodcock 1981). Fecal coliform and bacteria parameters obtained from samples from Hopukani Spring were found to be negligible (NASA 2005). Similar investigations into well water found at much lower elevations were also found to be negligible (NASA 2005).

2.1.3.5         Threats to the Hydrology

Threats to the hydrology of Mauna Kea include those associated with human presence and activity on the mountain and climate change. Human activities that have the potential to impact water resources quality, and to a lesser degree quantity, include any actions that add to the current wastewater volume or that change in-situ patterns of water movement. Examples are: leaking facility pipes; accidental spills of contaminants; and improperly filtered wastewater. These contributions may affect the quality of water seeped to springs along Mauna Kea’s flanks, as well as the fresh water aquifers beneath the mountain. Potential threats from changes in climate involve alteration of current weather patterns, such as changes in rainfall or wind. While the exact impacts of climate change to the MKSR are unknown, results of some general climate circulation model runs suggest that the trade wind inversion will be more persistent. This scenario is expected to result in a reduction of the number of storms that frequent the islands annually and subsequently lower precipitation levels at the upper slopes and summit area of Mauna Kea. Such a change may impact the volume of the annual snowpack and its persistence; the thickness of permafrost and its extent and persistence; and annual precipitation regimes. See Sections 2.2.1.3.6 and 3.2.11.

2.1.4          Climate

2.1.4.1         Overview of Mauna Kea’s Climate

Climate refers to the average of recorded weather variables over some period of time, which is then used to represent meteorological condition. At the upper elevations of Mauna Kea, the prevailing conditions are dry, windy, and cool, with high visibility and low surface albedo; it has been designated as semi-arid, barren alpine desert tundra (Ugolini 1974).

Climatic Influences: The atmospheric feature that most strongly influences the climatic regime of Mauna Kea, as in other parts of the Hawaiian Islands, is the North Pacific Anticyclone. This semi-permanent  high pressure ridge is located some 2,000 miles (3,219 km) north and east of the Hawaiian Islands, shifting its center from lat 30º N, long 130º W, in the winter, to lat 40º N, long 150 W, in the summer. The anticyclone is formed as warm air from the equatorial zones rises and moves north toward lat 30º N. where the air cools and sinks back toward the earth’s surface. This system is commonly referred to as a Hadley Cell, named after the first western scientist who described it. A result of the sinking air is the trade winds that blow outward from the center of the cell, and in this case, toward the Hawaiian Islands. As the

20 See Mauna Kea Weather Center: http://www.jach.hawaii.edu/weather/

warm air sinks and blows from the northeast, it encounters rising air from the ocean surface that cools as it rises, and at the point of contact between the two air parcels the layer of warm air overlies the cool air. This atmospheric feature is termed an inversion; in Hawai‘i it is commonly called the trade wind inversion. In vertical profile, the air column around Hawai‘i under this climatic regime can be described as comprising three layers: from sea level to 2,000 ft (610 m) is the marine layer, where evaporation from the ocean lifts water upwards; from 2,000 ft to 7,000 ft (610 m to 2,133 m) is the cloud layer, where water in the air parcel condenses, forming clouds; from 7,000 ft (2,133 m) to approximately 20,000 ft (6,096 m) is the dry inversion zone, where the atmosphere is dry and stable. Figure 2.1-15 depicts a typical inversion capping of the clouds at approximately 7,500 ft.

A second significant factor governing the weather patterns of the Hawaiian Islands is their position on the earth and the fact they are surrounded by a large thermal control—the ocean. Another factor in the climatic regime is the amount of incoming solar radiation, which, due to the island’s position in the tropical belt, results in only small annual shifts.

Seasons: There are two meteorological seasons in Hawai‘i, winter, (October–April) and summer (May– September), with the trade winds blowing approximately 80 percent of the time in the summer and 50 percent of the time in the winter (Giambelluca and Sanderson 1993). Pre-contact Hawaiians recognized these two seasons as the cool (winter) season (ho‘oilo) and the warm (summer) season (kau). Rainfall associated with the trade winds results from winds encountering the side of the islands almost perpendicular to the incident angle, which forces the air parcels upwards, cooling it, whereupon the moisture in the air condenses and forms clouds which often generate rain. On the windward sides of the islands, trade wind showers are common; however, from sea level to 7,000 ft (2,133 m), the amount and frequency of the rainfall received in a given location is strongly correlated with elevation.

The highest trade wind rainfall rates occur on the windward sides of the islands, in an elevation band of 2,500 to 7,000 feet (762 to 2,133 m). At 7,000 ft, (2,133 m) the trade wind inversion caps upward migration of the clouds, and thereafter, rainfall decreases elevation. As a result, when the trade wind inversion is present, Mauna Kea remains dry from roughly 7,000 ft ( 2,133 m) upwards (da Silva 2006). Average annual rainfall totals, represented by isohyetal lines, show a significant decrease from 7,000 ft, (2,133 m) to the summit elevations at the top of Mauna Kea (see Figure 2.1-16).

Storms: The Hawaiian Islands are subjected to other weather patterns when the trade winds break down due to a shift in the high pressure ridge caused by perturbations in the atmosphere. In some cases ridges of high pressure set up over the islands, causing winds to be light and allowing local land sea breezes to develop. Other shifts from the normal trade wind regime are due to the storms that frequent the islands, including cold front storms, upper-level and surface low-pressure systems (including kona lows) tropical depressions, and hurricanes. Storms caused by cold fronts generally occur in the winter months and deliver high precipitation levels. Cold fronts of varying intensities arrive from the northwest, causing varying amounts of rainfall and moderately strong winds. Following the passage of the leading edge of the front, skies clear and temperatures reach seasonal lows during the night time hours. Storms caused by kona lows also visit the islands during the winter months, arriving from the southwest. Kona storms are less variable in rainfall levels, are more frequent, and contribute the highest percentage to annual rainfall levels in the leeward and high-elevation areas of the islands. Storms created by upper-level low-pressure systems reach the island periodically, again mostly during the winter months. These storms often bring intense rainfall and strong damaging winds.

Hurricanes and tropical storms are rare occurrences in Hawai‘i, but when they reach the islands, they can cause widespread damage. Hurricanes and tropical storms occur during the summer months and can cover entire islands. Such storm systems bring most of the annual precipitation to Hawaii’s leeward areas and the mountainous zones above 7,000 ft (2,133 m), including Mauna Kea (Giambelluca and Sanderson 1993). The number of winter storms that reach the islands varies year to year, with the northwest islands of the archipelago receiving more events than the southeast islands. No records were located documenting the number of non-trade-wind storms that affect Mauna Kea annually, but it is presumed to be highly variable, with a range of two to ten storms a year.

El Niño Southern Oscillation: Generally occurring every four to seven years, El Niño events are associated with a general warming of the ocean’s surface water near the eastern Pacific, off the coast of Peru. The associated moist, warm air rises, destabilizing the upper atmosphere. This encourages the development of thunderstorms over the equator, and significantly reduces wind flow and precipitation everywhere in Hawai‘i (Juvik and Juvik 1998), while increasing winds at higher elevations, including the Mauna Kea summit region (da Silva 2006). El Niño conditions are generally associated with a greater number of tropical cyclones (Juvik and Juvik 1998). La Niña events are the opposite phase of the El Niño Southern Oscillation cycle and are associated with a cooling of the ocean’s surface temperatures (Juvik and Juvik 1998). The associated cooler conditions create weather patterns opposite those of El Niño events, resulting in frequent storms to the Islands (Juvik and Juvik 1998), including the summit region (da Silva 2006).

Climate Change: A comprehensive discussion of various hypotheses concerning how climate change may affect the Mauna Kea climate regime is presented in Section 2.2.1.3.6.

2.1.4.2         Climatic Variables

Wind: Approximately 80 percent of the time, the wind blows from the west at the upper elevations of Mauna Kea. This typically changes during warmer months, and for the remaining 20 percent of the time, wind comes from the east (Juvik and Juvik 1998; da Silva 2006). On occasion, southerlies will form, due to unstable upper atmospheric conditions. Southerlies bring in storm fronts and large amounts of rain (Birchard 2008). Average wind speeds at 8,530 ft (2,600 m) at Pu‘u La‘au range between 2.7 to 3.6 miles per hour (1.2 to 1.6 meters per second) (Nullet et al. 1995). Average wind speeds at Mauna Kea’s summit normally vary between a maximum of 23 miles per hour (10 meters per second) in January and a minimum of 11 miles per hour (5 meters per second) in September (da Silva 2006); however, higher speeds have been noted during storms (NASA 2005). The dry and breezy conditions facilitate high rates of evaporation at the summit and maintain the cool, dry atmosphere (da Silva 2006; Birchard 2008). The pan evaporation rate for the summit area is reported as 70 in (178 cm) per year (Ekern and Chang 1985), and Nullet et al. (1995) observed rates of evaporation ranging between 0.16 and 0.6 in/day (4.1–15.2 mm/day) at their Pu‘u La‘au station or 27–57 in/year (693–1,460 mm/yr).

Wind vectors (direction and speed) across the summit area play a large role in the aeolian environment, transporting small debris including bugs from lower elevations up to the summit area. Obstructions to wind flow such as at the crests of the pu‘u can redirect the wind or slow it, creating eddies or small vortexes that reduce the energy, or holding capacity, of the wind, allowing debris in the air parcel to fall out. The aeolian environment of the summit area is unique, the persistent wind forcing resident fauna to adapt (see Section 2.2.2.2). A literature search did not find any studies investigating the effects of the observatories on the wind vectors; it is logical to assume, however, that there are some effects on a micro scale. The nature of these effects on deposition and transport of wind-borne debris is unknown. The observatories have been designed, however, to minimize turbulent drag on the air stream, which may reduce their effects on the summit aeolian regime. The observatory support buildings are, for the most part, conventional boxes both at the summit and Hale Pōhaku.

Research by Businger and others has begun to measure wind variability within one of the summit area pu‘u (Businger 2008). Part of the study’s goal is to measure wind vectors on the inside and outside of the pu‘u to better understand how the physical structure of the area affects deposition of food supplies for the wēkiu bug.

Temperature: Due to its latitude, annual temperature flux in Hawai‘i is small, with a mean daily temperature difference of only 7.5° F (4°C) at the summit of Mauna Kea between the coldest month and the warmest month (da Silva 2006). During winter, between October and April, the mean daily minimum temperature is 32.5°F (0.28°C); during the summer, between May and September, the mean daily maximum is 40°F (4.4°C) (da Silva 2006). Mean monthly temperatures above the inversion layer generally range between 24.8°F and 32.9°F (-4ºC and 0.5º C) in January, one of the coldest months, and between 38.3°F and 42.8°F (3.5ºC and 6.0ºC) in September, considered a warm summer month (da Silva 2006). Even though variability between annual mean lows and highs is minimal, temperature ranges recorded at the summit area are quite large, ranging from 2°F to 61°F (-16.6°C to 16.1°C). Average temperatures at Hale Pōhaku, at 9,000 ft (2,743 m), range between 30°F and 70°F (-1°C and 21ºC) throughout the year (Group 70 International 1999).

Precipitation: The longest period of record of statistical data representative of the summit area climate is from the National Weather Service (NWS) station Mauna Kea Observatory 1, at an elevation 13,780 ft (4,200 m).²¹ The data set represents years 1969–2000, a 31-year period of record. For this period, average precipitation is reported as 7.41 in (188 mm). It is unknown if this precipitation value includes the contribution of water from snowfall. The Subaru Telescope recorded precipitation data for a period of seven years from 1999 to 2005. Mean annual precipitation was estimated at 15.5 in (393 mm) by interpolating annual precipitation from a cumulative plot for 1999–2003 (Miyashita et al. 2004). This value includes the contribution from snowfall, although the efficiency of snow capture by the recording instrument is unknown. Ehlmann et al. (2005) reports annual precipitation as a range of 4.7 to 17.7 inches (12 to 45 cm) recorded at the VLBA, located below the summit area. It is obvious from these numbers that the mean precipitation is variable year to year.

Average relative humidity for Mauna Kea was found to stay relatively constant, at approximately 36 percent throughout the year, with the highest values occurring during November (41 percent) and the lowest during April (30 percent) (da Silva 2006), therefore its effect on local precipitation may be minimal. The dew point was also observed to stay relatively consistent, having an annual mean value of 4.1º F (-15.5º C), with the coldest month being December at -1.3º F (-18.5°C) (da Silva 2006).

The amount and duration of snow and ice covering the summit during the months of November–March is variable (Laws and Woodcock 1981). Snowpack volumes fluctuate from year to year (da Silva 2006) as does, most likely, the formation of ice. No data on average snowfall, snowpack volumes, or patterns of ice formation for the MKSR was found in the literature; however, based upon precipitation occurrence, associated relative humidity, and average temperatures, da Silva (2006) calculated that snowfall was more likely to occur at the MKSR in January than in any other month.

2.1.4.3         Climate Studies at the Summit of Mauna Kea

Nullet et al. (1995) performed a two-year study along a cross-section of the Island of Hawai‘i to document differences in climate patterns between the leeward and windward sides of the island. In her dissertation, da Silva (2006) contributes to the understanding of prevailing weather conditions at the summit. She compiled meteorological datasets from the Canada-France-Hawai‘i Telescope between September 1994 and March 2006 and from the United Kingdom Infra-Red Telescope between May 1991 and March 2005. She then analyzed numerous attributes, including pressure, temperature, relative humidity, and snowfall as model-derived proxies.

2.1.4.4         Threats to Climate

The impacts and threats associated with global warming on the climate regime of Mauna Kea are discussed in Section 2.2.1.3.6, Climate Change.

2.1.4.5         Climate Information Gaps

The following information gaps regarding the climate at high elevation areas of Mauna Kea have been identified through literature review and consultation with experts:

1. Data and Analysis While it is evident from the literature that the meteorological processes over the Island of Hawai‘i are fairly well understood, climate data such as precipitation associated with snowfall and analysis of the spatial distribution of precipitation specific to the summit region and upper slopes of Mauna Kea is lacking. Specifically, snowpack depth and its snow-water equivalent is not measured or recorded. In addition to providing information important for understanding local conditions and dynamics, collection of these data over the long term will be valuable to any  future studies investigating climate change

21 It is unknown at which observatory this station is located, since its metadata data does not contain a name corresponding to an observatory. It is possible that it is the UH 0.6 m telescope, because its installation coincides with the initial date of the climate data collection.

2.1.5          Air Quality and Sonic Environment

While it may be a truism that air quality is a significant factor affecting astronomers at Mauna Kea, religious practitioners, recreational users, the tourist industry, and local residents all value the clear views and clean air at the summit. Increased noise levels at MKSR affect all visitors to the summit, albeit in varying degrees. Because the summit area is used by many visitors as a place for worship and silent contemplation, the natural quiet associated with the higher elevations of Mauna Kea is considered a natural resource.

2.1.5.1         Air Quality

The quality of the air at the summit of Mauna Kea is well known throughout the astronomy community. Contributors to air pollution at the summit include vehicle exhaust and fugitive dust from road grading, construction, and other activities conducted on unpaved surfaces. Although there is no active monitoring for air quality at the Mauna Kea summit, the National Oceanic and Atmospheric Administration (NOAA) Mauna Loa Observatory has collected air quality data for the summit of Mauna Loa since its construction in 1956 (Juvik and Juvik 1998; Barnes 2008). These data indicate that for the air pollutants considered by the Hawai‘i State Department of Health (DOH) to be of greatest concern (ozone, carbon monoxide, and sulfur dioxide), the air quality at Mauna Loa is excellent. Given the similarities between the two  locations, it has been suggested that the overall air quality at Mauna Kea is excellent as well (NASA 2005; Barnes 2008). Five DOH monitoring stations do exist at other locations on the island, including at Hilo and Kona and at three locations in the Puna District; however, all of these monitor air quality below the trade-wind inversion layer.

Another potential source of air-borne particulates and sulfur dioxide is Kīlauea volcano. Early 2008 volcanic activity from Halema‘uma‘u Crater at Kīlauea Volcano released record amounts of the gas, as much as 4.4 million pounds/day (2,000 tonnes/day) and ambient air concentrations were found to exceed 40 ppm along the road neighboring the crater’s rim (U. S. Geological Survey 2008b). This far exceeds the DOH and federal air quality standards for this pollutant, which limits sulfur dioxide concentrations to 0.14 ppm based on a 24-hour averaging period.²² Sulfur dioxide releases from the Kīlauea summit prior  to late December 2007 have typically been 330,693–440,925 pounds/day (150–200 tonnes/day) (U. S. Geological Survey 2008b). Kīlauea also generates ash emissions. One such emission created an ash plume 0.5 to 1.0 mile above ground level (0.8–1.6 km) (U. S. Geological Survey 2008a). The new Kīlauea vent sits at nearly 4,000 ft elevation (1,219 m), but gas and ash debris emitted from it are most likely kept below the inversion layer when it is present. However, even during periods of southerly winds this may

22 http://hawaii.gov/health/environmental/air/chart.pdf

2.1.6          Visual Environment

The attributes ascribed to the appearance of Mauna Kea are also considered one of the mountain’s natural resources, a resource that has been valued for generations.

2.1.6.1         Viewshed

There is an ancient Hawaiian saying: Mauna Kea kuahiwi ku ha‘o i ka mālie (Mauna Kea is the astonishing mountain that stands in the calm). Many Hawaiians consider the summit region of Mauna  Kea a “sacred landscape.” Some draw a sense of inspiration, well-being, and security by looking at it from a distance; Maly (1999) writes, “Simply looking at Mauna Kea from afar, seeing it standing there reaching to the heavens, gave the Hawaiian spiritual strength.” Others are drawn closer, and ascend the mountain to view its features.

Famous for its dominating presence and its smooth, shield-like silhouette, Mauna Kea has been a beacon for centuries to travelers coming to the islands. Similar in significance is the view from Mauna Kea. Descriptions of sweeping views were often recorded by early Western visitors; those looking up to the summit and those looking down from the top of Mauna Kea:

Friday, April 25. The appearance of Hawaii, this morning was exceedingly beautiful. We were within a few miles of the shore; and the whole of the eastern and northern parts of the island were distinctly in view, with an atmosphere perfectly clear, and a sky glowing with the freshness and splendor of sunrise. When I first went on deck, the gray of the morning still lingered on the lowlands, imparting to them a grave and somber shade; while the region behind, rising into a broader light, presented its precipices and forests in all their boldness and verdure. Over the still loftier heights, one broad mantle of purple was thrown; above which, the icy cliffs of MOUNA-KEA…blazed like fire, from the strong reflection of the sun-beams striking them long before they reached us on the waters below. As the morning advanced, plantations, villages, and scattered huts were distinctly seen along the shore…

In the evening Hawaii and Mouna-kea again, at a distance, afforded another of the sublimest of prospects;—while the setting sun and rising moon combined in producing the finest effects on sea and land. The mountains were once more unclouded, and with a glass we could clearly discern immense bodies of ice and snow on their summits… [Observations of C.S. Stewart sailing into Hilo Bay in April 1823 (Maly and Maly 2005).]

The view from the summit was sublime beyond description, embracing, as it did, the three other great mountains of Hawaii, and the grand old “House of the Sun [Haleakala],” 75 miles distant, looking up clear and distinct, above a belt of clouds. [‘The Ascent of Mauna Kea, Hawaii’, Report of W.D. Alexander on the Mauna Kea Trip of 1892, (Maly and Maly 2005).]

Today, visitors to the summit can more easily experience the vistas, including breathtaking sunrises and sunsets. Residents from around the island value the changing colors of Mauna Kea throughout the day, with people from the eastern side describing the mountain’s beauty at sunrise, while those on the northwestern side experience the sunsets (Maly 1999).

On a cloud-free day, views from the summit region include Mauna Loa to the south, Hualālai to the west, the flanks of summit cinder cones to the east, and other islands in the Hawaiian chain to the north- northwest. Due to persistent cloud cover, Hilo usually cannot be seen during the day, and due to a lighting ordinance, Hilo’s street lights use low-pressure sodium lamps to reduce night-time glow from populated areas (Wainscoat 2007). To reduce thermal impacts, the observatories are painted white and when skies are clear, the summit region and observatories can be seen from Hilo, Honoka‘a, Waimea, Kilauea summit, sections of the Mauna Kea Summit Access Road and much of Puna. Views from Hilo are of the southern and eastern flanks, while views from Waimea are of the northern flanks of Mauna Kea. During warmer months, the formation of an inversion layer between 5,000 and 9,000 ft (1,524–2,740 m) may obstruct views of the summit from lower elevations, as well as views of the lower elevations from the summit. Due to topography, Hale Pōhaku is not visible from the summit, while views of the summit region and the observatories from other portions of the Mauna Kea Summit Access Road and Hale Pōhaku are blocked from view, in many places, by cinder cones.

As mountain topography is an integral part of the local viewscape, it must be considered during planning for any proposed development, redevelopment, or decommissioning of facilities in the summit region. Existing observatories have impacted the viewscape in some locations, both from the summit and of it, and they do obscure portions of the 360-degree view from the summit area. Section IX of the 2000 Mauna Kea Science Reserve Master Plan provides a physical planning guide that contains guidelines for future development of the summit and support facilities, including siting and design criteria to reduce visual impact of new facilities (Group 70 International 2000). This included designation of the Astronomy Precinct, to consolidate astronomical development (see Figure 1-3). The plan seeks to minimize the visual impact from significant cultural areas by respecting views from the pu‘u and archaeological sites in the siting of any potential future facilities, along with avoiding interference with the visual connections between the major shrine complexes and pu‘u. Trails that become etched into the cinder from repeated use are another consideration for local viewscapes. As described in Section 3.2.1, these footpaths can be visually distracting and disturb habitat.

2.1.6.2         Light Quality

In addition to striking views from and of Mauna Kea, the “seeing” ability from the summit region as it relates to astronomy is very high. It has been well documented that the MKSR is a premier location for astronomical activities (Walker 1983; Businger et al. 2002; Wainscoat 2007). This is in part because the atmosphere above the higher elevations of Mauna Kea is so stable (Walker 1983; Businger et al. 2002). Dark skies, generally favorable weather, and clean, clear air permit almost year-round un-obscured conditions for optimal night seeing. These attributes of seeing ability are affected directly and indirectly by four primary factors: the site’s remote location, its elevation, its topography, and the climate (Businger et al. 2002). Managing these attributes for optimal influence on night sky viewing will be essential to the continued success of astronomy at the MKSR.

One of the main issues found within the literature is the impact of night glow from populated areas, which is affected by the types of lighting used in residential, business and industrial districts at night (Wainscoat 2007). The light emitted from these sources has increased sky brightness by as much as 30% above natural levels in some areas (Wainscoat 2007). A strong lighting ordinance enacted in 1989, on the island of Hawai‘i, has helped maintain optimal darkness by requiring that existing street lights be retrofitted with fully enclosed, low-pressure sodium bulbs (Copman 2007). The yellowish hue of the sodium lights provides effective illumination and is expected to save money, because they use less energy than older bulb types (Copman 2006; Wainscoat 2007). In addition, the light fixtures prevent light scatter, reducing the glare that negatively affects seeing quality both on the street and at the MKSR. Light from as far away as Maui is also affecting viewing quality at the summit, as is light from neighboring PTA (Wainscoat 2007). Approximately 6.2 miles (10 km) away from the summit of Mauna Kea, PTA is “the single largest source of light pollution for the observatory[ies]” (Wainscoat 2007). Impacts from this source may be lessened however; older style lights also reduce the effectiveness of night-vision equipment used during night training events, and the Army is slowly retrofitting (Wainscoat 2007).

2.2    Biotic Environment

Mauna Kea, the tallest mountain in Polynesia, has the greatest diversity of biotic environments anywhere in the Hawaiian Archipelago (Juvik and Juvik 1984). Ecosystems on Mauna Kea range from the highly modified fertile lowlands to an alpine stone desert located at the summit at 13,796 ft (4,205 m). For the Mauna Kea Natural Resources Management Plan, the ecosystems under consideration are those found above approximately 9,000 ft (~2,700 m), beginning at Hale Pōhaku and rising to the summit. High elevation ecosystems on Mauna Kea can be divided into two basic types: the subalpine ecosystem, which occurs from approximately 5,600 ft to 9,800 ft (1,700 m to 3,000 m) elevation, and the alpine ecosystem, which occurs above 9,800 ft (3,000 m) (Gagné and Cuddihy 1990). The shift from subalpine to alpine ecosystems is determined by the elevation of the nocturnal ground frost line (Mueller-Dombois and Fosberg 1998). The subalpine and alpine ecosystems can be further subdivided by vegetation community, as described in Section 2.2.1. The following sections (Sections 2.2.1 through 2.2.4) discuss the plant, invertebrate, bird, and mammal species found in the subalpine and alpine ecosystems of Mauna Kea (with the focus being on Hale Pōhaku and the Mauna Kea Science Reserve (MKSR)). Each section also reviews previous research for each group (especially biological surveys) done at Hale Pōhaku and the MKSR, as well as information gaps, and threats to native populations of plants and animals.

In addition to the general descriptions of the flora and fauna, more-detailed discussions of federal and state Threatened and Endangered species, Candidate Species, and Species of Concern are presented in each section. Threatened and Endangered Species are species that are legally required to be protected (under either federal or state law) (see Section 1.4.3). Candidate species are those species not yet listed but for which there exists sufficient evidence on biological vulnerability and threats to support a proposal to list as Endangered or Threatened. There is no legal mandate to protect Candidate Species, but generally it is in the best interest of land mangers to protect them in order to prevent the need for listing. Species of Concern are those species that might be in need of conservation action, but that are not currently Listed or Candidate species. Species of Concern receive no legal protection and use of the term does not  necessarily imply that a species will eventually be proposed for listing. The numbers of federal and state listed species that occur, or potentially occur, in the subalpine and alpine regions of Mauna Kea covered by this management plan are presented in Table 2.2-1. Many of the species in the State of Hawai‘i lists are also included in the federal lists, so the two lists overlap. The total number of species that are protected by either federal or state laws found (or formerly found) in the areas covered by this plan are:  12 Endangered, one Threatened, two Candidate, and 16 Species of Concern (two of which are also listed as state Endangered on islands other than Hawai‘i). These species are listed in Table 2.2-2. See the glossary for more detailed definitions of the terms Endangered, Threatened, Candidate, and Species of Concern.

Table 2.2-1. Number of Federal and State Listed Species, Candidate Species, and Species of Concern found or potentially found at Hale Pōhaku and MKS Group

& Snails Federally Endangered 4 0 5 1 Federally Threatened 1 0 0 0 Federal Candidate for Listing 1 1 0 0 Federal Species of Concern 0 6 6 0 State Endangered 4 0 5 1 State Threatened 1 0 0 0 State Candidate for Listing 1 0 0 0 State Species of Concern 4 3 0 0

Table 2.2-2. List of Federal and State Threatened, Endangered, Candidate, and Species of Concern found, or potentially found, at Hale Pōhaku and MKSR

Group                                 Scientific Name                                              Common Name                            Legal Status2 Endangered Species
Plant	Argyroxiphium sandwicense sandwicense	‘Ahinahina, Mauna kea silversword	FE, SE
Plant	Asplenium fragile var. insulare	Diamond spleenwort	FE, SE
Plant	Phyllostegia racemosa var. racemosa	Kiponapona	FE, SE
Plant	Vicia menziesii	Hawaiian vetch	FE, SE
Bird	Branta sandvicensis	Nene (Hawaiian goose)	FE, SE
Bird	Buteo solitarius	‘Io	FE, SE
Bird	Hemignathus munroi	‘Akiapola‘au	FE, SE
Bird	Loxioides bailleui	Palila	FE, SE
Bird	Pterodroma sandwichensis	‘Ua‘u (Hawaiian petrel)	FE, SE
Mammal	Lasiurus cinereus semotus	‘Ope‘ape‘a (Hawaiian hoary bat)	FE, SE
Threatened Species
Plant	Silene hawaiiensis	Hawai‘i catchfly	FT, ST
Candidate Species
Plant	Ranunculus hawaiiensis	Makou	FC, SC
Arthropod	Nysius wekiucola	Wekiu bug	FC
Species of Concern
Plant	Chamaesyce olowaluana	‘Akoko	HSOC
Plant	Cystopteris douglasii	Douglas’ bladderfern	HSOC
Plant	Dubautia arborea	Mauna Kea dubautia, na‘ena‘e	HSOC
Plant	Sanicula sandwicensis	Hawaii black snakeroot	HSOC
Arthropod	Agrotis melanoneura	Black-veined agrotis noctuid moth	FSOC, HSOC
Arthropod	Coleotichus blackburniae	Koa bug	FSOC
Arthropod	Hylaeus difficilis	Difficult yellow-faced bee	HSOC
Arthropod	Hylaeus flavipes	Yellow-footed yellow-faced bee	FSOC, HSOC
Arthropod	Micromus usingeri	Flightless brown lacewing	FSOC
Snail3	Succinea konaensis	Succineid snail	FSOC
Snail	Vitrina tenella	Zonitid snail	FSOC
Bird	Asio flammeus sandwichensis	Pueo	FSOC, SE4
Bird	Chasiempis sandwichensis	‘Elepaio	FSOC
Bird	Hemignathus virens virens	‘Amakihi	FSOC
Bird	Himatione sanquinea	‘Apapane	FSOC
Bird	Pluvialis fulva	Kolea (Pacific golden plover)	FSOC
Bird	Vestiaria coccinea	‘I‘iwi	FSOC, SE5

2.2.1    Botanical Resources

The following review of botanical resources focuses on the conditions at Hale Pōhaku (and surrounding areas), the Summit Access Road (from Hale Pōhaku to the summit), and the Mauna Kea Science Reserve (MKSR). Information on the plants found in these areas was gathered primarily from a small number of botanical accounts of high elevation habitats on Mauna Kea (Hartt and Neal 1940; Smith et al. 1982; Char

2 Legal Status: FE = Federally Endangered, FT= Federally Threatened, FC = Federal Candidate for listing, FSOC = Federal Species of Concern, SE = State Endangered, SC = State Candidate for Listing, HSOC = Hawaii State Species of Concern, ST = State Threatened.

3 It is unknown whether snails are present at Hale Pōhaku – no surveys for snails have been completed at this elevation.

4 State Endangered on Oahu only.

5 State endangered on Oahu, Lanai, and Molokai only.

A list of vascular plants occurring at Hale Pōhaku and the MKSR is presented in Table 2.2-3. Lichen species are presented in Table 2.2-4, and mosses in Table 2.2-5. Threats to the subalpine and alpine plant communities of Mauna Kea are discussed in Section 2.2.1.3, and information gaps are discussed in 2.2.1.4. Photos of common native species found in the subalpine and alpine zones are presented in Figure 2.2-2. Photos of rare plants (Threatened, Endangered, Candidate, Species of Concern) are presented in Figure 2.2-3. Photos of common invasive species are presented in Figure 2.2-4.

2.2.1.1       Subalpine Plant Communities (Hale Pōhaku and Lower Summit Access Road)

The subalpine community on Mauna Kea can be divided into three major types: open dry forest (or woodlands) dominated by māmane (Sophora chrysophylla) trees, tussock grassland, and subalpine dry shrublands. Tussock grasslands were once an important vegetation community on Mauna Kea. These grasslands were made up of Deschampsia nubigena, Panicum tenuifolium, Poa sandvicensis, Trisetum glomeratum, Agrostis sandwichensis, and Eragrostis atropioides (Mueller-Dombois and Fosberg 1998). However, overgrazing by feral and domesticated sheep and goats, and establishment of invasive weed species, has virtually eliminated these grasslands (Mueller-Dombois and Fosberg 1998).

Subalpine dry shrublands are dominated by pūkiawe (Leptecophylla tameiameiae), ‘ōhelo (Vaccinium reticulatum) and an occasional ‘ōhi‘a tree (Metrosideros polymorpha) (Cuddihy and Stone 1990). The dry shrubland community may also be found above treeline up to 9,800 ft (3,000 m) and grades into the alpine dry shrubland community (Gagné and Cuddihy 1990). Because of the similarity between the subalpine and alpine dry shrublands, these communities are discussed in more detail in 2.2.1.2 (Alpine Communities).

Subalpine woodlands are dry most the year, with annual rainfall ranging from 15 to 39 inches (380 to 1,000 mm), most of which falls between December and March. Fog drip from clouds that form in the afternoons is an important source of moisture in this region (Gilbertson et al. 2001). Understory plants tend to be concentrated under māmane trees, where they receive fog drip (Gagné and Cuddihy 1990).

6 The nocturnal ground frost line is the elevation above which frost form at night. Below this elevation frosts seldom form.

Māmane occurs in almost pure stands on the eastern, northern, and western slopes of Mauna Kea, and in a narrow band at tree line on the southern slope (Scott et al. 1984). Other tree species, such as pilo (Coprosma montana) are scarce, and naio (Myoporum sandwicense) is absent in these areas (Scowcroft and Conrad 1992). Naio trees are co-dominant with māmane on the southwestern slopes of Mauna Kea (Scott et al. 1984).

Māmane woodlands once stretched from sea level on the leeward side of Mauna Kea to the tree line, but have been greatly reduced due to habitat alteration at lower elevations (for grazing, agriculture, and development) and uncontrolled grazing at the higher elevations by feral sheep (Ovis aries), mouflon  sheep (O. musimon), goats (Capra hircus), and historically, cattle (Bos taurus) and horses (Equus caballus) (Giffin 1982; Scowcroft and Giffin 1983; Hess et al. 1999). The lower elevation for the māmane-naio forest type is currently approximately 6,000 ft (1,800 m) (Aldrich 2005). Although feral grazer abundance was greatly reduced in the area in the 1980s, and is currently low⁷, the forest has not fully recovered, due to continued browsing and the presence of invasive plant species that inhibit māmane regeneration (Williams 1994; Hess et al. 1996). The understories of most māmane forests are now dominated by invasive grasses such as orchardgrass (Dactylis glomerata), common velvetgrass (Holcus lanatus), sweet vernalgrass (Anthoxanthum odoratum), and Kentucky bluegrass (Poa pratensis) (Hess et al. 1996), although native grasses can still be found in some areas (see below). The heavy growth of the invasive grasses suppresses germination of māmane seeds and increases the likelihood of fires in the dry woodland (Hess et al. 1996). Māmane regeneration in these degraded woodlands is highest in the higher elevation areas (such as at Hale Pōhaku), where grass densities are low (Hess et al. 1996).

Prior to human disturbance, dry forests and shrublands were some of the most diverse plant communities in Hawai‘i (Cuddihy and Stone 1990; Aldrich 2005). Māmane forests are thought to have always been fairly open, and historically had an understory community with many herbaceous species and abundant shrubs such as pūkiawe, ‘ōhelo, and ‘āheahea (Chenopodium oahuense) (Mueller-Dombois and Fosberg 1998). Although the understories of the māmane woodlands on Mauna Kea are currently dominated by invasive grasses and shrubs, native understory species can still be found in the region. Native plant species commonly found in māmane forests (historically and/or currently) are listed below (Skottsberg 1931; Hartt and Neal 1940; Gagné and Cuddihy 1990; Mueller-Dombois and Fosberg 1998; Char 1999a; Aldrich 2005; Bishop Museum 2007b, a). A list of plant species found in the subalpine region (at and  near Hale Pōhaku) on Mauna Kea is presented in Table 2.2-3.

Native grasses and sedges found in māmane woodlands include Hawai‘i bentgrass (Agrostis sandwicensis), alpine hairgrass (Deschampsia nubigena), lovegrass (Eragrostis sp.), mau‘u lā‘ili or Hawaii blue-eyed grass (Sisyrinchium acre), pili uka (Trisetum glomeratum), two sedge species (Carex macloviana and C. wahuensis), and Hawai‘i wood rush (Luzula hawaiiensis). Alpine hairgrass and pili uka are the two most common grasses in this community (Cuddihy and Stone 1990; Char 1999a). Native herbs found in the māmane woodlands include Hawai‘i stinging nettle (Hesperocnide sandwicensis), ‘ena‘ena (Pseudognaphalium sandwicensium), makou (Ranunculus hawaiensis), and Hawaii black snakeroot (Sanicula sandwicensis). In addition, botanical surveys of Mauna Kea done in the 1820s and 1830s indicated that the native strawberry, or ‘ōhelo papa (Fragraria chiloensis) was abundant in the subalpine and alpine regions (Hartt and Neal 1940). It has declined in abundance on the island of Hawai‘i, possibly due to a pathogen introduced with the naturalized woodland strawberry (Fragraria vesca) (Wagner et al. 1990).

7 Sheep, and evidence of browsing, continues to be observed in the subalpine and alpine zones of Mauna Kea. A flock of approximately 60 sheep was observed in February 2008 in Pohakuloa Gulch within the Ice Age Natural Area Reserve (Hadway 2008).

−        ‘akoko (Chamaesyce olowaluana)

−        ‘āheahea (Chenopodium oahuense)

−        ‘aiakenēnē (Coprosma ernodeoides)

−        alpine mirror plant (Coprosma montana)

−        ‘a‘ali‘i (Dodonaea viscosa)

−        three species of na‘ena‘e (Dubautia arborea, D. ciliolata ciliolate, and D. scabra)

−        nohoanu (Geranium cuneatum hololeucum)

−        pūkiawe (Leptecophylla tameiameiae)

−        ‘ūlei (Osteomeles anthyllidifolia)

−        ‘ākala (Rubus hawaiensis)

−        alpine catchfly (Silene struthioloides)

−        alpine tetramolopium (Tetramolopium humile humile)

−        ‘ōhelo (Vaccinium reticulatum)

Of these, pūkiawe is the most common in the higher elevation reaches of the subalpine community (Gagné and Cuddihy 1990; Mueller-Dombois and Fosberg 1998). Native vines and lianas commonly found in māmane woodlands include two species from the mint family (Lamiaceae): littleleaf Stenogyne (Stenogyne microphylla) and mā‘ohi‘ohi (Stenogyne rogosa), and a large climbing liana or sprawling shrub, pāwale (Rumex giganteus) (Cuddihy and Stone 1990; Gagné and Cuddihy 1990).

Non-native species commonly found in the māmane woodlands include the invasive grass species discussed above and several herbs and shrubs including telegraph plant (Heterotheca grandiflora), hairy cat’s ear (Hypochoeris radicata), Virginia pepperweed (Lepidium virginicum), and common mullein (Verbascum thapsus) (Gagné and Cuddihy 1990). Common mullein is an invasive species and is listed as a Hawai‘i State Noxious Weed (Division of Plant Industry 1992; DOFAW n.d.). Other state and federal noxious weeds found in the subalpine community include the federally listed Kikuyu grass (Pennisetum clandestinum), and the state listed fountain grass (Pennisetum setaceum) and fireweed (Senecio madagascariensis). Common mullein and telegraph plants were very abundant in the vicinity of Hale Pōhaku in October 2007 (personal observation). Invasive species are discussed further in Section 2.2.1.3.3.

Plant Communities at Hale Pōhaku

Char (1999a) describes māmane woodlands at Hale Pōhaku as clumps of māmane trees, 16 to 18 ft tall, interspersed with open areas of bare soil or rocky outcroppings. She describes understory plants at Hale Pōhaku as tending to be denser under and around the clumps of māmane, with groundcover plants being primarily mixed bunch grasses forming upright tussocks. The most abundant grasses are two native grasses, alpine hairgrass (Deschampsia nubigena) and pili uka (Trisetum glomeratum), and an introduced needlegrass, Nassella cernua (called Stipa cernua in Char’s 1999 report and all older references). Common non-native grasses and herbaceous species found at Hale Pōhaku include ripgut brome (Bromus diandrus), orchardgrass (Dactylis glomerata), hairy cats-ear (Hypochoeris radicata), alfilaria or pin clover (Erodium cicutarium), sheep sorrel (Rumex acetosella), common groundsel (Senecio vulgaris), and common mullein (Verbascum thapsus). Patches of non-native California poppy (Eschscholzia californica) are locally common near the cabins. Char (1999a) does not mention the high density of common mullein or fireweed (Senecio madagascariensis) currently found at Hale Pōhaku. In fact, fireweed is not mentioned at all in Gerrish (1979) or Char (1985, 1999a), suggesting this population increase is a recent development.

In addition to the māmane woodland found at Hale Pōhaku, there is a small grove of Eucalyptus trees above the information station parking lot. A few shrubs of non-native tagasaste, or broom (Cytisus palmensis), also occur here.

Subalpine Fungal Communities

There have been relatively few studies of the higher elevation fungal communities on Mauna Kea, and no fungal surveys have been conducted at Hale Pōhaku itself. Despite the dry conditions on Mauna Kea’s upper elevations, there are a wide variety of fungal species that inhabit the subalpine and alpine habitats found there. A survey of higher fungi⁸ conducted in the māmane-naio forests on Mauna Kea between elevations of 6,000 and 9,000 ft (1,828 and 2,743 m) found 71 species of Ascomycetes (cup fungi such as yeast, mildew, morels and truffles) and Basidiomycetes (club fungi such as mushrooms, toadstools, earthstars, stinkhorns, brackens, rusts, and smuts) (Gilbertson et al. 2001). Desert stalked puffballs and earthstars are characteristic fungi found in higher elevation areas on Mauna Kea and commonly appear after rains (Hemmes and Desjardin 2002). Some of the more common ground-dwelling species that occur in māmane-naio woodlands include the salt-and-pepper shaker earthstar (Myriostoma coliforme), partially-buried puffballs such as Disciseda anomala and Disciseda verrucosa, fornicate earthstars (Geastrum fornicatum), hygroscopic earthstars (Geastrum corollinum and G. campestre), desert stalked puffballs (Battarraea phalloides), and stalked puffball (Tulostoma fimbriata var. campestre) (Hemmes and Desjardin 2002). Hemmes and Desjardin (2002) report Tulostoma fimbriata var. campestre growing above the treeline, at 9,842 ft (3,000 m), often in association with plants such as the silversword (Argyroxiphium sandwicense ssp. sandwicense). Some of the more common fungi that appear on trees and downed tree-branches include Heliocybe sulcata and Hypoxylon submonticulosum, conks such as Phellinus robustus, and bracket fungi such as Gloeophyllum trabeum (Gilbertson et al. 2001; Hemmes and Desjardin 2002). A new species of witch-broom-forming fungus (Botryosphaeria mamane) has been discovered growing on māmane trees, generally causing death of the branches it infects (Gardner 1997). Other newly discovered species include four white-rot associated fungi, Hyphodermella maunakeaensis, Phanerochaete crescentispora, and Radulomyces kama‘aina, and Radulomyces poni (Gilbertson et al. 2001).

An important group of fungi for the functioning of native ecosystems are the mycorrhizal fungi, which form symbiotic associations with the roots of plants (Habte 2000). The plants provide the fungi with carbohydrates (from photosynthesis) and in return, the fungi greatly increase the surface area of the roots for better absorption of water and mineral nutrients such as phosphates (Gemma and Koske 2001). The presence of the fungi may also improve plant resistance to disease (Habte 2000). Plants grown in areas where the mycorrhizal fungi have been eliminated (such as disturbed, eroded, or denuded areas) often do

8 Higher fungi are those that produce complex fruiting bodies and release spores (for example, mushrooms). Lower fungi include the Zygomycotina and the Chytridiomycotina. Chytrid fungi are important saprophytes and parasites in both aquatic and terrestrial habitats and are biodegraders of materials such as chitin, keratin and cellulose. They also play a role in nutrient recycling. Chytrid fungi have been implicated in the global reduction of frog populations. Zygomycetes are mostly terrestrial fungi and live in decaying plant or animal matter. Bread mold (Rhizopus stolonifer) is an example of zygomycotinid fungi.

very poorly (Habte 2000; Gemma and Koske 2001). The most common mycorrhiza are the arbuscular mycorrhiza (AM). The fungi found in arbuscular mycrorrhizae are generally not plant-specific, and are difficult to study because they cannot be grown without the host plant (Gemma and Koske 2001). Mycorrhizae are especially important to plant growth in nutrient-poor soils, such as in Hawai‘i where soils tend to have low amounts of available phosphorous (Gemma et al. 2002). Over 90% of endemic Hawaiian plants regularly form arbuscular mycrorrhizae in the field, and most of these species require AM to grow in low-fertility soils (Gemma and Koske 2001; Gemma et al. 2002). AM fungi are found in most Hawaiian soils, even in high altitude areas and on young lava flows (Gemma et al. 2002; Koske and Gemma 2002). Many native plants in the subalpine māmane woodlands and shrublands that have been tested were found to form associations with AM fungi. Native species found to form AM include māmane (Sophora chrysophylla), pūkiawe (Leptecophylla tameiameiae), ‘ōhelo (Vaccinium reticulatum), ‘aiakenēnē (Coprosma ernodeoides), ‘a‘ali‘i (Dodonaea viscosa), na‘ena‘e (Dubautia ciliolata ciliolata, and D. scabra), ‘ōhi‘a (Metrosideros polymorpha), and ‘ūlei (Osteomeles anthyllidifolia) (Koske et al. 1990; Gemma and Koske 2001). As research into AM fungi continues, there is no doubt that additional native species will be found to form associations with AM fungi. Many non-native invasive species also form associations with AM fungi (Koske et al. 1992). The relationships between invasive plants and mycorrhizal communities are discussed further in Section 2.2.1.3.3 (Invasive Plants). Because mycorrhizal fungi are easily eliminated in disturbed and barren areas, any restoration or transplanting attempts made in the subalpine and alpine zones on Mauna Kea should be done with seedlings that have been inoculated with AM fungi in the greenhouse, in order to increase the chance of establishment of the plants in the field (Habte 2000; Gemma and Koske 2001).

2.2.1.1.1      Threatened and Endangered Species

Endangered plant species (federal and state) found (historically and/or currently) in the subalpine community include the Mauna Kea silversword (Argyroxiphium sandwicense subspecies sandwicense), diamond spleenwort (Asplenium fragile var. insulare), kiponapona (Phyllostegia racemosa var. racemosa), and Hawaiian vetch (Vicia menziesii). The only Threatened plant species found in the subalpine community is Hawaiian catchfly (Silene hawaiiensis).

Historical records indicate that the endangered Mauna Kea silversword grew abundantly as low as 6,000 ft (1,800 m) above sea level (Hartt and Neal 1940). The Mauna Kea silversword is found in a Department of Land and Natural Resources (DLNR)-maintained enclosure near Hale Pōhaku and in the MKSR. This spectacular but extremely rare species is discussed in more detail in Section 2.2.1.2.4.

Diamond spleenwort, a fern, is currently found in scattered populations on Hawaii Island between 5,250 and 7,800 ft (1,600 and 2,380 meters) elevation, including Hawaii Volcanoes National Park, Hilo, Pu‘u Hualalai, Pu‘u Wa‘awa‘a, 1823 lava flow, Hualālai summit, Keauhou Ranch, Pu‘u Huluhulu, Kapāpala Forest Reserve, and Pu‘u Moana and Pōhakuloa Training Area (Shaw 1997; USFWS 1998a). It was previously found on Mauna Kea as high as 9,600 ft (2,926 m) (Hartt and Neal 1940). This species has not been observed at Hale Pōhaku (Char 1999a).

Kiponapona is a vine normally found in mesic to wet forests on the windward slopes of Mauna Kea and Mauna Loa. It was recorded by Cuddihy in 1979 (Bishop Museum 2007a) as occurring in a subalpine community at Shipman Ranch, above Maulua and below Keanakolu Road, on the northeast slope of Mauna Kea. This species has not been observed at Hale Pōhaku (Char 1999a). Hawaiian vetch is a climbing herb that was previously found in the subalpine communities on Mauna Kea and Mauna Loa (Skottsberg 1931) but is currently found only at lower elevations (Wagner et al. 1990). This species has not been observed at Hale Pōhaku (Char 1999a).

Hawaiian catchfly is a sprawling shrub found in open, dry areas up to approximately 9,880 ft (3,011 m) in elevation (USFWS 2002). It is closely related to Silene struthioloides (Wagner et al. 1990). S. hawaiiensis was recorded at Hale Pōhaku by Char, in 1985. However, in her 1999a summary report, she observed only S. struthioloides (no species of Silene were recorded in her 1990 survey of Hale Pōhaku). It is possible that the Silene species at Hale Pōhaku are all Silene struthioloides, but this would need to be confirmed with a comprehensive vegetation survey.

All of the Threatened and Endangered plant species listed above have been impacted by grazing, habitat alteration, and invasive plant species.

Māmane woodlands are critical habitat for the endangered Palila (Loxioides bailleui), a bird now found only in māmane woodlands on Mauna Kea (Juvik and Juvik 1984). More information about the Palila can be found in Section 2.2.3. Information on the fauna found in the subalpine woodlands is presented in Sections 2.2.2 through 2.2.4.

2.2.1.1.2      Candidate Species and Species of Concern

The only federal and state Candidate species found in the subalpine community on Mauna Kea is makou (Ranunculus hawaiensis). Makou, an endemic buttercup, was once very plentiful in subalpine and alpine communities (Rock 1913; Hartt and Neal 1940). Makou populations have decreased due to predation by slugs and feral animals such as pigs, goats, cattle, and sheep, and competition with invasive plant species (USFWS 2006).

State Species of Concern in the subalpine community include ‘akoko (Chamaesyce olowaluana), Douglas’ bladderfern (Cystopteris douglasii), Mauna Kea dubautia (Dubautia arborea), and Hawaii black snakeroot (Sanicula sandwicensis).

‘Akoko, a small tree in the family Euphorbiaceae, was once common in the subalpine forest, but has been reduced in abundance, primarily due to fire and grazing of small trees and saplings by feral ungulates (Shaw 1997). Feral sheep and goats also girdle larger trees by stripping bark from their trunks (Shaw 1997).

Douglas’ bladderfern is an endemic fern found in low densities in both subalpine and alpine communities. It was not recorded as occurring at Hale Pōhaku by Char (1985, 1999a) or Gerrish (1979). However, it was recorded by Smith et al. (Smith et al. 1982) as occurring on the summit. This species is discussed further in Section 2.2.1.2.5.

The Mauna Kea dubautia is a large shrub or small tree found in subalpine and alpine communities on Mauna Kea. Dubautia are closely related to silverswords (Argyroxiphium), and often form hybrids with other Dubautia species and with members of the genus Argyroxiphium (Carr 1985).

Hawaii black snakeroot is an herb in the Apiaceae family. It is restricted to subalpine woodland and shrublands on Maui and Hawai‘i (Wagner et al. 1990). Little information is available about this species.

Most of these species have been greatly reduced in abundance due to grazing by feral animals, habitat alteration, and competition with introduced plants (Wagner et al. 1990).

2.2.1.1.3      Vegetation Surveys at Hale Pōhaku

Since 1979, there have been four qualitative⁹ botanical surveys at Hale Pōhaku: a 1979 study of the Hale Pōhaku area and two other locations by Grant Gerrish (Gerrish 1979), a 1985 study of the proposed construction camp site and staging areas by Char (Char 1985), a 1990 study of the proposed dormitory area for the Subaru Japan National Large Telescope (JNLT), also conducted by Char (Char 1990), and a 2004 survey of a small area in the construction staging area at the lower limit of the Hale Pōhaku facility (Pacific Analytics 2004). There have been no surveys at Hale Pōhaku for fungi or lichens.

Gerrish (1979) surveyed two areas: The area now occupied by the upper buildings at the Mid-Elevation Facilities (between approximately 9,260 ft and 9,330 ft elevation) is termed Zone 1. The second area immediately to the south (Zone 2 or “proposed park”), is now the parking lot, stone cabins, and storage buildings for the lower Mid-Elevation Support Facilities. Zone 2 comprises an area from approximately 9,200 ft to 9,260 ft elevation. Gerrish does not discuss his methodology other than to state that the area was explored by foot and that “each part of the site was visited several times.” No quantitative data were recorded, except for a rough count of māmane trees on the site, although locations of several plants of interest (Geranium cuneatum, Stenogyne rugosa, and Stenogyne microphylla) were recorded on a figure (see Figure 2.2-5 for a reproduction of this figure).

In 1985, Winona Char and an assistant conducted botanical surveys of three areas proposed for the location of the temporary construction camp housing at Hale Pōhaku. The three areas consisted of an area northeast of the existing Mid-Elevation Support Facilities (Area II in Char 1985), and two areas immediately south of the Visitor’s Information Station (Areas IA and IB in Char 1985). See Figure 2.2-6 for a reproduction of Char’s survey transect figure. Char states that “an intensive walk-through survey method was used.” Char recorded no quantitative data on species abundances; however, she noted species composition at the three areas surveyed, and presented these data in the species-list table included in her report. Thus it is possible to determine how widespread a given species was in the surveyed areas at Hale Pōhaku during that time period, and where areas of higher diversity were. For example, Area II of her report had 37 of the 42 species found, while area IA had 30, and Area IB had only 18.

Char’s 1990 study consisted of an “intensive walk through survey” of the Hale Pōhaku Dormitory area, as part of an assessment conducted for the Subaru (JNLT) telescope mid-level facilities (Char 1990). The region covered was similar to that of Gerrish (1979), although she covered a little less area. See Figure 2.2-7 for a reproduction of Char’s (1990) survey area. Much of the groundcover in the area of the actual dormitories had previously been removed for the construction of the Keck dormitory. No quantitative data on species abundances were recorded. No Threatened or Endangered species were observed during the survey, and Char does not mention the presence of Silene hawaiiensis, recorded in her earlier survey at Hale Pōhaku (Char 1985). Char mentions in the report that two weedy species previously not recorded from Hale Pōhaku were found during this survey: rabbit-foot clover (Trifolium arvense) and telegraph weed (Heterotheca grandiflora).

9 A qualitative botanical survey identifies the plant species in an area and may estimate abundances (e.g., common, rare) based on the observer’s opinion, without recording actual data on population sizes or distributions. A quantitative study records the species and provides a measure of population sizes (or densities), usually by counting individuals in a given area such as a transect. Most of the botanical surveys conducted at Hale Pōhaku and in the MKSR have been qualitative.

In 2004, a botanical survey was conducted in the 0.5 acre (0.2 ha) construction staging area at the lower limits of the Hale Pōhaku facility. The survey was conducted to determine if using the staging area for construction of additional Keck telescopes would impact the native vegetation (and the endangered Palila habitat) (Pacific Analytics 2004). The survey area has been used for construction staging since 1990 and is also used for overflow parking at the Visitor Information Station. The survey covered the entire staging area and a buffer of 100 ft. (31 m) around the staging area (see Figure 2.2-8). Survey methodology is not described, and the survey found no māmane trees within the staging area (and, in fact, it found very little vegetation at all), but did find māmane in the 100 ft. buffer area. Groundcover at the site consisted mainly of the invasive ripgut brome (Bromus diandrus), scattered native alpine hairgrass (Deschampsia nubigena) and pili uka (Trisetum glomeratum), and invasive needlegrass, Nassella cernua (called Stipa cernus in the report). Other species found in the survey include common groundsel (Senecio vulgaris), pin clover (Erodium cicutarium), common mullein (Verbascum thapsus). Pacific Analytics also recorded the presence of evening primrose (and included a photo), but mistakenly gave the scientific name for willow herb (Epilobium billardierianum ssp. cinereum), another non-native herb in the same plant family as evening primrose. The correct scientific name for evening primrose is Oenothera stricta ssp. stricta. Both willow herb and evening primrose are present at Hale Pōhaku. Other than the māmane trees and scattered native grasses, no native plants were observed within the surveyed area (Pacific Analytics 2004).

2.2.1.2 Alpine Plant Communities (Summit Access Road and MKSR)

Alpine plant communities on Mauna Kea begin just above the treeline, at approximately 9,500 ft (2,900 m), and rise to the summit of the mountain at 13,795 ft (4,205 m). The alpine plant communities can be divided into three basic types: shrublands, grasslands, and stone desert. There are no sharp lines of delineation between the plant community types; the three communities grade into one another, beginning with the alpine shrubland at the treeline, which grades into the alpine grasslands, and culminates with the alpine stone desert, at the summit (Mueller-Dombois and Fosberg 1998; Char 1999b; Conant et al. 2004; Aldrich 2005).

There have been few detailed studies of the alpine plant communities on Mauna Kea, although there are some useful descriptive historical accounts (Hartt and Neal 1940, and references therein). The three community types are all characterized as being predominantly barren rock and cinder with sparse vegetation (Aldrich 2005). Plant density decreases with increasing elevation, with the result that there are only scattered plants at the higher elevations. The alpine shrublands are inhabited mainly by low-lying shrubby species, while the upper elevations are inhabited by grasses and herbaceous species (Mueller- Dombois and Fosberg 1998). Heavy grazing by feral ungulates has decimated the plant communities in the alpine shrublands and grasslands (Hartt and Neal 1940; Mueller-Dombois and Fosberg 1998), and invasive plant species now compete with native plants for limited resources such as water and sheltered growing locations. The three plant communities are described in further detail in Sections 2.2.1.2.1 through 2.2.1.2.3. Threats to the alpine plant communities are described in Section 2.2.1.3, and information gaps are discussed in Section 2.2.1.4.

The alpine shrublands on Mauna Kea are dominated by pūkiawe (Leptecophylla tameiameiae), and are often referred to as Leptecophylla shrublands¹⁰ or scrub desert (Mueller-Dombois and Fosberg 1998;  Char 1999b; Aldrich 2005). Leptecophylla shrublands are the dominant plant community from the treeline at 9,500 ft (2,900 m) to around 11,150 ft (3,400 m) above sea level (Mueller-Dombois and Fosberg 1998). These shrublands are also found below the treeline in the subalpine zone, as mentioned in Section 2.2.1.1. The density and diversity of plant species found in the Leptecophylla shrublands decreases with increasing altitude, from the subalpine region to the alpine region. At the upper elevations of its range, the Leptecophylla shrublands consist mainly of scattered pūkiawe shrubs and tufts of native grasses (Mueller-Dombois and Fosberg 1998).

Native herbs and shrubs commonly found in Leptecophylla shrublands include ōhelo (Vaccinium reticulatum), alpine catchfly (Silene struthioloides), and Mauna Kea dubautia (Dubautia arborea). Native ferns found in this community include Douglas’ bladderfern (Cystopteris douglasii), kalamoho (Pellaea ternifolia), ‘olali‘i (Asplenium trichomanes), and ‘iwa‘iwa (bird’s nest ferns, Asplenium adiantum- nigrum). Native grasses found in Leptecophylla shrublands include Hawaiian bentgrass (Agrostis sandwicensis), and pili uka (Trisetum glomeratum). Species historically common, but now uncommon, found in this community include ‘āhinahina (the Mauna Kea silversword, Argyroxiphium sandwicense ssp. sandwicense), lava dubautia (Dubautia ciliolata ssp. ciliolata), ‘ōhelo papa (Hawaiian strawberry, Fragraria chiloensis), ‘ena ‘ena (Pseudognaphalium sanwicensium)¹¹, nohoanu (Geranium cuneatum  ssp. hololeucum) and alpine tetramolopium (Tetramolopium humile ssp. humile var. humile). See Section 2.2.1.2.4 for more information about the silversword and other rare species found in the alpine shrublands.

There are several non-native plant species that have taken hold in the alpine shrublands on Mauna Kea. Non-native herbs found in this community include hairy cat’s ear (Hypochoeris radicata), sheep sorrel (Rumex acetosella), common mullein (Verbascum thapsus), fireweed (Senecio madagascariensis), and the common dandelion (Taraxacum officinale). Historically recorded non-native herbs include big chickweed (Cerastium fontanum), bull thistle (Cirsium vulgare), hairy horseweed (Conyza bonariensis), and woodland groundsel (Senecio sylvaticus). Although they were not recorded in the MKSR by Char (1999), who did not survey below 12,000 ft (3,650 m), these species are likely still found in the alpine shrubland community on Mauna Kea. Non-native grasses found in the Leptecophylla shrublands include Kentucky bluegrass (Poa pratensis), and historically, annual bluegrass (Poa annua), and velvet grass (Holcus lanatus).

Common mullein (Verbascum thapsus) and sheep sorrel (Rumex acetosella) were observed to be abundant along the Summit Access Road in the lower regions of the alpine shrubland plant community in October 2007 (J. Garrison, personal observation), and have been found at the summit near the observatories (Ansari 2008). These species are discussed in more detail in Section 2.2.1.3.3.

10 Formerly called Styphelia shrublands in older references, due to a name change for pūkiawe from Styphelia tameiameiae to Leptecophylla tameiameiae. Some scientists further divide the shrublands into Leptecophylla alpine-scrub (9,500–10,500 ft/2,900–3,200 m) and Leptecophylla low-scrub desert (10,500–11,150 ft/3,200–3,400 m) (Mueller-Dombois and Fosberg 1998). The Leptecophylla Alpine-scrub is composed of primarily of tall densely growing pūkiawe shrubs. The Leptecophylla Low-scrub Desert is composed of scattered, low-growing pūkiawe shrubs, two native grasses (Agrostis sandwicensis and Trisetum glomeratum), three native fern species (Pellaea ternifolia, Asplenium adiantumnigrum, A. trichomanes), two native composites (Tetramolopium humile, Pseudognaphalium sandwicensium), and one invasive weed, hairy cat’s ear, Hypochoeris radicata (Mueller-Dombois and Fosberg 1998; Conant et. al 2004).

11 Called Gnaphalium sandwicensium in older references.

Very little information is available regarding the fungal communities present in the alpine regions on Mauna Kea. The stalked puff-ball (Tulostoma fimbriata var. campestre) can be found growing above the treeline, often in association with plants such as the silversword (Hemmes and Desjardin 2002). A study of the endangered silversword (Argyroxiphium sandwicense ssp. macrocephalum), and na‘ena‘e (Dubautia menziesii) on Haleakalā, Maui, found that both species formed associations with AM fungi, including Entrophospora infrequens and several unidentified species in Glomus, Scutellosopra, and Acaulospora (Koske and Gemma 2002). It can be assumed that similar relationships can be found between AM fungi and Argyroxiphium and Dubautia species found on Mauna Kea.

2.2.1.2.2      Alpine Grassland

Alpine grasslands replace Leptecophylla shrublands around 11,000 ft in elevation (3,400 m), although Leptecophylla (pūkiawe) shrubs can be found in all habitats, clear to the summit (Mueller-Dombois and Fosberg 1998). The alpine grasslands on Mauna Kea, which occur up to 12,800 ft (3,900 m) in elevation, are dominated by two native grasses, Hawaiian bentgrass (Agrostis sandwicensis), and pili uka (Trisetum glomeratum) (Mueller-Dombois and Fosberg 1998). Char (1999b) recorded that the Hawaiian bentgrass was more abundant than pili uka, although both are found at very low densities. Other native species found in the alpine grassland community include those found in the alpine shrubland communities, although at much lower densities.

Very few good stands of alpine grassland currently exist due to overgrazing by feral and domestic sheep and goats.

2.2.1.2.3      Alpine Stone Desert

The alpine stone desert plant community is found above 12,800 ft (3,900 m) on Mauna Kea (Mueller- Dombois and Fosberg 1998). This plant community consists of several species of mosses and lichens, an unknown number of species of algae, and a limited number of vascular plants, predominantly the same species found in the alpine shrublands and grasslands (Hartt and Neal 1940; Char 1999b; Aldrich 2005). Most of the species of plants found in the region are endemic (occurring only in Hawai‘i) or indigenous (native to Hawai‘i but occurring elsewhere). A few non-native plant species have also become established here, even at the summit (Hartt and Neal 1940; Char 1999b). The composition of this plant community is discussed in more detail below in Sections 2.2.1.2.3.1 and 2.2.1.2.3.2.

High wind speeds, high solar radiation, regular freezing and thawing cycles, low precipitation, high rates of evaporation, and the porosity of the substrate all limit the development of the plant and animal communities in this zone (Aldrich 2005). Plant density is extremely low in this high elevation climate, and plant distribution is determined primarily by substrate type (Smith et al. 1982). Cinder cones do not provide suitable growing habitat for most plants because of the instability of the surface material, which is destructive to plant root systems, and the high porosity of cinders, which allows for rapid water drainage (Hartt and Neal 1940; Char 1999b). Additionally, the absence of organic matter in the soil further decreases its ability to hold water (Hartt and Neal 1940), making water and available nutrients limiting resources in this region.

Mosses and lichens are found in protected areas on andesite (Hawaiite-mugearite) lava flows, in pits, fissures, small caves, overhangs and shaded pockets and crevices (Char 1999b). Vascular plants are found mainly at the base of rock outcrops where there is an accumulation of soil and moisture, and some protection from wind (Char 1999b). Aeolian and colluvial material found scattered throughout the lava flows in low-lying swale areas provide poor habitat for plants (Char 1999b).

2.2.1.2.3.1    Algae, Lichens, and Mosses

Algae species have not been extensively surveyed in the alpine stone desert on Mauna Kea. Several species of algae and diatoms are found in Lake Waiau (Massey 1978), and one species of algae (Haematococcus sp.) is known to occur on snow banks, staining the snow red (Smith et al. 1982; Aldrich 2005). There are undoubtedly species of algae present in the soils of Mauna Kea (Smith et al. 1982).

Lichens are a symbiotic relationship between a fungus (generally an Ascomycete) and a green alga, a blue green bacterium, or both (Hemmes and Desjardin 2002). A survey of lichens found on Mauna Kea was conducted in 1982 by Smith, Hoe, and O’Conner. They identified 21 species of lichens and five possible other species that could not be collected because they were crustose species imbedded in the andesite flows. A complete list of lichen species observed on Mauna Kea is presented in Table 2.2-4. Around half of the lichen species found on Mauna Kea are endemic, two of which (Pseudephebe pubescens and Umbilicaria pacifica) are limited to Mauna Kea alone (Smith et al. 1982; Char 1999b). Pseudephebe pubescens, a species primarily found in high altitude and alpine regions of the world (Smith et al. 1982), has not been recorded anywhere else in Hawai‘i or on any other tropical island. The remaining species were indigenous to the Hawaiian Islands. Lecanora muralis is the most abundant lichen on Mauna Kea, and is found throughout the summit, on all substrate types, including cinders and colluvial material on the cinder cones up to the summit of Pu‘u Wēkiu (Smith et al. 1982). Other common species on the summit are Lecidea skottsbergii and Candelariella vitellina, both of which are found on rocks “larger than a small fist” (Smith et al. 1982).

Lichens are found throughout the summit of Mauna Kea, but the highest densities and diversity of lichens tends to be found on andesite rocks, in north- and west-facing protected locations, away from direct exposure to the sun (Smith et al. 1982). Areas to the west of the major cinder cones have a low density and diversity of lichens, most likely due to a rain shadow effect created by the cinder cones (Smith et al. 1982).

Two areas of high lichen concentration and unique assemblages were identified by Smith et al. (1982): the southern slope of Pu‘u Wēkiu, just below the Switchback Road (Intensively Studied Area 7), and the lava flows north of Pu‘u Poli‘ahu (Intensively Studied Areas 2, 3, and 4) (Smith et al. 1982). The  southern slope of Pu‘u Wēkiu has many large rocks, and it supports the “highest substantial colony of lichens in the state” (Smith et al. 1982). The lava flows north of Pu‘u Poli‘ahu are characterized by a high diversity of lichens, including Pseudephebe pubescens (Smith et al. 1982).

Using information from Smith et al. (1982), Char (1999b) identified four lichen communities on the summit of Mauna Kea, based on species composition, substrate, and orientation (north-south). These lichen communities include: 1) nearly vertical north-facing andesite rocks characterized by an association of Umbilicaria hawaiiensis, Pseudephebe pubescens, and Lecanora muralis; 2) vertical west-facing andesite rocks characterized by a mixed association of Acarospora depressa, Candelariella vitellina, Lecanora muralis, Lecidea skottsbergii, Lecidea vulcanica, Physcia dubia, Rhizocarpon geographicum, and Umbilicaria hawaiiensis; 3) south-facing rocks characterized by an association of Umbilicaria pacifica, Physcia dubia, Lecanora muralis, Candelariella vitellina, and Lecidea skottsbergii; and 4) cinder cones, deposits of aeolian or colluvial material on lava flows, and scattered rocks and cobbles. Diversity of species was low on cinder cones and on aeolian and colluvial materials on lava flows, with only the most common lichen species present, such as Lecanora muralis. Candelariella vitellina and

Lecidea skottsbergii are found on small rocks or cobbles scattered throughout the cinder and colluvial material (Char 1999b). In addition, there are numerous small caves throughout the summit region that are colonized by Lepraria species. Lepraria can tolerate deep shade and can be found up to three meters deep in some of the larger caves (Smith et al. 1982).

Mosses in the alpine stone desert occur in protected places where water is more consistently available, such as under overhanging rocks and in shaded crevices or caves where snow melts slowly (Smith et al. 1982). Mosses are predominantly found on the north-northeast and south-southeast facing sides of rocky mounds, generally in association with runoff channels from snow melt (Smith et al. 1982). Moss cover was much lower in the rain-shadow region west of the summit cone, due to the more arid conditions (Smith et al. 1982). Mosses have not been observed in loose cinders or on the aeolian or colluvial fields (Char 1999b).

Smith et al. (1982) conducted a survey of the mosses on the Mauna Kea summit area (above 13,000 ft, 3,960 m) and found approximately 12 species (some could not be identified with certainty to the species level), most of which are indigenous to the Hawaiian Islands. Two species, Bryum hawaiicum and Pohlia mauiensis are endemic (Smith et al. 1982). All the moss species found there are related to temperate species. The most common species of moss were a previously undescribed species of Grimmia and  Pohlia cruda (Smith et al. 1982).

Grimmia are silvery-gray mosses that form clumps in run-off channels and semi-exposed rock faces. Members of this genus are the mosses most often seen at the summit (Smith et al. 1982). Pohlia cruda is  a bright green moss found in well-protected, deeply shady locations. Pohlia species are so well hidden they are unlikely to be seen by the causal observer (Smith et al. 1982). The remaining moss species were not as abundant and tended to occur in habitats intermediate between the somewhat exposed Grimmia habitats and the protected Pohlia habitats (Smith et al. 1982). A complete list of mosses observed on the summit of Mauna Kea is presented in Table 2.2-5.

2.2.1.2.3.2    Vascular Plants

Very few species of vascular plants are found within the summit area (Smith et al. 1982; Char 1999b). The most abundant native vascular plant species found at this elevation are two grass species, Hawaiian bentgrass (Agrostis sandwicensis) and pili uka (Trisetum glomeratum), and two fern species, ‘iwa‘iwa (Asplenium adiantum-nigrum) and Douglas’ bladderfern (Cystopteris douglasii). Of these four species, Hawaiian bentgrass is the most common. The grasses tend to be found at the bases of large rock outcroppings where fine substrate and moisture accumulate (Char 1999b). The native fern, ‘iwa‘iwa, is found on cinder plains and lava flows from the summit down to approximately 2,000 ft (610 m) (Valier 1995; NASA 2005). Douglas’ bladderfern grows on weathered rocks up to 13,400 ft elevation (4,084 m) (Char 1999b). Historically, the Mauna Kea silversword (Argyroxiphium sandwicense ssp. sandwicense), pūkiawe (Leptecophylla tameiameiae), ōhelo (Vaccinium reticulatum), and alpine catchfly (Silene struthioloides) have been observed at or near the summit (Hartt and Neal 1940; Mueller-Dombois and Fosberg 1998). Some of these plants may still be present in more remote, unsurveyed areas.

Non-native species found in the alpine stone desert include Hairy cat’s ear (Hypochoeris radicata) and common dandelion (Taraxacum officinale), both of which are temperate weed species with a world-wide distribution (Smith et al. 1982; Char 1999b). Non-native species historically observed in the alpine stone desert include annual bluegrass (Poa annua), Kentucky bluegrass (Poa pratensis), big chickweed (Cerastium fontanum ssp. vulgare), bull thistle (Cirsium vulgare), hairy horseweed (Conyza bonariensis), sheep sorrel (Rumex acetosella), and common chickweed (Stella media) (Hartt and Neal 1940). Individuals or populations of these species may still be present in the area.

Smith et al. (1982) observed fragments from other vascular plant species, including one grass and one legume species. As they were unable to locate the source of these fragments, they postulated that these species were blown up to the summit by wind. Wind-borne seeds and plant fragments from lower elevations may act as sources for invasive plant species to the alpine regions of Mauna Kea, although many lowland species will not be able to grow there due to the harsh conditions.

2.2.1.2.4      Threatened and Endangered Species

‘Āhinahina (the Mauna Kea silversword, Argyroxiphium sandwicense ssp. sandwicense) is the only federally Endangered species found in the alpine vegetation communities on Mauna Kea. The Mauna Kea silversword is a subspecies of silversword found only on Mauna Kea, and historically occurred from 8,500 ft (2,700 m) to 12,300 ft (3,750 m) (Wagner et al. 1990; Robichaux et al. 2000). Hartt and Neal (1940) describe the silversword as being found as low as 6,000 ft (1,830 m) in elevation in historical times. ‘Āhinahina is a spectacular plant, with thick, sword-shaped, shiny, silvery-green leaves growing in a giant rosette. When it flowers, the Mauna Kea silversword grows a large stalk, up to nine feet tall, that is covered with up to 600 pink to wine-red flowers (Wagner et al. 1990).

Although they are now extremely rare, the Mauna Kea silversword was once so common on Mauna Kea that the dry leaves and stems were used as fuel for campfires (Cuddihy and Stone 1990). The population size of the Mauna Kea silversword has been drastically reduced through grazing by feral sheep (Ovis aries), goats (Capra hircus), mouflon sheep (Ovis musimon), and cattle (Bos taurus) (Hartt and Neal 1940; USFWS 1994; Robichaux et al. 2000). Their numbers began decreasing after the introduction of grazing animals, and the species was already rare as early as 1892, only 99 years after the introduction of the first grazing animals on Hawai‘i (Hartt and Neal 1940). By the 1970s there were only 34 individual silversword plants known to exist on Mauna Kea (Forsyth 2002). Although the impact of grazing ungulates on the silversword and other vegetation on Mauna Kea was recognized early on (Hartt and Neal 1940), the efforts to control feral ungulates on the mountain have waxed and waned over time, and grazing animals have never been eliminated from Mauna Kea (Juvik and Juvik 1984).

Recovery efforts for the Mauna Kea silversword are underway through the efforts of the U.S. Fish and Wildlife Service (USFWS), the Division of Forestry and Wildlife (DOFAW), and University of Arizona plant biologist Dr. Rob Robichaux. The recovery effort comprises an outcrossing program¹² in the field, greenhouse propagation of seeds, and outplanting seedlings into the wild (Aldrich 2005). To date over 4,000 seedlings have been outplanted in protected areas in the wild. There are currently five active,  fenced outplanting exclosures of the Mauna Kea silversword in the alpine shrubland and grassland areas on Mauna Kea, and one naturally occurring population at Waipahoehoe gulch (USFWS 1994; Aldrich 2005). Recently, a small population of Mauna Kea silverswords was discovered in the MKSR (Nagata 2007; Tomlinson 2007).

Due to the drastic reduction in population size, and early propagation attempts using only three individual plants as founders for outplanted populations, the silversword has gone through a genetic bottleneck and lost some genetic diversity (Robichaux et al. 1997; Friar et al. 2000). Adding to the problem, there are

12 Outcrossing is the process whereby the pollen from the flower of one plant is placed, by hand, on to the receptive area of the flower of another plant, usually some distance away. Because the Mauna Kea silversword is self-incompatible (meaning that it cannot pollinate itself), the outcrossing program ensures that each plant receives pollen from an unrelated (or at least less closely related) plant. This protects the genetic diversity in the species and ensures a higher output of viable seed from individual plants.

2.2.1.2.5      Candidate Species and Species of Concern

There are no federal or state Candidate species found in the alpine regions of Mauna Kea. There are two state Species of Concern found in this region, Mauna Kea dubautia (Dubautia arborea) and Douglas’ bladderfern (Cystopteris douglasii).

Dubautia arborea, or na‘ena‘e is a small tree or shrub found in subalpine and alpine communities on Mauna Kea. Dubautia are closely related to silverswords (Argyroxiphium spp.), and often form hybrids with other Dubautia species and with species of Argyroxiphium (Carr 1985). Its numbers have been reduced due to grazing by feral animals, habitat alteration, and competition with introduced plants (Wagner et al. 1990; World Conservation Monitoring Centre 1998).

Cystopteris douglasii is a small, endemic bladderfern that grows on weathered rocks exposed to trade winds (Char 1999b). C. douglasii on Mauna Kea is unusual because other members of this genus grow in more-protected microclimates (Char 1999b). It is found only from high elevation areas on Maui and Hawai‘i. Char (1999b) believes that the Mauna Kea Cystopteris douglasii may represent a new variety or even a new species of Cystopteris. This already rare species is threatened by habitat alteration, invasive species, and grazing animals (Hawaii Biodiversity and Mapping Program n.d.).

2.2.1.2.6      Vegetation Surveys in MKSR

There have been no quantitative vegetation surveys in the MKSR. There are many descriptive historical accounts of the vegetation on Mauna Kea, dating back to 1826, and one of the more detailed historical vegetation accounts was conducted by Hartt and Neal in 1935 (Goodrich 1826; Baldwin 1890; Alexander 1892; Douglas 1914; Hartt and Neal 1940). The Hartt and Neal study lists all plant species collected on Mauna Kea during the 1935 botanical survey of the mountain, including the highest elevation at which each species was seen. This study provides valuable information on historical presence of species on the mountain that can serva as a baseline with which to compare modern day surveys.¹³

13 Plant species recorded by Hartt and Neal (1940) can be identified in Tables II.2-3 through II.2-5 by the number 3 in the Reference (Ref.) column.

C.W. Smith, W.J. Hoe, and P.J. O’Conner conducted a thorough descriptive vegetation study in a limited region at the summit of Mauna Kea. Figure 2.2-9 shows the locations of their surveys, which were limited to “only those regions considered for future telescope construction to the year 2000 as described in the MKSR Master Plan (July 1982)” (Smith et al. 1982). This vegetation survey covered seven “intensively studied” areas, which were carefully searched to get a detailed record of the species of lichens, mosses, and vascular plants present. The report does not provide information on type of survey methods used (e.g., transects, random sample locations, wandering searches, systematic searches) in the intensively studied areas. Figure 2.2-9 also shows the “reconnaissance areas” included in the study, but provides no detail on what level of effort was put into detailing plant species found in these areas. The report does state that “no formal quantitative sampling was undertaken because the amount of cover was too low for conventional techniques” (Smith et al. 1982). Most of the information on moss and lichen species in the MKSR presented in this Natural Resources Management Plan comes from this report.

The other three recent plant surveys in the MKSR were conducted by Winona Char. In 1988, Char conducted a survey of the proposed site for the Very Long Baseline Array (VLBA) antenna facility, between 12,200 and 12,400 ft (3,720 and 3,780 m) elevation, and an for alternative site at 11,800 ft (3,600m) elevation (MCM Planning 1988). A “walk-through” survey method was used in this study. The report provides a species list, recording present/absence of species at the proposed site, the alternative site, and the Summit Access Road near the site. Species recorded were a subset of those found by Smith et al. (1982). In 1992, Char conducted a rapid survey for lichen species in the future location of the Smithsonian Radio Telemetry Facility, to aid in placement of the pads to avoid areas of high lichen abundance (MCM Planning 1994). No data on species abundance or composition are presented in this report. In 1999, Winona Char produced another report on the plant communities of the summit area of Mauna Kea. Most of the information in the 1999 report came from the Smith et al. (1982) vegetation survey. Information was also gathered by Char on June 21, 1999, during a “reconnaissance-level field survey” of the “slope beyond the summit ridge and to the northwest of the summit ridge”, in the areas proposed for the “Next Generation Large Telescope and the Optical Interferometer Array Site” (Char 1999b). No information on survey methodology is provided in the report, and no information is provided on the species located in these specific areas. Unfortunately, although the report states there is a map showing survey locations, no map was present in the copy of the reports provided with the Master Plan. A list of species observed, and their relative abundance is provided, although there is no information on how relative abundance was established.

There have been no studies of vegetation communities on Mauna Kea between the upper edge of Hale Pōhaku (9,340 ft/ 2,850 m) and 11,800 ft (3,600 m). No formal surveys have been conducted in the Ice Age Natural Area Reserve (NAR), although records are opportunistically kept when species of interest (particularly native species, or expansion of invasive species ranges) are noted.

2.2.1.3       Threats to Botanical Communities on Mauna Kea

Threats to the subalpine and alpine botanical communities on Mauna Kea include habitat alteration for development, agriculture and livestock grazing, fire (in the subalpine and lower alpine communities), invasive plant species, non-native animals (such as feral goats, sheep, rats and arthropods), human uses, and climate change.

Habitat alteration threatens native plant communities by changing the growth environment to the extent that the species can no longer survive there. Examples of habitat alteration on Mauna Kea include agriculture, livestock grazing (in the subalpine zone), and development (buildings and infrastructure such as roads, parking lots, etc.). Invasive species may also alter habitat to make it unsuitable for native plant species. Invasive plant species are further discussed in Section 2.2.1.3.3. For Hale Pōhaku and the MKSR, most habitat alteration occurs through development such as building of new telescopes and associated facilities, use of unpaved areas for parking lots, off-road vehicle use, the spread of invasive plants, and grazing by feral ungulates. The effects of non-native animal species on the plant communities are further discussed in Section 2.2.1.3.4.

2.2.1.3.2      Fire

Subalpine communities on Mauna Kea are susceptible to fire because of the dry conditions there. Alpine communities are not as susceptible because of the low density of plants. Many native Hawaiian plants such as pūkiawe (Leptecophylla tameiameiae) are not fire tolerant (Hughes et al. 1991; Smith and Tunison 1992). Fires in subalpine woodland are rare natural events (Hess et al. 1999). However, fires from military training activities at Pohakuloa Training Area and accidental wildfires set along roadsides, near developments, and in recreational areas pose a threat to the subalpine dry forest and shrubland communities (Gagné and Cuddihy 1990). The presence of invasive grass species in the subalpine communities increases the risk of fire by providing a source of continuous fine fuels in areas that previously had naturally discontinuous fuel beds due to the patchy nature of the subalpine communities (Smith and Tunison 1992; Hess et al. 1999). Several species of invasive grasses also increase greatly in abundance after fires (Hughes et al. 1991), effectively inhibiting germination of native species such as māmane (Hess et al. 1999). Velvet grass (Holcus lanatus) and sweet vernalgrass (Anthoxanthum odoratum) are two species that increase rapidly after fires and provide fuels for further fires (Smith and Tunison 1992). Fountain grass (Pennisetum setaceum) is another extremely fire-prone species that grows in dense clumps and can alter the natural fire regime of an area (Smith and Tunison 1992; Benton 2006). This species, normally found at lower elevations, was recently discovered (and removed) at 9,000 ft (2,740 m) in Pohakuloa Training Area on Mauna Loa (Higashino 2008).

2.2.1.3.3      Invasive Plants

Non-native, invasive plant species can impact native plant communities by altering the environment, for example, by lowering groundwater table, changing fire regimes, increasing or decreasing shade, smothering plant growth. They also compete with native plants for limited resources such as nutrients, water and light, and can attract or support increased populations of herbivores and disease or parasite organisms. Invasive plants may also affect the mycorrhizal fungi that native Hawaiian plants rely on, and conversely, the presence of mycorrhizal fungi can either enhance or reduce an invasive species’ success in colonizing a new area (Stampe and Daehler 2003). One study found that some non-mycorrhizal invasive plants release antifungal chemicals that destroy or weaken the mycorrhizal soil communities and thus negatively impact the native species that rely on these fungi (Stinson et al. 2006). Another recent study found that the presence of invasive plant species can alter the diversity and composition of mycorrhizal fungal communities, which in turn could impact native plant communities (Hawkes et al. 2006). Other studies have suggested that the presence of mycorrhizal fungi may enhance the ability of some non-native plants to invade and compete with native species, and in some cases actually aid in the transfer of nutrients from the native plants to the invasive ones (Marler et al. 1999; Carey et al. 2004).

There are 151 recorded species of non-native plants in the Hawaiian Islands that grow above 6,500 ft (2,000 m), of which around 14% (21 species) are reported as being disruptive to native plant communities (Daehler 2005). Invasive plants currently found in the subalpine and alpine plant communities at Hale Pōhaku and MKSR include the non-native grasses described in Section 2.2.1.3.3 and invasive herbs such as common mullein (Verbascum thapsus) and fireweed (Senecio madagascariensis). The most common invasive plant species found in the subalpine and alpine regions of Mauna Kea are discussed below. See Figure 2.2-4 for photos of common invasive species found at Hale Pōhaku and MKSR.

Grasses: Invasive grasses such as needlegrass (Nassella cernua), ripgut brome (Bromus diandrus), orchardgrass (Dactylis glomerata), velvet grass (Holcus lanatus), rye grass (Lolium sp.), Kentucky bluegrass (Poa pratensis), and sweet vernalgrass (Anthoxanthum odoratum) are common in the subalpine regions of Mauna Kea. As discussed in Section 2.2.1.3.2, dense growth of invasive grasses increases the risk of fire in the dry subalpine zone by providing a continuous fuel source. In addition to increasing the risk of fire, invasive grasses compete with native species for nutrients and water, and directly impede regeneration of native plants by smothering seedlings (Hess et al. 1999). A few grass species, including annual bluegrass (Poa annua), Kentucky bluegrass, and sweet vernalgrass were historically recorded in the alpine plant community (Hartt and Neal 1940). It is unknown whether these species are still present and if so, whether they are impacting native plant species in the alpine community. Even at low densities there remains the possibility that non-native grasses are competing with native plant species for limited resources and protected growth areas.

Common mullein: Common mullein (Verbascum thapsus) is a Hawai‘i State Noxious Weed that is native to the temperate zone of Europe, and is adapted to disturbed dry and rocky sites (Juvik and Juvik 1992). It is a stout plant with thick, silvery, woolly or hairy leaves that grow in a rosette (somewhat similar to the Mauna Kea silversword). Common mullein produces a tall flowering stalk that produces thousands of seeds. Although bees often pollinate common mullein flowers, they are also able to self- pollinate (Ansari and Daehler 2000). This allows the spread of the plant in areas where pollinators are scarce. Like the Mauna Kea silversword, common mullein flowers once and then dies. However, it takes a little less than two years to reach maturity, while the silversword takes three to fifty years (USFWS 1994; Ansari and Daehler 2000). Mullein seeds can remain dormant in the soil for 100 years or more (Juvik and Juvik 1992). In an extreme example, mullein seeds from an archeological dig in Denmark dated 1300 AD were still viable in the 1960s (Odum 1965; Ansari and Daehler 2000). This means that even if adult plants are removed from an area, seedlings will continue to sprout and need to be removed for many years to come. Mullein is currently abundant at Hale Pōhaku and is present on roadsides and remote upland areas on Mauna Kea along the Summit Access Road, up to 12,460 ft (3,800 m) (Juvik and Juvik 1992; Ansari and Daehler 2000). No biocontrol insects or pathogens have been introduced to Hawai‘i to control this species (Ansari and Daehler 2000). Mullein appears to be unpalatable to grazing ungulates, due to the density of leaf hairs (Juvik and Juvik 1992; Ansari and Daehler 2000). Mowing or clipping of the flowering stalk causes mullein to produce more flowering stalks. Chemical control can also prove difficult, although there are a few chemicals, such as a 10% Roundup solution, that can be used (Ansari and Daehler 2000). Removing the entire plant before it flowers, or cutting the taproot appear to be the most effective means of control, although care must be taken to remove most of the taproot, or resprouting can occur (Ansari and Daehler 2000; Loh et al. 2000).

Telegraph weed: Telegraph weed (Heterotheca grandiflora) is a weed of dry, disturbed areas that is native to California and the southwestern United States and Mexico (Wagner et al. 1990). Not much information is available on the impacts of telegraph weed in Hawai‘i. Like mullein, telegraph weed has hairy, grey-green leaves and produces a long stalk. However, it is easily distinguished from mullein by the fact that it branches at the top of the stalk and has bright yellow, daisy-like flowers (Weed Society of

Fireweed: Fireweed (Senecio madagascariensis) is a Hawai‘i State Noxious Weed that originates from South Africa and was accidentally introduced to Hawai‘i in the 1980s, possibly in contaminated fodder imported from Australia (Division of Plant Industry 1992; Le Roux et al. 2006). Fireweed competes with other plants for limiting resources such as nutrients and water, and is a heavy invader of pasturelands (Le Roux et al. 2006). Fireweed is poisonous to livestock (Le Roux et al. 2006). Although it was not recorded as present at Hale Pōhaku or MKSR in previous plant surveys (Gerrish 1979; Char 1985, 1990, 1999a), it is now common at Hale Pōhaku and can be found along the Summit Access Road (Fox and IfA 2007). Fireweed has also been observed in the Ice Age NAR up to 12,000 ft (3,660 m) elevation (Cole 2007). Currently the Hawai‘i Department of Agriculture is working on a biological control program for this  weed (Hawaii State Office of Environmental Quality Control 2008).

Hairy cat’s ear: Hairy cat’s ear (Hypochoeris radicata) is a widely distributed weed originating from Eurasia (Wagner et al. 1990). Its leaves grow in a rosette at the base of the plant. Yellow daisy-like flowers are found on the tips of leafless branching flowering stems. It is similar in appearance to the common dandelion (Taraxacum officinale), but can be distinguished by the fact that leaves of hairy cat’s ear are covered with hairs and are shaped differently. The taproot is a popular food item for feral pigs, which may dig up large areas looking for them (Smith 1985). The plant is also a preferred forage item for grazing animals (Ohio Agriculture Research and Development Center n.d.). Hairy cat’s ear, or gosmer, is found both at Hale Pōhaku and in the MKSR (Smith et al. 1982; Char 1999a). Little information is available about the impacts of this species on native plant communities, but as it attracts foraging feral ungulates and competes with other species for water and nutrients, it most likely has a negative impact.

Common dandelion: Common dandelion (Taraxacum officinale) is a cosmopolitan weed of temperate climates, that is generally found in higher elevation, wet, disturbed areas in Hawai‘i (Wagner et al. 1990). On Mauna Kea it was found above 13,000 ft (3,900 m) by Smith et al. (1982), and was historically observed growing on the shores of Lake Waiau (Hartt and Neal 1940). Hartt and Neal (1940) also record it as occurring in the subalpine zone, down to 6,800 ft (2,000 m) on Mauna Kea, although it was not recorded as being present at Hale Pōhaku by Char (1985, 1999a) or Gerrish (1979). It is unknown what impact, if any, this weedy species has on native plant communities on Mauna Kea.

Future Invasions: A further threat to high elevation environments on Mauna Kea exists in invasion by new plant species not currently found there. Posing a particular threat are species that are adapted to subalpine, alpine, or arid environments. These may be introduced though deliberate introduction (plantings in landscaping), natural expansion by lower-elevation invasive species, or accidental introduction through human activities (such as seeds stuck to vehicles or visitors’ shoes). Introductions of non-native species continue in Hawai‘i, despite growing education about their destructive nature. Around 9% of non-native species found growing at high elevations in the Hawaiian Islands were first recorded in the past 30 years (Daehler 2005).

Over half (52%) of non-native species growing in high elevation areas in the Hawaiian Islands originate from Europe (Daehler 2005). While the number of non-native species drops off exponentially with increasing altitude, the proportion of temperate species increases linearly with elevation, up to 9,800 ft (3,000 m), at which point all the non-native species found are temperate in origin (and 80% are native to Europe or Eurasia). The vast majority (93%) of non-native species found in high elevation areas on the Hawaiian Islands are herbaceous (either grasses or herbs), and about one third (27%) are grasses (Daehler 2005). This may be due in part to the fact that many of the non-native plants in higher elevation zones are associated with ranching—a major source of introductions, as contaminants in feed and seed and as purposeful introductions for forage (Daehler 2005). However, despite the dominance of herbaceous species in the overall counts of high-elevation, non-native species, it is the woody species that make up the majority (73%) of disruptive invaders to high-elevation native plant communities (Daehler 2005). This information is important because it gives resource managers a tool to predict which new species may become invasive in the subalpine and alpine zones on Mauna Kea, and can help prioritize eradication efforts for those species. For example, resource mangers at Hale Pōhaku may prioritize eradication of a new species of shrub or tree originating from high-elevation areas in Europe over that of an herb originating from a low-lying tropical island. However, one must be careful not to over generalize, as there are some tropical introductions, such as fountain grass (Pennisetum setaceum), which are extremely harmful and aggressive invaders of high elevation areas (see below for more information on fountain grass).

There are several invasive plant species that may become established in the subalpine and alpine zone in the future, particularly if anthropogenic climate change affects the rainfall regimes in the Hawaiian Islands. One species which may pose a future threat to the subalpine communities on Mauna Kea is gorse (Ulex europaeus), an invasive shrub (and State Noxious Weed) currently found between 1,400 ft (450 m) and 7,870 ft (2,400 m) on Mauna Kea (Markin et al. 1988). Gorse thrives on soils derived from volcanic ash and does well in disturbed areas with low fertility (Leary et al. 2005), where it forms impenetrable thickets and smothers native plant growth (Daehler 2005). This species may be able to colonize the subalpine community through natural dispersal, accidental introduction of seeds, or through environmental changes brought about by climate change or by habitat alteration brought about by other invasive plant or animal species.

A second species that may invade the subalpine zone at Hale Pōhaku is fountain grass (Pennisetum setaceum). Although fountain grass grows and reproduces better at lower elevations, it is capable of surviving in the subalpine areas on Mauna Kea (Williams et al. 1995). It has already been observed at Pohakuloa Training Area at 9,000 ft (2,740 m) elevation (Higashino 2008), and it may just be a matter of time before it spreads further. Fountain grass is native to northern Africa, and in Hawai‘i it occurs in dry open places such as barren lava flows and cinder fields (Wagner et al. 1990). It exhibits a broad ecological tolerance which enables it to survive at a variety of temperatures, although it does appear to be susceptible to freezing (Williams and Black 1993; Williams et al. 1995). The upper limit of fountain grass on Mauna Kea may be determined by freezing temperatures (rather than by drought) – as global climate change increases temperatures on the mountain, this species is likely to increase its elevational range. It is considered a serious pest in dry areas, because it alters the natural fire regime and because it is an aggressive colonizer that out-competes native species (Wagner et al. 1990; Tunison 1992; Daehler 2005; Benton 2006). Fountain grass seeds are primarily wind dispersed but can also be spread by water, livestock, humans, vehicles, and possibly birds (Benton 2006). The seeds may remain viable in soil for  six years or longer (Tunison 1992), making control difficult.

It is impossible to accurately predict the exact plant species which will invade the subalpine and alpine zones on Mauna Kea in the future, but managers must be especially wary of plant species that are adapted to dry climates, early successional habitats, high elevation climates, have wind-dispersed seeds, and/or that originate from the temperate zone.

Introduced animals, ranging from insects to mammals to birds, can have a detrimental effect on native plant communities. This is demonstrated especially well by the impacts of feral ungulates on the subalpine woodland community on Mauna Kea. Many of the native plant populations have been reduced, and some (such as the Mauna Kea silversword) brought to the very brink of extinction through browsing pressure from introduced goats, sheep, and cattle. The threat from feral ungulates is not limited to the subalpine environment: damage to native plants such as ōhelo (Vaccinium reticulatum) has been recently observed at 12,600 ft (3,840 m) in the Ice Age NAR, adjacent to the MKSR (Hadway 2007). Interactions between non-native animals and plants may also negatively affect native subalpine plant communities. For example, sheep on Mauna Kea prefer māmane and native perennial grasses over introduced perennial grasses such as sweet vernalgrass (Scowcroft and Conrad 1992). This selective browsing not only directly reduces native plant abundance but also indirectly reduces them through increased competition and smothering of seedlings by the invasive grasses, which then have the competitive advantage as the less preferred food materials. More information on feral ungulates is provided in Section 2.2.4.

Non-native animals such as birds and mammals can also negatively impact native plant communities through dispersal of invasive plant seeds, and in some cases through direct predation of native seeds and seedlings (Bruegmann 1996; Cabin et al. 2000). Rodents and invasive insects are known to eat native plant seeds. Rodent predation of native plant seeds is implicated in the failure of native forest  regeneration in the dry forests at Kaupulehu, Hawai‘i, in areas where ungulates have already been excluded (Cabin et al. 2000), and in the dry forest of Kanaio Natural Area Reserve, on the leeward side of East Maui (Chimera 2004). On one positive note, Cabin et al. (2000) found that rodents did not forage on māmane seeds.

Non-native birds are thought to play an important role in the dispersal of invasive plant species in Hawai‘i (Stone 1985; Woodward et al. 1990). Invasive and native bird species likely disperse different species because of differences in diet and foraging behavior (Woodward et al. 1990). Birds disperse seeds on their feet and feathers, in nesting material (Dean et al. 1990), and most commonly via their digestive systems as a result of fruit consumption (Stiles and White 1986; Wunderle 1997). Birds may either pass seeds through the digestive tract and excrete them or regurgitate them before they leave their stomach or gizzard. While most seeds are not carried for long distances (generally less than 100 m), a small fraction of seeds may be moved much longer distances (up to several kilometers) by birds (McDonnell and Stiles 1983; Stiles and White 1986; Debussche and Isenmann 1994; Wunderle 1997). Birds tend to retain smaller seeds longer than larger seeds (which they often regurgitate); thus, small seeds tend to be moved greater distances than large seeds (Levey 1986; Stiles and White 1986). Studies of seed dispersal by native and invasive birds in the Hawaiian Islands reveal that non-native birds are effective dispersers of invasive plant species (Stone 1985; Woodward et al. 1990; Garrison 2003; Chimera 2004). For example, in disturbed mesic forest and tree plantations on O‘ahu, Japanese white-eyes were found to disperse seeds from most of the fruiting invasive plants in the area, and conversely, did not disperse seeds from native species (perhaps in part due to low density of native species) (Garrison 2003). In less disturbed native vegetation, non-native birds will also disperse native plant seeds, and may be important dispersers of native plants in areas where native bird populations are reduced (van Riper 1980b; Cole et al. 1995a; Chimera 2004). In native dry forests on Maui, Chimera (2004) found that Japanese white eyes dispersed seeds of several species of native plants, as well as non-native.

More information on non-native animals can be found in Sections 2.2.2 through 2.2.4.

Human use can impact an area in many ways including wear and tear (e.g. increased erosion or soil compaction in areas that are frequently walked or driven on); direct reduction in plant and/or animal density (through the picking or collecting of plants or hunting of animals); introduction of new species of plants and animals (accidentally or on purpose); pollution (e.g. air pollution, chemical spills, oil dripping from vehicles, improper disposal of trash); habitat alteration (e.g. conversion of native habitat to buildings, roads, parking lots, and to agricultural areas or grassland for livestock foraging); and accidents (e.g. fires, landslides caused by construction activities and road building). Air pollution and dust can impact vascular plants in several ways, including greatly reducing photosynthesis, transpiration, and efficiency of water use; increasing leaf temperatures (with potentially serious effects during periods of high temperatures); and lowering primary production (growth) (Sharifi et al. 1997). Air pollution is also known to impact lichen and moss growth and community diversity (Hutchinson et al. 1996).

Human use impacts to the native plant communities in high elevation areas of Mauna Kea include (but are not limited to):

• Increased instability of the cinder areas caused by off-road vehicles and skiers (Smith et al. 1982)
• Soil erosion at Hale Pōhaku due to building construction (Gerrish 1979)
• Soil compaction and erosion on trails found on the summit and at Hale Pōhaku
• Habitat alteration through the development of telescopes and telescope facilities
• Habitat alteration through introduction of invasive species from ranching activities and landscaping (subalpine zone)
• Habitat alteration and reduction in native plant diversity and abundance, resulting from the introduction of ungulates (goats, sheep, mouflon, and cattle)
• Pollution from accidental oil spills, chemical spills, and vehicle leaks and exhaust
• Habitat degradation through improper disposal of trash by recreational users
• Increased dust from road grading and vehicles driving on dirt roads

The impacts of human use of Hale Pōhaku, the Summit Access Road, and the MKSR are discussed in more detail in Section 3.

2.2.1.3.6      Climate Change

Studies of ancient pollen from soil cores in the Hawaiian Islands suggest that Hawaiian plant  communities responded to past climate changes with changes in community composition and in plant densities (Hotchkiss and Juvik 1999; Benning et al. 2002). Thus, there is little doubt that current plant communities will also respond to future changes in temperature, rainfall, and cloud cover that occur in the islands. However, there is currently a great deal of discussion about what effects climate change will have on trade wind and rainfall regimes in the Hawaiian Islands (Giambelluca and Luke 2007; Hamilton 2007). Although several climate models have been developed to study global climate change, most of the models are at too large a scale to accurately predict what will occur in Hawai‘i, given the islands’ steep topography, which has a strong effect on the weather patterns (Hamilton 2007).

Recent advances in climate modeling have allowed for a more fine-scale rainfall model, and the results from this model are currently under investigation (Hamilton 2007). Some of the early results from the work with the fine scale model include the prediction of an overall warming of the islands, leading to increased moisture in the air, an overall increase in rainfall, and possibly an increase in snowfall in the higher elevation areas. An additional finding is that the intensity of warming is positively related to altitude (Hamilton 2007). This means that the higher altitude areas on the islands will see greater gains in temperature than lower altitude areas. These findings suggest that high altitude areas may become wetter and warmer in the future, with greater snowfall on the summit.

Rainfall and cloud-base climate change scenarios for the neotropics developed in the late 1990s for montane cloud forests predict an increase in the height of cloudbanks, resulting in reduced cloud contact at the current elevation of most cloud forests (Pounds et al. 1999; Still et al. 1999; Benning et al. 2002). Cloud forests rely on contact with clouds to receive moisture, as do the māmane forests on Mauna Kea (Gagné and Cuddihy 1990; Gilbertson et al. 2001). Thus the raising of the cloud layer could seriously impact cloud forests. Provided there are no significant barriers to upward migration of plant species, the cloud forest should respond by moving to a higher elevation. These cloud base scenarios also predicted an increase in temperature in higher elevation areas, leading to faster melting of glaciers, a phenomenon that has been observed worldwide.

In opposition to the above predictions, other climatologists predict that conditions in high elevation areas in the Hawaiian Islands will become much drier (Giambelluca and Luke 2007). This prediction is based on changes in the trade wind inversion that have been observed in the last several decades. The dominant source of rain in Hawai‘i is orographic lifting, by which air is forced up the mountain where trade winds meet the windward slopes (Giambelluca and Luke 2007). The trade wind inversion caps the upward motion of the wind, limiting cloud development in higher elevation areas. If the frequency of occurrence, or the height of the trade wind inversion is altered by climate change, this will have profound effects on rainfall in areas at or above the elevation of the trade wind inversion (Giambelluca and Luke 2007). Climate research over the past few decades suggests that the trade wind inversion has and will continue to become more persistent and lower in height, leading to a drier climate in Hawai‘i, in particularly at high elevation areas (Cao 2007; Cao et al. 2007; Giambelluca and Luke 2007). The climatic changes  associated with the changes in the trade wind inverstion include decreasing rainfall and streamflow, with given streams declining in annual flow by 50% in the last 90 years (Oki 2004). However, it is uncertain whether the changes in the trade wind inversion observed over the last few decades are part of the warming trend (climate change) or are a result of natural multi-decadal variability in rainfall and trade wind inversion occurrence (Giambelluca and Luke 2007).

Under the assumptions of the above models and scenarios, the potential effects of the various aspects of climate change on high elevation plant communities on Mauna Kea are discussed below.

1. Increase in temperature AND rainfall:
   1. Upwards movement of treeline, due to upwards movement of frost line (Flenley 1998; Benning et al. 2002; Kullman 2006; Baker and Moseley 2007)
   2. Movement of subalpine community into alpine community (Flenley 1998; Kullman 2006)
   3. Decrease in the area covered by alpine vegetation (Flenley 1998; Kullman 2006)
   4. Expansion of shrublands (Cannone et al. 2007)
   5. Increased plant growth by certain species (Danby and Hik 2007; Erschbamer 2007)
   6. Change in composition of plant communities (including local extinctions) due to differing response to changes in temperature, rainfall, etc. (Kullman 2006; Erschbamer 2007; Kazakis et al. 2007; Van de Ven et al. 2007)
   7. Invasion by new non-native species from lower elevations that were previously kept out by freezing temperatures (Benning et al. 2002; Weltzin et al. 2003)
   8. Shorter duration of snow pack before it melts in the higher elevation areas, leading to longer periods of time without snow pack, and drier soil conditions between periods of rain and snowfall (Kullman 2006; Bjork and Molau 2007)
   9. Higher plant densities in subalpine woodland and alpine shrubland and grassland
   10. Increased invasion by non-native species previously kept out by low availability of moisture (Weltzin et al. 2003; Erschbamer 2007)
   11. Increased growth rates of plants
   12. Increased competitive edge by fast growing invasive plants (Weltzin et al. 2003; Erschbamer 2007)
2. Increase in temperature and decrease in rainfall:
3. Increase in CO₂ concentration
   1. Fertilization of all plants leading to increased growth (Weltzin et al. 2003
   2. Further competitive edge by fast growing invasive species (Weltzin et al. 2003)

Although it is not yet possible to accurately predict what will occur on Mauna Kea, it seems likely that under the influence of climate change, the alpine communities on Mauna Kea will decrease in extent. The sub-alpine communities will either move upwards in elevation due to increased temperature and rainfall, or will be lost at upper elevations due a drier climate. Finally, the abundance of invasive species and their diversity may increase (especially under the higher rainfall scenario), leading to shifts in plant community composition in all regions. Drought resistant invasive species will be the primary invaders of high elevation areas.

2.2.1.4       Botanical Community Information Gaps

The following information gaps regarding the condition of the subalpine and alpine plant communities at Hale Pōhaku and the MKSR have been identified through review of the literature and consultation with local experts:

1. Quantitative botanical surveys

a)       Hale Pōhaku: Although several plant surveys have been conducted at Hale Pōhaku (Gerrish 1979; Char 1985, 1990, 1999a; Pacific Analytics 2004), no quantitative botanical studies documenting population size and distribution of native and non-native species have been conducted there. The last survey that involved more than a brief examination of field conditions was conducted by Char in 1990.

b)      Summit Access Road: No botanical surveys have been conducted along the Summit Access Road between Hale Pōhaku and MKSR.

c)       Mauna Kea Science Reserve: Limited botanical surveys have been conducted in the MKSR. Smith et al. (1982) surveyed only the plant species found above 13,000 ft (3,960 m) and only in areas considered for future telescope construction (as described in the 1982 Master Plan). A figure showing the areas covered by this study is included as Figure 2.2-9. Although the study area was thoroughly searched, no quantitative sampling was conducted. Other studies conducted there were very limited in scope.

2.2.2    Invertebrates

Invertebrates are animals lacking a backbone. This enormous group of organisms covers a wide range of terrestrial and marine forms such as the arthropods (insects, spiders, crustaceans), mollusks (snails, bivalves, squid, octopus), annelids (segmented worms such as earthworms), echinoderms (starfish, sea urchins, sea cucumbers), lampshells, bryozoans, sponges, cnidarians (jellyfish, coral, sea anemones), ctenophores (comb jellies), and many phyla of worms (priapulid worms, flatworms, roundworms, nematodes, horsehair worms, velvet worms, and acorn worms). Invertebrates constitute approximately 97% of all known species on earth. New species are still being discovered regularly. Because of their sheer numbers, wide diversity of forms and functions, and (often) small sizes, invertebrates are generally poorly known and even more poorly understood. There are undoubtedly many hundreds (or even thousands) of species of invertebrates that await discovery in the Hawaiian Islands.

Invertebrate species known from the subalpine and alpine regions of Mauna Kea are presented in Table 2.2-6. This table was compiled from a variety of sources, including the review of invertebrate species found in high elevation areas of Mauna Kea presented in Aldrich (2005) and searches of scientific literature and databases. This table does not represent a complete list of species found in the area. There have been relatively few studies of invertebrates done in the region. Because of the sheer number of species, and wide diversity of forms, a detailed survey of invertebrates on Mauna Kea would take many years (or even decades), and would no doubt fill several volumes. Because of this diversity and complexity, this plan focuses primarily on the arthropods (primarily insects and spiders) found in the upper elevations of Mauna Kea. A second important group of invertebrates, the land snails, are also discussed. Arthropods comprise more than 75% of the native Hawaiian biota, and include some of the world’s best known species radiations (Roderick and Gillespie 1998). Discoveries about this group of animals are still being made on Mauna Kea (Brown 2008; Medeiros 2008). For example, the wēkiu bug, the now-famous insect found at the summit of Mauna Kea, was only discovered in 1979 (Howarth and Montgomery 1980), and is still being studied. Photos of selected native invertebrates are presented in Figure 2.2-10 and photos of common invasive invertebrates are presented in Figure 2.2-11.

Arthropods: The māmane forests on Mauna Kea have high arthropod diversity—more than 200 species have been collected there, and many more are likely to be found should additional studies be done (NASA 2005).

Lepidoptera (Moths and Butterflies). An important group of arthropods found in the subalpine māmane forests are the Lepidoptera (moths and butterflies), including several moth species that feed on māmane (Sophora chrysophylla) seeds (NASA 2005). Although moths and butterflies have been intensively studied world-wide, there is still much to learn about the species that inhabit the higher elevation areas on Mauna Kea. Recently a new species of flightless Thyrocopa moth was discovered above the treeline on Dubautia ciliolata near Hale Pōhaku by Matt Medeiros (Medeiros 2008; Oboyski 2008). This new species is diurnal (most moths are nocturnal), appears to forage on dead leaves of shrubs and clumps of grass, and has lost the ability to fly (Medeiros 2008). It moves around by jumping, and could easily be mistaken for a grasshopper by the casual observer. So far, it appears that this species is limited to Mauna Kea, but more research is needed (Medeiros 2008). Other Thyrocopa species that can be found in the subalpine zone at Hale Pōhaku include Thyrocopa indecora and T. adumbrata (Medeiros 2008). Other moth species found the subalpine area includes moths in the genus Mestolobes. These are small brown moths that are thought to be endemic to the Hawaiian Islands (Zimmerman 1958). Not much is known about these moths, including diet and habitat preferences (Medeiros 2008).

The māmane-feeding Lepidoptera include moths from the genus Cydia (of which there are at least seven species on Mauna Kea), Peridroma, and Scotorythra. These moths are the most important prey items for the endangered Palila (Loxioides bailleui; see Section 2.2.3, Birds), and are likely an important protein source for developing Palila chicks (Brenner et al. 2002). Parasitism of Cydia moths by several wasp species may be reducing moth abundance in the māmane woodlands. Parasitic wasps have been implicated in the decline or extinction of at least 16 Lepidopteron species in Hawai‘i (Oboyski et al. 2004). Brenner et al. (2002) found four common parasitoid wasps that attack larval Cydia moths: Calliephialtes grapholithae, Diadegma blackburni, Pristomerus hawaiiensis, and Euderus metallicus. The first three species appear to be accidental introductions to Hawai‘i (including the deceptively named P. hawaiiensis), while the fourth (E. metallicus) appears to be native to the islands (Brenner et al. 2002). However, the actual origins of the latter three species are still under debate (Oboyski et al. 2004). In their study of parasitism of Cydia larvae, Oboyski et al. (2004) found an additional common parasitic species, Brasema cushmani, which is an introduced biological control agent for the pepper weevil, Anthonomus eugenii (Oboyski et al. 2004).

Brenner et al. (2002) found that parasitism rates were lower in the high-elevation populations of the Cydia moths than in the lower elevation populations: only 20% of Cydia larva were parasitized at 8,860 ft  (2,700 m), while 94% were parasitized at 5,900 ft (1,800 m). Cydia larva abundance in māmane pods increased with elevation, peaking at around 8,695 ft (2,650 m) (Banko et al. 2002). However, a subsequent study found no difference in parasitism rates for Cydia species at differing elevations (Oboyski et al. 2004), although this study did not include Cydia larvae from below 6,889 ft (2,100 m), where the highest rates of parasitism occurred in the Brenner et al. (2002) study. Although overall parasitism rates did not differ with elevation in their study, Oboyski et al. (2004) found that parasitism rates by native and introduced wasp species differed with elevation. Parasitism by the native wasp, Euderus metallicus, increased with elevation, while parasitism by Calliephialtes grapholithae (non- native) and Pristomerus hawaiiensis (origin unknown) decreased (Oboyski et al. 2004). Parasitism rates by two other species, Diadegma blackburni and Brasema cushmani, did not vary significantly with elevation.

Another native moth species, Uresephita polygonalis virescens, was previously a common prey item for the Palila but is no longer observed to be part of the Palila diet. Banko et al. (2002) suggest that this species has been reduced in abundance by parasitism. Finally, the black-veined Agrotis noctuid moth (Agrotis melanoneura) is known to reside on Mauna Kea (Bishop Museum 2007c). Very little  information is available regarding this species. It has been observed at light traps at Hale Pōhaku in recent years, and is uncommon but widespread on Mauna Kea (Giffin 2009).

Hymenoptera (Bees, Wasps, and Ants). There are no native ants (or social insects of any kind) in the Hawaiian Islands. However, other members of the hymenoptera are present, and represent a diverse group that has undergone much radiation in the islands. Native bees, such as those found in the family Colletidae, are important pollinators, while most of the native wasps are arthropod parasites, often helping to keep herbivorous insect populations in check (Mitchell et al. 2005a). The yellow-legged yellow-faced bee (Hylaeus flavipes) is the only Hylaeus observed at high elevations on Mauna Kea (Aldrich 2005), where it is found associated with māmane (Magnacca 2008). It is also thought to be a potential pollinator of the Mauna kea Silversword (Aldrich 2005). Other native bees that may be found in the subalpine zone (but which have not been confirmed for Hale Pōhaku) include H. ombrias, H. difficilis and H. volcanicus (Magnacca 2008). Invasive hymenoptera found in the subalpine zone on Mauna Kea include the five parasitoid wasp species and one parasitoid fly species, ants, honeybees (Apis mellifera) and yellowjackets (Vespula pensylvanica) (Banko et al. 2002; Oboyski 2008). These are discussed in more detail in Section 2.2.2.1.2.

True bugs (Heteroptera): A new species of plant bug, Orthotylus sophorae, was recently discovered in association with māmane woodlands from 3,200–9,000 ft (1,000–2,750 m) above sea level on Hawai‘i. It is often found in association with other māmane-associated Heteroptera species, including the endemic nabid Nabis kahavalu and endemic lygaeid Nesius (Icteronysius) ochriasis (Polhemus 2004). Other lygaeid bugs (relatives of the wēkiu bug, which lives at the summit) found in the subalpine region include Neseis nitida comitans, Nysius coenosulus, Nysius palor and Nysius terrestris (Englund et al. 2002).

Other arthropod species of interest found in the subalpine region include the Hawai‘i long-horned beetle (Plagithmysus montgomeryi), koa bug (Coleotichus blackburniae), and wolf spiders (Lycosa species).

Snails: The Hawaiian Islands has an impressive diversity of land snails, with at least 779 species found in ten families (Cowie et al. 1995; Hadway and Hadfield 1999). Many of these species are endemic (found no where else in the world). Land snail abundance and diversity has been greatly impacted by the arrival of humans on the islands, due to habitat destruction, introduction of predators and diseases, and overcollecting. Up to 90% of the species are now thought to be extinct (Hadway and Hadfield 1999; Rundell and Cowie 2003). Introduced predators, including rats (Rattus rattus), rosy wolfsnail  (Euglandina rosea), garlic snail (Oxychilus alliarius), and the predatory flatworm Platydemus manokwari, have heavily impacted native snail populations (Meyer 2006). The highest diversity of land snails is found in wetter forests below the subalpine zone on the Island of Hawai‘i. Even so, there are several species of land snail that occur, or once occurred, in the subalpine māmane woodlands on Mauna Kea (Hadway and Hadfield 1999).

No surveys for snails have been conducted in the subalpine regions as high as Hale Pōhaku. However, a survey for snails at Pu‘u La‘au Forest Reserve from 6,200 to 8,600 ft (1,890 to 2,621 m) elevation conducted in 1995–1997 found four species of snails: two endemic, one of unknown origin, and one invasive species. The endemic snails found at Pu‘u La‘au include Succinea konaensis and Vitrina tenella. The snail of unknown origin was an unidentified species in the genus Striatura. The non-native snail found was the garlic snail, Oxychilus alliarius (Hadway and Hadfield 1998; Hadway and Hadfield 1999). This species is discussed further in Section 2.2.2.1.1. Historically, Partulina confusa, a tree-dwelling snail endemic to the Island of Hawai‘i, was found in māmane-naio forests such as those found at Pu‘u La‘au Forest Reserve. However, none were located during the survey of this area, and this species may be extinct (Hadway and Hadfield 1998; Hadway and Hadfield 1999).

Succinea konaensis and Vitrina tenella are federal Species of Concern and are discussed in Section 2.2.2.1.1. There are three Striatura species of snail on the federal Species of Concern list, but it is unknown whether the species found at Pu‘u La‘au is one of them.

2.2.2.1.1      Threatened and Endangered Species, Candidate Species and Species of Concern

There are no federal or state listed Threatened or Endangered Species of invertebrates known to be present at Hale Pōhaku or in the subalpine zone of Mauna Kea.

Federal Species of Concern include the koa bug (Coleotichus blackburniae), the flightless brown lacewing (Micromus usingeri), the black-veined Agrotis noctuid moth (Agrotis melanoneura), several species of native Hylaeus bees including H. flavipes, H. difficilis, and H. ombrias, and two species of snails (Succinea konaensis and Vitrina tenella). The black-veined Agrotis noctuid moth and the Hylaeus bees are also listed as Hawai‘i state Species of Concern.

The koa bug is the only native herbivorous stink bug in Hawai‘i (Roderick and Gillespie 1998). It was quite common until the 1960s, when several parasites were released in Hawai‘i to control Nezara viridula, a pest stinkbug. These parasites have decimated koa bug populations, and it is now rare in the wild (Asquith 1995). Higher elevation areas may provide a refuge for koa bug from introduced biological control agents (Oboyski 2008). The flightless brown lacewing has recently been collected on Dubautia arborea on Mauna Kea (Tauber et al. 2007). The black-veined Agrotis noctuid is uncommon but widespread on Mauna Kea, and has been observed at Hale Pōhaku. The current status of the native bee populations at high elevation areas on Mauna Kea is unknown, as no formal surveys have been conducted there. Hylaeus flavipes has been observed foraging on māmane (Sophora chrysophylla) trees at Hale Pōhaku (Aldrich 2005; Magnacca 2008). The other species of bees listed above are thought to be found in dry forests and shrublands but have not been studied at Hale Pōhaku or the vicinity.

Succinea konaensis and Vitrina tenella, both listed as federal Species of Concern, are ground dwelling snails. In the survey conducted at Pu‘u La‘au on Mauna Kea, both of these species were found beneath rocks at approximately 8,500 ft (2,590 m) (Hadway and Hadfield 1998). Predators of these high elevation snails include ground foraging birds such as Ring-necked pheasants and rodents, primarily rats (Schwartz and Schwartz 1951; Hadway and Hadfield 1999). Ring-necked pheasants may eat the snails mainly  during breeding season to provide calcium for eggshells (Schwartz and Schwartz 1951). Other than for some snails in the family Achatinellinae, very little is known about the life history of Hawaii’s endemic terrestrial snails (Rundell and Cowie 2003), and little information is available regarding Succinea konaensis and Vitrina tenella.

2.2.2.1.2      Invasive Invertebrate Species

Invasive invertebrates are a serious threat to Hawai‘i. The Hawai‘i Invasive Species Council estimates that two or more serious arthropod pests arrive in the islands every year. Infamous new arrivals to Hawai‘i include the little fire ant, which has a very painful sting; the Erythrina gall wasp, which is destroying native wiliwili trees; and the Varroa mite, which is a threat to the multimillion-dollar queen bee, honey, and pollination industries (Wilson 2008).

Invasive arthropods found in the subalpine region of Mauna Kea include (at a minimum) the five parasitoid wasp species and one parasitoid fly species, European earwig (Forficula forficularia), ants, honeybees (Apis mellifera) and yellowjackets (Vespula pensylvanica) (Banko et al. 2002; Oboyski 2008; Englund et al. 2009). Both ants and yellowjackets are known to have detrimental affects on native arthropod populations, which in turn can affect the native plant and bird communities.

Honeybees (Apis mellifera) were introduced to the Hawaiian Islands in 1875 (Barrows 1980). They are thought to compete with native nectarivorous insects such as native bees, but their impact on native pollinators in Hawai‘i has not been fully studied (Magnacca 2007). In areas where native pollinators are few or missing, honeybees may provide pollination services to some native plant species.

Yellowjackets were first introduced to Kaua‘i in 1919, and have since spread to all the other major Hawaiian Islands except Kaho‘olawe and Ni‘ihau (Gambino et al. 1990; Gruner and Foote 2000). Yellowjackets were found by Banko et al. (2002) at 9,186 ft (2,800 m) on Mauna Kea. There appears to be no relationship between yellowjacket numbers and elevation on Mauna Kea, suggesting that this species is able to survive equally well in the subalpine zone as in lower elevations (Banko et al. 2002). Currently yellowjacket densities are low on Mauna Kea (Banko et al. 2002). However, yellowjackets are known to seriously impact native arthropod communities (Gambino et al. 1987; Stone and Anderson 1988; Gambino et al. 1990; Aldrich 2005), and they could pose a threat in the subalpine woodlands and shrublands if their densities increase. On Maui, yellowjacket nests in high elevation areas were primarily found beneath pūkiawe (Leptecophylla tameiameiae) bushes, which also support a honeydew producing mealybug, Pseudococcus nudus, a food source for the yellowjackets (Gambino et al. 1990). Pūkiawe are fairly abundant in the subalpine zone on Mauna Kea, and the mealybug species is also found on the island of Hawai‘i. Therefore it is possible that yellowjackets may enjoy the same accommodations and food source in the subalpine zones on Mauna Kea as they do at Haleakalā.

There are no native ants in Hawai‘i (Loope et al. 2001). Wetterer et al. (1998) conducted a survey of ant species on the western flank of Mauna Kea, from 5,500 to 10,300 ft elevation (1,680 to 3,140 m). They found that ants were abundant up to 6,600 ft (2,010 m), and were found at low densities above that (Wetterer et al. 1998). Five species of invasive non-native ants have been found on Mauna Kea: Linepithema humile, Cardiocondyla venustula, Pheidol megacephala, Tetramorium bicarinatum, and Monomorium pharaonis. Another study of Mauna Kea ant species, conducted in 1999 by Banko et al., found similar species to Wetterer et al, but at even higher elevations (Banko et al. 2002). The species with the highest elevational range and highest densities are Cardiocondyla venustula (8,038 ft/2,450m) and Linepithema humile (9,186 ft/2,800 m) (Wetterer et al. 1998; Banko et al. 2002). Pheidol megacephala, Tetramorium bicarinatum, and Monomorium pharaonis were found in fewer locations and at lower densities (Wetterer et al. 1998). Pheidol megacephala are found up to 6,725 ft (2,050 m), Tetramorium bicarinatum are found up to 5,970 ft (1,820 m), and Monomorium pharaonis are found up to 6,332 ft (1,930 m) (Wetterer et al. 1998; Banko et al. 2002). A study of invasive invertebrates present at Hale Pōhaku and the MKSR conducted in 2007-2008 by Bishop Museum entomologists indicate that there are no ants currently established at Hale Pōhaku or in the MKSR (Englund et al. 2009).

Linepithema humile, or the Argentine ant, was first discovered at Fort Shafter, O‘ahu, in 1940 (Zimmerman 1941) and has since spread to the other islands. While it has not yet been found at Hale Pōhaku, it is known to occur at similar elevations on other parts of Mauna Kea (9,186 ft/2,800 m) and at 9,450 ft (2,880 m) on Haleakalā, Maui (Cole et al. 1992; Wetterer et al. 1998), and is able to colonize dry upland areas (Krushelnycky et al. 2005). The Argentine ant is a serious threat to native flora and fauna because of its appetite for arthropods, seeds, and nectar (Aldrich 2005). It is a predator of many endemic arthropods, including noctuid moths and Hylaeus bees, which are the pollinators of rare subalpine plants such as the Haleakalā silversword, Argyroxiphium sandwicense macrocephalum (Stone and Anderson 1988; Cole et al. 1992). Cole et al. (1992) found that many invertebrate populations on Haleakalā were smaller in areas infested with Argentine ants than in areas not infested. As Mauna Kea silverswords (Argyroxiphium sandwicense sandwicense) are thought to be pollinated by Hylaeus bees, the establishment of a colony of Argentine ants in the subalpine zone on Mauna Kea could further inhibit recovery of the small population of silverswords found there.

In 2007-2008, Bishop Museum scientists observed European earwigs (Forficula forficularia) in high numbers around the Onizuka Visitor Information Station at Hale Pōhaku (Englund et al. 2009). It appears to be restricted in elevation and has not become established above the Visitor Information Station (Englund et al. 2009). This species is predatory, and could potentially impact native invertebrate species in the subalpine zone (Englund et al. 2009). Monitoring of the distribution and impact of this species on native invertebrates should be conducted.

The garlic snail, Oxychilus alliarius, is an introduced terrestrial snail that was first recorded in the Hawaiian Islands in 1937 (Hadway and Hadfield 1998). It can be very abundant, especially in moist ground in forested areas (Hadway and Hadfield 1998). This species is an omnivore and opportunistic predator, and appears to negatively impact native snail populations (Howarth 1985). Garlic snails consume other snails with shells less than 0.11 inches (3 mm) in length, including native succineid snails (Meyer 2005, 2008). It has been found at 8,600 ft (2,621 m) elevation on Mauna Kea, but its true elevational limit is unknown.

2.2.2.1.3      Invertebrate Surveys at Hale Pōhaku

There have been no quantitative studies of invertebrate communities at Hale Pōhaku. Englund et al. (2002) conducted a brief visual survey of Lygaeid bugs found at Hale Pōhaku and the Summit Access Road in September 2002. In 2007-2008, Englund et al. (2009) conducted qualitative (presence/absence) sampling for invasive invertebrates at Hale Pōhaku. This important study increased understanding of the species of invasive invertebrates present in the subalpine region of Mauna Kea. Interest in invertebrate communities in the subalpine zone on Mauna Kea is increasing and several researchers have recently collected specimens at Hale Pōhaku (Medeiros 2008; Oboyski 2008).

2.2.2.2       Alpine Invertebrate Communities (MKSR and Upper Summit Access Road)

There is little information available regarding invertebrate communities in the alpine shrublands and grasslands of Mauna Kea, as very few studies have been conducted in this region. In the summers of 2007 and 2008, Bishop Museum entomologists conducted surveys for invasive invertebrate species along the Summit Access Road and in the MKSR, providing data on presence of several new species in the region (Englund et al. 2009). Other research is currently underway on the insect communities in the shrublands found in the upper subalpine and lower alpine zones on Mauna Kea, and more information should be available in the near future (Oboyski 2008). J.M. Brown is conducting a study of Trupanea arboreae, a native Tephritid fruit fly that is associated with Dubautia and other members of the silversword alliance on Mauna Kea (Brown 2008). Tephritid flies are herbivorous flies that feed on plant material and form galls in plant tissues (Brown et al. 2006). Since so little information is available on the alpine shrublands and grasslands on Mauna Kea, the remainder of this section will focus on the invertebrate community found on the summit of Mauna Kea.

Invertebrate communities in the alpine stone desert have received a fair amount of attention since the discovery of the wēkiu bug and other resident species at the summit of Mauna Kea, in 1980. The arthropod community on the summit of Mauna Kea can be divided into two parts: those species that are blown up the mountain by the wind and die there in the cold (referred to as aeolian drift), and those cold- adapted species that are permanent residents and that feed on the dead and dying arthropods found in the aeolian drift or on one-another (Howarth and Montgomery 1980; Howarth and Stone 1982). All species that have been found on the summit are listed in Table 2.2-6, and the aeolian drift species are distinguished from the resident species in the 8^th column of the table. Although the aeolian-drift species provide an important food source for the resident species, they are not discussed in detail here, because so long as they continue to blow up the mountain in large numbers, their exact species composition is probably not important to the survival of the residents. Through the various studies conducted at the summit of Mauna Kea, 21 resident species and 21 species of undetermined status (unknown if they are resident or aeolian) have been recorded as occurring in the alpine stone desert. An additional 67 species (47 non-native, 12 native, and eight of unknown origin) have been recorded in the aeolian drift, although this number will no doubt continue to climb over time as more collecting is done.

The 21 resident species include 12 native species, five species of unknown origin, and four non-native species. Of the 21 species with unknown status (whether they are resident or aeolian), four are native species, seven are unknown, and ten are non-native species. These numbers are approximate because of the uncertainty of many species identifications.

Native resident (and potential resident) species include the wēkiu bugs (Nysius wekiuicola), a noctuid moth (Agrotis sp.), a hide beetle (Dermestes maculatus), a large wolf spider (Lycosa sp.), two sheet web spiders (Erigone species), an unidentified Linyphiid sheet web spider (Family Linyphiidae), two  unknown Entomobryid springtails (Family Entomobryidae), a Collembola springtail (Class Collembola, family and species unknown), two species of mites (Families Anystidae and Eupodidae), a bark louse (Palistreptus inconstans) and a centipede (Lithobius sp.). The wēkiu bug (Nysius wekiuicola) is the best- studied invertebrate at the summit – there is little information available regarding the habits of most of the other summit species. The wēkiu bug is discussed in more detail in Section 2.2.2.2.1. The remainder of the native resident species are discussed below.

Lycosid spider: Invertebrate surveys at the summit discovered a large (up to 2 cm body length), black wolf spider (Lycosa sp.). This wolf spider is thought to be endemic to the Hawaiian Islands, although its distribution elsewhere is not known (Howarth and Montgomery 1980; Howarth and Stone 1982). Many lycosid species are capable of ‘hang gliding’ or long-distance dispersal by wind (Howarth and Montgomery 1980). The wolf spider is an ambush predator, hiding under large rocks until an active prey comes within range (Howarth and Stone 1982; Howarth et al. 1999). It likely preys on any actively moving arthropod including the wēkiu bug (Englund et al. 2002). The female wolf spider builds nests of silk and earth under rocks, and remains with the nest to protect the developing eggs (Howarth and Stone 1982). The wolf spider is found in low densities across the summit in a wider variety of areas than the wēkiu bug (Howarth et al. 1999).

Other spiders: Three presumably native Linyphiid spiders (Erigone sp.) were collected in 1982, but were not seen in 1997–1998 surveys (Howarth et al. 1999). One Erigone species (Species A in Howarth and Stone 1982) is described as being a “small, brown, sheet web spider which builds its sheet-like web  across vesicles and other indentations on the undersides of rocks in the summit area” (Howarth and Stone 1982). This species makes small (2–3 mm diameter) flat, white, circular egg cases that are placed on the undersides of rocks, and was abundant wherever there were suitable rocks (Howarth and Stone 1982).  The second Erigone species (Species B in Howarth and Stone 1982) was a single distinctive male located near 13,000 ft (3,960 m) on the northwest slope of the surveyed area (See Figure 2.2-12). The third species belonged to an unknown genus in the Linyphiidae family, and had similar range and habitats to the Erigone Species A.

Centipede: A small black centipede in the genus Lithobius, presumed to be endemic, occurs primarily on lava flows with large outcrops of andesitic rock. The centipede burrows in the silt and aeolian debris in cracks and under rocks at the base of lava cliffs (Howarth and Stone 1982). Like many of the other species encountered on Mauna Kea’s summit, the centipede is thought to feed on aeolian drift (Richardson 2002). Few individuals of this species have been collected or observed, and little is known of its ecology.

Agrotis moth: This undescribed species, originally identified as an Archanarta species in 1982 (Howarth and Stone 1982), is a black moth whose larvae feed on foliose lichens, dead arthropod remains, and even the remains of larger animals (including the skin of mummified sheep) (Howarth et al. 1999). Adults have been observed from approximately 10,000 ft (3,048 meters) to the summit (Howarth and Stone 1982). Very little is known about this species.

Resident (and possible resident) species of uncertain origin include an unidentified rove beetle (Staphylinidae), an unidentified Hydrophilid beetle (family Hydrophilidae), a moth fly (Psychoda species), an unidentified scuttle fly (family Phoridae), a fungus gnat (Sciara sp.), an unidentified ichneumonid wasp (family Ichneumonidae), unidentified micro-hymenoptera, and several unknown species of mites (Families Bdellidae, Laelapidae, Phytoseidae, and one unknown family). No information is available regarding the distribution of these species, their abundance, or behavior at the summit.

Non-native resident (and potential resident) species include: a book louse (Liposcelis divinatorius), big- eyed bug (Geocoris pallens), a hunting spider (Meriola arcifera), a sheet web spider (Lepthyphantes tenuis), and an unidentified jumping spider (family Salticidae). One non-native species of fly, the blue bottle fly (Calliphora vomitoria), a predatory carabid beetle (Agonum muelleri), and two species of diving water beetle (Rhantus pacificus, which is endemic to the Hawaiian Islands, and an undetermined Hydrophilid of unknown origin), were recorded as occurring in Lake Waiau (Englund and Preston 2008; Englund et al. 2009). Non-native species are discussed further in Section 2.2.2.2.2 (Invasive Species).

2.2.2.2.1      Threatened and Endangered Species, Candidate Species, and Species of Concern

The wēkiu bug (Nysius wekiuicola) is a federal Candidate species. No other Species of Concern, Candidate, Threatened or Endangered species are known to reside in the MKSR. Currently a Candidate Conservation Agreement (CCA) is being developed by the U.S. Fish and Wildlife Service in cooperation with OMKM, the University of Hawai‘i Institute for Astronomy, and other agencies and organizations involved in astronomical activities on Mauna Kea (Richardson 2002; Wada 2008). The goal of the CCA is to provide long-term protection to endemic arthropods at the summit of Mauna Kea (including the wēkiu bug). CCA activities would include monitoring of species status and habitat quality, removing some of the known threats, educating personnel, habitat restoration, and incorporation of species conservation measures into planning and management activities. If completed and properly implemented, the CCA may remove the need to list the wēkiu bug under the Endangered Species Act, and should help protect other endemic invertebrates at the summit as well (Richardson 2002).

The wēkiu bug (Nysius wekiuicola) was first recognized as a new species in 1979 and was formally described by Ashlock and Gagne in 1983 (Ashlock and Gagne 1983). It is a true bug in the family Lygaeidae (order Heteroptera), and is approximately the size of a grain of rice (Ashlock and Gagne 1983; Richardson 2002). Wēkiu bugs reside under rocks and cinders on the summit of Mauna Kea, where they feed diurnally (during the day) on dead and dying insects blown up the mountain from lower elevations (Howarth and Montgomery 1980; Ashlock and Gagne 1983; Howarth 1987). Wēkiu bugs use their straw- like beaks to suck the hemolymph (a fluid comparable to blood) from other insects (Richardson 2002). They do not appear to feed on healthy, living individuals of the other resident arthropod species (Ashlock and Gagne 1983). The wēkiu bug (Nysius wekiuicola), and its sister species, Nysius aa, which resides on the summit of Mauna Loa, differ from other species in the genus Nysius in being scavengers and predators of dead and dying arthropods. All other known species in the genus are seed and/or plant feeders (Ashlock and Gagne 1983; Polhemus 1998). Food resources alone probably do not greatly influence the distribution of wēkiu bugs, as arthropod diversity and abundance in the aeolian drift was found to be similar in areas where wēkiu bugs are found and those where they are not (Howarth et al. 1999). However, it is possible that abundance of flies and other weak-flying aeolian waifs is higher along ridge crests and in areas where wind eddies drop their particulate loads (Howarth et al. 1999). Snowfields may chill and store insects for consumption by resident scavengers such as the wēkiu bug, and the bugs can often be seen foraging on the edge of snow banks (Englund et al. 2006). Permafrost is believed to be a critical source of moisture for the wēkiu bug, however there is no evidence suggesting that permafrost availability or distribution is different between areas inhabited by wēkiu bug and those not inhabited by them (Howarth et al. 1999). In addition, Howarth et al. (1999) found moist substrates at most sites studied, especially within the sandy ash layer below the surface scoria.

Habitat type and abundances: Howarth and Stone (1982) found that wēkiu bugs were abundant above about 13,450 ft (4,100 m) on undisturbed areas on Pu‘u Wēkiu and Pu‘u Hao Oki, on stable accumulations of loose cinders and tephra rocks, where the interstitial spaces are large enough to allow the bug to migrate downwards to moisture and shelter. These habitat types were found on the ridges and craters of the cinder cones (Howarth and Stone 1982). Areas that had accumulated aeolian dust and silt, such as Pu‘u Poliahu, had fewer wēkiu bugs. Howarth and Stone (1982) did not survey areas outside of Pu‘u Wēkiu, Pu‘u Hau Oki, and Pu‘u Poliahu (see Figure 2.2-12 for survey area). Howarth et al. (1999) had high trap capture rates on Pu‘u Hau Oki, where the inner crater walls and crater bottom had been modified by observatory construction activity. This suggested that observatory construction and other human activities had not impacted wēkiu bug distributions at the summit outside of the immediate  vicinity of paved and covered areas (Howarth et al. 1999). Polhemus (2001) found a high density of wēkiu bugs on Pu‘u Hau Kea in the Ice Age NAR. He found that the bugs were most abundant on the rim and inner crater of the pu‘u. Englund et al. (2002) found that wēkiu bugs are restricted to the rims and inner craters of each alpine cone, and, with only one exception, were found within 150 ft (50 m) of the peak elevation of each pu‘u. Englund et al. (2002) found wēkiu bugs on Pu‘u Hau Kea, Pu‘u Māhoe, Pu‘u Poepoe, Pu‘u Ala, Pu‘u Mākanaka, and an unnamed Pu‘u near VLBA (at 11,920 ft/3,630 m). In their study, the highest capture rate of wēkiu bugs occurred on Pu‘u Poepoe (33 bugs), and the second highest capture rate occurred on Pu‘u Hau Kea (nine bugs); capture rates were low on all other pu‘u (Englund et al. 2002). In 2004, wēkiu bugs were found only on Pu‘u Hau Oki and Pu‘u Pōhaku (Englund et al. 2005). In 2005, the wēkiu bug research team found additional wēkiu bug populations on Pu‘u Hau Kea, Pu‘u Pōhaku, Pu‘u Wēkiu, Pu‘u Poli‘ahu, and Pu‘u Lilinoe. This final location represented a significant extension of known wēkiu bug core habitat (Englund et al. 2006).

In 2006, Porter and Englund published a report further clarifying habitat preferences by wēkiu bugs. This study found that wēkiu bugs mainly reside on or near the crater rims of cinder cones that formed nunataks (ice free areas rising above the surrounding glacier) or that lay at the glacier limit during the last glaciation, and that the bug is most abundant on the north- and east-facing slopes (and on slopes shaded by local topography), where seasonal snow remains the longest (Porter and Englund 2006). Crests of glacially overridden cones and inter-cone expanses of glacial till appear to lack suitable wēkiu bug habitat (Porter and Englund 2006). Wēkiu bug surveys conducted in 2006 seem to support this theory (Englund et al. 2007). This study identified several new wēkiu bug populations, including a significant population around cinder cones immediately adjacent to the VLBA facility and at Pu‘u Ko‘oko‘olau, and at a nunatak southeast of the VLBA facility. This latter nunatak was approximately 0.4 ha in size and was located at the very small outlying cone southeast of the VLBA facility. Wēkiu bugs appear to be restricted to non-glaciated habitats as they were not caught in traps in the glaciated regions of the same 11,910 ft (3,630 m) cone, even though such areas were less than 20–30 m away (Englund et al. 2007). Surprisingly, wēkiu bugs were not found in what appeared to be suitable habitat at the remote Pu‘u Māhoe cone area.¹⁴

Jesse Eiben, a PhD candidate at the University of Hawai‘i, has been researching wēkiu bug genetics and natural history since Fall 2005 and has discovered that wēkiu bugs are found not only on the summits of the pu‘us, as described by Englund et al., but also on the flanks and at the bases of the cones where cinders have accumulated to sufficient depths (Eiben 2008). Figure 2.2-20 shows the potential and known wēkiu bug habitat in the Mauna Kea Science Reserve, as determined by Eiben.¹⁵

There has been some discussion about whether wēkiu bug populations have decreased, increased, or remained the same over time since the first survey in 1982 (Howarth et al. 1999; Polhemus 2001;  Englund et al. 2002). Many insect populations naturally undergo cycles of low and high abundance over long periods of time (Howarth et al. 1999). Most of the studies were not designed to calculate population densities of wēkiu bugs, and instead measured activity levels. Trap methodologies differed between studies, which no doubt affected capture rates. Perhaps most importantly, wēkiu bug capture rates appear to be heavily influenced by climactic conditions such as presence of snow (Englund et al. 2006; Porter and Englund 2006; Englund et al. 2007), and thus it may not be appropriate to compare capture rates across studies that were conducted during different conditions or time of year. Because of these reasons it is difficult to draw any firm conclusions regarding changes in abundance of wēkiu bugs, and to subsequently identify any cause. Changes in population size, if they have occurred, could be due to a variety of causes including weather patterns, habitat disturbance, presence of invasive species, and long- term population cycling (Howarth et al. 1999). However, ten years of study following the 1997–1998 surveys suggest that wēkiu bugs are still found in all locations from the original studies, and more, and that they are able to live in both undeveloped and developed areas at the summit that have the appropriate cinder type and depth (Polhemus 2001; Englund et al. 2002; Englund et al. 2005; Englund et al. 2006; Porter and Englund 2006; Englund et al. 2007; Eiben 2008).

14 The absence of wekiu bug in traps from one trapping season does not necessarily indicate that the species does not occur in an area. Additional efforts would need to be made to truly determine if the bug was present or absent in any particular area.

15 Known wēkiu bug habitats are locations where wēkiu bugs have been captured in the high elevation regions of Mauna Kea by Jesse Eiben or the Bishop Museum teams. Potential habitats are areas that contain the correct type of cinder for wēkiu bugs, but where the bugs have not yet been captured.

Two spiders, Lepthyphantes tenuis and Meriola arcifera have invaded the Science Reserve since 1982. The first (L. tenuis) is a sheet web spider from Europe that may compete with the native sheet web spiders (Howarth et al. 1999). The second (M. arcifera) is a non web-building, ground hunting spider native to Argentina, Bolivia, and Chile (Howarth et al. 1999). This species was first collected in Hawai‘i in 1995 and is limited to upper elevations on the Saddle Road to the summit of Mauna Kea (Howarth et al. 1999). It is possible this species may prey on or compete with the wēkiu bug and other arthropods at the summit (Howarth et al. 1999).

Hippodamia convergens, a non-native beetle introduced in 1896 as a biological control agent of aphids, has been recently discovered at Pu‘u Pōhaku in the Ice Age NAR (Ramsdale 2004; Englund et al. 2005). This species is tolerant of alpine conditions, and in addition to feeding on aphids can feed on dead insects. It therefore may compete directly with the wēkiu bug for food (Englund et al. 2005). Englund et al. also found several other non-native beetle species known to eat dead invertebrates in 2004. These species (which include Aleochara verna, Creophilus maxillosus, Tachyporus nitidulus, Sphaeridium scarabaeoides Necrobia rufipes, and Dermestes frischii) may also compete with wēkiu bug for food, although there remains some question as to whether these species feed on isolated dead insects in a similar way to wēkiu bugs (Ramsdale 2004; Englund et al. 2005).

In a study of invasive invertebrates conducted by the Bishop Museum in 2007-2008, a non-native species of predatory carabid beetle, Agonum muelleri, was discovered around Lake Waiau (Englund et al. 2009). Englund et al. (2009) state that it appears to be restricted to the region immediately around Lake Waiau. As this is not favorable wēkiu bug habitat, it is unlikely this species is currently impacting the wēkiu bug. However, this predatory beetle may be impacting other native invertebrates found in the area (Englund et al. 2009).

2.2.2.2.3      Invertebrate Surveys at MKSR

Although there have been sporadic mentions of arthropods occurring at the summits of Mauna Kea and Mauna Loa over the last 110 years, the first comprehensive arthropod inventory at the summit of Mauna Kea was not conducted until 1982, following the discovery of the wēkiu bug (Nysius wekiuicola) in 1979 (Guppy 1897; Bryan 1916; Meinecke 1916; Bryan 1923, 1926; Swezey and Williams 1932; Howarth and Montgomery 1980; Gagne and Howarth 1982; Howarth and Stone 1982). Since then, there has been a fairly steady stream of research on the arthropods at the summit, with the focus being on the activity, distribution, and abundance of the wēkiu bug (Ashlock and Gagne 1983; Howarth 1983; Gagne 1986; Edwards 1987; Howarth 1987; Edwards 1988; Duman and Montgomery 1991; Howarth et al. 1999; Polhemus 2001; Brenner 2002; Englund et al. 2002; Smith 2003; Englund et al. 2005; Englund et al. 2006; Porter and Englund 2006; Englund et al. 2007; Englund et al. 2009). In addition to these studies (discussed below), several project specific wēkiu bug mitigation and monitoring plans have been created (Pacific Analytics 2000, 2001a, b). Management recommendations made in these studies and plans are incorporated into Section 4 (Component Plans) of this Mauna Kea Natural Resource Management Plan.

Invertebrate Studies:

1. Howarth, F. G. and F. D. Stone. 1982. An assessment of the arthropod fauna and aeolian ecosystem near the summit of Mauna Kea, Hawai‘i. Submitted to Group 70. Bernice P. Bishop Museum, Honolulu. 18 p.
2. Montgomery, S. L. 1988. A report on the invertebrate fauna found on the proposed NRAO VLBA antenna facility site, Mauna Kea Science Reserve, Mauna Kea, Hāmākua, Hawai‘i. Appendix G in: MCM Planning. 1988. Final supplemental environmental impact statement VLBA Antenna Facility, Mauna Kea, Hāmākua, Hawai‘i, September, 1988. Amendment to the Mauna Kea Science Reserve Complex Development Plan. Prepared for The National Radio Astronomy Observatory, Socorro, New Mexico. 4 p.
3. Howarth, F. G., G. J. Brenner and D. J. Preston. 1999. An arthropod assessment within selected areas of the Mauna Kea Science Reserve. Final Report. Prepared for the University of Hawai‘i Institute of Astronomy. Appendix J in Group 70. 2000. Mauna Kea Science Reserve Master Plan. Bishop Museum and Pacific Analytics, Honolulu. 65 p.
4. Polhemus, D. A. 2001. A preliminary survey of wēkiu bug populations at Pu‘u Hau Kea, in the Mauna Kea Ice Age Natural Area Reserve, Hawai‘i Island, Hawai‘i. Smithsonian Institution, Washington, D.C. 4 p.
5. Englund, R. A., D. A. Polhemus, F. G. Howarth and S. L. Montgomery. 2002. Range, habitat, and ecology of the wēkiu bug (Nysius wekiuicola), a rare insect species unique to Mauna Kea, Hawai‘i Island. Final report. Prepared for Office of Mauna Kea Management, University of Hawai‘i. Hawai‘i Biological Survey Report 2002–023. Bishop Museum, Honolulu, Hawai‘i. 49 p.
6. Englund, R. A., A. Ramsdale, M. McShane, D. J. Preston, S. Miller and S. L. Montgomery. 2005. Results of 2004 wēkiu bug (Nysius wekiuicola) surveys on Mauna Kea, Hawai’i Island. Final Report. Prepared for Office of Mauna Kea Management. Hawai‘i Biological Survey, Bishop Museum, Honolulu, Hawai‘i. 37 p.
7. Englund, R. A., A. E. Vorsino, H. M. Laederich, A. Ramsdale and M. McShane. 2006. Results of 2005 wēkiu bug (Nysius wekiuicola) surveys on Mauna Kea, Hawai’i Island. Final Report. Prepared for Office of Mauna Kea Management. Hawai‘i Biological Survey Report 2006–010. Hawai‘i Biological Survey, Bishop Museum, Honolulu, Hawai‘i. 63 p.
8. Porter, S. C. and R. A. Englund. 2006. Possible geologic factors influencing the distribution of the wēkiu bug on Mauna Kea, Hawai‘i. Hawai‘i Biological Survey Report 2006-031. Prepared for the Office of Mauna Kea Management. Hawai‘i Biological Survey, Bishop Museum, Honolulu, HI. 29 p.
9. Englund, R. A., A. E. Vorsino and H. M. Laederich. 2007. Results of the 2006 wēkiu bug (Nysius wekiuicola) surveys on Mauna Kea, Hawai’i Island. Final Report. Prepared for Office of Mauna Kea Management. Hawai‘i Biological Survey Report 2007-003. Hawai‘i Biological Survey, Bishop Museum, Honolulu, Hawai‘i. 66 p.
10. Englund, R.A., D.J. Preston, A.E. Vorsino, S. Meyers, and L. L. Englund. 2009. Results of the 2007–2008 Invasive Species and Wēkiu Bug (Nysius wekiuicola) Surveys on the Summit of Mauna Kea, Hawai‘i Island. April 2009 Draft Report. Hawaii Biological Survey Report prepared for the Office of Mauna Kea Management. 73 pp.
11. Eiben, J.A. In progress. The life history and genetics of the wēkiu bug (Ph.D. research project). Plant and Environmental Protection Sciences, College of Tropical Agriculture and Human Resources, University of Hawai‘i at Manoa.

Howarth and Stone (1982) conducted the first general survey of arthropods in the summit area of Mauna Kea. Methodology used in the survey included visual surveys (walking transects, turning over rocks) and placement of 100 pitfall traps at 88 different sites within the study area. Figure 2.2-12 shows the location of their survey points, which were limited to part of the plateau between Pu‘u Poli‘ahu and Pu‘u Wēkiu, Pu‘u Hau Oki, and the hillock sites near the end of the road to the north slope, at approximately the 13,000 ft (3,960 m) elevation (Figure 2.2-12). The pitfall traps were baited with fermented fish or shrimp paste (to attract scavenger species such as the wēkiu bug), buried so that their lips were flush with the ground surface, and filled with an insect preservant (ethylene glycol) that killed all invertebrates entering the traps (e.g. death traps). A cover rock was placed over the trap. The traps were left open for a long period of time, ranging from three to eight weeks. This study provided a list of species found at the site, identified to the lowest taxonomic level possible; distribution and abundance data for the wēkiu bug and lycosid spider; and notes on different habitat types (based on substrate type, size, and terrain) on the summit. Observations on species behavior and habitat types were noted when observed. Nearly 12,000 wēkiu bugs were collected in traps during this study (Porter and Englund 2006).

In 1988, Montgomery completed a visual survey, conducted on foot, of the proposed site for the VLBA antenna facility, between 12,200 and 12,400 ft (3,720 and 3,780 m) elevation; the nearby Summit Access Road; and an alternative site at 11,800 ft (3,600 m) (Montgomery 1988). No pit traps were utilized for  this project. The undersides of rocks, and the areas beneath rocks were examined for the presence of arthropods, moisture, and debris. Only three native species were observed at the proposed VLBA site: the Agrotis moth, a sheet-web spider (Erigone sp), and a springtail in the family Entomobryidae. Three non- native species were also observed at the site: two species of flies (Hydrellia tritici and Pollenia rudis), and one parasitoid wasp in the family Chalcidoidea. No wēkiu bugs were observed during the survey. The native wolf spider (Lycosa species) was observed only at the alternative site. Because these sites were flat and uniformly bedded with ash and cinders, they were not considered prime habitat for the native arthropods found at the summit (Montgomery 1988).

In 1997–1998, Howarth, Brenner and Preston conducted the second general arthropod survey for the summit area (Howarth et al. 1999). Sampling techniques utilized in the study were different from the 1982 survey. Live pitfall traps (where food, water, and shelter were provided) were used during sampling activities, and traps were left open for only three days at a time. Pitfall traps were placed on Pu‘u Wēkiu, Pu‘u Hau Oki, Pu‘u Māhoe, Pu‘u Kea Ridge, Pu‘u Lilinoe, and the north slope plateau road. Sampling locations are presented in Figures 2.2-13 through 2.2-15. The 1997–1998 sampling activities increased the area surveyed from that of the 1982 study, in order to determine if wēkiu bugs were found in other areas and habitat types, and to further study the distribution of lycosid spiders and other invertebrates of interest. A much lower number of wēkiu bugs (0.16 wēkiu bugs per trap) were captured in this study than in the 1982 study (60 bugs per trap). Because of the differences in trap methodology it is not possible to directly compare wēkiu bug abundances found in the 1982 study to the 1997–1998 study. Among the reasons for this, live traps leave open the possibility of escape and, also, predation of arthropods within the traps by predators such as the lycosid spider. Also, shorter trap times are less likely to reflect arthropod responses to variations in weather (and other factors) (Howarth et al. 1999). This study was useful, nonetheless, in refining information known about the distribution of wēkiu bugs, and added to the list of species known to reside at the summit and to arrive as aeolian drift.

In 2001, due to concern about possible reduction in wēkiu bug habitats due to astronomy-related development, and to learn more about their distribution elsewhere on the summit, Dan Polhemus conducted a preliminary survey of wēkiu bug populations at Pu‘u Hau Kea, in the Mauna Kea Ice Age NAR. Polhemus used ethylene glycol pit fall traps (death traps) similar to those used in 1982. Ten traps were distributed across the outer slope, rim, and inner crater, and were left open for a period of four days. A large number of individual wēkiu bugs were caught (47 bugs/trap). Of these, 20% were adults, and the remainder were instars¹⁶ of various stages. Most of the bugs were caught on the rim and inner crater. This study raised interest in conducting a comparison of various trapping methods (live vs. death) to help determine which method would be most productive for use in future studies.

In response to Polhemus’ 2001 findings, Englund, Polhemus, Howarth and Montgomery conducted further investigations into the elevational distribution of wēkiu bugs and their presence/absence on various unstudied pu‘u (cinder cones) on Mauna Kea (Englund et al. 2002). Areas surveyed during this study included Pu‘u Mākanaka, Pu‘u Māhoe, Pu‘u Ala, Pu‘u Poepoe, Pu‘u Keonehehe‘e, and adjacent unnamed cones, and several unnamed cones near the VLBA facility. Sample locations are shown in

16 Instars are the life stages of insects that occur between molts.

The Englund et al. (2002) study also compared trapping methodologies used in previous studies to determine if information obtained from studies using different methodology is comparable. To assess the effectiveness of various trapping methods on wēkiu bug capture rates, Englund et al. (2002) placed eight live pitfall and eight ethylene glycol (death) traps in a pair-wise fashion in windblown areas near the upper rim of Pu‘u Hau Kea. This allowed direct comparison of trapping methods during the same time period. Both sets of traps were left undisturbed for three nights. Unfortunately only nine wēkiu bugs total (five in live traps, four in death traps) were captured during the study. The capture rates were essentially the same for both trap types; however, one important note from the study was that the live traps actually had a 56% mortality rate. This finding was also true for previous studies using live traps (Brenner et al. 1999). For the following reasons, Englund et al. determined that for baseline monitoring of wēkiu bugs, and for general assessments of invertebrate diversity, ethylene glycol (death) traps were preferable over live traps because:

1. Large predators such as lycosid (wolf) spiders can consume items within live traps. This makes it difficult to determine whether traps with no wēkiu bugs were truly empty or whether the wēkiu bugs had been consumed.
2. Because insects cannot escape the ethylene glycol traps, they provide a better measure of wēkiu bug presence and the presence of other species of interest, including invasive species.
3. Judicious use of ethylene glycol traps will have little or no negative long-term impacts on wēkiu bug populations.

In April and July 2004, Englund, Ramsdale, McShane, Preston, Miller and Montgomery returned to Mauna Kea for additional wēkiu bug surveys, and to continue research begun in the 2002 study (Englund et al. 2005). Areas included in the survey included Pu‘u Kanakaleonui (9,594 ft/2,925 m), Pu‘u Pōhaku, Pu‘u Hau Oki, Lake Waiau, Red Hill, and Pu‘u Hau Kea. The investigators placed 55 pitfall traps (similar to those used in 2002) in the field: five were placed at Pu‘u Hau Oki in April and the remainder (50 traps) at Pu‘u Kanakaleonui, Pu‘u Pōhaku, Lake Waiau, Red Hill, and Pu‘u Hau Kea in July. Figure 2.2-17 shows the locations of the survey efforts. Only ten bugs were captured at Pu‘u Hau Oki in April and only one was captured at Pu‘u Pōhaku in July. No wēkiu bugs were captured at Pu‘u Hau Kea (previously known to have a high number of wēkiu bugs), Lake Waiau, Red Hill, or Pu‘u Kanakaleonui in July 2004. As in 2002, a comparative test of five ethylene glycol (death) and five shrimp pitfall (live) traps was conducted at Pu‘u Hau Kea in July 2004. However, no wēkiu bugs were captured in either set of traps in July 2004. Although this study did not provide much information about wēkiu bugs (except that they were not active in July 2004), it did provide information on the presence of eight new species of introduced beetle unknown to reside on Mauna Kea (three of which were new records for the island of Hawai‘i). These are described in Section 2.2.2.2.2.

In addition to the above work, Englund et al. (2005) placed temperature and humidity loggers in the field to obtain microhabitat data on wēkiu bug habitats. These were placed in 1) areas known to have high wēkiu bug densities, 2) areas disturbed by development that previously had high wēkiu bug densities, and 3) areas adjacent to high-quality habitat that lack wēkiu bugs. A total of eight data loggers (two subsurface data loggers, and six surface data loggers) were installed in July (at Pu‘u Pōhaku and Pu‘u Hau Kea). In December, 18 data loggers (nine pairs of subsurface and surface) were installed along a “transect starting at the bottom of the northwest rim and extending in a southeasterly direction to the cinder cone and down the slope to the bottom of the Pu‘u Hau Kea cinder cone” (Englund et al. 2005). In addition to the transect line, several data loggers were placed in a variety of areas throughout the Mauna Kea Summit (for a total of 45 data loggers placed in December). Initial results from this study include recording of the extreme fluctuations of temperature and humidity that occur at the surface of the substrate on Mauna Kea. The one subsurface probe that was discussed showed a much lower variation in temperature and humidity than the surface probes, thus indicating a more stable environment and reinforcing the idea that wēkiu bugs can find refuge in subsurface areas during great fluctuations in the surface conditions.

In 2005, Englund, Vorsino, Laederich, Ramsdale and McShane continued investigations into wēkiu bug distribution on Mauna Kea, expanding the total area of Mauna Kea surveyed and greatly increasing the field time and number of traps placed (Englund et al. 2006). Some of the most remote areas of the Mauna Kea summit were sampled in 2005. Sampling locations (using the names given in the report, (Englund et al. 2006)) included:

−        Pu‘u Hau Kea

−        Pu‘u Wēkiu

−        Red Hill

−        Pu‘u Hoaka

−        “Below sub millimeter array”

−        Pu‘u Pōhaku

−        Unnamed N. Pu‘u VLBA

−        Unnamed S. Pu‘u VLBA

−        Pu‘u #1 S.E. of VLBA

−        Pu‘u #2 S.E. of VLBA

−        Horseshoe Crater (very large unnamed crater downslope of Pu’u #2 S.E. of VLBA)

−        The plateau where the 30 m telescope has been proposed (northeast of Pu‘u Poliahu)

−        Pu‘u Lilinoe and the cinder cones surrounding it

−        Skiing area called “Poi Bowl” that is down slope of Pu‘u Hau Oki

−        “11,672 ft Pu‘u” at northwest summit

−          Unnamed pu‘u at northwest summit

−        Cone at terminal moraine.

Sampling locations are shown in Figure 2.2-18. The investigators used live traps baited with shrimp paste and a smaller number of ethylene glycol (death) traps, similar to those described in Englund et al. (2002). A total of 153 traps were placed in sampling locations (90 in May and 63 in June). In addition, the investigators tested a new survey technique involving timed visual surveys, mainly in areas around snow banks. A total of 70 wēkiu bugs were observed or collected in 2005, with nearly equal amounts collected in traps in May and June (17 in May, 16 in June; the remaining 37 bugs were collected during visual observations). Many wēkiu bugs were observed during timed visual collections along snow banks at the summit of Pu‘u Hau Kea (May) and Pu‘u Poliahu (June), indicating that the populations there remain robust (Englund et al. 2006). An important finding of the 2005 study was that visual collection along snow banks was more time-efficient than either shrimp (live) or ethylene glycol (death) traps. For example, 13 wēkiu bugs were observed in 20 minutes of observation time in May, while the traps in the nearby areas collected only 11 wēkiu bugs in a 10-day period. One limitation of visual surveying along snow banks is that snow banks are patchily distributed in time and space, and they are not present in all locations where wēkiu bugs live. Therefore this method may not be appropriate as a sole survey methodology. However, it is a useful and quick way to survey for the presence of bugs during the right climatic conditions.

Another important finding of the 2005 study was that sampling efforts confirmed that wēkiu bugs were found only in cinder cone areas and were never in glacial floor areas between cinder cones, such as the proposed location of the 30-meter telescope facility and the Poi Bowl ski area (Englund et al. 2006).¹⁷ Pu‘u Lilinoe was the only area where new populations of wēkiu bugs were found. However, this represents a significant extension of their known core habitat. Wēkiu bugs were also found at Pu‘u Hau Kea, Pu‘u Pōhaku, Pu‘u Wēkiu and Pu‘u Poliahu. No wēkiu bugs had been observed at Pu‘u Poliahu between 1982 and 2005, but in 2005 a large snow bank was found along the northeastern side of Pu‘u Poliahu during sampling activities, which probably increased wēkiu bug activity and trapping success. No wēkiu bugs were trapped or observed around “Horseshoe Crater” despite the presence of a large snow bank and large amounts of aeolian drift, or at the western summit cones above Pu‘u La‘au cabin, indicating these areas may either not be good wēkiu bug habitat or that they have yet to be colonized by wēkiu bugs.

As in 2002 and 2004, a comparative test of ethylene glycol (death) and shrimp pitfall (live) traps was conducted at Pu‘u Hau Kea in May and June 2005. Five wēkiu bugs were captured in glycol traps and three in shrimp traps in May, indicating no difference between the effectiveness of the two. However, in June, eleven bugs were collected in glycol traps, while only one was collected in the shrimp traps. Thus the results from June suggest that perhaps glycol traps are more efficient, at least during times of reduced snow cover (June). However, overall the results from the trap efficiency studies remain inconclusive due to low capture rates.

The collection of temperature and humidity data begun in 2004 continued through 2005. Initial analysis of the thermal data indicated that areas inhabited by wēkiu bugs seem to be both colder and have more stable temperatures than those not inhabited by wēkiu bugs. Englund et al. (2006) also found that spring months when wēkiu bugs are most active exhibited dramatic daily temperature shifts, with temperatures dropping below freezing every night. Another finding was that the telescopes do not seem to have warmed (or affected the thermal regimes of) nearby Pu‘u Hau Oki. These initial findings will be further investigated over time.

In addition to the above study conducted by Englund et al., Porter and Englund conducted another important study of wēkiu bug distribution in 2005. This study was a collaboration between a geologist (Steven Porter, University of Washington) and wēkiu bug expert Ron Englund (Bishop Museum), to determine whether wēkiu bug distribution is influenced by the geology of the summit. By analyzing historical wēkiu bug capture data, Porter and Englund found that glaciation that occurred around 20,000 years ago seems to have influenced current wēkiu bug distribution. Wēkiu bugs are predominantly found on or near the crater rims of cinder cones that formed nunataks (ice-free areas rising above a surrounding glacier) or that lay at the glacier limit during the last glaciation (Porter and Englund 2006). Porter and Englund (2006) determined that the bugs prefer to reside in cinder and spatter near the unmodified crests of cinder cones, and that they are concentrated in areas where seasonal snow remains the longest (on the north- and east-facing slopes, and on slopes shaded by local topography). Snow patches appear  to increase in area and number with increasing altitude above the limit of glaciation. Aeolian drift is concentrated along the margins of snow packs. Snowpacks thus provide an important source of food and moisture for the wēkiu bugs. Crests of glacially overridden cones and inter-cone expanses of glacial till appear to lack suitable wēkiu bug habitat, and seasonal snow tends to disappear quickly in these areas

17 However, see discussion of Jesse Eiben’s work at the end of this section. Eiben has consistently found wēkiu bugs on crater flanks and bases, such as at Poi Bowl.

(Porter and Englund 2006). Porter and Englund recommend that further wēkiu bug surveys be conducted in geologically promising unstudied pu‘u.

Porter and Englund (2006) also conclude that wēkiu bugs are capable of living on the unmodified snowy slopes near the telescope facilities, and that telescopes sited on glacially modified lava flows (unsuitable wēkiu bug habitat) would not likely affect wēkiu bug populations on the summit. This information can be used to help plan the locations of future telescope development or modifications of existing telescopes on the mountain.

In 2006, Englund, Vorsino, and Laederich continued the Bishop Museum study of wēkiu bugs on the summit of Mauna Kea (Englund et al. 2007). This study continued along the line of the 2002–2005 studies and included visual surveys, shrimp bait (live) pitfall traps, and ethylene glycol (death) traps. Sampling was conducted in April and May 2006, and the trap effort was increased above previous year’s efforts (158 traps, total; 70 in April and 88 in May). Areas sampled include: Pu‘u Hau Kea, Pu‘u Waiau, Pu‘u Ko‘oko‘olau, Pu‘u Hoaka, Pu‘u Poli‘ahu, “11,989 ft Pu‘u” (near Pu‘u Hoaka), Glacier Cone, Horseshoe Crater, pu‘u near Māhoe, pu‘u north of VLBA, pu‘u south of the VLBA, far pu‘u beyond VLBA, 1^st pu‘u beyond VLBA, “below Keck & Subaru,” and Pu‘u 11,605. Locations of traps are presented in Figure 2.2-19. Englund et al. found that the snowy conditions in 2005–2006 provided favorable conditions for wēkiu bugs, and a large number of individuals (114) were trapped or observed around snowbanks. Interestingly, Englund et al. trapped no wēkiu bugs during a period of heavy snowfall and snow cover in April, suggesting that wēkiu bugs may remain inactive for some time during and after heavy snowfall.

During the 2006 field season, the “nunatak hypothesis” put forth by Porter and Englund (2006) was tested by conducting sampling for wēkiu bugs at the various locations suggested as promising in the 2006 report (Porter and Englund 2006). Several significant new bug populations were found using these predictions, most notably in areas adjacent to the VLBA facility and areas around the Adze Quarry (Englund et al. 2007). Surprisingly, wēkiu bugs were not found in what appeared to be suitable habitat at the remote Pu‘u Māhoe area.

As in 2002–2005, the 2006 study included a comparative test between shrimp (live) pitfall traps and ethylene glycol (death) traps. This year, ethylene glycol traps were found to be much more effective than shrimp pitfall tests (43 wēkiu bugs captured in ethylene glycol traps and only one captured in shrimp traps). Because of the effectiveness of the ethylene glycol traps over two years of testing, Englund et al. state that no further pitfall-trap testing will be done (Englund et al. 2007)¹⁸.

Englund et al. (2007) also conducted a brief survey of Nysius aa, the Mauna Loa bug, which is the sister species to the wēkiu bug (Nysius wekiuicola). They found that the Mauna Loa bug, although it too resides in cinder habitats, appears to be much more abundant within a broader range of habitat types than the wēkiu bug.

In 2007 and 2008, Englund et al. continued the wēkiu bug sampling studies from previous years, although with some modifications to the methods. The number of trap hours was reduced, fewer locations were included, and no ethylene glycol kill traps were deployed (Englund et al. 2009). Despite lower effort, a large number of wēkiu bugs were captured in 2007, indicating a robust population. Comparison of wēkiu bug density and mean annual temperatures were conducted to help understand wēkiu bug distribution. Results suggest that the wēkiu bug is most abundant in the areas with coldest annual temperatures, and

18 Despite this, live trapping continues to be the preferred method utilized by Bishop Museum.

An important study of the life history and genetics of the wēkiu bug is currently being undertaken by Jesse Eiben and Dr. Dan Rubinoff, at the University of Hawai‘i. They are rearing the bug in  the laboratory in order to study the developmental stages of the bug, and to develop a temperature-dependent growth curve for the bug. The egg and immature stages have never been described in the literature. One interesting observation made is that the wēkiu bug is much more “heat-tolerant” than previously described and in some ways does better (i.e., grows faster) at warmer temperatures than commonly found on the mountain. The temperature of wēkiu bug microhabitat on the mountain varies widely over the course of each day, with surface temperatures of up to 110 degrees F commonly seen in full sun, even when the air temperature is at freezing. The growth-rate observations support the idea that the bug takes advantage of solar heating in some form at the microhabitat level to aid in growth and egg development. The genetics component of the project was devised to study population-level issues, such as whether bugs regularly migrate between various pu‘u, or if movement between the pu‘u is limited to a very few bugs (effectively changing the amount of “inbreeding” that occurs in each small population of bugs). The first round of genetic analysis showed very little variation within the species, suggesting the possibility of a recent genetic bottleneck. Further investigations (using nuclear DNA microsatellite analyses) are underway to further study population genetics in the wēkiu bug (Eiben 2008).

Based on results of his fieldwork, and those of the Bishop Museum Study, Jesse Eiben has prepared a map of potential and known wēkiu bug habitat in the Mauna Kea Science Reserve. This map is reproduced in Figure 2.2-20. The map shows the extent of wēkiu bug habitat across the summit of Mauna Kea, and highlights areas of existing disturbance (infrastructures such as observatories and roads). Further investigation of the distribution of wēkiu bugs on the summit is needed to increase the accuracy of this map. However, the map is useful in delineating areas that should be considered for protection from further development.

2.2.2.3       Threats to Invertebrate Communities on Mauna Kea

2.2.2.3.1      Habitat Alteration

Habitat alteration threatens native invertebrate communities by directly removing habitat (through development) or changing it to the extent that the invertebrates are no longer able to live there (for example, by changing host-plant abundances). At both Hale Pōhaku and the MKSR, habitat alteration occurs through development of astronomy facilities and support structures (such as parking lots), every- day use, and (primarily in the subalpine zone) introduction of invasive species.

A prime example of habitat loss through development is the loss of wēkiu bug habitat on the summit through construction of telescope facilities. In some cases, site preparation for telescope facilities  included bulldozing the cone summit to create a level surface, the removal or flattening of rims, and filling of craters with debris (Porter and Englund 2006). For example, during the construction of the Keck Observatory, the top ten meters of Pu‘u Hau Oki was removed and the crater was filled with approximately 12 meters of pyroclastic debris (Porter and Englund 2006). All told, the development of W. M. Keck and Subaru Observatories filled or buried about one-third of the within-crater habitat of Pu‘u Hau Oki (Pacific Analytics 2000).

Wēkiu bug habitat is easily altered by vehicular traffic and construction activity, as the tephra cinders preferred by the bug are easily crushed into dust-sized particles. Prime habitat can be quickly degraded to compacted silt and mud by use of off-road vehicles (Richardson 2002). It has been estimated that since 1963, approximately 62 acres (25 hectares) of potential arthropod habitat have been lost to astronomy- related development on the summit (Richardson 2002; Smith 2003).¹⁹ Wēkiu bug habitat may also be altered by dust blown up from road grading and other construction activities on the summit (Pacific Analytics 2000). This dust can reduce surface porosity and fill pockets between cinders, inhibiting movement by arthropods and perhaps affecting wēkiu bug food sources by decreasing the accumulation of aeolian drift (dead and dying insects blown from lower elevations) in these surface pockets (Pacific Analytics 2000). The true level of impact from dust is unknown at this time, as it has not been studied.

Grazing by introduced mammals has heavily altered habitats in the subalpine woodlands, by changing the composition of plant species in favor of invasive weed species. Many of the native plants previously used by native invertebrates, such as Hylaeus bees, have been reduced in abundance to the point that the small and widely dispersed native plant populations are no longer able to support pollinator populations (USFWS 1994; Aldrich 2007). Regarding the situation with native bees, Magnacca (2007) states that

 “The strong dependence of all species of Hylaeus on native plants to the near-complete exclusion of exotics (with the sole exception of Tournefortia among the latter) means that to conserve them native forests and shrublands must be preserved. Conservation of the plants and bees is a reciprocal situation, given the likely status of Hylaeus species as primary pollinators of many important plants. Based on current distributional information, it is clear that for bees to be present, at least some level of vegetation diversity is required. This is probably due to a combination of temporal (i.e., year-round availability of floral resources) and nutritional factors.”

Thus, habitat alteration through removal of plant species can seriously impact populations of pollinators and other animals that rely on the plants as source of food or shelter. The destruction of their pollinators can, in turn, can make it difficult or even impossible for these plant species to repopulate the area. Once started, this vicious cycle can be difficult to eliminate.

2.2.2.3.2      Invasive Species

As described above, grazing mammals and invasive plant species can reduce populations of native plants, effectively reducing food and shelter for native pollinators and the other arthropods that rely on them, and thus indirectly reducing native arthropod populations. Other invasive animals, such as rats and non-native birds, can impact arthropod populations directly through predation. In fact, several species of flightless insects no longer occur on the main Hawaiian Islands where rats are common (Gagne and Christensen 1985). However, invasive invertebrates are perhaps the greatest threat to native invertebrates in Hawai‘i. Invasive invertebrates threaten native invertebrate populations through competition, predation, habitat alteration, and parasitism. Nearly 4,000 species of invertebrates have been introduced to the Hawaiian Islands, 75% of which are arthropods (including 2,400 insects and 500 other arthropods) (Loope et al. 2001). Most non-native invertebrates arrive unintentionally, and the remainder (about 1/5) are introduced as biological control agents (Loope et al. 2001).

Invasive parasitoid wasps and flies are likely reducing Cydia moths and other moth species that live in the subalpine māmane woodlands (Brenner et al. 2002; Oboyski et al. 2004). The larvae of these moths are an important source of protein for the endangered Palila (Brenner et al. 2002). Thus the parasitoid wasps not only directly affect the moths they attack but also indirectly affect predators of the moths such as the

19 Actual habitat loss is unknown. The 62 acres represents the total area at the summit disturbed through observatory and infrastructure development.

Invasive yellowjackets and ants are also likely affecting native invertebrate populations in the māmane woodlands. Cole et al. (1992) found that many invertebrate populations on Haleakalā were smaller in areas infested with Argentine ants than in areas not infested, and Gambino et al. (1987) found that yellowjackets prey on native arthropods. Wetterer et al. (1998) conducted bait surveys for ants along the “Observatory Road” on Mauna Kea from 8,005 ft (2,440 m) to 9,250 ft (2,820 m) elevation. One Argentine ant, Linepithema humile, was trapped at Kalepeamoa at 8,497 ft (2,590 m).

The predatory European earwig (Forficula forficularia) was found in high numbers around the Onizuka Visitor Information Station at Hale Pōhaku (Englund et al. 2009). This could potentially impact native invertebrate species in the subalpine zone (Englund et al. 2009).

Invasive snail species such as the garlic snail (Oxychilus alliarius) have a direct negative impact on native snails through predation. Native terrestrial snails are also impacted by invasive mammals such as rats and birds such as Ring-necked pheasants.

At the summit of Mauna Kea, the greatest threat to the native arthropod populations is introduction of invasive arthropods that are adapted to alpine conditions. In recent decades two alpine-adapted spider species have invaded the summit of Mauna Kea. One, a sheet web spider from Europe, Lepthyphantes tenuis, may compete with the native sheet web spiders, although this has yet to be studied. The other, greater, threat is Meriola arcifera, a ground hunting spider native to Argentina, Bolivia, and Chile that may prey on or compete with the wēkiu bugs and other arthropods at the summit (Howarth et al. 1999). There are also several non-native beetle species that may have become established on the summit (Englund et al. 2005; Englund et al. 2009). The primary beetle species that may impact native arthropods are Agonum muelleri and Hippodamia convergens. Agonum muelleri is a predatory carabid beetle that likely preys on native invertebrates in the summit region. It appears to be currently limited to Lake  Waiau, and may not have an impact on wēkiu bugs unless it spreads to other regions of the summit (Englund et al. 2009). Hippodamia convergens is a non-native alpine-adapted beetle introduced in 1896 as a biological control agent of aphids. H. convergens is known to feed on dead insects and therefore may compete directly with the wēkiu bug for food. The impact of these species, if any, has yet to be determined. If food (in the form of aeolian drift) is not limiting, then scavengers such as H. convergens are not likely to pose as great a threat as do predators such as Agonum muelleri and Meriola arcifera.

There is always potential for the introduction of new invasive species to both Hale Pōhaku and the summit through importation of goods from similar climates (such as astronomical equipment); accidental transport on vehicles, clothing, and equipment; and the inevitable spread of biological control agents. Perhaps the most threatening introduction to the summit would be a predator or parasitoid that is small enough to fit into the interstitial spaces in the cinders that are used by the wēkiu bugs to escape such predators as the native wolf spider (Pacific Analytics 2000).

Human use can impact native invertebrate populations by altering their habitats, introducing invasive species, increasing pollution (through spills of hazardous material, vehicle emissions, etc.), improper disposal of trash, and collecting of plant and animal specimens for both scientific research and hobbies.

The tephra habitats utilized by wēkiu bugs are easily degraded to compact, silty habitats during construction activities, off-road vehicle use, and trampling (Howarth and Stone 1982). Erosion is a common problem in construction and heavy use areas at Hale Pōhaku (Gerrish 1979). Erosion and soil impaction can limit native vegetation relied upon by native arthropods in the subalpine zone, and can also directly impact soil-dwelling species.

Chemical spills can occur both at Hale Pōhaku and the summit area. Several oil spills have occurred at the summit (Smith et al. 1982), and hazardous chemicals are used to clean the mirrors at several observatories (Pacific Analytics 2000; NASA 2005). Spills of petroleum products and hazardous chemicals will most likely kill most of the native ground-dwelling arthropods in the immediate vicinity. Oil spills, in particular, will take a long time to biodegrade at the summit because of the cold and dry conditions found there (Howarth and Stone 1982).

The impacts of human use of Hale Pōhaku, the Summit Access Road, and the MKSR are discussed in more detail in Section 3.

2.2.2.3.4      Climate Change

For a discussion of potential ramifications of climate change on weather patterns and native plant communities see Section 2.2.1.3.6.

It is very difficult to predict just what impacts global climate change will have on Mauna Kea in the future. Currently it is believed (and this is supported by recent climate data) that temperatures will increase and that the greatest increases will occur at Mauna Kea’s highest points. Climate change scenarios variously predict an increase in rainfall associated with the increase in temperatures (Hamilton 2007), or a decrease in rainfall at high altitudes due to a depression of the inversion layer (Loope and Giambelluca 1998; Calanca 2007). Either rainfall scenario will spell change for the invertebrate communities on Mauna Kea.

An increase in temperature at the summit may lead to:

1. Upwards movement into the subalpine and alpine zone of both native and non-native invertebrate species from lower elevations that were previously kept out by freezing temperatures, possibly increasing competition with and/or predation on the current subalpine and alpine arthropods (Benning et al. 2002; Weltzin et al. 2003).
2. Shorter duration of snow pack before it melts in the higher elevation areas, leading to longer periods of time without snow pack. This could affect wēkiu bugs because they often forage on the edges of snow banks and tend to do well in years of heavy snowfall (Englund et al. 2006). However, if warming temperatures lead to an increase in snowfall at the summit (Hamilton 2007), this may have a beneficial effect on arthropods such as the wēkiu bug that feed on the edges of snow banks.

If wind patterns change, the summit of Mauna Kea could see a change in the amount of aeolian drift deposited. This could have serious impacts on the arthropod community at the summit, which is almost entirely dependent on the aeolian drift as a food source.

2.2.2.4       Invertebrate Community Information Gaps

There is still much to be learned about the invertebrate communities at both Hale Pōhaku and the MKSR. Of the two areas, the MKSR summit has had more intensive research conducted, with most of the research focusing on the wēkiu bug. Through research conducted over the past ten years, a clearer picture of wēkiu bug distribution, habitat preferences, and biology is beginning to appear. However, very little information is available on the other arthropods found at the summit, and next to no information is available about arthropod communities at Hale Pōhaku and in the MKSR below the summit.

The following information gaps regarding the condition of the subalpine and alpine arthropod communities at Hale Pōhaku and the MKSR have been identified through review of the literature and consultation with local experts:

1. Quantitative invertebrate surveys

a)       Hale Pōhaku: No comprehensive quantitative studies have been conducted at Hale Pōhaku documenting the composition of the invertebrate community, population sizes or distribution of native and non-native species. However, the Bishop Museum has completed a presence/absence study of invasive invertebrates in this area (Englund et al. 2009). Additionally, a brief visual survey for other species of Lygaeid bugs (relatives of the wēkiu bug) was conducted at Hale Pōhaku in 2002 (Englund et al. 2002). Some species- or genus- specific work has recently been conducted on lepidopteron and tephritid fly species (Brown 2008; Medeiros 2008).

b)      Summit Access Road: No comprehensive invertebrate surveys have been conducted along the Summit Access Road between Hale Pōhaku and MKSR, with the exception of the invasive invertebrate survey, and the brief surveys for lygaeid bugs and ants described above.

c)       Mauna Kea Science Reserve: Two comprehensive arthropod surveys have been conducted in the MKSR (Howarth and Stone 1982; Howarth et al. 1999). Both of these surveys were limited in the area covered, but produced a good baseline list of species found in the community. Since then, research has been ongoing into the status and biology of the wēkiu bug (Englund et al. 2002; Englund et al. 2005; Englund et al. 2006; Englund et al. 2007; Eiben 2008; Englund et al. 2009). Since ten years have passed since the last comprehensive survey, it may be time to conduct another. If available, information on by-catch from the last six years of Bishop Museum wēkiu bug studies and the 2007-8 invasive species surveys, may provide a relatively accurate species list.

1. Status of Invasive Species

Limited information is available regarding density, distribution, and effects of established invasive invertebrate species at Hale Pōhaku and the MKSR. Wetterer et al. (1998) conducted bait surveys for ants along the “Observatory Road” from 8,005 ft (2,440 m) to 9,250 ft (2,820 m) elevation and found one Argentine ant (Linepithema humile) at Kalepeamoa at 8,497 ft (2,590 m). The recent survey of invasive arthropods completed by the Bishop Museum (Englund et al. 2009) has provided an update on invasive species present at Hale Pōhaku and the MKSR, although it does not provide densities or distribution maps. There is a continuing need for ongoing inspections for and monitoring of invasive invertebrate species on the properties and identification of environmental problems they may be causing. There currently exists no information on whether any of the invasive invertebrate species present are impacting native species of plants and animals. Species of concern include, at a minimum, yellowjackets, ants, parasitoid wasps, predatory beetles, and the two invasive spiders at the summit.

2.2.3        Birds

Hawai‘i has an incredible diversity of birds, a great number of which are endemic species (meaning they are found only here). This amazing diversity of species evolved from a few different species of birds that managed to colonize the islands. For example, the Hawaiian honeycreepers, a diverse group of approximately 50 species (including extinct forms), are thought to have evolved from a single species of cardueline finch that arrived 15 to 20 million years ago, on the islands that now make up the northwestern Hawaiian island chain (Freed et al. 1987; James 2004). Hawaiian birds have been greatly affected by the arrival of man and his associated animal species on the islands – in fact, only 35 of more than 100 species present before man arrived are still alive today (Olson and James 1982; Pratt et al. 1997). Most Hawaiian bird species went extinct before European contact, but 27% of endemic Hawaiian birds have gone extinct since 1778 because of human activities (Fancy and Ralph 1998). A large percentage of extant native bird species are endangered due to habitat loss, non-native predators (cats, rats, and mongoose), disease (avian malaria and pox), hunting and over-collection (historically for feathers, meat, or specimens), and competition with non-native birds and insects for food (Scott et al. 1986). There are numerous non-native species of birds in the islands. The first avian introduction came with the arrival of the early Polynesians, who brought the Red junglefowl (Gallus gallus) for food, but most introductions came after Western contact, either as intentional introductions or escaped pets (Moulton et al. 2001).

2.2.3.1       Subalpine Bird Communities (Hale Pōhaku and Lower Summit Access Road)

The māmane woodlands have a fairly diverse bird community, including frugivores, nectarivores, insectivores, and two raptor species. The māmane trees themselves are the primary food source for birds in the region, providing nectar and seeds on a seasonal basis (Hess et al. 2001). Several bird species also prey in the insects that utilize the māmane trees. Māmane woodlands have been severely degraded by non-native browsing animals (cattle, sheep, goats), and consequently native bird populations have declined in this forest type (Juvik and Juvik 1984). See Section 2.2.3.3 for more information on the threats to avian communities on Mauna Kea and Sections 2.2.1.3.4 and 2.2.4 for more information about the effects of non-native mammals on the māmane woodland and its inhabitants. Photos of native bird species found in the subalpine zone are presented in Figure 2.2-21 and photos of non-native bird species are found in Figure 2.2-22.

Native bird species found in māmane woodlands on Mauna Kea include the Palila (Loxioides bailleui), ‘Amakihi (Hemignathus virens), ‘Apapane (Himatione sanguinea), ‘Elepaio (Chasiempis sandwichensis sandwichensis), ‘Akiapola‘au (Hemignathus munroi), ‘I‘iwi (Vestiaria coccinea), ‘Io (Buteo solitarius), Kolea (Pluvialis fulva) and Pueo (Asio flammeus sandwichensis) (Scott et al. 1986). Each of these species is discussed in more detail below, in Sections 2.2.3.1.1 and 2.2.3.1.2. The Hawaiian petrel or ‘Ua‘u (Pterodroma sandwichensis), has been observed in subalpine lava flows on Mauna Loa, at 8,000–9,200 ft (2,440–2,800 m), and occasionally in subalpine and alpine habitats on Mauna Kea (Conant 1980; Kjargaard 1988; Hu et al. 2001). The ‘Ua‘u is discussed further in Section 2.2.3.2. Of the above species only the Palila, ‘Amakihi, ‘Apapane and ‘I‘iwi have been observed at Hale Pōhaku in recent times (see Section 2.2.3.1.4).

Non-native birds found in māmane and māmane-naio woodlands on Mauna Kea include Black Francolin (Francolinus francolinus), Erckel’s Francolin (Francolinus erckelii), Chukar (Alectoris chukar), Japanese Quail (Coturnix japonica), Ring-necked Pheasant (Phasianus colchicus), Wild Turkey (Meleagris gallopavo), California Quail (Callipepla californica), Eurasian Skylark (Alauda arvensis), Melodious Laughing-thrush (Garrulax canorus), Red-billed Leiothrix (Leiothrix lutea), Northern Mockingbird (Mimus polyglottos), Common Myna (Acridotheres tristis), Japanese White-eye (Zosterops japonicus), Northern Cardinal (Cardinalis cardinalis), House Finch (Carpodacus mexicanus), House Sparrow (Passer domesticus), Warbling Silverbill (Lonchura malabarica), Nutmeg Mannikin (Lonchura punctulata), and Yellow-fronted Canary (Serinus mozambicus) (van Riper 1978; Scott et al. 1986). Of these, only eight species (Erckel’s Francolin, California Quail, Eurasian Skylark, Red-billed Leiothrix, Japanese White-eye, House Finch, House Sparrow, and Yellow-fronted Canary) have been recorded as occurring at Hale Pōhaku during limited survey work conducted there. However, it seems likely that the most of the non-native species listed above can be found at or near Hale Pōhaku, at least seasonally. See Section 2.2.3.1.3 for information on invasive bird species at Hale Pōhaku.

Birds found in the subalpine community at Hale Pōhaku are listed in Table 2.2-7. Species that are likely to occur, but have not been observed (as recorded in literature or observed by experts), are indicated with a “??” under the Hale Pōhaku column.

2.2.3.1.1      Threatened and Endangered Species

Federally listed Endangered species that occur in māmane woodlands on Mauna Kea include the Palila (Loxioides bailleui), ‘Akiapola‘au (Hemignathus munroi), ‘Io (Buteo solitarius) and Nēnē (Branta sandvicensis). The latter three species have not been recorded at Hale Pōhaku. There are no federally listed Threatened species known to be found at Hale Pōhaku.

‘Akiapola‘au (Hemignathus munroi) are honeycreepers with a strongly decurved upper bill and a stout, woodpecker-like lower bill that can be used to drill holes in trees and loosen bark (Scott et al. 1986). The ‘Akiapola‘au then uses its upper bill as a tool to pick out insects (primarily moth larvae and beetles) from under the bark (Pejchar and Jeffrey 2004). ‘Akiapola‘au are primarily insectivorous, but also supplement their diet with sap from ‘ōhi‘a trees (Pejchar and Jeffrey 2004). Prior to disturbance by man and deforestation by introduced grazing mammals, mesic and dry forest cover was nearly continuous from eastern Mauna Kea to Hāmākua. During this time, ‘Akiapola‘au were most likely common and widespread (Scott et al. 1986). In the 1970s, ‘Akiapola‘au were still found in low numbers in māmane and māmane-naio woodlands on Mauna Kea from 6,200 to 9,500 ft (1,900 to 2,900 m) elevation (Scott et al. 1986). Currently ‘Akiapola‘au very rare in (and perhaps even extirpated from) the subalpine communities on Mauna Kea (VanderWerf 2008), and are primarily found in koa-‘ōhi‘a forests (Pejchar 2005).

Palila (Loxioides bailleui) are seed-eating finches with stout beaks and a yellow head and breast (Hawaii Audubon Society 1997). The Palila is one of three remaining seed-eating honeycreepers in the Hawaiian archipelago, and the only one left on the main islands—the other two species are limited to the Northwestern Hawaiian Islands of Nihoa and Laysan (Banko 2006). They are also the only remaining species of Hawaiian bird that relies solely on dry forest for habitat (Pratt et al. 1997). Palila feed on the green seedpods of māmane trees, eating the seeds inside and preying on caterpillars of Cydia and other moth species that also feed on the seeds. Palila also eat naio fruits as well as māmane flowers, buds, and young leaves (Hawaii Audubon Society 1997; Banko 2006). Palila were once common in lowland dry forests on several of the Hawaiian Islands, but due to habitat alteration, first by humans, and subsequently by grazing mammals, the Palila’s range has decreased to a small band around Mauna Kea, in the last remaining stands of māmane woodlands. Most Palila are found in the southwestern portion of the mountain (Banko 2006). Given their reliance on māmane, the main threat to current Palila populations is habitat degradation and loss, caused by grazing of māmane seedlings by non-native mammals; smothering by invasive plant species (such as grasses); increased frequency and intensity of fires (fueled by invasive grasses); and development (Banko 2006). Availability of māmane seeds is an important limiting factor, and Palila may not breed during drought years when fewer māmane seedpods are produced (Pratt et al. 1997). Predation by non-native mammals is also a threat to Palila, although predators are not as abundant in the subalpine zone on Mauna Kea as they are in lowland areas. Predation rates (of all birds) may be higher around developed areas, such as Hale Pōhaku, where human activities provide an addition source of food and water for predators. Invasive parasitoid wasps are also thought to impact the moth species upon whose caterpillars Palila feed, thus reducing an important food source for Palila adults and chicks (Brenner et al. 2002; Oboyski et al. 2004). An additional threat to the Palila is the presence of avian malaria at lower elevations. Even if māmane woodlands were restored in lowland areas, Palila would most likely remain limited to high elevation sites, due to their lack of resistance to avian diseases spread by invasive mosquitoes. Mosquitoes are rare or absent at high elevation sites on Mauna Kea (Banko 2006). Palila were recognized as endangered in 1966 and were included in the original Endangered Species Act, in 1973 (Banko 2006). Hale Pōhaku falls within the critical habitat of the Palila (Figure 2.2- 23), which extends to 10,000 ft on Mauna Kea (Stemmerman 1979). Habitat restoration, translocation of Palila to suitable habitats that are currently unoccupied, and captive breeding are all part of the management activities currently directed at helping bolster Palila populations and keeping the species from going extinct (Banko 2006).

‘Io (Buteo solitarius), or Hawaiian hawk, are territorial, monogamous raptors that feeds on birds, mammals, insects, and spiders (Scott et al. 1986; Mitchell et al. 2005b). They occur from sea level to approximately 8,500 ft (2,600 m) on the Island of Hawai‘i and are known to utilize a broad range of forest habitats. ‘Io avoid unforested areas and are most abundant in native forests (Klavitter et al. 2003). They have been observed in subalpine māmane-naio woodlands in the past (Scott et al. 1986), but recent survey work suggests that ‘Io do not utilize māmane-naio forests much, if at all (Klavitter et al. 2003). There is no evidence that avian malaria, introduced predators, or environmental contaminants are seriously affecting the ‘Io population (Griffin et al. 1998). Survey work indicates that ‘Io populations are stable, and the species may be a candidate for down-listing from Endangered to Threatened, or removal from the Endangered Species list altogether (Klavitter et al. 2003).

The Nēnē (Branta sandvicensis) is the only remaining species of goose in the Hawaiian Islands from the seven or more species that existed prior to the arrival of Polynesians (Olson and James 1982). Nēnē historically inhabited grasslands, grassy shrublands, and dryland forest, from sea level to the subalpine and alpine zones (USFWS 2004). They likely inhabited high-elevation sites such as the māmane woodlands in the subalpine zone on Mauna Kea during the non-breeding season (USFWS 2004). Nēnē feed on leaves, buds, flowers and seeds of grasses and herbs, and the fruits of ‘ōhelo (Vaccinium reticulatum), ‘aiakenēnē (Coprosma ernodeoides), and other plants (Scott et al. 1986). Nēnē are ground nesting birds and their numbers have been greatly reduced by non-native mammalian predators (USFWS 2004). Nēnē populations are small and are currently sustained by a captive breeding program. They may suffer from inbreeding depression (USFWS 2004). Their present distribution reflects locations of release sites of captive-bred birds (Banko et al. 1999). On the Island of Hawai‘i, Nēnē are currently found in a number of areas from sea level to 7,900 ft (2,400 m). Population centers of Nēnē in the wild include: Hakalau Forest National Wildlife Refuge; Hawai‘i Volcanoes National Park; Kahuku; Keauhou area including Kulani; Kipuka ‘Ainahou area including Pu‘u ‘O‘o Ranch and Pu‘u 6677; Pohakuloa area including Saddle Road, and the Pu‘uwa‘awa‘a area including Pua Lani and Pu‘uanahulu (USFWS 2004). Nēnē were not observed at Hale Pōhaku during the 1979 and 1985 bird surveys, and no evidence suggests they are currently using the area (Stemmerman 1979; Stemmerman Kjargaard 1985).

2.2.3.1.2      Candidate Species and Species of Concern

There are no federal Candidate species of birds found at Hale Pōhaku. Federal Species of Concern found at Hale Pōhaku include the Hawai‘i ‘Amakihi (Hemignathus virens), ‘Apapane (Himatione sanguinea) and ‘I‘iwi (Vestiaria coccinea). Other federal Species of Concern that may occur at Hale Pōhaku (but have not been recorded there) include the Hawai‘i ‘Elepaio (Chasiempis sandwichensis sandwichensis), Kolea (Pluvialis fulva), and Pueo (Asio flammeus sandwichensis).

‘I‘iwi (Vestiaria coccinea), are bright vermillion (red with a touch of orange) honeycreepers with a long, strongly curved salmon-colored bill and black wings, and have a squeaky call that sounds like “a rusty hinge” (Hawaii Audubon Society 1997). ‘I‘iwi wings produce a distinctive whirring noise in flight (Fancy and Ralph 1998). They feed primarily on nectar and secondarily on insects (especially butterflies and moths). They were once one of the most common forest birds in the islands, present in forests from sea level to the tree line (Fancy and Ralph 1998). ‘I‘iwi are thought to be monogamous during the breeding season, and will defend small breeding territories. The breeding season coincides with peak ‘ōhi‘a flowering, with most breeding occurring between February and June (Fancy and Ralph 1998). During the non-breeding season, they can be found foraging in flocks, or may defend a territory in areas of intermediate flower density (Fancy and Ralph 1998). ‘I‘iwi abundance in subalpine forests is tied to nectar availability, as measured by māmane flower abundance (Hess et al. 2001). Hess et al. (2001) found that while there is a small resident population of ‘I‘iwi in the subalpine māmane woodlands most ‘I‘iwi move between māmane woodlands and their primary habitats, mesic to wet koa and ‘ōhi‘a forests. ‘I‘iwi are mostly likely uncommon visitors to Hale Pōhaku, and are most likely to be observed there while māmane are flowering (VanderWerf 2008). ‘I‘iwi are highly susceptible to avian malaria (with mortality rates over 90%) and viable populations of these birds persist only in high elevation areas where mosquitoes are rare or absent. Japanese white-eyes compete with ‘I‘iwi for food, and studies have found a negative relationship between the abundance of ‘I’iwi and Japanese white-eyes (Fancy and Ralph 1998). In addition, non-native mammalian predators (rats and cats) are though to impact ‘I‘iwi populations.

‘Apapane (Himatione sanguinea) are bright crimson honeycreepers with black wings and tail. They have a long decurved bluish-black bill, and feed primarily on nectar, but also take insects and spiders (Fancy and Ralph 1997). Like the ‘I‘iwi, ‘Apapane make a whirring noise during flight. ‘Apapane breed in mesic and wet ‘ōhi‘a forests. They make seasonal and daily movements from wet forest to subalpine woodland and leeward dry woodlands when nectar is available there, mainly in September through November (Fancy and Ralph 1997; Hess et al. 2001). In the breeding season, ‘Apapane maintain small breeding territories, are monogamous, and lay clutches of one to four eggs (three on average). Breeding activity begins October-November and peaks in February through June. During the non-breeding season, ‘Apapane forage together in small flocks, or in mixed flocks with other species of honeycreeper (Fancy and Ralph 1997). ‘Apapane are susceptible to avian pox and malaria, and have the highest prevalence of malaria (Plasmodium) parasites of any native or alien bird species in the Hawaiian Islands (van Riper et al. 1986; Fancy and Ralph 1997). However, some believe that ‘Apapane may be developing an immunity to malaria (Atkinson et al. 1995). Birds that survived initial malaria infections seemed to be resistant to new infections (Yorinks and Atkinson 2000). Other factors that likely impact ‘Apapane populations are habitat loss and degradation (including habitat alteration by invasive plants) and predation by non-native mammalian predators. It is unknown whether competition with non-native birds and insects is affecting ‘Apapane populations (Fancy and Ralph 1997).

Hawai‘i ‘Amakihi (Hemignathus virens virens) are yellowish-green honeycreepers with a thin, slightly decurved beak that feed on insects, nectar, and fruit (Hawaii Audubon Society 1997). They are the most common native birds remaining, ranging from 2,100 ft to 9,840 ft (650 m to 3,000 m) elevation, and have a strong association with dry and mesic forests, including māmane and māmane-naio woodlands (Kern and van Riper 1984; Scott et al. 1986). ‘Amakihi in subalpine woodlands nest primarily in māmane trees and choose trees that are taller than average (Kern and van Riper 1984). Because of their varied diets, ‘Amakihi populations in subalpine woodlands do not fluctuate as greatly on a seasonal (or daily) basis as do ‘Apapane and ‘I‘iwi populations (Hess et al. 2001). However, ‘Amakihi are highly dependent on nectar availability, especially during the breeding season, and will not breed in areas that do not have sufficient densities of māmane flowers (van Riper 1984). ‘Amakihi retain mates for more than one  season, are territorial, and breed from November through July, with the most nesting occurring in March through May (van Riper 1987). Generally two to three eggs are laid during a breeding attempt. Overall reproductive success of the ‘Amakihi is average for open-nesting passerines (around 35%), with the greatest causes of failure being nest desertion and failure of the eggs to hatch (van Riper 1987). High survival rates and relatively long life of adult birds (van Riper 1987) may aid in population stability. Hawai‘i ‘Amakihi are susceptible to avian malaria, and low elevation populations have fairly high infection rates (Woodworth et al. 2005). Despite this, ‘Amakihi populations have recently been increasing in lowland areas on Hawai‘i. It is currently unknown whether these populations are being supplemented by movement of ‘Amakihi from higher elevation populations, or if the ‘Amakihi is developing some level of resistance to avian malaria and is thus able to reproduce and maintain stable populations despite high infection rates (Woodworth et al. 2005).

‘Elepaio (Chasiempis sandwichensis) are insectivores that often catch their prey in the air (Conant 1977). They were once very abundant in forested areas of O‘ahu, Kaua‘i, and Hawai‘i, and are still widespread, but not abundant, in many areas. They are found primarily in koa-‘ōhi‘a forests. The Hawai‘i ‘Elepaio (Chasiempis sandwichensis sandwichensis) is a federal Species of Concern. Some authorities recognize the Mauna Kea subalpine ‘Elepaio as a separate subspecies, Chasiempis sandwichensis bryani (Pratt 1980). Habitat degradation has reduced their densities in māmane woodlands, and they are most abundant in this habitat type on the southwestern slope of Mauna Kea, with highest densities near Pu‘u La‘au (Scott et al. 1986). In subalpine environments, ‘Elepaio nest primarily in māmane trees, preferring taller trees than average (van Riper 1995). Nest predation by feral cats and rats is less common in ‘Elepaio nesting on Mauna Kea than in ‘Elepaio that nest in other habitats, due to the low density of predators in this habitat type (Amarasekare 1993; van Riper 1995). ‘Elepaio are territorial and monogamous, and stay in their territories year-round (van Riper 1995). Nesting occurs from February to August (van Riper 1995). C. van Riper (1995) found that the Mauna Kea ‘Elepaio eggs had unusually high hatching failure (25% vs. around 7% for other passerines), which may limit their productivity. Even so, overall  reproductive success was fairly high, with 80% of nests fledging at least one young (van Riper 1995), suggesting that perhaps ‘Elepaio populations in subalpine māmane forest on Mauna Kea may be limited primarily by lack of adequate habitat. ‘Elepaio are also negatively affected by the presence of invasive birds such as the Japanese white-eye (Mountainspring and Scott 1985).

Kolea, or Pacific Golden Plovers (Pluvialis fulva), are migratory shorebirds that spend the winter in Hawai‘i and the summer in the arctic, where they breed. Generally Kolea arrive in August or September and leave by early May (Hawaii Audubon Society 1997). While they are in Hawai‘i they maintain foraging territories, which most birds return to year after year (Johnson et al. 2001). Some non-breeding birds will stay for the summer. Kolea forage on lawns, fields, and grassy mountain slopes for invertebrates and occasionally eat leaves and flowers (Hawaii Audubon Society 1997). They are found up to approximately 10,000 ft (3,048 m) elevation, and utilize open areas in the subalpine zone on Mauna Kea (Hawaii Audubon Society 1997; Pratt 2008).

Pueo, or Hawaiian Owls (Asio flammeus sandwichensis), are ground-nesting owls found on all the major Hawaiian Islands, in shrublands, grasslands, and montane parklands (Scott et al. 1986). Pueo hunt at  dawn and dusk (and sometimes during the day) and feed on small mammals (mostly rodents), birds (native and non-native), and insects (Snetsinger et al. 1994; Hawaii Audubon Society 1997). Breeding occurs throughout the year and three to six eggs are laid (Snetsinger et al. 1994). Because Pueo build their nests on the ground, they are extremely susceptible to predation by non-native mammals such as cats and mongoose, and habitat alteration (development, agriculture). Non-native barn owls and feral cats may also compete with Pueo for food (primarily small rodents and birds). Pueo nests have been observed at 9,022 ft (2,750 m) elevation on eastern Mauna Kea, in māmane woodlands at Kanakaleonui and above Pu‘u La‘au Cabin, on the western slope, at the bases of māmane trees (Snetsinger 1995).

2.2.3.1.3      Invasive Species

Black Francolin (Francolinus francolinus), Erckel’s Francolin (Francolinus erckelii), Chukar (Alectoris chukar), Japanese Quail (Coturnix japonica), Ring-necked Pheasant (Phasianus colchicus), Wild Turkey (Meleagris gallopavo), and California Quail (Callipepla californica) are all game birds that were introduced and managed for hunting in grasslands, shrublands, and open woodlands. Most of the game birds are generalists and feed on plants, invertebrates (especially insects), fruits, and seeds. The impacts  of these non-native birds are both positive and negative. On the positive side, Cole et al. (1995) found that Chukar and Ring-necked Pheasants on Haleakalā, Maui were at least partially filling the ecological role of extinct and rare native birds (such as the Nēnē) as the primary dispersers of seeds of native plants such as pūkiawe (Leptecophylla tameiameiae), ‘ōhelo (Vaccinium reticulatum), nohoanu (Geranium cuneatum), ‘aiakenēnē (Coprosma ernodeoides), pilo (Coprosma montana), and a native sedge, Carex wahuensis. All these species are found at Hale Pōhaku or in the subalpine zone on Mauna Kea. Pūkiawe seeds are notoriously difficult to germinate without treatment, yet those found in game bird droppings had high germination rates. This suggests that these birds may play an important role in maintaining pūkiawe populations in upland areas (Cole et al. 1995b). Although māmane seeds were eaten by the introduced game birds in the Haleakalā study, there were no māmane seedlings germinated from game bird droppings, suggesting that the birds do not aid in the regeneration of māmane through seed dispersal, and in fact, may reduce māmane regeneration if enough seeds are consumed. In addition, invasive plant parts in Chukar and Ring-necked Pheasant diets consisted mainly of flowers and leaves rather than fruits and seeds. Arthropods (primarily non-native species such as ladybugs) made up a relatively small portion of the game bird diets. On the negative side, these birds did disperse some seeds of invasive species, including common velvetgrass (Holcus lanatus), hairy cats-ear (Hypochoeris radicata), mouse ear chickweed (Cerastium vulgatum), common catchfly (Silene gallica), and common evening primrose (Oenothera biennis). All these plant species, or closely related ones, are found at Hale Pōhaku. However, native seed germinations from Chukar and Ring-necked pheasant droppings outnumbered invasive species five to one in the Haleakalā study (Cole et al. 1995a). Introduced game birds may well be spreading both native and invasive species at Hale Pōhaku, and the extent of their impacts there is unknown. Studies conducted in other locations have found that non-native birds are often the vectors of invasive plant seeds (Vitousek and Walker 1989; Garrison 2003; Chimera 2004). For example, another game bird, the Kalij Pheasant (Lophura leucomelana), which is not present at Hale Pōhaku, is a major disperser of banana poka (Passiflora molissima), an invasive weed in wetter forests on Hawai‘i (Lewin and Lewin 1984). See Section 2.2.1.3.4 for more information of the spread of invasive plants by non- native animals.

Japanese White-eyes (Zosterops japonicus) were introduced to Hawai‘i in 1929, and are now the most abundant land birds in the Hawaiian Islands. They occur in almost every habitat type, up to 10,170 ft (3,100 m) on Hawai‘i (Scott et al. 1986). They are most abundant at forest edges and open forests, and prefer habitats with invasive species in the understory (Scott et al. 1986). They consume nectar, fruit, and invertebrates and may compete directly with most of the native honeycreepers found in māmane woodlands. ‘Elepaio, ‘Amakihi and ‘I‘iwi abundances are negatively related to abundance of Japanese White-eyes in Hawai‘i (Mountainspring and Scott 1985). Japanese White-eyes are present at Hale Pōhaku, although their current status is unknown.

Red-billed Leiothrix or Pekin Nightingale (Leiothrix lutea) feed primarily on fruits, but also eat arthropods and seeds. They are found in low densities throughout Mauna Kea, reaching high densities only in denser woodlands with naio or water sources, up to 9,500 ft (2,900 m) (Scott et al. 1986). Their spread above 9,500 ft is probably limited mainly by the availability of water. They are generally not  found in high densities in māmane forest without naio, and in general, tend to be more common in areas with a higher abundance of fruit. Red-billed Leiothrix were observed at Hale Pōhaku in 1979 but not in 1985 (Stemmerman 1979; Stemmerman Kjargaard 1985). Their current status at Hale Pōhaku is unknown. However, an OMKM Ranger, Pablo McCloud, found a dead Red-billed Leiothrix in the summit area of MKSR in November 2006, indicating that they are still found at lower elevations in the region (Nagata 2007).

House Finches (Carpodacus mexicanus) are omnivorous, but feed extensively on seeds, buds, fruits  (Scott et al. 1986). They are common in developed and agricultural areas, high elevation ranchlands, disturbed wet forest, and māmane-naio forest. They are found in low densities at Hale Pōhaku and reach greatest numbers in naio woodlands and areas with available water. The highest House Finch densities on Mauna Kea are associated with water seeps (Scott et al. 1986). They chiefly inhabit forest edges, pastures, open woodland and scrub. House Finch nest in māmane and naio trees in subalpine woodlands, and will also nest in introduced tree species (van Riper 1976). They may actively disperse invasive plant seeds, but also eat fruits of pūkiawe (Leptecophylla tameiameiae), ‘aiakenēnē (Coprosma ernodeoides), pilo (Coprosma montana), ‘ōhelo (Vaccinium reticulatum), and naio (Myoporum sandwicense) (Scott et al. 1986). House finches are found at Hale Pōhaku, primarily in the developed areas (Stemmerman 1979; Stemmerman Kjargaard 1985).

House Sparrows (Passer domesticus) are omnivorous, and will eat almost anything edible (Scott et al. 1986). They are common on all islands especially in urban and agricultural areas. House sparrows are almost always found in association with human disturbance (houses, buildings, ranches paddocks, feedlots, campgrounds), and at Hale Pōhaku they are associated with buildings such as the Visitors Information Station (Stemmerman 1979; Stemmerman Kjargaard 1985).

Northern Cardinals (Cardinalis cardinalis) eat seeds, fruits, and arthropods. They are widely dispersed (primarily in introduced and disturbed native forests) but are not as abundant as Japanese White-eyes. They are limited to the western and southwestern portions of Mauna Kea, and do not occur in māmane forests that have bare or open understories. They seem to prefer forests with introduced shrub and

20 Passerines are birds in the order Passeriformes, usually called perching birds or songbirds. This order contains more than one- half of all bird species.

Eurasian Skylarks (Alauda arvensis) are insectivores that inhabit dry scrub, savanna, and woodlands, primarily degraded, fragmented and deforested habitats (Scott et al. 1986; Hawaii Audubon Society 1997). Skylarks primarily forage in grass and on the ground. They were observed at Hale Pōhaku during the last bird survey there, in 1985 (Stemmerman Kjargaard 1985). Their current status at Hale Pōhaku is unknown.

Melodious Laughing-thrush, or Hwamei (Garrulax canorus), eat arthropods and fruits (Mountainspring and Scott 1985). On Mauna Kea they are restricted primarily to woodlands with naio (whose fruit they eat), from sea level to 9,500 ft (2,900 m), and are more common at low elevations. They seem to prefer brushy understories with structural and floristic diversity (Scott et al. 1986). They were not recorded as occurring at Hale Pōhaku during the 1979 and 1985 surveys (Stemmerman 1979; Stemmerman Kjargaard 1985).

Northern Mockingbirds (Mimus polyglottos) are primarily insectivores, but also feed on fruits and seeds (Scott et al. 1986; Chimera 2004). They are well established in dry areas on Hawai‘i, and occupy a wide range of elevations and vegetation types, including in naio forest and māmane woodlands. The Mauna Kea population was established after 1978 (Scott et al. 1986). They were not recorded as occurring at Hale Pōhaku during the 1979 and 1985 surveys (Stemmerman 1979; Stemmerman Kjargaard 1985), and their current status at Hale Pōhaku is unknown.

Common Mynas (Acridotheres tristis) eat insects (Hawaii Audubon Society 1997), and are found in association with forest edges, pastures, and disturbed areas, up to 7,550 ft (2,300 m). They are cavity nesters (Scott et al. 1986). They were not recorded at Hale Pōhaku during the 1979 and 1985 surveys (Stemmerman 1979; Stemmerman Kjargaard 1985), and it seems unlikely that they will become established at Hale Pōhaku unless they expand their elevational range.

Warbling silverbill (Lonchura malabarica) feed almost exclusively on seeds. They were first discovered in Hawai‘i in 1972, are common in coastal mesquite woodlands, and range up to 10,170 ft (3,100 m) on Mauna Kea (Scott et al. 1986). In their native Africa they are found in dry savannas, thorn-scrub, grasslands, and desert areas near water. Their habitat in Hawai‘i is similar (Scott et al. 1986). They were not recorded as occurring at Hale Pōhaku during the 1979 and 1985 surveys (Stemmerman 1979; Stemmerman Kjargaard 1985), but it seems likely they could inhabit the area.

Nutmeg Mannikin (Lonchura punctulata), also known as Ricebirds or Spotted Munia, are very small seed eating estrildid finches, and can often be seen perched on a sturdy stalk of grass. They are highly nomadic, and occupy very open or disturbed sites and forest edges. Nutmeg Mannikins are mainly associated with introduced trees in low elevation areas, but can occur up to 9,500 ft (2,900 m) on Mauna Kea (Scott et al. 1986). They often travel in large flocks. They were not recorded at Hale Pōhaku during the 1979 and 1985 surveys (Stemmerman 1979; Stemmerman Kjargaard 1985), and their current status there is unknown. However it is likely that Nutmeg Mannikins are found at Hale Pōhaku.

Yellow-fronted Canary (Serinus mozambicus) forage in the grass and on the ground for seeds and insects (Hawaii Audubon Society 1997). They were first reported in 1964 on O‘ahu and 1977 on Hawai‘i, on the upper slopes of Mauna Kea (van Riper 1978; Scott et al. 1986). Yellow-fronted Canaries are common in low elevation forests, but can go as high as 9,200 ft (2,800 m) in māmane woodlands. They prefer open, dry parkland habitat and open woodlands (Hawaii Audubon Society 1997). Their native habitat in Africa is lightly wooded country, savanna, brush and cultivated areas (Scott et al. 1986). Yellow-fronted canaries were observed at Hale Pōhaku in 1977, but were not recorded in the 1979 and 1985 bird surveys (van Riper 1978; Stemmerman 1979; Stemmerman Kjargaard 1985). Their current status at Hale Pōhaku is unknown.

In addition to competing with native birds for food, invasive birds can also act as a source and reservoir of avian diseases, such as avian malaria and pox, and act as a food source for invasive mammalian predators, which also prey on native species. Section 2.2.3.3.2 has more information on avian diseases and non- native predators.

2.2.3.1.4      Bird Surveys at Hale Pōhaku

In 1975, van Riper et al. conducted a comprehensive census of Palila habitat (māmane and naio forests in the subalpine zone on Mauna Kea) to determine the distribution of Palila on the mountain (van Riper et al. 1978). This survey area included Hale Pōhaku. Although no specific numbers are mentioned for Hale Pōhaku, Palila did occur in the region and were found primarily near the tree line, in large trees (van Riper et al. 1978).

In 1979, prior to construction of the mid-elevation facilities, Maile Stemmerman conducted bird surveys in three areas: Hale Pōhaku, Hawaiian Homes Commission Lands, and Humu‘ula Sheep Station (Stemmerman 1979). Of the three areas, Stemmerman found that the undeveloped areas of Hale Pōhaku presented some of the best bird habitat, primarily because of the relatively intact nature of the māmane woodlands. Both Hawaiian Homes Commission Lands and the Humu‘ula Sheep Station were considered marginal bird habitat. Eight species of birds were observed during the Hale Pōhaku survey: two native species (‘Apapane and ‘Amakihi) and six non-native species (House Finch, House Sparrow, Japanese White-eye, Red-billed Leiothrix, Erckel’s Francolin, and California Quail). The House Finch and House Sparrow were primarily associated with the developed areas. Stemmerman speculated that Hale Pōhaku habitat may be suitable for Pueo (Asio flammeus sandwichensis), ‘Io (Buteo solitarius) and Hawaiian Petrel (Pterodroma sandwichensis), but did not observe these species. However, Stemmerman noted that known nesting sites for the Hawaiian Petrel were considerably higher on Mauna Kea, and it seemed unlikely this species would use the area for nesting. Stemmerman also noted that while Nēnē (Branta sandvicensis) are known to roost on Mauna Kea, none had been observed at Hale Pōhaku. Although Stemmerman did not observe ‘I‘iwi (Vestiaria coccinea) and Palila (Loxioides bailleui) at Hale Pōhaku, she states they most likely use the area when the māmane trees are flowering (‘I‘iwi) or have green seed pods (Palila), and that Palila had been observed at Hale Pōhaku regularly over the previous five years. Finally, Stemmerman commented that although Chukar (Alectoris chukar) were not observed during the survey, they are most likely at Hale Pōhaku.

In May 1985, Maile Stemmerman Kjargaard conducted an additional bird survey at Hale Pōhaku (Stemmerman Kjargaard 1985). Species observed during this survey included ‘Apapane, ‘Amakihi, House Finch, House Sparrow, California Quail, Japanese White-eye, and Skylark.

In addition to the above bird surveys, several observations of species at Hale Pōhaku have been recorded in the literature, or in field records by experienced birders. In 1977, Mae Mull and associates observed three Palila foraging on māmane seed pods in māmane trees “close to the restrooms and the United Kingdom Dormitories” (Mull 1977). Although this was not part of a formal bird survey of Hale Pōhaku, Mull was able to observe the birds up close for more than half an hour. In 1978 C. van Riper III observed Yellow-fronted Canary (Serinus mozambicus) at Hale Pōhaku, as well at as other locations in māmane forests on Mauna Kea (van Riper 1978). Eric VanderWerf observed ‘I‘iwi at Hale Pōhaku in 2006, and describes them as seasonal/uncommon visitors to the area (VanderWerf 2008).

2.2.3.2       Alpine Bird Communities (MKSR and Upper Summit Access Road)

The harsh conditions in the alpine zone on Mauna Kea make it difficult for most vertebrate species to make a living there. No birds are known to currently inhabit or regularly use the summit area or the alpine shrubland and grasslands. An occasional bird may be observed flying through the area, and sometimes birds are blown up the mountain during strong winds and die there (Munro 1945; Montgomery and Howarth 1980). Several mummified Red-billed Leiothrix have been found at or near the summit, one as recently as November 2006 (Montgomery and Howarth 1980; Nagata 2007).

2.2.3.2.1      Threatened and Endangered Species, Candidate Species and Species of Concern

There are no federal Threatened Species, Candidate Species, or Species of Concern known to inhabit the alpine community on Mauna Kea. There is one federal Endangered species, the Hawaiian Petrel (Pterodroma sandwichensis)²¹ or ‘Ua‘u (Banks et al. 2002), which may have historically utilized lower portions of the alpine zone on Mauna Kea.

The ‘Ua‘u is a pelagic seabird that historically nested in the mountains of all main Hawaiian Islands (Conant et al. 2004). ‘Ua‘u nest in underground burrows and feed at sea. Prior to human contact the ‘Ua‘u was widely distributed from sea level to at the least mid-elevations on all the main islands (Hu et al. 2001). On the Island of Hawai‘i, it was once abundant on the saddle area between Mauna Loa and Mauna Kea (Conant et al. 2004). A breeding colony of ‘Ua‘u is known from Hawai‘i Volcanoes National Park from 8,000 ft (2,440 m) to 9,200 ft (2,800 m) elevation (Hu et al. 2001). In 1954, Richardson and Woodside found five freshly dug ‘Ua‘u burrows at Pu‘u Kole, east of Hale Pōhaku, and in the 1960s and 1970s there were observations of ‘Ua‘u from Pu‘u Kole around the eastern flank of Mauna Kea to Pu‘u Kanakaleonui (Kjargaard 1988). Currently they are thought to be located on Mauna Loa along the summit trail, and on Mauna Kea above 9,850 ft (3,200 m) near Pu‘u Kanakaleonui (NASA 2005). Kjargaard (1988) notes that skeletal remains of ‘Ua‘u were found on the Mauna Kea at elevations up to 12,400 ft (3,780 m), possibly indicating presence of the birds in the alpine zone. However, Conant et al. (2004) point out that Hawaiian petrels were used as food by the ancient Hawaiians, and the presence of the bones at these high elevations could represent either petrel activity or the remains of an ancient Hawaiian meal. No ‘Ua‘u were observed during bird surveys conducted (in a rather limited area) on the summit of Mauna Kea in 1988 (Kjargaard 1988).

Modern day threats to the ‘Ua‘u include predation by non-native mammals, especially feral cats, trampling of colonies by feral ungulates such as sheep, and possibly avian malaria and pox (Hu et al. 2001; Day et al. 2003).

 

21 Although it is listed as the Dark-rumped Petrel (Pterodroma phaeopygia sandwichensis) under the Endangered Species Act, this species has recently undergone a name change and is referred to as the Hawaiian Petrel (Pterodroma sandwichensis) in recent literature.

Other than the mummified remains of several Red-billed leiothrix found near Lake Waiau and at the summit, no invasive bird species have been found at or near the summit (Montgomery and Howarth 1980; Nagata 2007).

2.2.3.2.3      Bird Surveys at MKSR

Very few bird surveys have been conducted in the MKSR. In 1988, Maile Kjargaard conducted a vertebrate (bird and mammal) survey at two proposed sites for the VLBA antenna, located between 11,800 ft and 12,400 ft (3,600 and 3,780 m) elevation (Kjargaard 1988). Both diurnal and nocturnal observations were made. The primary goal of the survey was to look for signs of Hawaiian Petrels. No birds or evidence of petrel burrows were observed.

There are no other records of bird surveys for the MKSR. No formal bird surveys have been conducted in the Ice Age NAR, however opportunistic sightings of birds are recorded in their GIS database (Hadway 2008).

2.2.3.3       Threats to Bird Communities on Mauna Kea

Because native birds are mainly found below the treeline on Mauna Kea, the following discussion of threats to bird communities on Mauna Kea will focus primarily on the subalpine zone, and in particular, on māmane woodlands.

2.2.3.3.1      Habitat Alteration

Habitat alteration is one of the primary causes of extinction of native birds in Hawai‘i, and remains one of the biggest threats to the survival of the remaining native species. In the subalpine forests of Mauna Kea, habitat alteration is primarily responsible for the current endangered status of the Palila (Loxioides bailleui), and the reduced population sizes of several other Hawaiian honeycreepers found there. Habitat alteration has occurred through the activities of man (e.g. clearing of land for ranching, and limited development); grazing by introduced ungulates on māmane seedlings, saplings, and mature trees (thus preventing forest regeneration); and invasion by non-native weeds and grasses (which compete with native plants for resources, smother native seedlings, and increase the risk of fire). Although time- consuming and expensive, it is feasible to reduce the threat of habitat loss to native birds in the subalpine forests on Mauna Kea through combined efforts of fencing, ungulate extirpation, and controlling invasive grasses. Spreading seeds or planting seedlings of native species would help speed the process of recovery.

2.2.3.3.2      Invasive Species

Invasive species can affect native bird populations through habitat alteration, competition, predation, and disease transmission. Native birds are threatened by a variety of invasive species, including plants, microbes, invertebrates, and vertebrates. While each of these groups damaged native bird populations outright, it is likely the combination of these threats, working together, that is driving current native bird populations toward extinction. Land managers face the challenge of addressing these multiple threats in an integrated manner, rather than piecemeal, to successfully protect native species. The piecemeal approach can lead to serious negative outcomes. For example, the removal of feral ungulates on Sarigan Island (Commonwealth of the Northern Mariana Islands), while allowing for native forest recovery, also instigated the rapid expansion of an invasive vine that had previously not been a problem. This vine now covers much of the native forest (Kessler 2002). This is not an isolated example. As Zavaleta (2002) points out, “the most common secondary outcome of a single-species eradication is the ecological release of a second (plant or prey) exotic species previously controlled by the removed species (herbivore or predator) through top-down regulation.” On Mauna Kea, removal of feral pigs (Sus scrofa) and sheep (Ovis aries) from exclosures allowed for some native plant regeneration but also increased competition between invasive and native plants (Scowcroft and Conrad 1992). Eradications of feral cats (Felis catus) on small islands in New Zealand have led to increased populations of introduced rats, and removal of the rats possibly lead to an irruption of crazy ants (Anoplolepis gracilipes) on Bird Island in the Seychelles (Zavaleta 2002). While these interactions between invasive species make management more difficult, being aware of them, planning accordingly, and using adaptive management techniques can help overcome these obstacles.

Invertebrates and Disease Organisms

Invasive invertebrates that can affect native bird populations include parasitic worms (Phyla Platyhelminthes, Acanthocephala, and Nematoda); parasitic and blood feeding species such as mosquitoes, mites, fleas, and flies; and nectarivorous and insectivorous species that compete with birds for food, such as honeybees, yellowjackets and ants (Loope et al. 2001). Parasitic and blood feeding species (such as mosquitoes) not only affect the host through the taking of blood or flesh, but also by spreading diseases (Loope et al. 2001).

Currently there are two avian diseases that are impacting native bird populations: avian poxvirus (Poxvirus avium) and avian malaria (Plasmodium relictum) (Freed 2005). Avian pox is a virus that causes skin lesions, and in more serious cases necrotic lesions in mucous membranes of the mouth and upper respiratory tract. In most cases avian pox does not kill the bird, although mortality rates are higher in birds that develop lesions in the oral or respiratory cavities (van Riper et al. 2002). Pox can be transmitted via biting insects (mosquitoes, midges, flies) or directly through contact with infected birds (or contact with a perch or other material touched by an infected bird). Native birds are more susceptible to avian pox than introduced birds, and avian pox is more common in mesic than dry forests (van Riper et al. 2002). High- elevation dry forests such as māmane woodlands may provide native birds a refuge from the avian pox virus.

Avian malaria is a disease caused by protozoan in the genus Plasmodium. Malaria cannot be transmitted directly between birds and requires a vector (mosquitoes) to move between hosts. The parasite uses the mosquito to reproduce and its offspring then infect a new bird host when it is bitten by the mosquito. Avian malaria was not present in Hawai‘i until mosquitoes were introduced in 1827 (Freed 2005). Native forest birds are extremely susceptible to infection with P. relictum, and in lab experiments, 65–90% of birds die after being bitten by a single infective mosquito (Woodworth et al. 2005). According to Woodworth et al., the effects of avian malaria include “severe anemia, the destruction of mature erythrocytes, declines in food consumption, and activity levels, and loss of up to 30% of body weight. Individuals that survive acute infection develop concomitant immunity to homologous strains of the parasite, but remain infective to mosquitoes, probably for life” (Woodworth et al. 2005). In Hawai‘i, malaria is spread mainly by the southern house mosquito (Culex quinquefasciatus), which is limited in elevation because of cold intolerance. However, recent evidence indicates that the mosquitoes are moving up the mountain, perhaps in response to a warming climate (Freed 2005). Native birds are more susceptible to avian malaria than are introduced birds, and the prevalence of avian malaria is higher in mesic and wet forests than dry forest (van Riper et al. 1986).

The vectors for avian malaria and pox are not found in the subalpine zone on Mauna Kea (Pratt et al. 1997), and avian malaria (P. relictum) has a threshold temperature of around 59° F (15°C), below which it is not transmitted to birds (Benning et al. 2002). However, birds such as ‘I‘iwi and ‘Apapane that frequently travel between lower elevation forests and the subalpine zone can be infected while in the lower elevation habitats. Protection and restoration of high elevation forests, including māmane woodlands, may allow individuals of these species to persist without being exposed to malaria, and in the face of global warming may provide the only disease free habitat for forest birds (Benning et al. 2002).

Invasive invertebrates with the potential to impact native bird populations include honeybees, yellowjackets, parasitoid wasps and ants. The latter three could impact bird populations by reducing native arthropod populations upon which the birds feed. See Section 2.2.2.1.2. Honeybees, and some ant species, may compete with native birds for nectar (Hansen et al. 2002; Traveset and Richardson 2006). Honeybees are present up to the treeline, but pollinator interactions have not been studied in māmane forests (Oboyski 2008).

Invasive Plants

Invasive plants such as grasses and vines can impact native bird populations on Mauna Kea through displacement of native subalpine forest and shrublands (Banko et al. 2002). Invasive grasses and weeds can prevent forest recovery by smothering the seedlings of māmane and other native plants. Invasive grasses can also change the fire regime. See Section 2.2.1.3.2 for more information on the relationship between invasive grasses and fire regimes. A large wildfire in the māmane forest would seriously reduce available habitat for the endangered Palila (Banko et al. 2002).

Invasive Predators

Invasive predators such as cats, rats, barn owls, and mongoose have a direct impact on native bird populations. Cats and mongoose eat both adult birds and chicks, while rats primarily consume eggs (and sometimes chicks). Although rats, cats, and mongoose are not abundant in māmane woodlands, they still impact Palila populations (Banko et al. 2002). Although rats (Rattus rattus) are rare in māmane woodlands, they do depredate Palila nests, and Banko et al (2002) state that their impact is out of proportion to their numbers. Feral cats (Felis catus) are thought to be the most serious predator of Palila, particularly at their nests (Banko et al. 2002). Mongoose (Herpestes auropunctatus) are thought to have less of an impact on Palila, because they do not climb trees (Banko et al. 2002). However, mongooses, along with feral cats, have had a serious impact on ground nesting birds such as the Pueo and Nēnē. Barn Owls (Tyto alba) prey primarily on rodents, but do consume a small number of native birds and insects (Snetsinger et al. 1994). Their status in the māmane woodlands near Hale Pōhaku is unknown. Although mice (Mus musculus) are present in māmane woodlands, they do not appear to depredate Palila nests. They do, however, eat seeds and seedlings of native plants and can therefore indirectly impact native bird populations by changing plant communities. Because of their toxic seed coat, māmane seeds do not seem to be a preferred food of mice (Banko et al. 2002).

Predators such as feral cats and rats may be more abundant in developed areas such as around Hale Pōhaku because of increased availability of food and water (in refuse, landscaping, etc). A large number of visitors come through the Visitor Information Station and many of them eat lunch there. Any food left out or improperly disposed of (left in thopen) will no doubt be consumed by rats, cats, and non-native birds.

Invasive Birds

Non-native birds can compete directly with native birds for resources such as food. Japanese White-eyes are likely to compete directly with insectivorous and nectarivorous honeycreepers for limited resources in māmane woodlands. Non-native birds also can act as a food base for predators, which will take native birds as prey in addition to the non-natives.

There are several human uses at Mauna Kea that impact native bird species. Introduction and maintenance of populations of non-native mammals for hunting and ranching activities impact native bird species that utilize māmane forest (such as the Palila) through habitat degradation by grazing feral and domestic ungulates. Sheep, cattle, and goats damage māmane trees and prevent regeneration of the forest, while at the same time enhancing the spread and establishment of non-native plant species. Hunting and ranching do not occur at Hale Pōhaku, proper, but both occur close by. Because Hale Pōhaku is not fenced, feral ungulates may still use the site. Access to hunting and hiking areas via trails and roads passing through Hale Pōhaku by both vehicles and hikers can also lead to introduction of invasive species and erosion. Other human uses, such as tourism and scientific research also have impacts, such as introduction of invasive plants and animals, providing food sources to invasive arthropods, mammals and birds, and (limited) trampling of forest habitat. Improperly disposed food items and water used in landscaping and cleaning activities may help sustain larger populations of invasive species than would otherwise occur in the subalpine environment.

Because birds do not occupy the summit regions, human uses in the astronomy district will not directly impact bird populations. However, astronomy support facilities at mid-elevation areas do impact bird habitat through habitat loss, limited contamination (small spills associated with such activities as vehicle maintenance), unintentional provision of food and water for invasive species, and general wear and tear. At Hale Pōhaku these impacts are present, but they are generally limited in scope due to the small size of the developed area.

Human uses of all kinds also increase the risk of accidental fires. Sparks from catalytic converters, improperly discarded cigarettes and matches, camping fires, military exercises, and other activities can all cause wildfires. Fires are further discussed in Section 2.2.1.3.2.

2.2.3.3.4      Climate Change

Climate change could impact bird populations in several ways. If climate change affects plant communities then it could change availability of food and habitat to the native birds that utilize the māmane woodlands. Depending on the change, this could have both negative and positive effects on the bird community. A change in the plant community (especially an increase in density of fruiting plants) in response to increased rainfall, as predicted by some climate models (Hamilton 2007), could lead to an increase in abundance of non-native vertebrates and invertebrates, which may negatively effect native bird populations. On the other hand, an increased availability of insects, nectar, and fruit (seed pods) due to increased rainfall may also have a positive benefit for species such as the ‘I‘iwi, ‘Amakihi, ‘Apapane, and Palila. However, if climate change leads to increased drought conditions in the subalpine environment of Mauna Kea, as predicted by other climate researchers (Cao 2007; Cao et al. 2007; Giambelluca and Luke 2007; Giambelluca 2008), this will no doubt have a negative impact on most of the native bird species that utilize māmane woodlands. Palila populations have been observed to decline during periods of dry weather (such as El Niño events), when food resources were limited (Gray et al. 1999). See Section 2.2.1.3.6 for more information on potential effects of climate change on plant communities.

Warming temperatures may also allow invasive species, including disease vectors, to move to higher elevations on the mountain. This would have a devastating effect on native birds, many of which are currently surviving only in areas free of malaria and its vector, the mosquito.

2.2.3.4       Bird Community Information Gaps

2.2.4    Mammals

Hawai‘i has very few native species of mammals. Most of the native mammals found in the Hawaiian Islands are marine mammals. The only native land mammal in Hawai‘i is the ‘Ope‘ape‘a, or Hawaiian hoary bat (Lasiurus cinereus semotus). Conversely, Hawai‘i has many non-native species of animals that were brought to the islands by humans, beginning with the arrival of the first Polynesians. Some of these were accidental introductions, but most were purposeful, either for food, pets, or biological control.

2.2.4.1 Subalpine Mammal Communities (Hale Pōhaku and Lower Summit Access Road)

Mammals found in the subalpine zone on Mauna Kea include the Hawaiian hoary bat (Lasiurus cinereus semotus), feral cats (Felis catus), black rats (Rattus rattus), mice (Mus musculus and Mus domesticus), domesticated sheep (Ovis aries), mouflon sheep (Ovis musimon), feral sheep/mouflon sheep hybrids, goats (Capra hircus), cattle (Bos taurus), feral pigs (Sus scrofa), and mongoose (Herpestes auropunctatus). The bat is discussed in Section 2.2.4.1.1, and the remainder are discussed in Section 2.2.4.1.2. Table 2.2-8 lists mammal species known to occur in subalpine and alpine habitats on Mauna Kea, including Hale Pōhaku and the MKSR. Figure 2.2-24 presents photos of mammal species found in high elevation areas on Mauna Kea.

2.2.4.1.1      Threatened and Endangered Species, Candidate Species and Species of Concern

The federally listed Endangered ‘Ope‘ape‘a (Lasiurus cinereus semotus) was once found on all the main Hawaiian Islands, but now is thought to be limited to Hawai‘i, Kaua‘i, and Maui. It was listed as a federally Endangered Species in 1970. ‘Ope‘ape‘a have been observed up to 13,500 ft (4,115 m) on Mauna Loa, and use a variety of both native and non-native vegetation types (Frasher et al. 2007). While the Hawaiian hoary bat typically roosts alone in foliage (as opposed to roosting in large colonies as many bats do), it has also been observed in lava tubes, man made structures, and rock crevices (Frasher et al. 2007). ‘Ope‘ape‘a are known to migrate, and their densities in high elevation areas are thought to be highest during the winter months (December through March) (Menard 2001; Bonaccorso 2008; Menard 2008). ‘Ope‘ape‘a have been observed in the māmane woodlands on Mauna Kea (NASA 2005), but the status of the bat at Hale Pōhaku and environs is unknown.

2.2.4.1.2      Invasive Species

Non-native mammals found at Hale Pōhaku include feral cats (Felis catus), black rats (Rattus rattus), mice (Mus musculus and Mus domesticus), mongoose (Herpestes auropunctatus), domesticated sheep (Ovis aries), mouflon sheep (Ovis musimon), goats (Capra hircus), cattle (Bos taurus), and feral pigs (Sus scrofa). Each of these has had a role in the degradation of māmane woodlands and/or their associated animal communities on Mauna Kea.

Invasive mammals have had a serious impact on native Hawaiian species, as predators, competitors, and agents of change in the structure and composition of plant communities. Invasive mammalian predators include cats, dogs, rats, mongoose, feral pigs, and mice. Cats, rats, and mongooses all prey on bird species found in the māmane woodlands. See Section 2.2.3.3.2 for more information on predation of native birds. Rats and mice eat insects, and may especially impact flightless species (of which there are several in the subalpine and alpine zones on Mauna Kea). Several species of flightless insects no longer occur on the main Hawaiian Islands where rats are common (Gagne and Christensen 1985). We may never know what impact mammalian predators have had in native invertebrates in the subalpine zone, as arthropod communities in these areas were not well documented historically.

Feral pigs, goats, sheep, mouflon, cattle, and even horses have impacted plant communities throughout the islands, and the māmane woodlands are one of the hardest hit areas. The damaging effects of grazing animals on native forests have been known for some time. In 1909, Augustus Knudson of Kaua‘i wrote in the Hawaiian Forester and Agriculturalist that “Cattle and goats are really the only enemy that Hawaiian forests have. Kill them off and prevent their return, and in ten years you cannot recognize the region again; in twenty years the forest is practically restored, though young.” (Berger 1975).

Non-native ungulates have been present on Mauna Kea for some time. Scowcroft and Conrad (1992) summarized the history of introduction of grazing animals to Mauna Kea as follows (Scowcroft and Conrad 1992):

Domestic livestock were introduced to the Hawaiian Islands late in the 18th century (Kramer 1971). Feral populations of cattle, sheep, and goats (Bos taurus, Ovis aries, Capra hircus) soon became established in forests. By 1825, feral sheep had established in Mauna Kea’s subalpine woodland. Lacking natural predators except for wild dogs (Canis domesticus), the sheep population reached about 40,000 animals by the early 1930s, one for every 5 a (2 ha) of habitat (Bryan 1937b). Sheep suppressed māmane and other tree reproduction over large areas, stripped bark from tree stems, and consumed herbaceous vegetation, thereby leaving the soil exposed to accelerated erosion (Warner 1960). Because damage to the ecosystem was severe and because feral sheep competed with commercial flocks (Judd 1936), Hawai‘i Territorial foresters built a stock-proof fence around the Mauna Kea Forest Reserve (Bryan 1937a) and reduced the population through sheep drives and hunter-guide programs. Fewer than 500 feral sheep were left by 1950 (Bryan 1950). Control efforts were then relaxed, and populations increased.

Sustained yield management for public hunting was started in 1955, with the population kept below 5,000 animals. During the 1970s, the population averaged 1,500 animals. Even at this relatively low level, vegetation continued to deteriorate where sheep concentrated, especially at tree line (Scowcroft 1983; Scowcroft and Giffin 1983; Scowcroft and Sakai 1983).

Ecosystem damage has also been caused by mouflon sheep (Ovis musimon), which were released in the Mauna Kea Forest Reserve starting in 1962 (Giffin 1982). Food preferences, grazing and browsing behavior, and herding habits are similar to those of feral sheep, and native plants are particularly susceptible to damage by mouflon. In 1986, the largest concentrations of mouflon were on the southeastern and northwestern flanks of the mountain, and animals were moving into areas formerly occupied by feral sheep. The mouflon population was estimated at 500 animals (R.E. Bachman, pers. comm. 1986).

Continued degradation of the māmane woodland in the late 1970s posed a significant threat to the Palila (Loxioides bailleui), an endangered endemic Hawaiian bird now found only in the subalpine woodland of Mauna Kea (Berger 1972; van Riper et al. 1978; Scott et al. 1984). Palila depend on māmane and, to a lesser extent, on naio trees for food and nest sites (van Riper 1980a). Critical habitat for Palila was designated in 1977 and included almost all State-owned māmane and naio-māmane woodlands on Mauna Kea (USFWS 1977).

Because of continued habitat degradation and the attendant threat to the Palila, a suit on behalf of the Palila was brought by conservationists against the state of Hawai‘i. Noncompliance with the U.S. Endangered Species Act of 1973 was charged. A detailed account of the initial Palila lawsuit and events that proceeded was given by Juvik and Juvik (1984). The suit was decided in favor of the bird, and the State was ordered to remove feral sheep and feral goats completely and permanently from those portions of the māmane forest designated as critical Palila habitat. The status of mouflon sheep was not affected by the court order.

The feral sheep and goat “eradication” effort was completed in 1981. Five years later, it was evident that a few feral sheep had escaped death, as expected. The authors have seen a flock of 20 feral sheep at tree line on the western side of Mauna Kea. Sheep tracks and signs of recent browsing were common in the vicinity of that sighting.

To date there are still feral ungulates roaming the subalpine and alpine zone on Mauna Kea, and they continue to negatively impact native plant communities there. Between 200-500 feral ungulates are shot each year as part of control efforts conducted in the Mauna Kea Forest Reserve. Feral sheep and mouflon are the most abundant ungulates, as feral goats have been greatly reduced through hunting efforts. Only 26 feral goats have been observed (and shot) during semi-annual helicopter hunting efforts conducted by DOFAW over the last ten years (Fretz 2008). Although a fence was built around Mauna Kea to protect  the forest reserve, funds for upkeep are not sufficient, and the fence has many holes that allow feral animals to continue to move across the fenceline, into and out of Mauna Kea Forest Reserve (Fretz 2008). DLNR is seeking funding to build and maintain a perimeter fence to protect the Mauna Kea Forest Reserve. If this funding is granted, the presence of the new (or repaired) fence, combined with funds for proper upkeep, will help prevent migration of feral ungulates from lower elevations, and allow for more successful control, and eventually, eradication of feral ungulates found in the upper elevations of Mauna Kea (provided that sufficient effort is made to eradicate the animals). Currently efforts are being made to fence important areas of Palila habitat, rather than the entire Forest Reserve.

Non-native mammals can also negatively impact native subalpine communities through dispersal of invasive plant seeds, and in some cases through direct predation of native seeds and seedlings (Bruegmann 1996; Cabin et al. 2000). Rodents often eat native plant seeds. Rodent predation of native plant seeds is implicated in failure of native forest regeneration lowland dry forests (Cabin et al. 2000; Chimera 2004). Although rodents are not known to forage on māmane seeds (Cabin et al. 2000), they are likely to impact regeneration of other native species in the māmane woodlands on Mauna Kea.

2.2.4.1.3      Mammal Surveys at Hale Pōhaku

There have been no surveys conducted specifically for mammals at Hale Pōhaku. During her 1979 bird survey, Stemmerman observed mice in the woodlands and developed areas at Hale Pōhaku, and feral goats less than ¼ mile down slope of Hale Pōhaku (Stemmerman 1979). DLNR-DOFAW conducts semi- annual helicopter shoots of all feral sheep, goats, and mouflon on Mauna Kea within the forest reserve and up to the summit. Therefore data exists on numbers shot each year, which can be used to track changes in relative abundances in feral ungulate numbers over time (Fretz 2008).

2.2.4.2       Alpine Mammal Communities (MKSR and Upper Summit Access Road)

The density of mammals in the alpine zone on Mauna Kea is low due to limited food resources. Sheep, goats, cattle, cats and mice have all been recorded in the alpine zone of Mauna Kea (Hartt and Neal 1940; Juvik and Juvik 1984; Forsyth 2002; Conant et al. 2004).

2.2.4.2.1      Threatened and Endangered Species, Candidate Species and Species of Concern

No Endangered, Threatened, or Candidate Species, or Species of Concern are known to reside in the alpine zone on Mauna Kea. It is possible that the federally listed Endangered ‘Ope‘ape‘a (Lasiurus cinereus semotus) may occasionally use the area, although no records regarding this are available. It seems unlikely that this species would roost here given the cold climate and lack of trees.

2.2.4.2.2      Invasive Species

Sheep, goats and cattle have been documented all the way up to the summit of Mauna Kea. Grazing ungulates will feed on almost any palatable plant not protected by rocky crevices or impassable topography on the summit (Conant et al. 2004). Prior to ungulate control efforts, feral ungulates decimated the once thriving silversword population in the subalpine and alpine zones on Mauna Kea and no doubt reduced abundances of other palatable native species (Hess et al. 1999). A flock of 60 feral sheep was recently (Feb. 2008) observed in Pohakuloa Gulch and scat was observed at Lake Waiau in the Ice Age Natural Areas Reserve (Hadway 2008). Feral goats are likely to be rare or absent from MKSR, but feral sheep and mouflon are expected to occur there (Fretz 2008).

Although densities of feral ungulates in the alpine zone on Mauna Kea are currently low, even a few animals can exert serious grazing pressure on the plants found in this community, and feral ungulates continue to threaten native plant communities. For example, a recently discovered, isolated population of Mauna Kea silversword (Argyroxiphium sandwicense sandwicense) at approximately 12,200 ft (3,700 m) elevation in MKSR showed signs of grazing by feral ungulates (Nagata 2007; Tomlinson 2007). Other impacts of feral ungulates on the alpine zone include soil/cinder compaction, addition of nutrients to nutrient poor soils, and seed dispersal. These issues have not been examined on Mauna Kea (Aldrich 2005).

Feral cats and rats may be present in the lower reaches of the alpine zone, at very low densities. If Hawaiian petrels utilize the alpine areas on Mauna Kea, mammalian predators may prey on eggs, nestlings and adult petrel. Mice have been observed within the observatories and along the road above 12,000 ft (3,660 m) (Conant et al. 2004). Mice are known to eat arthropods and seeds and could have a negative impact on native arthropod communities at the summit, especially around the developed areas. However, their overall impact on the alpine community is unknown.

2.2.4.2.3      Mammal Surveys at MKSR

In 1988, Maile Kjargaard conducted a vertebrate (bird and mammal) survey at two proposed sites for the VLBA antennae, located between 11,800 and 12,400 ft (3,600 and 3,780 m) (Kjargaard 1988). The primary goal of the survey was to look for signs of Hawaiian petrels. Evidence of use of the sites by sheep was present in the forms of droppings, and sheep remains. Because of the lack of food in the area, Kjargaard suggested that these sites were mainly used as refuges from hunting pressure at lower elevations.

No other records of mammal surveys conducted in the MKSR are available. No formal surveys for mammals have been conducted at the Ice Age NAR, but opportunistic sightings of mammals (or sign of mammal damage to vegetation) are recorded in the DNLR GIS (Hadway 2008). Records from semi- annual helicoptor shoots conducted in MKSR by DLNR provide a record of changes in relative abundance of feral ungulates over time (Fretz 2008).

2.2.4.3       Threats to Mammal Communities on Mauna Kea

Habitat loss and pesticide use to control insects are believed to be the primary threats to Hawaiian hoary bat (Lasiurus cinereus semotus) (USFWS 1998b). Not enough information is available about the bat’s use of subalpine habitats to speculate about what factors might affect its populations in high elevation areas of Mauna Kea. Although they have been observed at high elevation areas on Mauna Loa, their use of high elevation habitats on Mauna Kea is not well documented (Bonaccorso 2008; Menard 2008).

2.2.4.4       Mammal Community Information Gaps

There have been no mammal-specific surveys conducted in Hale Pōhaku and the MKSR, although Maile Stemmerman Kjargaard did look for signs of mammals during her bird surveys at Hale Pōhaku and MKSR. The following information gaps regarding the condition of the subalpine and alpine mammal communities at Hale Pōhaku and the MKSR have been identified through review of the literature and consultation with local experts:

1. Quantitative mammal surveys

a)       Hale Pōhaku: No quantitative studies documenting the composition of the mammal community, population sizes, or distribution of native and non-native species have been conducted at Hale Pōhaku. A brief (qualitative) inspection for mammals was conducted by Stemmerman in connection with her 1979 bird survey. DOFAW collects data on the number of sheep and mouflon shot during helicoptor shoots conducted twice a year. Although these numbers do not provide an estimate of population size, they can be used to track changes in relative abundance of sheep and mouflon on the mountain over time.

3.1.1        Astronomical Research and Facilities

The summit of Mauna Kea hosts the world’s largest ground-based astronomical observing site, considered to be the finest in the world. Physical characteristics that set Mauna Kea apart from other sites include: high altitude, atmospheric stability, minimal cloud cover (about 325 days per year are cloud free at the summit), low humidity, dark skies (because of its distance from urban development), minimal atmospheric pollutants, and the transparency of the atmosphere to infrared radiation. The trade wind inversion layer caps the upper layer of clouds at an approximate elevation of 7,000 ft (2,133 m) for most of the year resulting in a stable dry air mass above the inversion (see Section 2.1.4, Climate). Due to the location of the Hawaiian Islands within the northern hemispheric tropics, astronomers can observe the entire northern sky and nearly 80 percent of the southern sky.

In the 1960s, the University of Hawai‘i (UH) initiated an astronomical research program to attract global interest in constructing and operating telescopes in Hawai‘i in scientific collaboration with UH. Haleakalā, and subsequently Mauna Kea, were targeted as ideal locations. A small site-testing dome (with a 12½ inch telescope) was built on Pu‘u Poli‘ahu in 1964, initiating Mauna Kea as a modern-day astronomical site. The Institute for Astronomy (IfA) was founded in 1967, to manage the observatories and facilitate collaboration. The Board of Land and Natural Resources created the Mauna Kea Science Reserve (MKSR) in 1968, granting UH a 65-year lease (Lease No. S-4191) for a scientific complex including observatories (see Section 1.4). The MKSR includes all land within a 2.5 mile radius of the summit, above about 11,500 ft (3,505 m), except for the area within the Mauna Kea Ice Age Natural Area Reserve (NAR) (see Figure 1-3). Since the creation of MKSR, thirteen observatories have been built on Mauna Kea, operated by eleven countries¹, and used by scientists from around the world. The observatories include nine optical and infrared telescopes, two single-dish millimeter- and sub-millimeter- wavelength telescopes, a sub-millimeter array, and a very long baseline array antenna (see Table 3-1 and Figure 3-1). In addition to full access to the UH 2.2 meter telescope, UH astronomers are guaranteed 10 to 15% of observing time on all the other telescopes on the summit of Mauna Kea. A series of general maintenance and upgrade projects have been completed, piecemeal, in association with telescope construction, funded by the sponsoring observatories. These projects included paving of portions of Mauna Kea (MK) Summit Access Road and construction of the Subaru construction cabins.

Table 3-1. Mauna Kea Telescopes (2008)

Source: http://www.ifa.hawaii.edu/mko/telescope_table.htm

	Name	Mirror	Owner/Operator2	Year Built
Optical/Infrared
UHH 0.9m3	UHH 0.9-m telescope	0.9m	University of Hawai‘i, Hilo	2008
UH 2.2m	UH 2.2-m telescope	2.2m	University of Hawai‘i	1970
IRTF	NASA Infrared Telescope Facility	3.0m	NASA	1979
CFHT	Canada-France-Hawai‘i Telescope	3.6m	Canada/France/UH	1979
UKIRT	United Kingdom Infrared Telescope	3.8m	United Kingdom	1979
Keck I	W. M. Keck Observatory	10m	Caltech/University of California	1992
Keck II	W. M. Keck Observatory	10m	Caltech/University of California	1996
Subaru	Subaru Telescope	8.3m	Japan	1999
Gemini	Gemini North Telescope	8.1m	USA/UK/Canada/Argentina/	1999
			Australia/Brazil/Chile
Submillimeter
CSO	Caltech Submillimeter Observatory	10.4m	Caltech/NSF	1987
JCMT	James Clerk Maxwell Telescope	15m	UK/Canada/Netherlands	1987
SMA	Submillimeter Array	8x6m	Smithsonian Astrophysical	2002
			Observatory/Taiwan
Radio

VLBA                  Very Long Baseline Array                     25m  NRAO/AUI/NSF                              1992

1 U.S., Canada, France, the United Kingdom, Japan, Taiwan, Argentina, Australia, Brazil, Chile, and the Netherlands.

2 AUI: Associated Universities, Inc.; NASA: National Aeronautics and Space Association; NRAO: National Radio Astronomy Observatory; NSF: National Science Foundation

3 In 2008 the UH 0.6-m telescope (built in 1968) was replaced by the UHH 0.9-m telescope. Information detailed in Section 3 referring to production of solid waste, hazardous materials, and water use refers to the UH 0.6-m telescope facility since that was in operation at the time of data collection.

Infrastructure, in the form of buildings, roads, and utility lines, supports the existing observatories on Mauna Kea, both at the summit and at the mid-level Hale Pōhaku facility. The extent and installation of these facilities is described in detail elsewhere (e.g., construction documents, environmental impact statements). Most relevant for this plan is a general discussion of the existing facilities and their operations and use as they relate to actual and potential impacts on the natural resources of Mauna Kea, as the facilities currently operate, or as a result of changes (redevelopment, new facilities, decommissioning). Section IX of the 2000 Master Plan (Group 70 International 2000) provides a Physical Planning Guide, which contains guidelines for addressing future physical development of the MKSR (summit facilities and support facilities), including siting and design criteria, in the context of protecting natural and cultural resources. These guidelines defined a 525 acre (212 ha) Astronomy Precinct to consolidate astronomical development (see Figure 1-3). Specific site-development plans prepared by proponents of the facilities are subject to review and approval.

3.1.1.2.1      Buildings

The total disturbed area for the installation of the existing 12 observatories at the summit is approximately 17 acre (7 ha), of which 4 acre (2 ha) is impervious surface, and the remaining area being adjacent and mostly unpaved leveled areas and access roads or driveways (NASA 2005). As depicted on construction drawings, the foundation depths and sizes of the buildings vary, but can extend over a hundred feet below the ground surface and cover hundreds of square feet of surface area. Some of the building’s useable areas are also located below grade. The VLBA antenna is situated approximately 1,590 ft (485 m) below the summit. The dish antenna and control building are accessed by a dirt-road spur from the Summit Access Road. Buildings at Hale Pōhaku include a support facility for the observatories, construction camp facilities, and Visitor Information Station (VIS) facilities. There is also a 0.74 acre (0.3 ha) exclosure supporting research into silversword and māmane forest restoration, which was established in 1972 (Scowcroft and Giffin 1983). This exclosure is managed by the Department of Land and Natural Resources (DLNR), and is not part of the UH facilities (see Figure 3-2).

3.1.1.2.2      Roads and Parking

The Summit Access Road extends 16.3 mi (26.2 km) from its intersection with Saddle Road to the summit, with an average width, including cuts and fills beyond the main route, of 45 ft (14 m) (NASA 2005). The road is paved along its entire length except for a 4.6 mile unpaved, gravel section that extends from Hale Pōhaku to the summit area (see Table 3-2).

Table 3-2. Coverage of Summit Access Road

Data from (NASA 2005)

Road Section	Paved Length	Acres Covered	Unpaved Length	Acres Covered
Saddle Road to Hale Pōhaku	6.3 mi (10.1 km)	34 acre (14 ha)	—	—
Hale Pōhaku to the Summit	3.7 mi (6.0 km)	20 acre (8 ha)	4.6 mi (7.4 km)	25 acre (10 ha)
Summit loop	1.7 mi (2.7 km)*	9 acre (3.6 ha)	—	—
Total	11.7 mi	63 acre	4.6 mi	25 acre

* A portion of which is unpaved.

Although there is no road maintenance plan, the unpaved portion of the road is graded approximately three times a week by Mauna Kea Observatories Support Services (MKSS) to keep it drivable, and when necessary, cinder pieces fallen from the roadside are collected and used to fill in ruts (Koehler 2008). Both the grading and other vehicles generate dust and other emissions, and move cinder material onto the road shoulder and downslope areas. In the spring of 2008, MKSS brought in basalt gravel from a quarry at Pōhakuloa to use as a substitute for the cinder on the most severely washboarded areas. As recommended by the Mauna Kea Management Board (MKMB) Environment Committee, the material was inspected for cleanliness and ants (MKMB Environment Committee 2007; Koehler 2008). This was the first time outside gravel has been used to cover the road surface (Koehler 2008). In addition, a soil additive designed to control dust (Durasoil) has been approved by the MKMB Environmental Committee and will be applied to limited stretches of the unpaved road, well below the summit (Koehler 2008). Based on the initial results of these road improvement strategies, it will be determined whether to continue to use Durasoil, the imported gravel, or both, and how frequently (Koehler 2008).

Future plans may include paving the unpaved portion of the summit access road and the remainder of the summit spur road, from the SMA building, past the Subaru Telescope to the Keck Observatory; however, concerns related to cost, environmental impacts, and facilitating access to the summit need to be evaluated.

There are three visitor parking areas along the Summit Access Road: Parking Area 1, just after the paved road begins; Parking Area 2, near the trailhead to Lake Waiau; and Parking Area 3, just past the junction of the access road and the summit loop. These areas are depicted on the map included in the safety brochure made available to workers and visitors, but are not identified by signage on-site. At the summit many visitors park near the UH 2.2m telescope if they plan to hike the summit trail. During the winter, before roads are fully cleared of snow and there are large numbers of private vehicles in the summit area, parking becomes congested and visitors park their vehicles along the road, wherever there is space. Commercial tour vehicles usually park in the area around the UH 2.2m telescope and Gemini Telescope during the sunset viewing times. For evening stargazing, there are designated parking areas for tour vehicles on lower portions of the mountain. Observatory vehicles park in designated areas near their buildings. Most parking areas are graded but unpaved.

3.1.1.2.3      Electricity and Communication

Underground power and communication lines supply Hale Pōhaku and summit facilities. Installation of the underground system to transmit electricity to the summit facilities began in 1985 and was completed in 1995. Rather than on-site generators, the facilities are now powered from a sub-station below Hale Pōhaku that is connected by overhead lines to the Humu‘ula Radio Site. In the mid-1990s, underground fiber optic lines were installed to provide high speed communications capability to the observatories. One benefit of these lines was a reduction in personnel needed on-site at some of the observatories, as they can now be controlled remotely. Expansion of these systems would be needed if facilities are sited in new locations.

3.1.1.2.4      Solid Waste

Trash is generated and collected at summit observatories and Hale Pōhaku facilities. All trash containers are required to be covered and secured to prevent providing a food source for invasive fauna and to reduce the possibility of escaping debris, which can occur during periods of high winds that occur frequently. The observatories are responsible for removing their trash from the summit. Trash from Hale Pōhaku and the dormitories is taken off the mountain daily by the MKSS housekeeping staff and brough to the main Hilo office where it is removed by sub-contractors daily (Wilson 2008). Recent estimates are that approximately 4,400 gallons (16.7 kl) of solid waste per week are removed from the MKSR and Hale Pōhaku facilities for disposal at an off-site landfill (see Table 3-3) (NASA 2005). Additional material is generated over short periods during construction activities.

Table 3-3. Solid Waste Generated by MKSR Facilities

Data from (NASA 2005)

Facility                                                            Trash Produced
UH (0.6-m) (24-in) and 2.2-m (88-in))	Two to three 30-gal (114-liter) bags weekly
CFHT	Four bins, 2 yd3 (1.5 m3) each, generated monthly
NASA IRTF	Three 30-gal (114-liter) trash bags weekly
UKIRT and JCMT	About one 30-gal (114-liter) trash bag for both facilities weekly
CSO	About 2,000 lb (900 kg) generated yearly
VLBA	One 30-gal (114-liter) bag weekly
W.M. Keck	3 yd3 (2.3 m3) dumpster emptied 1 to 2 times weekly
Gemini North	Several 50-gal (190-liter) trash bags weekly
Subaru Telescope	40 lb (18 kg) generated daily
SMA	Two to four 50-gal (190-liter) drums weekly
Hale Pōhaku Mid-Elevation Support Facilities	0.9 to 1.5 yd3 (0.7 to 1.1 m3) daily

3.1.1.2.5      Water

MKSS contracts with a trucking company to deliver potable water from Hilo to Hale Pōhaku and the summit observatories in 5,000-gallon-capacity (18,900 l) tank trailers owned by MKSS. Each observatory stores its own water and is responsible for maintenance of their water tanks. Water at Hale Pōhaku is stored in two beige-colored 40,000-gallon (151,400 l) tanks. Data from MKSS indicates that the Hale Pōhaku facilities (food, lodging, VIS) currently require approximately 30,000 gallons (113,500 l) of water weekly (Nahakuelua 2008). Water is trucked to the summit about twice a week for an annual total of approximately 502,500 gallons (1,902,000 l) (Koehler 2008). The annual use-quantities have remained fairly consistent over time, with a slight downward trend for the past seven years, likely in association with researchers shifting to remote facilities (see Table 3-4). It has been suggested that during peak snow periods visitor numbers increase, resulting in increased use of potable water and restroom facilities. Water use and wastewater generation both increase during construction periods. Increased use associated with future conditions will need to be numerically evaluated for the specific project. Any future development would require additional water storage tanks, water delivery, and a wastewater disposal system. It is possible that improvements to remote viewing technology will further reduce the number of staff and scientists at the Mauna Kea facilities, which will reduce water use correspondingly.

Table 3-4. Water Delivered to the MKSR Facilities

Data from (Koehler 2008)

FY (July 1 – June 30)                Total (gallons)
2002-03	608,000
2003-04	630,000
2004-05	548,000
2005-06	508,650
2006-07	532,625
2007-08	502,500

Each observatory owns an individual wastewater system (e.g., septic tank, cesspool) that has been permitted by the Hawai‘i State Department of Health (DOH). Currently there are a total of eight septic systems and three small capacity cesspools (see Table 3-5). Site plans for individual septic systems are held by the respective facilities; however DOH files do include information for four summit facilities (Subaru, Keck I and II, SMA, and VLBA). Maintenance and inspections of observatory wastewater systems are the responsibility of the owner. Information as to how often these systems are inspected, what is inspected, and by whom is not centrally reported. Except for VLBA, UKIRT, and JCMT, these systems are periodically inspected and pumped to remove digested biosolids by private firms (NASA 2005; Koehler 2008). There have been no documented wastewater spills at the observatories since 1998 (see Section 3.2.3 and Table 3-11). There is no plan for the construction of a sewer collection system for the summit area (Group 70 International 2000).

Hale Pōhaku has three small capacity cesspools and six septic systems (see Table 3-6). The systems at dormitory A, the old construction camp, and the utilities buildings have not been upgraded and use the small capacity cesspools for wastewater disposal. At Hale Pōhaku’s main common building, dormitories B, C, and the VIS, the wastewater systems have been upgraded to septic tanks that use the old cesspools as overflow leach fields to capture effluent discharges. Dormitory D was constructed with a septic system and no modifications have been made. A septic tank with a leach field is used at the new construction camp. DOH has site plans on file for the septic systems for most of the buildings at Hale Pōhaku (VIS,  the main common building, dormitories B and C, and the construction camp housing cabins 1-4). The high-use septic tanks at the main common building, the VIS and dormitory B are checked weekly (Friday) by MKSS staff. All Hale Pōhaku systems are checked on a monthly basis (Koehler 2008). One leak was detected and corrected in 2008 (see Section 3.2.3).

Wastewater entering these systems is from domestic sources (toilets, sinks) and basic facilities cleaning water (e.g., mopping). As designed, biosolids settle out of solution and decompose within the septic tank. When liquid waste in a septic tank reaches the overflow, effluent discharges either into a cesspool or a leach field; both allow effluent to absorb into the substrate below the ground surface. Similar to septic tank systems, cesspools collect all waste; the solids slowly decompose in-situ and liquid effluent discharges from the holding chamber. Effluent discharges from both cesspools and septic systems likely contain contaminants, including nutrients such as phosphorus and nitrogen, as well as organic and inorganic by-products.

It can be conservatively approximated that all water trucked to the summit becomes wastewater with eventual subsurface disposal. Questions have been raised regarding the transport and fate of this effluent wastewater and the potential for its migration to and contamination of aquifers. The amount of water being discharged into the wastewater systems (~502,500 gallons/yr (1,902,000 liters/yr)) is a small percentage of the overall volume of water contained in the Waimea Aquifer that, at 2,969 ft (904 m) elevation, lies far beneath the Astronomy Precinct. Using the DLNR Commission on Water Resource Management (CWRM) value of the aquifer’s annual sustainable yield, which is approximately 4.74 billion gallons, a conservative estimate of the annual recharge to the aquifer is 30 percent of the sustainable yield, or 1.42 billion gallons. As a result, the effluent wastewater discharge is 0.035 percent of the aquifer’s annual recharge. See Section 2.1.3 for a discussion of the region’s hydrology

Table 3-5. Wastewater Treatment and Disposal Systems at MKSR Facilities

Data from (NASA 2005)

Average Observatory                 Wastewater Flow                Treatment and Disposal System Rate (gpd)
UH 0.6-m (24-in) & 2.2-m (88-in)	115	2,500-gal (5-kl) septic tank and leach field
CFHT	295	Septic tank and leach field
NASA IRTF	50	1,450-gal (5-kl), two-compartment septic tank and leach field (90 linear ft (27 m))
UKIRT	111	1,130-gal (4-kl), two-compartment septic tank and leach field (75 linear ft (23 m))
CSO	65	7-ft (2-m) diameter, 10-ft (3-m) deep cesspool
JCMT	109	8-ft (2-m) diameter, 13-ft (4-m) deep cesspool
VLBA	31	7-ft (2-m) square-shaped, 10-ft (3-m) deep cesspool
W.M. Keck I & II	399	1,000-gal (4-kl) septic tank and 12-ft (4-m) deep seepage pit
Gemini	122	1,000-gal (4-kl) septic tank and 10-ft (3-m) deep seepage pit
Subaru Telescope	360	1,250-gal (5-kl) septic tank and two seepage pits
SMA	118	1,000-gal (4-kl) septic tank and leach field (265 linear ft (81 m))

Table 3-6. Wastewater Treatment and Disposal Systems at Hale Pōhaku Facilities

Data from (Koehler 2008)

Location	Septic Tank Volume (gal)	Septic Tank Pumping Schedule	Number of Cesspools	Cesspool Use	Current Flow (gpd)4	Design Use (gpd)5
Main Common Building	2,500	Every 3 months	2	Overflow leach fields	1,200	1,500
VIS	2,000	Every 3 months	1	Overflow leach field	1,200	1,250
Dormitory A (cook staff only)	None	None	1	Primary waste	100	500
Dormitory B	1,500	Every 9 months	1	Overflow leach field	750	1,600
Dormitory C	1,250	Every 12 months	1	Overflow leach field	600	1,000
Dormitory D	1,250	Every 12 months	0	N/A	400	1,000
NEW Construction Camp	2,000	Every 1.5 years (due to low usage); has only been pumped once	0	N/A	150	1,600
OLD Construction Camp	None	None	1	Primary waste	0	1,000
Utilities	None	None	1	Primary waste	100	500
Totals	10,500				4,500	9,950

4 VIS, New Construction Camp and Old Construction Camp are the only metered flows, all others are estimates. Total flow is actual water delivered to Hale Pōhaku,

5 These are estimates based on building capacities.

3.1.1.2.7      Hazardous Materials

Solid and liquid hazardous materials are used in routine observatory operations and generate hazardous waste after their use.⁶ A detailed accounting of the types and amounts of hazardous materials used and stored at the observatories and at the Hale Pōhaku facility is presented in Table 3-7 (NASA 2005). Each observatory has a written procedure for safety, handling, and disposal of hazardous wastes and emergency procedures for attending to spills. Licensed contractors transport all hazardous waste to a State-approved, off-site disposal facility in Hilo. There have been no documented spills of hazardous materials since 2004 (see Section 3.2.3 and Table 3-11).

Telescope operations may require glycol coolants; diesel fuel for emergency generators; hydraulic fluid; lubricants; compressed gases (e.g., carbon dioxide, helium, oxygen, nitrogen); mercury; mirror decoating acids (e.g., hydrochloric acid, potassium hydroxide, copper sulfate, hydrofluoric acid); and paints and solvents. The amounts used vary by facility, although data shows the Keck Observatory to be using and storing the largest amount, by volume, of hazardous materials (NASA 2005).

Hale Pōhaku has three underground storage tanks; one housing 11,500 gal (43,532 liters) of diesel and two housing 2,000 gal (7,570 liters) and 4,000 gal (15,140 liters) of gasoline, respectively. Tanks are located underground in front of the maintenance utilities shop and are believed to be approximately 25 years old. Due to the lack of secondary containment, in 1997 the tanks were retrofitted with a 24-hour a day sensor monitoring system that is checked daily (Nahakuelua 2008).

3.1.1.2.8      Mirror Washing

Five observatories (Keck, CFHT, Gemini, Subaru, and UH 2.2m) have their own facilities to conduct mirror washing activities (stripping aluminum from the reflecting surface of the mirror) at the summit. The other observatories bring their mirrors to one of those five for washing and recoating activities (McNarie 2004). At the Subaru telescope, wastewater generated when mirrors are washed has always been contained for off-site disposal, but from 1971 until 2001, the other observatories either disposed of the wastewater in either their onsite domestic wastewater disposal systems (UH, Keck I & II, Gemini) or in an open drain leading to the ground (CFHT). As part of the mirror re-aluminizing process, telescope

6 In Hawai‘i, hazardous wastes are administered by the DOH Solid and Hazardous Waste Branch (SHWB). See http://hawaii.gov/health/environmental/waste/hw/index.html for details. Hazardous materials transition into hazardous waste either by materials reaching their expiration date or the product being used. Hazardous waste has a regulatory definition that is triggered by the concentration of chemicals in materials. Hazardous waste is classified by amounts generated, from small to large, and is defined by mass or volume per calendar month. Users are required to report to the Environmental Protection Agency (EPA) when they generate a combined waste total equal to or greater than 100 kg per calendar month. This level triggers EPA to issue an identification number, which among other things lists the waste type and the facility name and address. The identification number is sent to DOH SHWB, which then conducts unannounced inspections of the site. Facilities are required to store, use, and dispose of hazardous materials and waste per the Resource Conservation and Recovery Act (RCRA). They are also responsible for developing a contingency plan to address potential spills or accidents.

Concerns have been raised about potential impacts to natural resources that could have been caused by chemicals (e.g., aluminum, mild acid solution, alcohol, detergents) that may have been disposed of with mirror washing wastewater during 1971–2001. In 2001, wastewater management protocols were changed in response to concerns from community groups about the potential impact of this wastewater on the surrounding environment. All mirror washing effluent is now collected and trucked off the mountain for off-site treatment and disposal (McNarie 2004). It is estimated that the total amount of aluminum used on one 3-meter diameter mirror is approximately 15 grams (Koehler 2008). Limited analysis conducted on the fate and transport of metals contained in the effluent wastewater derived from mirror washing during the period it was discharged onsite found no substantial impacts (NASA 2005). However, it is our understanding that the fate, transport and potential impacts to substrate and downgrade waters from metals and other contaminate byproducts previously discharged into the septic systems, cesspools, and dry swales are unknown due to uncertainties regarding capture rates of byproducts in the waste systems and the hydrogeologic properties of the area (see Section 2.1.3).

3.1.1.2.9      Construction

Major construction activities at the summit, undertaken to build, redevelop, or deconstruct facilities, require approval, permits, and environmental analysis. Minor construction may be conducted as part of on-going facility maintenance, but these activities are subject to Conservation District Use Permit conditions and approval by MKMB. Construction could involve use of hazardous materials; generation of dust and debris; increased traffic and use of heavy equipment; noise and vibrations from jackhammers, wrecking balls and other equipment; excavation and disposal of excavated material; grading and filling; drilling and pouring concrete for piles, piers, footings and foundations; and installation of structures (e.g., antennas, buildings).

3.1.1.3       Off-site Support Facilities

In addition to the summit observatories, each of the telescopes is supported by off-site facilities. While most of these are located at the Science and Technology Park complex, at UH Hilo, the Keck Observatory and the Canada-France-Hawai‘i Telescopes are based in Waimea, and the VLBA is operated remotely from its headquarters in New Mexico. The installation of high-speed fiber-optic lines to the summit has facilitated an increase in remote operation of the facilities, and may reduce the number of astronomers traveling to the summit. The ‘Imiloa Astronomy Center of Hawai‘i, located at UH Hilo, opened in 2006 (http://www.imiloahawaii.org/) and provides visitors with an opportunity to discover the connections between Hawaiian cultural traditions and astronomical research conducted at Mauna Kea.

3.1.1.4       Future Land Uses

Proposed plans for future astronomical development on the summit are described in the 2000 Master Plan (Group 70 International 2000). In addition to the potential construction of new observatories, other possible changes to the astronomy facilities include redevelopment of existing sites (i.e., dismantling an existing facility and replacing it with a new one on the existing footprint), upgrades to or expansions of existing observatories, and removal of some obsolete observatories. Changes could also involve improving utility service. Any future observatory development would occur within the 525-acre Astronomy Precinct portion of the MKSR, as delineated in the 2000 Master Plan (Group 70 International 2000).⁷ The Master Plan recommends protecting all of the major undeveloped pu‘u and the intervening areas from development.

Other potential future land uses include projects to support the various other uses of Mauna Kea. Hale Pōhaku provides supporting infrastructure and services to support observatories, visitor use, and scientific research. Although no specific plans have been proposed, the 2000 Master Plan suggests some changes to these facilities including removal of some of the older construction camp buildings; use of the Subaru construction camp facilities to support education and research activities; expansion of the visitor center to include a larger interpretive center, an observatory, and ranger facilities; and expanded parking (see Figure 3-2) (Group 70 International 2000). Growing visitor numbers have prompted discussion about improved facilities to support recreational users, including a rest area in the snow play area at the base of ‘Poi Bowl’, designated scenic lookouts at the summit, designated visitor parking within the MKSR, and additional visitor parking at Hale Pōhaku (Group 70 International 2000). There are no current plans to pursue any of these changes.

3.1.2        Scientific Research

Mauna Kea is a tropical high altitude, alpine environment with unique biological, geological, and cultural features. Although there have been some ground-based scientific studies conducted, the main focus of scientific work on the mountain has been astronomy. Many of the previous natural resources studies have been conducted in association with project-based environmental analyses. A recent field-based cultural resources survey of the entire MKSR was conducted to document and map the locations of cultural resources (McCoy et al. 2009).

Although ecologists and biologists have long been interested in the high alpine environments on Mauna Kea (Goodrich 1826, 1833a, b; Lyons 1875; Alexander 1892; Mesick 1909; Baldwin 1915; Hitchcock 1917a; Hitchcock 1917b; Bryan 1918; Daingerfield 1922; Bryan 1923, 1926; Swezey and Williams 1932; Wentworth et al. 1935; Ueno 1936; Gregory and Wentworth 1937; Coulter 1939; Neal 1939; Hartt and Neal 1940; Bartram 1952; Warner 1960; Smith 1967; Mueller-Dombois and Krajina 1968; Landgraf 1973), it was not until the discovery of the wēkiu bug in 1979 that any in-depth and ongoing scientific research into the natural resources (biology) of the summit area began. The focus of most biological research at the summit has been on the wēkiu bug, and relatively little is known about the other species that reside there. Arthropods were intensively sampled at the summit in 1982, but in only a very limited area (Howarth and Stone 1982). Plants at the summit have been cataloged in detail (but only in small areas) on two occasions, once in 1935 (Hartt and Neal 1940), and again in 1982, in association with an environmental impact statement (Smith et al. 1982). Biological studies conducted specifically at Hale Pōhaku and the MKSR are listed in Section 2.2.2 within the various subsections (Plants, Invertebrates,

7 The Institute for Astronomy (IfA) provides guidance on research and science, while future development of telescope on Mauna Kea is reviewed by the MKMB, OMKM and Kahu Kū Mauna. It is envisioned that MKMB, OMKM and Kahu Kū Mauna will be taking the lead on reviewing and updating the current Master Plan.

Interest in the processes and products of Hawaiian volcanoes have prompted scientific inquiry for centuries; investigations of Mauna Kea compiling a substantial portion of these (Washington 1923; MacDonald 1945; Stearns 1946; Woodcock et al. 1966; Porter 1972a; 1975; Wood 1980; West et al. 1988; Dorn et al. 1991; Wolfe et al. 1997; Sherrod et al. 2007). At high elevations on Mauna Kea, a dry climatic regime and low rainfall persist, resulting in little surface runoff and low erosion rates. Disturbance to the geologic features, especially those caused by recent human presence, are generally well preserved and difficult to erase. Evidence of the formation and retreat of three glacial events believed to have occurred during the Pleistocene has been well documented (Porter 1975; 1979a; Dorn et al. 1991; Lockwood 2000), as has the unique presence of permafrost (Woodcock et al. 1970; Woodcock 1974). Other topographical focal points of past and on-going study include the geomorphologic evolution of the summit and flank areas such as Lake Waiau. Meteorological attributes of Mauna Kea such as precipitation, surface pressure, dew point, and wind regimes, as well as their influences on local hydrology, air quality, and viewscape have been the subjects of numerous studies (Stearns and Macdonald 1946; Blanchard 1953; Chen and Wang 1994, 1995a; Chen and Wang 1995b; Nullet et al. 1995; Arvidson 2002; da Silva 2006). Other climate investigations have included portions of Mauna Kea’s lower elevation flanks (Chen and Wang 1995a; Hotchkiss et al. 2000) and neighboring high elevation locations on Kauai and Maui (Billings 1979; Nullet and Giambelluca 1990; Loope and Giambelluca 1998).

The Office of Mauna Kea Management (OMKM) both funds and provides logistical support for scientific studies on the natural environment, including research to understand microhabitat and microclimate selection by the wēkiu bug, initiated in 2001, and analysis of meteorological data. Additional studies are planned in support of recommendations made in this NRMP (see Section 4.1). The existing facilities at Hale Pōhaku are occasionally used to support visiting scientists, other than astronomers, who are conducting research on the mountain, and they have the potential to provide greater support for such scientists in the future. As use of the mountain for ground-based scientific research grows, managers must consider the potential impacts of further studies, weighed against potential benefits. Recent scientific studies commissioned by OMKM give significant consideration to the potential impacts on natural and cultural resources in the MKSR and involve consultation with Kahu Kū Mauna (an advisory group on matters of Hawaiian cultural resources on Mauna Kea), the MKMB Environment Committee, and the MKMB. Rarely does the MKMB Environment Committee go against recommendations made by Kahu Kū Mauna; however, temporary weather stations were established at specific sites to collect data associated with wēkiu bug studies against Kahu Kū Mauna’s advice (Businger 2008). Considerations were given to cultural concerns, potential impacts on the physical resources (e.g., disturbance of cinder during installation and trail-carving through repeated site access), and minimizing visibility by painting the structure to blend into the background (to reduce visits by curious persons and potential vandalism) (Mauna Kea Management Board 2007).

Mauna Kea has long been a site for tourism and private recreational activities including hiking, hunting, snow-play and sightseeing. Visitors are drawn by the natural beauty, scenic vistas, and accessible high peaks. Most of the upper elevation land is under the jurisdiction of the State of Hawai‘i, including the MKSR, the Mauna Kea Forest Reserve, and the Mauna Kea Ice Age Natural Area Reserve (see Figure 1- 4). The rules governing allowable activities in these areas differ (see Section 1.4). The Mauna Kea State Recreation Area is located at a lower elevation of 6,500 ft (1,981 m) and provides facilities to support recreational campers and hunters.

Studies of the optimal use of the recreational resources of Hawai‘i in 1964 looked at the manner and extent to which the lands around Mauna Kea should be developed for outdoor recreation (McIntosh and Milstein 1964). Although at the time visitor use was approximately 14,000 persons per year (compared with 400,000 for nearby Hawai‘i Volcanoes National Park), the unique recreational advantages of Mauna Kea (e.g., big game and bird hunting, skiing, and viewing of features such as the Adze Quarry and evidence of Pleistocene glaciation) were considered potential draws (McIntosh and Milstein 1964). Even so, the study concluded that that development of the area’s recreational capacity or commercial development would yield minimal financial returns, due the limitations of the area (i.e., limited water supply, remote location, limited facilities, and minimal publicity) (McIntosh and Milstein 1964).

Tourism has increased over the past several decades due to easier access and a greater number of organized commercial and educational tours (see Section 3.1.4). Although there is no official registration system to track users, OMKM has been keeping detailed records on the number of people visiting the VIS and the summit since the ranger program began in 2001 (Nagata 2007). MKSS estimates that in 2002, 105,000 visitors stopped at the VIS (Good 2003). Byrne (2008) indicates similar estimates of greater than 100,000 visitors per year over the past few years. Over the past few years (2006-2008), the total of all types of summit visitations by vehicles ranged between approximately 30,000-32,000 (OMKM unpublished data).⁸ Observatory vehicles and visiting 4-wheel drive vehicles represent the largest percentage of total vehicles on the mountain, with over 11,000 of the former and over 10,000 of the later, in 2008 (OMKM unpublished data). It is possible that recreational visitors to Mauna Kea will increase with the completion of the Saddle Road realignment.⁹ Ranger estimates indicate an average of about 30- 40 non-commercial visitors a day to the summit, most of them staying less than 30 minutes (OMKM Rangers 2007). It is anticipated that as tourism on the Big Island continues to grow, and with the ongoing improvements to Saddle Road, more tourists and recreational visitors will visit Mauna Kea in coming years. Currently OMKM rangers estimate that most recreational visitors are from the mainland or overseas, but there is no official tracking of visitor demographics (OMKM Rangers 2007).

Most visitors to Mauna Kea know little or nothing about the unique natural resources of Mauna Kea. Many think “it’s only rock,” while a small proportion of visitors are aware of ‘the bug’ (wēkiu bug). About 10% of the general public is aware of Lake Waiau, though its recent inclusion in the “Big Island Revealed” guidebook has made it a more popular visitor destination (OMKM Rangers 2007).

8 The reference (OMKM unpublished data) refers to data from OMKM database on Ranger patrol reports, ongoing collection 2001–present. Data is housed in a Microsoft Access database at the OMKM main office. For the period of June 2001 to April 2005, observatory vehicle totals reported by Ranger staff may have been inadvertently double or even triple counted.

9 Providing numerical estimates would be pure speculation. Even if rental car companies allow people to drive on Saddle Road, it is not known whether driving on the unpaved portion of the Summit Access Road would be allowed or prohibited

3.1.3.1       Visitor Services

The Visitor Information Station of the Onizuka Center for International Astronomy, established in 1986 at Hale Pōhaku, provides information on safety and hazards, astronomy, the observatories, and natural and cultural resources for independent travelers (Good 2003). From 1986 until 2000 the VIS was a part-time venture. It is now open daily from 9am – 10pm, providing information to both daytime visitors and nighttime stargazers. The VIS, including the gift shop, is staffed by paid employees and volunteers (more than 170, many from UH Hilo) through MKSS. Both employees and volunteers receive training to enable them to provide interpretive information. Signs on the outside of the building describe the dangers associated with traveling to the summit (see Figure 3-4), and staff are present to answer questions and provide information. The VIS website (http://www.ifa.hawaii.edu/info/vis/) provides information for those planning a visit, including weather and road conditions, safety information, names of tour companies, VIS-sponsored events, and links to related information about Mauna Kea.

At the VIS people can read informational panels about the telescopes, geological resources, and the wēkiu bug. To educate visitors on Mauna Kea and its unique environment, there are interactive displays, handouts, and videos (derived from the First Light video (PBS Hawaii 2004)) (see Section 4.4). The VIS staff and rangers play the videos for visitors, including Japanese language versions, either upon request or if they are trying to garner interest from visitors. Mauna Kea: A Guide to Hawai‘i’s Sacred Mountain was published in 2005 as a guide for visitors on the cultural and natural history of the mountain, along with its current activities and uses (Lang and Byrne 2005). Many visitors focus their attention at the VIS on the gift shop and spend little time looking at the displays or watching videos (VIS Staff 2007). The proceeds from the store go into a revolving fund to support VIS activities. There are public restrooms at the VIS, a few outdoor picnic tables, self-serve hot beverages, and covered trash receptacles outdoors.

The VIS offers an evening stargazing program that has gained popularity in recent years and is now offered daily. Many visitors attend this program after driving to the summit to watch the sunset. The stargazing program is not available to those participating in commercial tours. The VIS also conducts weekly summit tours every Saturday at 1 p.m., to visit the Keck I and the UH 2.2m telescopes. Guests must provide their own 4-wheel drive transportation to the summit. Attendance varies, but on occasion the VIS will have up to 20 or 30 people participating in these tours (VIS Staff 2007). Other VIS events include monthly programs on current astronomical research occurring on Mauna Kea and presentations  on cultural aspects of Mauna Kea. Public stargazing programs and all VIS activities are provided free of charge using donations from the observatory facilities. Preliminary plans exist for expanding and improving the VIS facilities to provide an enhanced experience for visitors (Group 70 International 2000; Good 2003).

The Keck Observatory has a visitor gallery that is usually open to the public during the weekdays. The Keck Observatory has a 15-minute video, an interactive kiosk, two public restrooms and a viewing area with partial views of the Keck I telescope and dome. The Subaru Observatory offers guided tours of their facility for those who sign up on their website. In addition to the public restrooms at the VIS and the Keck Observatory (where availability is limited), there are some portable latrines in the summit vicinity.

3.1.3.2       Rangers

OMKM officially began its ranger program in 2002, after an initial trial program (June 2001) designed to help determine their duties and responsibilities. There are five full time ranger positions. The rangers are hired by MKSS using OMKM funds. Two rangers are on duty daily, with three working on Saturdays. Shifts are three days on (two thirteen hour days and one twelve hour day) per week, and four days off, with most rangers staying at the Hale Pōhaku facilities during their days on. The rangers typically have diverse backgrounds, from those with cultural ties to the land, to those drawn to the mountain because of astronomy, to those looking to share their knowledge about the important natural resources of the area. Rangers receive on the job training from other rangers.

OMKM Rangers use the VIS as a base when they are not out on patrol, providing an additional resource to visitors. A key function of the rangers is to ensure the safety of visitors to Mauna Kea. Rangers advise visitors of weather conditions, the potential hazards associated with ascending the mountain (e.g., altitude sickness, road conditions), and recommended approaches to safely visiting Mauna Kea. They provide emergency assistance when necessary, including oxygen and water. Education is another important component of the ranger’s daily activities. They distribute the safety brochure, provide information on the unique natural and cultural resources, identify the various observatories, direct visitors to established hiking trails, and educate visitors on prohibited or destructive activities. Rangers are made available, as needed, to support activities such as movie making, to ensure that impacts by film crews are minimal (e.g., by limiting climbing on hillsides and trampling vegetation). Rangers also perform site maintenance activities including coordinating litter removal (‘an ever present responsibility’) and trail maintenance to deter use of non-established trails.

Rangers conduct patrols by car to the summit four times daily, with the last patrol at sunset. A primary purpose of these patrols is to observe and document the activities of the general public, observatory personnel, and commercial tour operators. This is partially accomplished by monitoring the road and vehicle traffic and documenting the number and types of vehicles at the summit or in transit (e.g., observatory, visitor, commercial). The patrol reports document weather conditions; how many hikers visit Pu‘u Poli‘ahu, Pu‘u Kūkahau‘ula (Pu‘u Wēkiu, summit peak), and Lake Waiau; visitor type and activities; ranger activity (e.g., trail maintenance, litter pick-up); and research activity. These reports are faxed to the OMKM office daily and entered into a Microsoft Access database. This database has records dating back to 2001 and can be queried for information on a variety of topics. Patrols also permit rangers to interact with visitors at the summit who may want information or directions, to evaluate the health and safety of visitors, to educate people on various aspects of Mauna Kea, and to provide guidance on permitted and prohibited activities. Much of the interaction that rangers have with visitors is related to providing them with information that they did not know (this includes people engaging in prohibited activities and walking in areas they should not). The ranger records provide valuable data on use of the area.

The rangers wear uniforms and drive State-owned vehicles identified as ranger vehicles. Although it is not a park, many visitors liken Mauna Kea to national or state parks, and VIS staff report receiving inquires from people looking to get their national park passport book stamped (VIS Staff 2007). Some visitors also believe the rangers have law enforcement powers. Although this perception likely has the benefit of reducing the human impact of visitors (e.g., making them less likely to litter and to respond favorably to requests to stay on trails), the rangers do not have any enforcement authority if they observe misconduct or legal infractions. This lack of authority is linked to the lack of rule-making authority for the area (see Section 1.4.2.3) and reduces OMKM’s overall ability to protect both natural and cultural resources. DLNR Division of Conservation and Resources Enforcement (DOCARE) is tasked with providing enforcement, though personnel do not maintain a presence on Mauna Kea. Twice a year rangers conduct inspections of each observatory for compliance with their conservation district use permits.

3.1.3.3       Hiking

Native Hawaiians traveled to Mauna Kea for religious and healing purposes and to procure stone from the Adze Quarry. According to Maly and Maly (2005), “Travel across the ‘āina mauna (mountain lands) of Mauna Kea is documented in native traditions, which describe ala hele (trails) passing from the coastal lowlands through the forest lands; along the edge of the forests; across the plateau lands of the Pōhakuloa- Ka‘ohe region, and to the summit of Mauna Kea. These ala hele approached Mauna Kea from Hilo, Hāmākua, Kohala, Kona, and Ka‘ū, five of the major districts on the island. Only Puna, which is cut off from direct access to the mountain lands, apparently did not have a direct trail to the ‘āina mauna.” Historic trails, created in the nineteenth and early twentieth centuries, either followed old trails or cut across new areas. They were often traveled on horseback for purposes of forestry, ranching, hunting, and recreation (Maly and Maly 2005). These trails, originating from all directions, form part of Mauna Kea’s landscape, with access paths carved from the lower elevations to Lake Waiau and the summit region. The Kūka‘iau-‘Umikoa Trail served as a route from the Hāmākua area, on the north side of the mountain, to Waiau, on the south side of the summit. The Mauna Kea-Humu‘ula Trail provided access to the summit from the south, originating at the Humu‘ula Sheep Station in the saddle.

Hiking is currently a popular day-use activity for some visitors to Mauna Kea. There are no camping facilities within the UH Management Areas. There are a few established (but unmarked) trails in the summit region and other trails at lower elevations (see Figure 3-8). A general map of the area, with some trail designations, is included within one of the handouts distributed at the VIS: “Visiting Mauna Kea Safely and Responsibly” (developed by OMKM). For those unfamiliar with the area it is advisable to get recommendations and directions from either the VIS staff or the rangers. Ranger reports between 2002 and 2008 suggest that approximately five to six thousand hikers use the summit region trails every year (see Table 3-8) (OMKM unpublished data). This represents the number of hikers counted by the rangers for the five points of access most commonly used: the Mauna Kea Ice Age NAR trail (to Lake Waiau), the Mountain Trail, the trail up Pu‘u Poliahu, the Summit Trail, and the trail along the access road (see Table 3-8). In addition to those who hike the short trail to the summit (Pu‘u Wēkiu), many use a secondary trail to access it from the parking lot of the UH 2.2m telescope. This secondary trail is used approximately 20 times a month by visitors (OMKM Rangers 2007) and crosses documented wēkiu bug habitat. The main trail to the summit is about 300 ft (91 m) up the road, unmarked, and not easily seen. Hikers have also been observed off-trail in the summit region, where they may damage wēkiu bug habitat and disturb previously undisturbed cinder. Rangers will provide directions to people at the summit looking to find the trails, educate people about the sensitive landscapes, and try to discourage use of the secondary trail by sweeping and raking the disturbed track to make it less visible.

One off-road-vehicle trail, 900–1,200 ft (274–366 m) long, to the top of Pu‘u Poli‘ahu, was frequented in the past by drivers looking to explore. The trail was originally cut for the installation of the University of Arizona Site Test Telescope, but in 2001, as vehicle access was not required for any operational needs and because Kahu Kū Mauna was concerned about disturbance to cultural sites and disrespect to Hawaiian culture, OMKM closed the road. MKSS tore up and raked the old road bed and installed large boulders to block vehicle access. A sign denoting Pu‘u Poli‘ahu as a sacred site was also erected. A few hikers now climb the trail, but visits appear to be infrequent (OMKM Rangers 2007). Rangers discourage visitors from visiting the Hau‘oki Crater, documented wēkiu bug habitat, but this is often difficult since the footprint trails persist and attract more visitors. Although it is not part of the MKSR, some visitors to Mauna Kea hike about a mile (from the road) to visit Lake Waiau, and approximately 10 visitors a month walk off-trail up Pu‘u Hau Kea and to the Adze Quarry (OMKM Rangers 2007).

Table 3-8. Number of Hikers by Trail

Data from OMKM unpublished data

Trail	2002	2003	2004	2005	2006	2007	2008
Lake Waiau	719	1,235	1,003	1,076	1,215	1,271	785
Mountain Trail	100	371	347	365	352	441	433
Pu‘u Poliahu	102	142	75	241	77	53	57
Summit Trail	4,198	3,730	3,431	4,885	4,077	3,766	2,909
Access Road	91	258	196	227	166	188	195
Total	5,210	5,736	5,052	6,794	5,887	5,719	4,379

Other trails around Hale Pōhaku include the trail through the silversword exclosure and the dirt access- road/firebreak encircling much of the mountain, which is used primarily by hunters. There are also hiking trails, West Ridge and Pu‘u Kalepeamoa (Lang and Byrne 2005). Outside the UH Management Area boundary, these two trails are located directly across the Summit Access Road from the VIS and lead to Pu‘u Kilohana and Pu‘u Kalepeamoa, respectively. The silversword enclosure trail, West Ridge, and Pu‘ukalepeamoa are the more popular trails (Byrne 2008); however, how often any of these trails is used and by whom is not monitored.

Trails at lower elevations provide additional access points to the mountain. In mid 2007, the Division of Forestry and Wildlife (DOFAW) opened up two existing trails to off-highway recreational use (Kawashima 2008). Both trails are used by hikers, mountain bikers, ATVs, hunters, motorcycles, horses, and 4-wheel-drive vehicles. Two check-in stations have been built for these trails, which serve two primary functions, to document trail use over time and to aid in search and rescue teams when people become lost or injured. Check-in stations are marked by signs and have sign-in sheets for visitors to write their names and activity. The DOFAW Na Ala Hele Trail and Access Program maintains check-in stations for off-highway recreational trails. Check-in stations began collecting visitor data in 2007.

3.1.3.4       Snow-Play

In 1913, Mid-Pacific Magazine carried an image captioned “Wai‘au lake is the only lake in Hawai‘i on which ice skating may be enjoyed” (Frear 1913). The first recorded skier on Mauna Kea took to the  slopes in February 1936, as documented in an article in Paradise of the Pacific (Lewis 1937). Skiing expeditions continued (Dickie 1967), and snow-play (skiing, snowboarding, sledding) is now a common winter pastime on the Big Island when the conditions are right. Some visitors load up the beds of their pick-up trucks with snow and haul it down to lower elevations for “winter” fun. As described in Section 2.1.4, snowfall on Mauna Kea’s summit is sporadic, with the winter months of January–March most likely to have suitable ski days. Other than for plowing the roads (conducted by MKSS) and directing parking, there is no logistical support for snow operations on the summit and it is difficult to control use and access. Rangers close the road at Hale Pōhaku until they receive confirmation that conditions are safe for visitors to proceed up the mountain. Sometimes people wait overnight in their cars for the opportunity. The primary area used for snow play, known as the Poi Bowl, is located directly east of the Caltech Submillimeter Observatory—in part because it is accessible by road both at the top and bottom of the run (see Figure 3-6 and Figure 3-7). This area is utilized by the wēkiu bug (Eiben 2008), although it may not represent prime wēkiu bug habitat (Englund et al. 2006). Many visitors institute a shuttle system to ensure that there is a car and driver to pick them up at the bottom of the run and drive them back to the top. These north-facing slopes maintain the snow the longest. Heavy snowfall brings visitors to a location east of the summit, which is a longer trail that requires a hike from the bottom, back to the roadway. Because there are no designated trails or ski lifts, visitors often hike off-trail hiking to reach the ski runs, and if there is not enough snow they hike on open cinder between the snow-covered areas.

Vehicle and visitor traffic to the summit may be particularly high on snow days, especially when they fall on weekends. Many people (especially locals) visit the mountain only where there is snow (see Figure 3-7). As many as 600 vehicles have been recorded traveling to the summit on heavy snow days, and each of these is likely carrying several passengers (OMKM unpublished data).

As described in Section 2.2.4, mammals were introduced to Hawai‘i in the late 1700s as a local food source. Livestock populations thrived, but landscapes were permanently altered by over-grazing and subsequent erosion. Although animal-control activities were conducted in the early 1900s, maintenance of game animals for hunting was a management goal again by mid-century. By the late 1940s the population of game mammals on Mauna Kea was allowed to increase to enable sustained harvest by hunters. The State maintained facilities at Pōhakuloa to support recreational hunters on Mauna Kea, which was regarded as a “hunter’s paradise” (Anonymous 1948). According to the U.S. Fish and Wildlife Service, “The numbers of feral sheep and goats grazing on the ranges of the various islands also created problems in the loss of habitat—the destruction of cover and subsequent erosion of the soil. Today the goats, sheep, and pigs are classed as game and are hunted as ‘mainlanders’ hunt deer. Hunting, in some areas, has reduced this ‘game’ to such low numbers that seasons must be imposed to insure “future sport” (Department of the Interior: Fish and Wildlife Service 1950). Guides advertised big game hunting expeditions, touting “…as a general rule, shooting can be regarded as ‘guaranteed,’ with a limit of two sheep and two pigs per day” (Collins 1957). As a result of a lawsuit filed to protect designated critical habitat for the endangered Palila, the māmane-naio forest (see Sections 2.2.3.1 and 2.2.4.1.2), a Federal court ordered the eradication of sheep and goats from Mauna Kea, in 1979. Although this goal was nearly achieved by 1981 through State-conducted eradication efforts (Scowcroft and Conrad 1992), the animals are still present on the slopes of Mauna Kea, and hunting continues to be a popular recreational and subsistence activity with local residents.

DLNR has divided hunting areas into units, each with its own set of regulations (see Figure 3-8). On the Big Island, Hunting Unit A is the Mauna Kea Forest Reserve and Game Management Area. Unit K covers the Big Island Natural Area Reserve properties, including the Mauna Kea Ice Age Natural Area Reserve. It operates under the same regulations as Unit A. Adjacent units E and G span Saddle Road and include hunting access points via Pōhakuloa Training Areas 1, 2, 3, and 4. Hunters with vehicles access the mountain through existing trails. Hunters on foot use gates or jump the fence (Mellow 2008).

There are two check stations for hunters to document hunting use, the type and number of animals taken. The Pu‘u Huluhulu Hunter Check Station is located at the bottom of Summit Access Road, for hunters accessing the mountain via Hale Pōhaku, Mauna Kea State Recreation Area, the Pōhakuloa pipeline, and Pōhakuloa Training Area. The Kilohana Hunter Check Station is located on the southwest side of the mountain, off of Saddle Road, at the 43-mile marker. It is accessed via Pu‘u La‘au Road and is for hunters accessing the mountain via this point.

Signs mark the access routes to the hunting areas around the Hale Pōhaku and higher on Summit Access Road. An access/firebreak road often used often by hunters circles the east, north, and west sides of Mauna Kea for 32 mi (52 km), at about 9,000 ft (2,740 m), within the Mauna Kea Forest Reserve. It is a 4-wheel-drive road, unpaved and infrequently maintained, and is accessed from just below Hale Pōhaku or from the Kilohana Hunter Check Station.

In Unit A, DLNR regulations (Title 13, Chapter 123, Rules Regulating Game Mammal Hunting) limit hunting to wild pigs, wild sheep, and wild goats. Pig, sheep, and goat hunting is year-round, with a bag limit of one pig per hunter per day and no bag limit for sheep or goats, nor is there a requirement of evidence of sex or species. Although there are no statistics, bird hunting (Title 13, Chapter 122, Rules Regulating Game Bird Hunting, Field Trials, and Commercial Shooting Preserves) is likely minimal in the summit area because few birds are sighted; it is more common around Hale Pōhaku, as there are many game bird species in the subalpine area. Although hunters are known to start looking for animals as far up as 12,000 ft (3,660 m), mammal hunting typically takes place at lower elevations on Mauna Kea in the DLNR Mauna Kea Forest Reserve where the animals are more numerous.

Hunter check-in station data is collected following a fiscal year (July 1-June 30) and tallied by the DOFAW office in Hilo. The number of hunting trips to hunting areas on Mauna Kea increased from 394 in 2005 to 1,091 in 2007 and 1,356 in 2008; however the average mammal take remained the same with approximately 0.3 animals per hunting trip (DOFAW Game Mammal Harvest Report).¹⁰ Number of game bird hunting trips increased from 1,453 in 2004 to 2,765 in 2007 and was 1,909 in 2008. Bird harvest numbers remained relatively consistent with approximately 1-2 birds per the average 3.8-hour hunting trip (DOFAW Game Bird Harvest Report). Harvest data for the two Mauna Kea check-in stations for 2004- 2008 mammal and bird harvests are presented in Table 3-9 and Table 3-10. As funding permits, additional hunting data is also collected through hunter surveys conducted by DOFAW and the results of both data sets reported in the Pittman-Robertson Annual Performance Report for the Game Management  Program.¹¹ Result summaries reported within the FY2005 Annual Performance Report indicate a significant and systematic but uneven potential underestimate of hunter effort and harvest by check-in station data (Johnson 2008). As described in Section 2.2.4, DOFAW conducts feral ungulate control on Mauna Kea, during which 200 to 500 sheep, goats, and mouflon are removed from Mauna Kea each year

10 The game harvest reports for Mauna Kea are available from the DOFAW Hilo office.

11 The Federal Aid in Wildlife Restoration Act, commonly called the Pittman-Robertson Wildlife Restoration Act, was initially passed in 1937. The Act authorizes the Secretary of the Interior to provide federal aid to state fish and game departments for wildlife restoration projects.

Daytime hunting is a permitted use in the MKSR under the terms of the lease between UH and BLNR, “pursuant to the rules and regulations of the Board” (DLNR 1995). The lease stipulates that hunting “must be coordinated with the activities of UH.” Commercial hunting operations are prohibited in the MKSR under the 1995 Management Plan.

Table 3-9. 2004 – 2008 Mauna Kea Check-in Station Data: Mammals

Data from Hawai‘i Island Game Mammal Harvest Report

Year*	Feral Sheep Rams          Ewes		Mouflon Sheep Rams          Ewes		Pigs Boars        Sows		Goats Billys         Nannys		Annual Totals
2004	0	0	59	46	13	14	0	0	132
2005	7	5	48	28	12	9	0	0	109
2006	1	0	95	66	10	14	0	0	186
2007	0	0	215	116	13	15	0	0	359
2008	0	0	272	158	25	9	0	0	464
Total	8	5	689	414	73	61	0	0	1250

* Fiscal year, July 1-June 30

Table 3-10. 2004 – 2008 Mauna Kea Check-in Station Data: Birds

Data from Hawai‘i Island Game Bird Harvest Report

Year*	Quail	Pheasant	Chukar	Francolin	Turkey	Dove	Grouse	Peafowl
2004	791	66	1088	652	129	1	0	0
2005	1676	62	1349	1022	91	3	0	0
2006	1256	150	1338	1353	163	8	0	0
2007	1226	157	1574	1093	209	31	0	0
2008	493	78	627	659	22	0	0	0
Total	5442	513	5976	4779	614	43	0	0

* Fiscal year, July 1-June 30

3.1.4        Commercial Activities

3.1.4.1       Commercial Tours

Commercial tours are a popular way for out-of-town visitors, including cruise ship passengers, to journey to Mauna Kea. Since most rental car companies prohibit the use of their vehicles on Saddle Road, and a 4-wheel-drive vehicle is recommended for driving to the summit, many individuals choose to join an organized tour. DLNR was legally responsible for the commercial permits through 2005, when the UH Board of Regents (BOR) accepted official responsibility to regulate commercial activities on UH-leased lands on Mauna Kea. This change followed the establishment of OMKM and the presence of rangers on the mountain, a presence that DLNR did not have. The BOR gave OMKM the responsibility of issuing permits and collecting fees from the nine commercial operators conducting tours on Mauna Kea. OMKM met with the commercial operators in 2006 to discuss potential changes to the permitting process. Recommendations were reviewed by the MKMB and the Kahu Kū Mauna Council, presented to the BOR, and accepted in November 2006. OMKM revised the terms and conditions of the commercial permit system. It increased monthly fees from $2 per passenger with a minimum monthly fee of $54 to $6 per passenger or a minimum of $1,200/month. The revision also instituted requirements for insurance coverage, a security deposit, penalties for non-compliance, data reporting via daily, monthly, and annual reports, and attendance at periodic meetings. Each of the nine permitted operators is allowed two evening tours per day, with no minimum restrictions on the number of daytime or sunrise tours until further notice (UH Office of Mauna Kea Management Commercial Tour Use Permit Requirements). The maximum number of passengers per vehicle is 14 with a total capacity including the driver not to exceed 15. The number of commercial vehicles in or on the premises is not to exceed 18 at any time and no more than two standard commercial tour vehicles or one modified vehicle per tour operator are allowed in the VIS parking lot at any one time.

Although the frequency of non-permitted commercial tours on the mountain has decreased substantially (OMKM unpublished data), data shows the number of visitors to Mauna Kea via commercial tours is increasing. DLNR estimates that between 1999 and 2005, these numbers grew from 24,164 to 43,877. During this same period yearly fees collected by DLNR increased from $48,562 to $87,838. The fees now being collected by OMKM are expected to exceed $250,000 in FY2008. Unlike the funds collected by DLNR that went into the State’s General Fund, funds collected under OMKM from the permitting  process are deposited into a revolving fund used to support management of the mountain.

A typical evening tour picks up passengers from hotels in either Kona or Hilo and arrives at the VIS around 4pm, allowing time for their clients to eat a picnic lunch and acclimate to the altitude. All commercial vehicles must exit the VIS parking lot by 5pm in order to ensure there is enough parking for individual visitors participating in the free evening public stargazing program. After driving to and spending sunset at the summit, each of the tour operator vehicles descends to a pre-determined viewing location, where they set up telescopes for stargazing. Although the commercial tours generally allow little time for hiking, the short trail to the summit of Pu‘u o Kūkahau‘ula (Pu‘u Wēiku) may be an option. After a few hours of stargazing, the clients are returned to their hotels. The entire tour takes between seven and nine hours. Some companies have begun to offer sunrise tours in addition to the popular evening stargazing tours. Tour vehicles are allowed on the summit from ½ hour before sunrise until ½ hour after sunset. Although the tour guides are knowledgeable about the natural and cultural resources of Mauna Kea, and provide safety information to their clients about the potential dangers associated with rapid ascent to high altitude, there is currently no OMKM requirement that informational materials be presented to tour participants.

3.1.4.2       Other Commercial Activities

As the management body with responsibility for the MKSR, OMKM receives requests for various other commercial uses of Mauna Kea include filming, concessions, bio-prospecting, resource extraction, and special events. Filming is the most common request, and while all permits are initiated through the Hawai‘i Film office, OMKM has the responsibility for reviewing and approving the applications. OMKM currently receives about 30 requests for filming every year, most of which are granted. Ranger support is provided to film crews on Mauna Kea to educate them and minimize potential negative impacts on the mountain’s resources. Currently each use request is considered by OMKM staff for compatibility with the overall mission of the Master Plan. All film requests are reviewed by OMKM, which may consult with observatories and MKSS to ensure the proposed activity would not interfere with their operations.

3.1.5 Cultural and Religious Practices

3.2   Impacts and Threats

The Mauna Kea ecosystem is unique and easily disturbed. Many of the human use impacts stem from uneducated visitors (see Section 4.4, Education and Outreach) and loosely regulated and minimally managed access. Concerns related to access extend to all types of users, including those associated with the observatories and other scientific activities, recreational and commercial users, and those participating in cultural practices. Potential impacts include: pollution, construction activities (dust, traffic, water use), visual disruption, habitat alteration (including disturbance of previously undisturbed natural areas), disturbance of cultural sites, and use conflicts. Threats from various user groups will vary in type and intensity, factors that must be considered when developing management recommendations. Increased emphasis on educating all users through outreach and on-site programs, along with stricter access management has the potential to reduce the severity of threats and their impact on natural resources.

The following sections describe the threats and related impacts of human use to natural resources.¹² Since many of the impacts result from more than one user group, the discussion is organized by type of impact, with relative impact levels from the user groups described to the extent possible.

12 Many of these threats and related impacts also affect cultural resources. See McCoy et al. (2009) for details.

The surfaces of cinder cones and adjacent lava fields on Mauna Kea are vulnerable to geomorphologic alterations caused by direct human contact (see Section 2.1.2). Continued hiking and walking over the cones crushes small, individual pieces of cinder leaving trails and footpaths that may negatively affect the viewshed and create dust-sized particles that can be wind-blown. Fugitive dust generated off of trails, unpaved road sections, and other exposed areas, as well as from construction activities is an ongoing concern to resource managers (see Section 3.2.2).

Infrastructure impacts on wēkiu bug habitat. Since the 1960s, approximately 62 acres (25 hectares) of potential wēkiu bug and other arthropod habitat has been lost to infrastructure development, including roads, parking lots and telescope facilities (Richardson 2002). The first comprehensive assessment of arthropods inhabiting Mauna Kea’s summit documented the vulnerability of their habitat to construction activities (Howarth and Stone 1982). More construction has occurred since then, resulting in damage to the tephra cinder habitat from crushing, grading, obliteration by infrastructure, and dust generation. One particular development, the construction of the Subaru Telescope (completed in 1999), resulted in loss of habitat on Pu‘u Hau Oki, where high numbers of wēkiu bugs had previously been found. DLNR approved a construction grading plan that allowed the summit of this cinder cone to be cut and graded and sidecast material pushed into the crater—filling it to a depth of approximately 40 ft (12.2 m), and excavation of the crater rim, resulting in a horseshoe-shaped crater. The material was subsequently graded to reduce visual impact. A study conducted in 1999 demonstrated that wēkiu bugs were still fairly abundant on Pu‘u Hau Oki, in the areas of the inner crater walls and crater bottom that had been modified during construction of the observatory. This suggests that the wēkiu bugs are able to recolonize previously disturbed areas (Howarth et al. 1999).

The main activities that disturb cinder include

• Road grading and travel by vehicles
• Hiking and off-road vehicle use
• Activities associated with infrastructure, such as construction, decommissioning, and removal; installation and maintenance of utilities
• Scientific inquiry

Road grading and travel by vehicles. Road grading is conducted approximately three times per week on the unpaved portion of the Summit Access Road, to eliminate washboarded areas. Grading activities also spread cinder that is collected from road kickouts along the length of unpaved sections of the access road. An extremely porous and friable material, the cinder is easily crushed by vehicles traveling up and down the mountain. In 2008, gravel from neighboring Pōhakuloa Training Area Rock Quarry was brought in as substitute for the cinder (Koehler 2008). Potential secondary impacts from importing gravel include introduction of invasive species and the use of bonding chemicals of unknown fate and transport through the environment. Accidents are also possible, and disturbance of native cinder may result if the vehicle is pushed off of the road or from recovery efforts. Between mid 2001 and early 2008, OMKM rangers reported 52 accidents along the access road from Hale Pōhaku to the summit, 41 of which occurred above Hale Pōhaku (OMKM unpublished data).

Hiking and off-road vehicle use. Within MKSR there are several trails that could be considered established, however only four are monitored by OMKM rangers: the trail to Lake Waiau, the Mountain Trail, the trail to Pu‘u Poliahu, and the trail to the summit. There are no permanent markers identifying any of the trails. During times of no snow, the established trails are easily seen and provide well-defined paths guiding visitors to places of interest. When snow is present, hikers choose more random paths across the landscape. New trails are created when visitors or researchers opt to explore new terrain, compacting substrate, disturbing areas adjacent to the path, and expanding the existing trail network.

New trails are easily etched into the landscape making them obvious to other users and impacting the existing viewscape (see Section 2.1.2) (see Figure 3-9). Due to lack of signage and a maintained trail network, a faint trail used infrequently may be discovered by others and become more established and impacted. Trails exposed to high use that have not been designed to minimize impacts to viewshed or reduce vulnerability to erosion processes may accelerate local surface erosion and viewshed impacts. Such trails may become hazardous as the slope becomes steeper. Over time, many small irreversible impacts such as ground cover disturbance and compaction may have a negative effect on existing biological communities, such as the arthropod community that requires loose cinder. Except where the snow is deep enough to completely cushion the impacts of footsteps, these types of impacts to the ground surface most likely occur whether snow is present or not. In addition, unrestricted access during the winter months may give the false impression that similar activities are non-problematic when the snow is gone, which is not the case.

Recreational use of 4-wheel-drive and other all-terrain vehicles (ATVs) is known to occur on the lands surrounding the MKSR. Not permitted within MKSR property boundaries, instances of such activity have been very infrequent, and are promptly stopped by MKSS personnel and OMKM Rangers. In areas where cinder features are located, impacts from the vehicles are similar to those associated with foot trails: crushing of cinders, compaction of surfaces, and generation of dust. The primary differences between foot and vehicle trails are in the severity of the impacts due to heavier weight of ATVs and their larger “footprint” and in the potential for petrochemical spills.

Infrastructure. The cinder cones of the Mauna Kea summit region are some of the most pristine and well preserved cones anywhere in Hawai‘i. Of the more than 20 cinder cones in the MKSR, only five show signs of human modification from the construction of observatories and supporting infrastructure (Pu‘u Poli‘ahu – road; Pu‘u Kea, Pu‘u Hauoki and the unnamed cone immediately west of Pu‘u Hauoki – grading for observatory sites; south and west slopes of Pu‘u Wēkiu – road construction) (Lockwood 2000). Some of this change may be irreversible, as restoring sites to their pre-impact topographic geometries will prove nearly impossible, such as has occurred on cinder cones that have been flattened or craters filled (see Section 4.3). Siting of new or redeveloped telescopes on existing or previously disturbed sites can help minimize impacts to the biological and physical environments. However, future proposed facilities may impact previously undisturbed areas. Decommissioning activities will likely involve some earth movement when removing structures or re-grading sites, although most of the area will have been previously disturbed. Infrastructure installation and improvements associated with new or redeveloped facilities (e.g., power supply and communications infrastructure, wastewater treatment facilities) may necessitate sub-surface work, including excavating utility trenches and other structures. Construction activities can disturb cinders through crushing by heavy equipment; excavation and disposal of excavated material; grading and filling; drilling for piles, piers, footings and foundations; and destabilization or sidecasting of cinder associated with removal of retaining walls. Both construction and decommissioning will result in increased vehicular traffic and higher axel weights.

Scientific inquiry. Geological and hydrological research at Mauna Kea has included limited invasive investigations (e.g., excavating, drilling holes) that disturbed both surface and sub-surface features. The tools, machinery, equipment, or chemicals used as part of these investigations are potential sources of direct impacts to surface and sub-surface resources. Impacts to the geology and associated viewscape can also occur as researchers walk to and from specific sites, disturbing and crushing cinder and incising trails into the landscape.

3.2.2        Air Pollution

Currently the air quality at the summit of Mauna Kea is thought to be quite good (based on air quality measurements taken on Mauna Loa), although it is not actively monitored. Human-caused contributors to air pollution at the summit include vehicle exhaust, chemical fumes from observatory construction and maintenance activities, and fugitive dust from road grading and construction or other activities conducted on unpaved surfaces. Although air pollution is not now considered to be a pressing issue, as vehicular traffic to the summit increases, the impact of vehicles on air quality, from exhaust and dust generation, can be expected to increase as well.

Dust deposited on snowfields has the potential to decrease surface albedo, which accelerates snow melt, and increases thaw-depth of permafrost (Walker and Everett 1987). Additionally, fine, aerosol-sized particles that become suspended in the airshed above the telescopes may adversely affect cosmic viewing.

Air pollution (including dust) can impact vascular plants in several ways, including greatly reducing photosynthesis, transpiration, and water use efficiency; increasing leaf temperatures (with potentially serious effects during periods of high temperatures); and by lowering primary production (growth) (Sharifi et al. 1997). Air pollution and dust are also known to impact the growth of lichen and moss and community diversity (Hutchinson et al. 1996).

Dust impacts on wēkiu bug habitat. It is hypothesized that deposition of dust can adversely impact wēkiu bug habitat by filling interstitial spaces between cinder, one of the vital components of the bugs habitat (see Section 2.2.3) (Howarth et al. 1999). In addition to filling the interstitial spaces used by wēkiu bugs, dust can have a direct impact on some insects, by acting as a desiccant (Alstad et al. 1982). It is unknown whether wēkiu bugs, or other summit arthropods, are susceptible to desiccation by dust. Although these species are adapted to living in very arid conditions, a heavy dust layer could potentially desiccate wēkiu bug eggs, or make it more difficult for wēkiu bugs to obtain the moisture they need from substrate (Eiben 2008). Additionally, a heavy layer of dust could bury prey items, making foraging more difficult (Eiben 2008).

The main activities that have the potential to generate air pollution include

• Road grading and travel by vehicles
• Activities associated with infrastructure, such as construction, decommissioning, and removal; installation and maintenance of utilities

Road grading and travel by vehicles. Road grading and vehicle traffic are currently the most significant contributors to dust generation on Mauna Kea (see Figure 3-10). Vehicles crush and kick-up cinder as they travel on the unpaved portion of the Summit Access Road or the 4-wheel-drive roads, creating a great deal of airborne dust. The dust disperses along the road corridors, coating the ground surface. Vehicle exhaust is another potential contributor to air pollution. See Section 2.1.5.1 for more information on air pollution generators and potential impacts to air quality.

Infrastructure. Construction and maintenance activities contribute to air pollution through vehicle and heavy equipment exhaust, dust generation, and use of volatile compounds during building construction and routine maintenance (cleaning). The use of best management practices to control dust during these activities (e.g., using water to wet down sites) helps to limit the amount of airborne particles. The impacts to air quality from other pollutants are likely to be temporary because the nearly constant winds at the summit would quickly disperse the pollutants these activities generate.

3.2.3        Substrate and Groundwater Contamination

Contamination of soils, substrates, Lake Waiau, groundwater, and aquifers is a potential side effect of a variety of human activities on the mountain. If significant, contaminant releases may have adverse effects on biological and water resources, human health, and visual resources (e.g., discoloring). Spills can fill interstitial spaces or result in the disturbance of cinder substrate. Transport of contaminants through the substrate has the potential to impact the quality of both surface water and groundwater. Direct toxic impacts on flora or fauna are also possible. The highest probability of impact is from petroleum products (e.g., fuel for vehicles and backup generators, lubricants, and cleaning fluids) and human waste. The main activities that have the potential to impact substrate and water quality include

• Travel by vehicles
• Release of hazardous material and petroleum product use by observatories and support operations
• Transport of hazardous materials off-site
• Sewage generation

Travel by vehicles. Vehicles are a potential pathway for release of liquid petroleum products into the environment, primarily through leaking (e.g., fuel lines, break lines, coolant), but also as a result of accidents and spills. Little information exists on the current extent of this problem, but it is possible to make some informed predictions about how the range of users, vehicle types, and use-levels relate to the potential threat level. Four-wheel-drive vehicles are prone to under-carriage damage that may not be apparent to operators, resulting in fugitive releases of contaminants. Their use in off-road environments puts less-frequently visited and monitored areas at risk. Recreational users traveling to Mauna Kea are either tourists in rental cars or local residents. Although vehicles visiting the summit stay on the road, accident-related discharge of contaminants and generation of fugitive contaminants from leaks are potential impacts. Vehicles operated by staff (observatories, OMKM, MKSS) comprise a high percentage of traffic to the summit, but are regularly maintained to ensure reliable performance, which reduces potential of fluid leakage. The frequency of use of vehicles and machinery for construction and maintenance depends upon need and is variable. While these vehicles are also potential sources of petroleum leaks, they, like the staff vehicles, are subjected to regular inspections to minimize potential contaminants (Nahakuelua 2008).

Use of hazardous materials. Hazardous materials utilized and stored at the observatories are listed in Table 3-7. The presence and use of hazardous materials at the observatories and the transport of hazardous waste off-site introduces the possibility of spills or leaks. There have been documented incidents, beginning in 1979, involving spills or leakage of hazardous materials (e.g., mercury, diesel fuel, ethlyene glycol, and sewage) (see Table 3-11, (NASA 2005)). These incidents were reported to DOH, and in all cases, clean-up was conducted in accordance with emergency spill response procedures. Some spills occurred inside buildings or on concrete structures, limiting their impact on the outside environment and potential impact on natural resources. Elemental mercury is used in four of the telescopes and is of particular concern to human health. The best available information suggests that while mercury spills have occurred (NASA 2005), spilled amounts occurred during mirror re-aluminizing activities and were small (McNarie 2004). Backup generators use diesel fuel, which is stored on-site. Secondary sources of contamination from generator equipment include waste oil and coolant (e.g., ethylene glycol). Any discharge of lubricants or solvents into the sanitary waste stream or directly onto the landscape outside the buildings could be problematic. In the past, there have been instances in which cinder was contaminated and then excavated to contain the potential effects of the spill; this approach has the secondary effect of cinder disturbance. Impacts are minimized by adhering to approved plans for storage, transport, and emergencies.

Sewage generation. The cesspools, septic tanks, and associated leach fields at the summit and Hale Pōhaku have been designed to meet State DOH permit requirements for sanitary waste systems. With telescope facility upgrades, many of the original cesspools have been replaced with septic tanks. Currently there are eight septic tanks with leach fields or disposal pits and three cesspools (NASA 2005). Solid and liquid waste discharged into these approved systems should minimize direct discharge of solid waste in the effluent and into the ground and allow for physical and bio-processing. However, the fate and transport of the effluent after discharge, and its likely impact on groundwater (either shallow or deep) is almost entirely speculative because the hydrogeology of the summit is poorly understood (see Section 2.1.3).

Surface or sub-surface contamination as a result of sewage leakage is potentially problematic. A two- gallon sewage spill from an incorrectly installed septic line contaminated cinder and snow in wēkiu bug habitat in the Pu‘u Hauoki crater in 1998 (NASA 2005). Although the impacted material was removed and the leak repaired, other leaks outside the sewage systems could result in disturbance and damage to cinder substrate, including wēkiu bug habitat. In March 2008 approximately 500-1,000 gallons of sewage overflowed from the VIS septic tank onto the ground and was reabsorbed. The incident was reported to DOH (Koehler 2008).

Human waste from visitors using the area is another potential pathway for contamination of substrate and sub-surface water. Public restroom facilities at the summit include portable latrines at the UH 2.2m telescope parking lot and the parking lot just past the junction of the access road and the summit loop, and indoor facilities in the Keck Observatory visitor’s gallery. With limited public facilities available for use, the greatest possibility for impact is most likely during snow-play days when there are hundreds of people in the summit area. Portable latrines are pumped weekly and the biosolids trucked off the mountain (see Section 3.1.1.2.6). The potential for accidental spills during weekly transport of the biosolid effluent and associated flush chemicals off the mountain is also a concern.

3.2.4        Erosion

Erosion is a natural process whereby wind, water or ice detaches soil particles and transports them from their original location. In general when water, in either solid or liquid phase, is the eroding agent, movement of particles follows the force of gravity. However, wind transported particles can be lifted and carried to higher elevations in ascending air parcels, resulting in deposition either upslope or downslope. Erosion rates are a function of the erodibility of ground surface and erosivity of the agent inducing particle displacement and transport. Human activities that reduce groundcover and concentrate overland flow increase erodibility and erosivity respectively, resulting in increased erosion rates. The summit area of Mauna Kea is subjected to all three agents of erosion. Due to the prevalence of wind on the mountain, exposed areas, including roads and trails are vulnerable to erosion by wind nearly year round. Erosion rates by water at the summit area are regulated in part by the high porosity of the surface cover allowing infiltration of precipitation into the ground surface, and by limited precipitation. In areas where compaction has occurred infiltration is reduced, resulting in increased runoff and erosivity. Below the summit region, in areas such as Hale Pōhaku, where surface conditions are dominated more by soil than volcanic substrate, erosion rates are higher due to the greater erodibility of the soils (Gerrish 1979). Activities that increase the potential for accelerated erosion include

• Road grading and travel by vehicles
• Hiking and off-road vehicle use
• Activities associated with infrastructure, such as maintenance and construction
• Concentration of storm water runoff generated off buildings and impervious areas

Hiking and off-road vehicle use. Vehicle and foot traffic in unpaved areas can increase erosion, particularly at Hale Pōhaku where groundcover is comprised of finer particles than found in the summit region. Also, more people visit Hale Pōhaku, including visitors, residents, and staff. For example, two of the more popular hikes are the West Ridge and Pu‘ukalepeamoa trails located just outside the UH Management Areas boundary and immediately across from the VIS. These trails, in addition to being on steep slopes and consisting of poorly designed pathways, are at many points completely devoid of cinder groundcover, leaving only extremely fine particles and a compacted trail base. The compacted base of the pathway limits the amount of water that can percolate into the ground, increasing the volume of water flowing downhill and removing sediment along the way. On the steeper sections, the finer particles create a slippery surface and hikers step out onto adjacent vegetated and rocky areas with better footing. This results in an increased footprint of the path, accelerated erosion, and scaring of the landscape.

Like foot traffic, ad hoc vehicle paths can also result in the concentration of runoff and associated increases in erodibility and soil loss. While immediate impacts from vehicles are likely more severe than those from foot traffic, due to the greater weight and larger footprint of the vehicles, in both instances, reoccurring disturbance of any area will cause significant compaction and degradation, substantially increasing erosion potential and extent.

Infrastructure. As a result of the low amount of precipitation and the high porosity of the ground surface at the summit, there is no evidence of erosion by surface runoff at any of the existing observatory sites (NASA 2005). During preparation of this report the authors visited locations throughout summit area and did not observe significant erosion caused by runoff off observatory structures. Runoff is generated from the impervious sections of the Summit Access Road and is routed into drainage ditches aligned along the road shoulder. Spaced intermittently along the Summit Access Road, large culverts underneath the road drain water from the upslope or mountain side of the road to the downslope side. While the volume of water moving through these systems is often small and intermittent, over time the fragile ground cover where the water pours out of the culverts is impacted. Movement of the substrate is evidenced by rocks found in the drainage ditches and cinder-filled drainage holes of the retaining walls (see Figure 3-11). Silt, rocks and small boulders fill culvert inlets, and rills and gullies at culvert mouths gradually become larger and more incised as the water travels downslope (see Figure 3-12). These processes are intermittent and changes are gradual, but they have the potential to eventually undermine the integrity of the road and negatively impact the viewshed.

Road drainages are cleaned approximately every four to five years and cinder buildup behind retaining walls has necessitated emptying only twice within the past 18 years (Koehler 2008). Construction activities can increase the potential for erosion. At summit locations, construction-related disturbance and

13 Headcut is a location where a sudden change in ground elevation occurs, usually at the leading edge of a gully. Headcuts often result in rapid erosion and incision of the runoff channel.

crushing of cinder ground cover would most likely increase displacement of cinder to down-slope locations and increase the potential for dust generation and its movement offsite.

Due to greater visitor counts and concentrations at Hale Pōhaku than at the summit, erosion impacts appear to be more extensive. Etched pathways and blocked drainage culverts were noted in the vicinity of the VIS parking lot and there are currently no sidewalks, signs, or other infrastructure guiding the visitor to specific destinations and away from ‘natural’ areas. When the parking lot is full, visitors park along the Summit Access Road. This creates a potential safety issue for those walking along roadside drainages, as well as concern for impacts to the landscape due to trampling. The unsightly appearance of the area may also have a negative effect on how visitor perceive the need to care for the mountain (see Figure 3-13).

3.2.5        Solid Waste Generation

Litter and larger fugitive trash impacts the visual aesthetics of the MKSR and degrades the landscape. In addition, it may interfere with deposition of food resources in the aeolian ecosystem, shade out vegetation, and damage geological resources upon impact. Food waste may provide a resource to support pest species and predators of native biota. Collection of debris is also of concern as the removal activity may do more harm than the actual debris if people or vehicles crush cinder in sensitive habitats (Howarth et al. 1999). The main activities and users that produce solid waste include

• Observatories and support facilities (trash)
• Construction (materials)
• Recreational users (litter, snow-play debris)
• Commercial tour groups (litter)
• Cultural practices (offerings)

Trash generated by the observatories and Hale Pōhaku is contained, collected on a regular basis and transported off-site to approved sanitary waste facilities. The potential exists for some of this trash to escape collection or to be blown about as a result of high winds at the summit. Similar concerns exist for construction material and debris, though best management practices are implemented to reduce the extent. Recreational and commercial tourists also contribute to debris found on the mountain. Litter (e.g., cigarette butts, plastic bags, broken glass) may result from the active discarding of waste or inadvertent disposal (e.g., as a result of high winds). Rangers report pick-up trucks heading to the summit on snow days with loose trash in their beds, which is likely to blow out in the windy conditions at the summit (OMKM Rangers 2007). In the spring, when the snow melts, snow-play trash is found on the mountain, including broken skis, snowboards, and general litter. While not “trash,” cultural and spiritual offerings are another source of uncontained material often left in the summit area.

3.2.6        Noise Generation

The primary receivers that might be disrupted by excessive noise are the human users of the mountain (e.g., scientists, cultural practitioners, recreational users). There is also the potential that noise generated by certain activities or systems would have an impact on biological resources. The main activities that produce sound to levels above natural background resulting in the generation of noise include

• Travel by vehicles
• Observatory operations
• Construction operations (e.g., heavy equipment use, drilling, excavation)

Ambient sound levels at Mauna Kea are low, with vehicle traffic and wind providing the dominant background. Observatory operations create minimal noise, while construction activities create intermittent, though sometimes significant, disruptions (see Section 2.1.5). An example of one potential noise impact illustrates the types of considerations that are currently evaluated prior to implementing a project. The Subaru Telescope requested permission to install a Sound Detection and Ranging (SODAR) unit on the roof of their control building to monitor local wind speeds and the thermodynamic structure of the atmosphere. In response to concerns from the MKMB, studies were conducted to evaluate the transmission of sounds (called “pings”) from the system. Results of this test, combined with an expert opinion from a member of the MKMB Environment Committee, provided information allowing the conclusion that the frequency and decibel level of the pings would not pose a problem for the occasional bird in the vicinity or for resident insect life.

3.2.7        Invasive Species

The potential impacts of invasive species on the fragile ecosystems of Mauna Kea are described in detail in Section 2.2. Many of the mountain’s native ecosystems have already been impacted by introduced animals and plants. However, invasives remain a continuing threat. Virtually any user, vehicle, equipment, or material that comes to Mauna Kea can be an unintentional carrier. Although Mauna Kea’s higher elevations are somewhat insulated from invasives, due to its inhospitable environment, certain species have been able to survive. The main activities and users that may introduce invasive species include

• Construction and maintenance (materials, vehicles, equipment)
• Road grading (importing gravel)
• Landscaping (materials, at lower elevations)
• Observatories and support facilities (materials, vehicles, researchers)
• Recreational users (hikers, hunters; footwear and vehicles)
• Cultural practitioners (offerings)

Construction and maintenance. Construction and maintenance activities may introduce invasive species to Hale Pōhaku and MKSR through several pathways. Invasive species can be transported on footwear or tires, on heavy equipment, in fill material, or can be contained in shipments of materials. Most species introduced through these pathways will be small (seeds, insects), although larger species (such as rodents) may be found in shipping containers containing supplies or equipment.

Road grading. Like construction activities, road grading can introduce invasive species through contaminated materials (e.g., ants living in gravel brought in for the road) and on the equipment used to deliver gravel. The grading equipment is housed at Hale Pōhaku and can pick up “hitchhikers” at the storage yard and carry them up the summit road. Until 2008, all material used on the roads was cinder obtained from areas adjacent to the roadway (Koehler 2008). At the request of MKSS, the MKMB Environment Committee reviewed and concurred with a proposal to import gravel from a quarry located within the Army’s Pōhakuloa Training Area (PTA). Required inspection protocol includes thorough inspection of all gravel brought to Mauna Kea for ants by an entomologist and wash-down of the delivery trucks (MKMB Environment Committee 2007).

Landscaping (lower elevations). Landscaping materials, such as plants and mulch, used at facilities can also harbor invasive species, as can equipment, clothing, and the shoes of the landscaping staff. Currently, minimal amounts of landscaping materials are utilized at Hale Pōhaku and there is no outdoor  landscaping at the summit facilities. According to the terms of the lease between the Board of Land and Natural Resources (BLNR) and UH (Lease No. S-4191), “In order to prevent the introduction of undesirable plant species in the area, the Lessee shall not plant any trees, shrubs, flowers, or other plants in the leased area except those approved for such planting by the Chairman.” Invasive species such as eucalyptus trees and California poppies have been purposefully planted in the past at Hale Pōhaku as part of landscaping or reforestation projects.

Observatories and support facilities (materials, vehicles, researchers). Invasive species can be accidentally transported in the goods and belongings of visiting scientists, on and in equipment  transferred from other astronomical facilities or manufacturing plants, and via vehicles traveling to and from other parts of the island to the mid-level and summit facilities. There is already at least one high- elevation-adapted invasive species at the summit (a ground-hunting spider, Meriola arcifera) that  possibly arrived from another astronomical site in South America (see Section 2.2.2.2.2). Researchers (such as biologists) who use off road vehicles (or hike in muddy areas) at other locations on the Big  Island and then travel to Hale Pōhaku or the MKSR are also potential vectors for invasive species. However, it should be noted that a good proportion of these researchers are aware of the problem and clean their equipment and shoes before moving between areas.

Recreational users (hikers, hunters). Hikers and hunters can inadvertently function as vectors and import invasive species to Hale Pōhaku and MKSR through seeds stuck in the mud on their hiking shoes and vehicle tires, and through deliberate releases (to establish populations of game species). Most of the deliberate introductions of game species (birds, mammals) occurred in the past, but it is possible that new introductions could still occur illegally. Hunters operating 4-wheel-drive vehicles are likely to spread seeds and invertebrates from other portions of the island, due to their extensive use of unpaved hunting roads, which often have plants, especially weedy species, growing in or alongside of them (Thomas 2008). Anyone who has driven a 4-wheel-drive vehicle off-road is familiar with the large amount of material picked up from mud puddles, roadside weeds, and dusty areas. Although hunters may have an impact on the areas’ natural resources through general access and use, the removal of non-native vertebrates (e.g., sheep, mouflon) has a beneficial effect on the habitat by eliminating individual animals that damage native vegetation. However, current hunting rates are not high enough to eliminate the population of feral sheep on Mauna Kea, even in conjunction with current feral ungulate control efforts conducted by DOFAW (see Section 2.2.4 and Section 3.1.3.5). Tourists arriving at Hale Pōhaku and MKSR in rented vehicles are also capable of spreading invasive species, although to a lesser extent, due to the fact that they usually stick to paved roads and their vehicles are regularly washed at the rental agencies.

Cultural practitioners. Cultural practitioners can spread invasive species in ways similar to recreational users. In addition, many cultural practitioners bring items that are intentionally left on site as offerings. These items may harbor invasive species, unbeknownst to the practitioners, such as weed seeds and insects, or they may attract non-native birds or small mammals looking for food.

3.2.8        Habitat Alteration

Habitats on Mauna Kea are home to unique species, some found nowhere else in the world. Any discussion of habitat disturbance necessarily involves other threats, including cinder disturbance, invasive species, and pollution. The impacts of habitat alteration on resident species are described in Section 2.2. The main activities and users that cause habitat disturbance include

• Construction and infrastructure
• Off-road vehicles and off-trail hiking
• Recreational users (hikers, snow-players, and hunters [through off-trail use and feral ungulates])
• Cultural practitioners (off-trail use)
• Scientific inquiry (off-trail use, direct sampling)

Construction and infrastructure. Previous summit development has disturbed areas of wēkiu bug habitat, and construction at Hale Pōhaku has resulted in the removal of small areas of māmane woodlands. Building facilities in new areas at the summit may disturb the habitat of lichens and resident native arthropods in areas of new facilities and associated construction. Structures may alter wind patterns, change the pattern of snow drifts, and affect the natural deposition of aeolian drift that supports the wēkiu bug and other arthropod populations. The alteration of wind patterns could either enhance or reduce the quality of wēkiu bug habitat, depending on how windflow patterns are altered.

Off-road vehicles and off-trail hiking. Off-road vehicle use and off-trail hiking can impact habitats through crushing of cinder (MKSR) or increased erosion (Hale Pōhaku), and through direct damage to native flora and fauna (e.g., crushing, trampling). There are no permanent barriers preventing vehicles from leaving the Summit Access Road, although access points where off road driving is known to occur are blocked with rocks. It has been postulated that people may be less aware of the sensitive habitat when they visit Mauna Kea in the summertime, since they are used to having free range over the slopes in the winter (OMKM Rangers 2007).

Recreational, cultural and scientific uses. Hikers, hunters, cultural practitioners, and researchers can all alter habitat by trampling, removing plant and animal material, introducing invasive species, creating new trails and creating new structures, such as shrines. The past has seen the greatest and most devastating impacts to the subalpine and alpine plant communities on Mauna Kea through the intentional maintenance of feral ungulate populations for recreational hunting. These feral ungulates have nearly destroyed the māmane woodlands and have reduced the once abundant Mauna Kea silversword populations to near zero. The impact of feral ungulates on the natural resources of high elevation areas of Mauna Kea far outweigh any other impact from human activities within the subalpine and alpine¹⁴ environments. Other invasive species, introduced accidentally or on purpose, have also contributed to the destruction of the subalpine and alpine communities. Invasive plants (primarily grasses) work in conjunction with feral mammals to suppress regeneration of māmane woodlands. See Sections 2.2.1 and 2.2.4 for more information.

3.2.9        Sample Collection and Incidental Take

Human activities on Mauna Kea can result in the reduction of plant and animal populations through both sample collection and incidental take (the unknowing or accidental killing or removing of an organism). Sample collection occurs mainly as the result of scientific research conducted at Hale Pōhaku and the

14 Excluding the summit (alpine stone desert) ecosystem, where feral ungulates do not occur.

3.2.10    Fire

Although there are few vegetated areas susceptible to fire in the MKSR, fire is a potential threat to habitat in the subalpine zone at Hale Pōhaku. Prior to the introduction of invasive grass species, wildfires were most likely infrequent in the subalpine zone. Invasive grasses increase the risk of fire in the subalpine zone by providing a source of continuous fine fuels in areas that previously had naturally discontinuous fuel beds, due to the patchy nature of the subalpine communities (Smith and Tunison 1992; Hess et al. 1999). These risks have also become greater with the reduction in animal populations that once fed upon the invasive grasses, reducing their fuel load. Potential sources of ignition include vehicles from both accidents and malfunctioning exhaust systems (especially on unpaved hunting roads), improperly disposed cigarettes and matches, arson, camp fires at lower elevations, lightning, and military training activities at PTA. Three major fires have been documented by MKSS, all located on the southern slopes of Mauna Kea, five to ten miles east of the Summit Access Road and below 9,000 feet (Koehler 2008). Control efforts were provided by the County Fire Department, the State Department of Forestry and Wildlife, and Pōhakuloa Training Area. MKSS donated water to the State Department of Forestry and Wildlife to help control these fires.

3.2.11    Climate Change

It is hypothesized that global warming will alter the climate of the Hawaiian Islands by inducing changes to precipitation frequency and amounts. This in turn is expected to alter the spatial distribution and density of flora in both the subalpine and alpine ecosystems of Mauna Kea. The exact changes to the precipitation and temperature regime and subsequently the plant life are unknown due in part to the complexity of the climatic system and the data necessary to generate precise model outputs. It is unlikely that the human-use activities occurring on Mauna Kea are contributing proportionally more to climate change than those occurring at other elevations in Hawai‘i, or at other locations on the Earth. That is, all human activities that involve the consumption of fossil fuels are contributing to global climate change, and any activities that can reduce this consumption will help reduce the impacts of climate change. The potential impacts of climate change on Mauna Kea high-elevation ecosystems are discussed in detail in Section 2.2.1.3.6.

3.2.12    Cumulative Impacts

Each human use or activity that occurs on Mauna Kea may have multiple impacts. Table 3-12 shows the interrelationships between the activities that occur on Mauna Kea and the threats to natural resources identified in the above sections. This table demonstrates the need for a holistic approach to natural resources management—simply controlling one human use or activity is unlikely to eliminate the associated threats. When attempting to reduce the impact of a threat to natural resources, all sources of the threat must be examined and if found to be significant, addressed.

Although the threats to natural resources occurring from the various human uses of Mauna Kea are discussed separately above, in reality, the overall impacts of human activities (combined with those of natural events such as weather patterns) are often greater than the sum of their individual parts. Threats to the survival of the Palila offer a good example of this. The Palila faces the cumulative impact of habitat destruction, the spread of invasive plants, browsing by feral sheep, predation by rats and cats, and the gradual effects of climate change on the distribution of māmane woodlands. Any one of these threats, working alone, would probably not condemn the bird to extinction, or at least could be relatively easily addressed by natural resource managers. However, the combination of these threats, if left unchecked or if treated in a piecemeal manner, will likely result in the loss of this species. Making the matter more difficult, the control of one threat often leads to the worsening of another threat. For example, the removal of feral cats may lead to an increase in rat populations. The example of the Palila identifies the  importance of understanding the complexity of natural systems and the variety of factors that may play into the survival of any given species or resource. Thus it is in the best interest of natural resource managers to identify, prioritize, and attempt to control as many threats to a given resource at the same time as is possible, and to carefully monitor the results of their actions. It is also necessary for each user group to understand that their activities affect the overall status of the natural resources on Mauna Kea, and that no one type of user alone is responsible for the damage occurring there.

4       Introduction to Component Plans

Section 4 of the Mauna Kea Natural Resources Management Plan (NRMP) provides guidelines for the establishment of OMKM’s Natural Resources Management Program (NRM Program). The overarching goal of the NRM Program is to preserve, protect and enhance the natural resources within the UH Management Areas on Mauna Kea, in order to promote long-term sustainable use of the sites. Achieving this goal requires understanding and monitoring of the status of natural resources on Mauna Kea; preventing and controlling threats to natural resources; preserving, enhancing and restoring sensitive ecosystems; conducting education and public outreach; and managing information and natural resources data. Each of these natural resources management needs is addressed in separate component plans.

Section                                      Component Plan
4.1	Natural Resource Inventory, Monitoring and Research
4.2	Threat Prevention and Control
4.3	Natural Resources Preservation, Enhancement, and Restoration
4.4	Education and Outreach
4.5	Information Management

Each component plan explains why it is needed; details the goals and objectives of the component plan; provides a brief review of the current understanding of the natural resources and management needs addressed by the component; and provides recommended management actions to meet the stated goals and objectives. The component plans also identify areas where management needs overlap and resources can be shared while still accomplishing the goals of each component plan. In these cases, readers are referred to other component plans, providing a more accurate overall needs assessment, and enabling easy cross-referencing. Wherever possible, recommended management actions are prioritized according to need, based on our current understanding of the natural resources on UH Management Areas.

The management actions are provided as recommendations and a resource, and it is not the intention of this NRMP that all of the management activities be implemented by OMKM. A subset of the  management actions will be implemented depending on available levels of staffing and funding. Additionally, some management actions may be discovered to not be appropriate upon collection of additional information on the status of the natural resources, through baseline inventory and long-term monitoring efforts. If a recommendation is implemented and it results in an action that would require ground disturbance or alteration of the existing environment, a separate environmental analysis will be conducted in compliance with existing State law. Prioritization of the actions is intended to provide a means to determine which management actions would have the most impact on natural resources protection and management. A high rank indicates that the action would afford the highest level of resource protection, and/or is perceived as being an important management action given the current understanding of natural resources on UH Management Areas.

The management actions detailed in the component plans are based on the principles of ecosystem management¹ and are aimed at maintaining ecosystem integrity, diversity, and health. The description of each management action also considers the potential impacts of conducting the action on natural and cultural resources. Coordination with other agencies, adjacent landowners, and the public, along with development of collaborative initiatives are encouraged whenever possible. Further information on the

1 An ecosystem consists of the plants, animals, and microorganisms within an area, the environment that sustains them, and their interactions. An ecosystem can range in size from a tiny site containing only a few species, such as an isolated wetland, to a huge area containing thousands of species, such as a tropical rainforest.

4.1       Natural Resources Inventory, Monitoring and Research Program

4.1.1    Introduction

Science-based natural resource management depends on obtaining quality data about the status of biological and physical resources. Inventory, Monitoring and Research (IM&R) programs provide these data. Comprehensive and well-designed IM&R programs allow managers to determine the status of natural resources, track changes in resources over time, identify new threats before they become established, measure progress towards meeting management objectives, and plan future research and management (Elzinga et al. 1998; Oakley et al. 2003). Data collected from IM&R programs can assist managers with 1) identifying areas that can be restored and preserved and 2) prioritizing management actions based on geographic area and sensitivity. Results from the IM&R programs can also be used to inform stakeholders of management successes or issues of concern, and to increase public trust and support for management actions. Demonstrated success in implementing management actions may result in an increased likelihood of support and funding for future projects.

A baseline inventory, or initial survey, establishes the current status of the area under management at the beginning of a natural resources management program. Many of the decisions and paths taken by the management program will follow from the results of the baseline inventory. Monitoring begins after the completion of the baseline inventory and tracks selected resources over time. Decisions on what resources to monitor over the long term will be based on the results of the baseline inventory and the objectives of the management program. By design, the baseline inventory is more comprehensive and inclusive than the monitoring program, and therefore it is more labor-intensive and expensive.¹ Research programs may begin after the baseline inventory is completed, or at any time during long-term monitoring. The purpose of the research component is to answer questions and fill in data gaps that are beyond the scope of the inventory and monitoring programs, but are necessary to understand and manage the resources and advance the body of knowledge.

Monitoring and research should not be conducted in isolation, but, rather, integrated with management to allow for implementation of best informed management decisions. Although resource managers are continually required to make management decisions under conditions of uncertainty, monitoring and research data provide a basis for making informed decisions and for revisiting those decisions as new information is gained. The cyclic process of linking monitoring and research with management is called adaptive management (see Figure 1-1). Effective IM&R programs must provide information relevant to current management issues and also anticipate, where possible, future management issues based on what is currently known about the status of the natural resources and threats to these resources (National Park Service 2006). The IM&R programs must be based on the best scientific methods available, produce quality data, and be implemented in a timely manner. Data must be collected and entered regularly and it must be accessible to managers. In turn, managers must integrate new information into their decision- making processes (National Park Service 2006).

Successful IM&R programs require systematic planning, data collection, and analysis of data. The right types of data must be collected to determine if progress is being made towards management goals. Monitoring is repeated over time, and in some cases may extend over long periods before it can be determined that a particular action, or set of actions, has been successful, or that a particular management goal has been met. Because of this, measures of success should include recognition for positive progress

1 At this writing, a baseline natural resources inventory has not been completed for the UH Management Areas on Mauna Kea (Mauna Kea Science Reserve, the Access Road, and Hale Pōhaku), but a cultural resources (archeological) inventory has (McCoy et al. 2009).

A great deal of time, effort, and thought over the years has gone into the development of natural resource inventory and monitoring programs used by natural resource managers. The National Park Service (NPS) has developed excellent guidelines for the development of inventory and monitoring programs, and is in the process of developing standardized monitoring protocols² for many different types of natural resources (see the NPS monitoring page, at http://science.nature.nps.gov/im/monitor/index.cfm). The methodology used for the development of the inventory and monitoring program for UH Management Areas follows the guidelines developed by the NPS. See Section 4.1.1.3 for information on the methods used.

4.1.1.1       Choosing Focal Natural Resources

Inventory, monitoring, and research efforts targeted at gathering information to guide management decisions should, initially, focus on filling identified information gaps. Because an inordinate amount of time, money, and effort would be needed to inventory and monitor all the natural resources at the UH Management Areas, successful inventory and monitoring programs must focus on a subset of the natural resources present. The natural resources selected for inventory, monitoring, and research must be chosen carefully, and should represent the overall health of the ecosystem, be important indicator species or physical resources, be protected or rare species, or have important human values. Natural resources to choose from when developing the inventory, monitoring, and research programs include water, air, soil, geological resources, plants and animals, and the various ecological, biological, and physical processes that act on those resources (National Park Service 2006). In the case of this Natural Resources Management Plan (NRMP), inventory, monitoring, and research efforts will focus on 1) native species (or communities) of concern, 2) important or unique physical features, 3) stressors that are known or suspected to impact native species and communities (e.g., invasive species, human use, soil erosion), and 3)  basic properties and processes of ecosystem health (e.g., water quality).

The purpose of the baseline inventory is to provide a general snapshot of the ecological integrity of the UH Management Areas; therefore, the number of resources surveyed will be larger than the number monitored. A subset of the natural resources included in the inventory will be chosen for long-term monitoring. It is not necessary to monitor every plant or animal population, or every abiotic process on the properties. The research program will focus on an even smaller subset of natural resources, with the goal of filling data gaps identified during inventory and monitoring. The natural resources to be inventoried are listed in Table 4.1-2, and the resources suggested for long-term monitoring are listed in Section 4.1.2.2 and discussed in more detail in Section 4.1.4. Potential research projects are outlined in Section 4.1.4.

In addition to deciding what natural resources to include in the IM&R programs, it is also necessary to decide where IM&R projects will take place. The ecosystem approach to natural resources management does not recognize property lines or political boundaries. Because of this, it will sometimes be necessary

2 See Section 4.1.1.2 for information on monitoring protocols.

to conduct IM&R activities outside the boundaries of the UH Management Areas. For example, invasive species do not care whether a property is managed by a private landowner or a government agency. Invasive species management efforts will, in many cases, require cooperation or collaboration between adjacent landowners to ensure success. Where permissible, it is recommended that other surrounding land managers adopt IM&R protocols similar to or the same as those presented here or by the National Park Service for Hawai‘i Volcanoes National Park and Haleakalā National Park. This will ensure comparability of results of natural resource monitoring data and research projects across property boundaries.

4.1.1.2       Program Protocols

Protocols are the methods used and the step-by-step instructions needed to conduct inventory, monitoring, and research projects. Protocols should

1. Document the questions being asked
2. Describe how the project will answer the questions
3. Describe the sampling framework and survey design
4. Provide step-by-step procedures for collecting, managing, and analyzing the data
5. Provide guidance on how the data will be presented (e.g., frequency and type of reports)
6. Allow for a testing period and evaluation of the effectiveness of the procedures before they are accepted for long-term monitoring or research (Oakley et al. 2003).

It is necessary to develop detailed protocols to be used in IM&R programs, to ensure that changes in the status of natural resources observed during monitoring and research are real, and not simply artifacts of differing methods of collecting data by different people (Oakley et al. 2003). According to Oakley et al. (2003), “protocols are 1) a key component of quality assurance for monitoring programs to ensure that data meet defined standards of quality with a known level of confidence, 2) necessary for the program to be credible, so that data stand up to external review, 3) necessary to detect changes over time and with changes in personnel, and 4) necessary to allow comparisons of data among places and agencies.” Protocols should include a narrative section that explains the rationale for the monitoring and research, and provides measurable objectives, a sampling design, field methodology, data analysis and reporting, personnel requirements, training procedures, and operational requirements for the program; a set of standardized operating procedures that provides step-by-step instructions on how to carry out all aspects of the narrative; and any supplementary material needed to support the protocol (e.g., maps, photographs, previous reports, and data). The contents of the protocol narrative, as adopted by the NPS are presented in Table 4.1-1. See Oakley et al. (2003) for more information.

Although inventory, monitoring, and research program goals and protocols are outlined in this management component, it will be the task of the Natural Resources Coordinator (NRC)³ to determine, at the time of development and implementation of the IM&R programs, whether the suggested protocols represent best available scientific knowledge and technology (both of these are subject to rapid advancements), and to fill in protocol details, such as sampling locations, and timing. This includes the task of writing the protocol narrative and developing the standardized operating procedures.

3 See Section 4.1.3.1 for more information on the Natural Resources Coordinator.

When possible, recommended inventory and monitoring protocols (e.g., methodology, effort, time of year) should be consistent with past studies and surveys in order to simplify comparison of the results and identification of trends. However, in some cases, there are no previous protocols. This is the case for many of the resources at Mauna Kea Science Reserve (MKSR) and Hale Pōhaku, where, either data have never been collected or data collection has not been done in a quantitative or systematic manner.⁴ In these cases, use of established monitoring protocols from other agencies managing similar ecosystems is recommended, as it will allow for comparison of data between land managers faced with similar ecological conditions and challenges.

For natural resources on UH Management Areas with no established protocols, monitoring protocols should be based on those used by other agencies in Hawai‘i (e.g., DLNR, NPS, USGS, and USFWS). As DLNR did not have any monitoring protocols available at the time of creation of the first draft of this NRMP, protocols produced by the National Park Service Pacific Network,⁵ and in particular those being developed for other high elevation locations in Hawai‘i such as Haleakalā and Hawai‘i Volcanoes National Park, were used as guidelines. These protocols are currently under development by the NPS and are available for use for other natural resource managers (HaySmith 2008). If DLNR protocols are not available at the time of implementation of inventory and monitoring activities, it is recommended that the OMKM NRC check with NPS to determine if the protocols for Haleakalā and Hawai‘i Volcanoes parks have been finalized, before beginning the inventory and monitoring programs. In the absence of monitoring protocols from DLNR or the National Park Service Pacific Network, this plan uses monitoring protocols developed for similar ecosystems found in the mainland United States and Canada. When no protocols were available from established monitoring programs (as was the case for arthropods), survey methodologies were obtained from the scientific literature or other reputable sources. The NRC should review current scientific literature and consult with local experts before implementing these methods, to ensure that the best available methodologies are used.

To the greatest extent possible, the methodologies used for the baseline inventory should be the same as the long-term monitoring protocols, unless otherwise specified, or decided upon by the NRC upon implementation of the inventory and monitoring programs. Any reasons for changing protocols should be documented and included in the next NRMP update. Care should be taken that the same units of measurement (e.g., plant density or pitfall trap capture rates [bugs/day]) are used in the baseline inventory and in long-term monitoring. This ensures that data from the baseline inventory can be compared to the data from long-term monitoring.

Research Project Protocols

As with the inventory and monitoring protocols, the protocol for each research project should be well thought-out and developed, describing the methods used and the step-by-step instructions for conducting the project. This is especially important in long-term research projects that may be conducted over several years, and by different staff. Ideally, the methodologies used in the research projects should be compatible with those used for the baseline inventory and long-term monitoring. Care should be taken to ensure that units used are similar so that direct comparison between the research project results and monitoring can be easily achieved.⁶

4 An exception to this is the ongoing study of the wēkiu bug at the summit by OMKM and Bishop Museum, where investigators have tried to standardize the trapping techniques and the timing of survey work.

5 Publicly available monitoring protocols developed by the NPS Inventory and Monitoring Program for a variety of natural resources are posted at: http://science.nature.nps.gov/im/monitor/protocoldb.cfm.

6 For example, it would be far more useful to record plant densities as number of plants per square meter in both research project plots and long-term monitoring plots than to have density recorded in the monitoring plots and percent cover recorded in the research project plots.

Research projects must be carefully designed to ensure they answer the questions posed. All research projects should have clear and testable hypotheses (predictions), and the methodologies chosen for each study must be able to test the hypotheses.

Table 4.1-1. Guidelines for long-term monitoring protocols: recommended protocol narrative content Background and objectives Background and history; describe the resource issue being addressed Rationale for selecting this resource to monitor Measurable objectives Sampling design Rationale for selecting this sampling design over others Site selection Criteria for site selection; define the boundaries or population being sampled Procedures for selecting sampling locations; stratification, spatial design Sampling frequency and replication Recommended number and location of sampling sites Recommended frequency and timing of sampling Level of change that can be detected for the amount and type of sampling being instituted. Field methods Field season preparations and equipment setup (including permitting and compliance procedures) Sequence of events during field season Details of taking measurements, with example field forms Post-collection processing of samples (e.g., lab analysis, preparing voucher specimens) End-of-season procedures Data handling, analysis, and reporting Metadata procedures Overview of database design Data entry, verification, and editing Recommendations for routine data summaries and statistical analyses to detect change Recommended reporting schedule Recommended report format with examples of summary tables and figures Recommended methods for long-term trend analysis (e.g., every 5 or 10 years) Data archival procedures Personnel requirements and training Roles and responsibilities Qualifications Training procedures Operational requirements Annual workload and field schedule Facility and equipment needs Startup costs and budget considerations References Source: Oakley et al. 2003

The methodology used to develop this IM&R component plan included

• Compilation of current information on Mauna Kea subalpine and alpine ecosystems and existing data and knowledge gaps through literature research, consultation with local experts, and stakeholder input (see Section 2)
• Review of past research and monitoring activities on Mauna Kea
• Review of available spatial data (Geographic Information System (GIS) database) and determination of spatial data needs for a successful monitoring program
• Review of the literature on monitoring program development and monitoring protocols
• Review of other successful monitoring programs and monitoring protocols and discussions with monitoring program managers and developers (e.g., Inventory and Monitoring Program at NPS)
• Identification of important abiotic and biotic resources (physical features, species, communities, ecosystems) and threats to these resources through literature research and consultation with subject matter experts and stakeholders
• Consultation on management and monitoring priorities with OMKM staff and the Environment Committee, local experts, and stakeholders
• Prioritization of natural resources to be inventoried, monitored, and researched based on the above information.

Materials to support the development of the monitoring program and monitoring protocols by the OMKM NRC are maintained in the EndNote library (see Section 4.5). These include pertinent guidance, articles, and reports on development of monitoring plans and protocols. Sample monitoring protocols and a copy of the NPS Pacific Islands Network Monitoring Plan (HaySmith et al. 2005) have been provided to the OMKM librarians for use by OMKM natural resources staff.

The overall frameworks for the inventory, monitoring, and research programs are described in this component. It is beyond the scope of this NRMP to develop complete and ready-to-implement inventory, monitoring, and research programs, in part because many of the long-term monitoring objectives and research projects will depend on the results of the initial baseline inventory that needs to be conducted on the UH Management Areas, and thus cannot be anticipated or described here. However, the steps needed to develop the inventory, monitoring, and research programs, currently known monitoring and research goals and objectives, and examples of useful monitoring and research protocols are presented. The first task of the NRC will be to complete the process of developing the IM&R program. To ensure success, the program must then be followed carefully over time. Future updates to the NRMP should modify management actions as deemed appropriate, following interpretation of the collected data. Any changes to the overall IM&R, and to individual monitoring and research protocols need to be thoroughly documented in future updates.

The first step in developing the IM&R program is to establish the goals and objectives of the program.⁷

The goals and objectives of the Mauna Kea natural resources IM&R program are:

Program Goals and Objectives                                   Section
Goal IMR-1	Determine baseline status of the natural resources (Baseline Inventory).	4.1.2.1
Objective 1	Establish baseline inventory survey protocols that are compatible with long-term monitoring protocols.
Objective 2	Collect baseline inventory data.
Objective 3	Refine long-term monitoring and research priorities based on results of baseline inventory.
Goal IMR-2	Conduct long-term monitoring to determine the status and trends in selected resources to allow for informed management decisions.	4.1.2.2
Objective 1	Determine which resources to monitor.
Objective 2	Establish monitoring protocols and write monitoring plans.
Objective 3	Conduct regular monitoring efforts.
Objective 4	Identify new data gaps.
Goal IMR-3	Conduct research projects to fill natural resource knowledge gaps that cannot be addressed through inventory and monitoring.	4.1.2.3
Objective 1	Identify and prioritize research projects.
Objective 2	Develop research protocols and obtain funding.
Objective 3	Conduct research projects and present results.
Objective 4	Evaluate the information obtained and adjust management actions as necessary.
Goal IMR-4	Create efficient, cost effective IM&R programs.	4.1.3.1
Objective 1	Complete as much inventory, monitoring, and research as possible using in- house staff and resources, interagency collaboration, and volunteer labor.
Objective 2	Streamline monitoring and research efforts, to minimize expenses and impacts to natural resources.
Objective 3	Carefully choose natural resources to inventory, monitor and research.
Objective 4	Use scientifically and statistically sound sampling protocol to ensure that the data collected is usable and can be successfully analyzed.
Goal IMR-5	Measure progress towards performance goal of preserving, protecting, and restoring Mauna Kea ecosystems.	4.1.3.2
Objective 1	Collect the right data and ensure data quality.
Objective 2	Analyze data to identify trends in natural resources status and to answer specific management questions.
Goal IMR-6	Increase communication, networking, and collaborative opportunities to support natural resource management and protection.	4.1.3.3
Objective 1	Produce reports to inform stakeholders, public, and collaborating agencies about the status of the natural resources.
Objective 2	Identify opportunities for collaborative data collection and resource management.

 

7 A goal is a brief statement of the overall purpose of a program. An objective is a more detailed statement that provides additional information about the purpose or desired outcome of the program (NPS 2006). Monitoring objectives should be realistic, specific, and measurable. An example of a monitoring objective is to “detect new localized populations of invasive non- native plants before they become established at high densities.”

4.1.2    Program Specifics

This section addresses the establishment of the baseline inventory, long-term monitoring, and research programs for the UH Management Areas. It provides general information on the goals and objectives of each of the programs. Detailed information on baseline IM&R projects for specific natural resources found on UH Management Areas (e.g., birds, plants) is provided in Section 4.1.4.

4.1.2.1 Baseline Inventory

Goal IMR-1: Determine baseline status of the natural resources (Baseline Inventory)

Establishing a solid baseline data set describing the distribution, abundance⁸, condition⁹, and diversity of natural resources is time and labor intensive, but necessary if natural resource personnel wish to understand the resources they are to manage and determine if management actions are having the desired effects. The initial labor-intensive survey is referred to as the baseline inventory. In the following years (during long-term monitoring), only subsets of the natural resources will be monitored or surveyed, in order to reduce costs and minimize impacts on the natural resources being monitored. Because the long- term monitoring is by necessity focused on a reduced number of natural resources, it is recommended that baseline inventories be conducted every twenty years (or as needed, depending on the resource being managed and conditions of the UH Management Areas). It is also recommended that additional detailed baseline inventories be conducted, as needed, in areas proposed for development. These baseline inventories should be conducted at the time the development is proposed, within the footprint of the area to be developed. It is recommended that these inventories also include a buffer of at least 1,640 ft (500 m) around the project footprint. This is especially important if the proposed area has not previously been included as an actual survey point in another baseline inventory. The purpose of conducting baseline inventories in areas of proposed development is to determine if the area contains sensitive resources such as protected species or unique geological resources, which need to be protected or mitigated for. However, without conducting baseline inventories in other portions of similar habitat on the mountain, it is difficult to know whether the proposed project area is more or less important or unique than surrounding areas. Thus, it is important to understand the distribution of natural resources over a larger area, rather than simply studying the area of proposed impact.

Baseline inventories have multiple purposes. The first is to record locations and abundances of species and abiotic natural resources found on the properties, so that the current status of the site may be understood. The second is to identify any problem areas or areas of special concern that may need additional attention in the future. These may include areas with rare, threatened, or endangered plant populations, patches of invasive plants that should be removed, and areas of physical hazards. The third is to collect quantitative data that can be used to compare future monitoring results against, and identify trends and detect changes. Fourth is to continue building the spatially-linked GIS database of the natural resources on Mauna Kea to facilitate and support monitoring, planning and management efforts (see

8 Abundance is used to define a species’ or feature’s population size, absolute count, or density.

9 For biological resources condition indicates the health of an individual or population, and/or reproductive status. For abiotic resources, condition refers to physical attributes.

Section 4.5). Geospatial maps (GIS layers) can be created that depict species locations, abundance and habitats, and to delineate physical features of historic and scientific value, culturally important sites, and use areas. These geospatial layers can be overlain to define management areas that may be considered off- limits for future development or suitable for future management activities. Conversely, the spatial  analysis can help identify best-site alternatives for future development, both at the summit and at Hale Pōhaku, while minimizing impacts to cultural and natural resources.

The natural resources that should be considered for inclusion in the baseline inventory, along with the priority of need for inclusion, are presented in Table 4.1-2. For information on how resources and activities were prioritized, see Section 4.1.4. Although the table detailing those resources to include in the baseline inventory is presented separately from the list of resources to include in the long-term monitoring program, there are many overlaps between the two, especially in the areas of field survey protocols and ideas for making the programs cost effective and efficient. Many of the cost-saving measures and ideas for such things as inter-agency collaboration and methods of obtaining data that are presented in Section

4.1.3   are applicable for the baseline inventory program, as well as for long-term monitoring. Please note that while all the natural resources found on UH Management Areas are included on the table below, it is not recommended that OMKM attempt to conduct baseline inventories on all of them, because to do so would be prohibitively expensive and time consuming. The resources to include in the inventory will have to be selected by the OMKM NRC (see Section 4.1.3) and the Mauna Kea Management Board (MKMB) Environment Committee. It is recommended that the focus of the baseline inventories be on high priority items, with medium and low priority items added in as time and funding allows.

Table 4.1-2. Natural Resources to Consider for Inclusion in Baseline Inventory¹⁰

Resource                             Hale Pōhaku                                              MKSR                              Priority        Background Category                                                                                                                                                                       Section
Physical
Geology	Geological features	Geological features	Medium	2.2.1
Surficial Features and Soils	Hiking trails / Off-road trails	Hiking trails / Off-road trails	Medium	2.2.2
	Ground condition in and around stormwater systems	Ground condition in and around stormwater systems	Low/ Medium
	Soils	Soils	Low
Hydrology	N/A	Lake Waiau	Low	2.2.3
	Seeps and streams	Seeps and streams	Low
	Locations of known chemical discharges	Locations of known chemical discharges	High
Climate and Weather	Meteorological parameters	Meteorological parameters	Low	2.2.4
	N/A	Snow water equivalent and snow pack depth	Low
Air Quality and Sonic Environment	Dust distribution/concentration	Dust distribution/concentration	Low	2.2.5
	Ambient background noise	Ambient background noise	Low
Biological
Plants	T&E* Species	T&E Species	High	2.2.1
	Māmane woodlands	Alpine stone desert	High
	Invasive plants	Invasive plants (along road)	High
	Subalpine shrublands	Alpine shrublands	Medium
	Subalpine grasslands	Alpine grasslands	Medium
Invertebrates	Native pollinators (bees, moths)	Summit arthropods	High	2.2.2
	Invasive wasps and ants	Invasive arthropods	High
	Snails	Alpine arthropods (below summit)	Medium
	Other native arthropods	Other native arthropods	Low
Birds	All birds	Hawaiian Petrel	High	2.2.3

10 This table presents a summary of the general categories of natural resources found in UH Management Areas, with details on what could be inventoried at Hale Pōhaku and in the MKSR, along with the perceived priority, and the pertinent background section of this NRMP that provides more information on the natural resource category.

Resource Category	Hale Pōhaku	MKSR	Priority	Background Section
Mammals	T&E Native Species (Hawaiian Hoary Bat)	N/A (No native species found in MKSR)	High	2.2.4
	Herbivores (sheep, goats)	Herbivores (sheep, goats)	High
	Predators (cats, mongoose, rats)	Arthropod Predators (rats, mice)	Medium
	Seedeaters (mice, rats)	Seedeaters (mice, rats)	Medium

*Threatened and endangered

Objective 1: Establish baseline inventory survey protocols that are compatible with long- term monitoring protocols

Actions

1. Establish baseline survey protocols based on best available monitoring protocols (e.g., NPS or USGS protocols applicable to high-elevation areas in Hawai‘i).¹¹ Protocols should  be documented in report format.
   1. Quantitative rather than qualitative methods should be used whenever possible, because data on abundance and distribution is more useful than simple presence/absence data.¹²
   2. Survey methodologies used should, when possible, be similar to those promulgated, or already employed, by federal and state agencies (such as NPS, USGS, DLNR).¹³ In this plan, NPS methodologies are applied or recommended whenever possible, because their monitoring programs are already under development.
   3. For some groups of organisms with extremely high diversity and abundance, such as invertebrates¹⁴, it may be necessary to conduct the baseline inventories in two tiers: the first would be a list of all species (or functional groups) found on the properties, while the second would be a more quantitative survey of the abundances of species of interest (such as keystone¹⁵ species or protected species). The findings from the first tier will help identify what would be studied in more detail in the second tier. This may be necessary to limit the overall costs associated with detailed survey work.

Objective 2: Collect baseline inventory data

Actions

1. Hire consultant teams or contractors to complete the baseline inventories. Contracting out the baseline inventory work is necessary due to the inherent broad scope, sizeable acreage, and quantity of data to be reviewed and analyzed. The amount of work required would be beyond the abilities of an in-house team to complete within a reasonable time period (one to two years).
   • Provide methodologies to be followed in the Request for Proposal (RFP), to inform bidders of the scope and requirements of the project.

11 Although the baseline inventory surveys are more intensive (cover a broader variety of species and a larger number of  sampling plots) than the long-term monitoring surveys, the data should be compatible and comparable.

12 Qualitative methods can be used when quantitative methods are not possible or are prohibitively expensive or time consuming. 13 DLNR does not have monitoring protocols for the Mauna Kea Ice Age NAR or other properties on Mauna Kea. If DLNR does develop inventory and monitoring protocols, an attempt should be made to ensure that protocols are similar to and compatible

with those being used by OMKM. Alternatively, DLNR could adapt protocols recommended within this report.

14 Study of the wēkiu bug has been ongoing for a decade. Therefore some baseline data exists for the summit invertebrate community, but is lacking for the remainder of UH Management Areas.

15 A keystone species is a species that plays a pivotal role in an ecosystem and upon which a large part of the community depends. Māmane trees are an example of a keystone species in the subalpine zone of Mauna Kea.

• • Require, as part of the Scope of Work, that the contractors follow the established protocols (to prevent contractors from utilizing their own methodologies, which may not be compatible with future monitoring efforts). ¹⁶
  • Require Global Positioning System (GPS) data to be collected at survey point locations.
  • Require that all data (including measurements and survey results) be entered into and submitted in a GIS dataset consistent with OMKM GIS protocols, ensuring compatibility with OMKM GIS (see Section 4.5).
  • Request that contractors search historical archives for data, collections, and images pertinent to the natural resources of Mauna Kea. Although unlikely to be quantitative in nature, this information would be useful for determining historical conditions on Mauna Kea, to use as a reference point for resource management and restoration efforts (see Section 4.3).

1. Conduct baseline inventories for plants, invertebrates, vertebrates, and physical resources. A baseline survey of cultural and archaeological resources has already been conducted for the MKSR (McCoy et al. 2009). Surveys for different groups of natural resources should be conducted simultaneously and on the same plots, to the greatest extent possible, to allow for exploration of relationships between groups. Proposed survey methodologies are included in the OMKM EndNote library (see details in Section 4.1.4). The following actions are recommended as part of the baseline inventory
   1. Obtain high-resolution aerial photos of the UH Management Areas to pre-screen for areas of interest (e.g., māmane woodland; isolated silversword populations; or areas heavily infested by invasive plants), identify problem areas and issues (such as erosion), and record visual conditions at time of baseline inventory.
   2. Determine locations for transects and plots
      1. Divide Hale Pōhaku, Access Road, and MKSR into grids.
      2. Randomly locate transects or sample points within each grid.
      3. Add additional areas of interest that are not covered by the randomly located survey points (e.g., silversword population in MKSR).
      4. Plot out proposed survey locations on GIS software to identify potential  problems (e.g., access, sensitive cultural resources). Remove points with potential problems and replace with another randomly selected location within the same grid. Visiting some grid spaces may not be feasible, and these may be excluded from the surveys.

1. Go into the field and record actual plot and transect locations with GPS. Mark monitoring plots with permanent plot markers, where permissible and feasible.¹⁷
2. Conduct all survey work in the same location within each grid cell, so that plant, invertebrate, and vertebrate surveys would all occur at the same locations. The size of the sampling plot, or frequency of sampling along the transect will depend on organism being studied.
   1. Multitask where possible; for example, a survey team could record bird species on one pass of the transect and then record a different element, such as plant density or diversity, on the return).
   2. Conduct surveys at appropriate times (both daily and seasonally) for organisms being surveyed (e.g., peak activity during early morning for birds, during peak flowering of

16 If no protocols are publicly available at the time of development of the RFP, it would be possible to require that proposals include methodologies to be used. The proposals would then have to be reviewed by unbiased (not related to the proposer) experts in the natural resource to be inventoried for scientific merit. At a minimum, OMKM should determine the number of plots and the areas to be included in the inventory.

17 Installing permanent plot markers may not be possible on the summit, where cultural considerations may prevent driving permanent plot markers into the substrate.

māmane trees for nectarivorous birds, peak fruiting period for seed-eating birds). This may necessitate multiple visits to each site at different times of day and during different seasons.

1. To minimize the impact on the landscape, through cinder compaction or soil erosion, and prevent leaving visible trails, field crews should attempt to access survey plots from different directions or different trails each time it is visited during a single survey season. Whenever possible, field crews should attempt to walk on larger rocks and boulders when accessing particularly vulnerable wēkiu bug habitat.
2. Create a GIS database with the results from baseline inventory (see Section 4.5, Information Management).
3. Update species-list tables presented in Section 2.2, Biotic Resources of this NRMP.

Objective 3: Refine long-term monitoring and research priorities based on results of baseline inventory

Actions

1. Using results of surveys, identify areas of concern or interest for future management, monitoring, and research.
2. Use baseline inventory results to further clarify long-term monitoring goals and to target resources of concern.
   1. Note any problems with data or survey techniques that are discovered during the baseline inventory efforts, and adjust the long-term monitoring protocols to overcome these problems. If problems are serious enough that they may preclude use of the baseline inventory data for future comparisons, it will be desirable to repeat the particular baseline inventory using more appropriate methodology.

4.1.2.2       Long-Term Monitoring

Goal IMR-2: Conduct long-term monitoring to determine the status and trends in selected resources, to allow for informed management decisions

Long-term monitoring is the only scientifically valid way to measure trends in the condition of natural resources on UH Management Areas. Data from long-term monitoring is necessary to assess the efficacy of management and restoration efforts, to provide early warning of impending threats, and to provide a basis for understanding and identifying meaningful change in complex natural systems (National Park Service 2006). Conducting long-term monitoring is integral to the process of adaptive management. Designing a long-term monitoring program requires 1) determining which resources to monitor, 2) establishing monitoring protocols and writing monitoring plans, 3) conducting monitoring efforts and analyzing data, and 4) identifying new data gaps to be filled by future monitoring. The monitoring program should be reviewed and updated every five years to ensure that it is addressing new data gaps. Details regarding monitoring needs for specific natural resources, and appropriate monitoring methodologies, are presented in Section 4.1.4. Detection of threats and observation of undesirable short- term and long-term changes to Mauna Kea ecosystems by means of long-term monitoring are discussed further in Section 4.2, Threat Prevention and Control.

Objective 1: Determine which resources to monitor

In many ways the resources chosen for long-term monitoring will depend on the results of the baseline inventory, and thus it is not possible to identify the exact resources that should be monitored on the UH Management Areas in this first draft of the NRMP. However, based on a review of scientific literature, interviews with local experts, and input from the OMKM Environment Committee and the public, a preliminary list of resources to be considered for long-term monitoring efforts is presented in Table 4.1-14 and in Section 4.1.4. This is not an exhaustive list, as there will likely be new issues raised by the baseline inventory and day-to-day experiences of the NRC that will need to be addressed by monitoring. Conversely, resources that are identified here as being worthy of monitoring may later be determined to be less important and may be dropped from the monitoring program. As an example, it may be discovered that another agency or group is collecting the same data as OMKM, and it may be possible to use this data rather than have OMKM staff continue to collect it.

It should be noted that while the majority of the natural resources found on UH Management Areas are included in Table 4.1-14, it is not recommended that OMKM attempt to monitor all of these resources, because to do so would be prohibitively expensive and time consuming. The resources to include in long- term monitoring will depend in part on which resources were included in the baseline inventory, and will have to be selected by the OMKM NRC (see Section 4.1.3) and the MKMB Environment Committee.

Actions

1. Complete baseline inventory work and data analysis.
2. Determine which species of concern (Threatened or Endangered Species, rare species, Candidate Species) occur on UH Management Areas. Protected species not previously recorded on UH Management Areas will need to be added to the monitoring plan and may require special monitoring protocols.
3. Identify resource monitoring efforts that would best aid decision-making by management or contribute to adaptive management; for example, monitoring invasive species targeted for elimination or native species or communities targeted for restoration efforts.
4. Determine a final list of natural resources to be monitored, using information obtained from baseline inventory, as well as information from other local agencies on what they are monitoring.
5. Prioritize resource monitoring efforts by urgency of need. For example, geologic features may not need to be monitored yearly, while populations of invasive plants or protected native species may need yearly, or more frequent, monitoring, depending on their population dynamics.

Objective 2: Establish monitoring protocols and write monitoring plans

Section 4.1.1.2 discusses the development of protocols for the monitoring plan. To the extent possible, the monitoring protocols should include

• Evaluation of potential ecological impacts of sampling protocols (e.g., crushing cinder, incidental take, dust, introduction of non-native species) and recommendations for low-impact sampling techniques
• Support for adaptive management, including evaluation of the effectiveness of management actions
• Identification of potential collaborations with existing programs such as Department of Land and Natural Resources (DLNR) Natural Area Reserves System (NARS), DLNR Division of Forestry and Wildlife (DOFAW), U.S. Fish and Wildlife Service (USFWS), U.S. Geological Survey (USGS), and National Park Service (NPS)
  • Evaluation of potential conflicts between cultural considerations and natural resources monitoring, and solutions to these conflicts

1. Check with federal and state agencies regarding the availability of monitoring protocols for high elevation areas in the Hawaiian Islands. The NPS protocols are currently under development but will likely be completed by the time OMKM is ready to begin long-term monitoring efforts.
   1. If protocols are not yet available, ask DLNR, USGS, and NPS personnel to recommend sampling methodologies.
   2. Using the above protocols as guidelines, develop monitoring protocols appropriate for the conditions at Hale Pōhaku, the Summit Access Road, and MKSR. Develop monitoring protocols for the following major groups and resources:
      1. Physical resources and abiotic conditions
         1. Soil (stability and natural and human induced erosion)
         2. Soil condition in and around waste water systems
         3. Water (Lake Waiau and at sites of seep and spring discharges)
         4. Climate and weather (rainfall, snowfall, temperature)
         5. Air quality
2. Plant communities
   1. Māmane woodland
   2. Alpine Shrubland
   3. Alpine Grassland
   4. Alpine Stone Desert
3. Invertebrates (native and invasive)
   1. Arthropods
   2. Snails (if present)
4. Birds (native and invasive)
5. Mammals (native and invasive)
6. If the monitoring protocols are substantially different from those used in the baseline inventory, determine their compatibility with baseline data and conduct trials to work out any problems with methodologies employed.
7. Consult a statistician to ensure that monitoring protocols are statistically sound and that data will be usable.
8. Enlist monitoring experts, such as personnel from USGS or NPS, to review monitoring protocols, if they are substantially different from protocols developed by other agencies for use in Hawaiian high elevation ecosystems.
9. Finalize monitoring protocols and put together final monitoring program plan.
10. Periodically review and evaluate the success of the monitoring program.
    1. Is monitoring occurring on schedule?
    2. Are the monitoring personnel able to keep up with the required field and office work?
    3. Are the data collected useful for tracking the state of the natural resources?
    4. Are the field data collection and data entry protocols easy to implement and do they  result in quality data?
    5. Any problems with data collection or processing should be reviewed, and the monitoring protocols updated to correct the problem.
    6. A thorough review of the monitoring program by the completion of the second year of implementation is recommended, to catch and correct problems early.

As discussed later under Section 4.1.3.1, it is not necessary to monitor every plot within the UH Management Areas annually. Annual monitoring will still occur, but a portion of the monitoring will shift locations according to a predetermined annual rotation. The balance of the monitoring will occur annually at special areas of concern on the UH Management Areas, and every two to five years at other plots, depending on how frequently the resources in these plots need be monitored. An example schedule is presented below.

Actions

1. The plots to be monitored for a given year will be determined by the monitoring protocols.
2. Develop an annual calendar showing monitoring dates and locations. This will ensure that the field personnel and NRC schedules are kept open for these periods and that no conflicting events are scheduled.
3. Analyze previous year’s monitoring data before starting the current year’s monitoring. This will allow for detection of problems with the previous year’s data and for the collection of replacement data if the problems are insurmountable. This step is critical to ensuring that flawed methodologies are not used unknowingly, year after year, to collect data that is unusable for future analyses.
4. Review monitoring protocols well before beginning fieldwork, to ensure that personnel are familiar with the procedures to be used, and that equipment is in working order.
   1. Enter monitoring data into database immediately after collection, store backup hard and  electronic copies in secure, fireproof and waterproof locations.
   2. Analyze the current year’s data, ideally immediately following collection.
   3. Produce the annual report for delivery to MKMB and Environment Committee.
   4. Share monitoring data with other agencies collecting similar data, and with public, as appropriate.

Sampling monitoring schedule (based on HaySmith et al. 2005)
Plot Set	Year 1	Year 2	Year 3	Year 4	Year 5	Year 6	Year 7	Year 8	Year 9	Year 10
A	X	X	X	X	X	X	X	X	X	X
B	X	X				X	X
C		X	X				X	X
D			X	X				X	X
E				X	X				X	X
F	X				X	X				X

Objective 4: Identify new data gaps

Identifying new data and knowledge gaps is part of the cycle of adaptive management. To identify new gaps, the NRC will use the results from the baseline inventories and long-term monitoring efforts, as well as articles in the current scientific literature, and will communicate with local experts and other land managers in Hawai‘i. The gaps identified can then be addressed in future years, through new long-term monitoring surveys and additional research projects.

Actions

1. Analyze data obtained from the baseline inventory and long-term monitoring as soon as it is available.
2. Stay abreast of current scientific literature that pertains to natural resources that occur on the UH Management Areas. Other scientific studies may identify knowledge gaps or relationships pertinent to the management of UH Management Areas.
   1. Collaborate, cooperate, and communicate with other state and federal agencies and programs that deal with land management on Mauna Kea and in other high-elevation systems in Hawai‘i. These agencies and people may have information pertinent to UH Management Areas that is not available in the scientific literature.
   2. Look for relationships between the natural resources (biotic and abiotic). For example, native plant density at Hale Pōhaku may be found to be negatively correlated with invasive plant density, or it may be found to be unaffected by invasive plants, but instead highly correlated with annual rainfall.
   3. Use the spatial analysis capabilities of GIS system to identify areas of interest, and look for relationships between biotic and abiotic factors, and changes in population boundaries.
   4. Update monitoring plan and monitoring protocols regularly, to address newly identified data gaps.
      1. Review the monitoring plan yearly, checking for needed changes or deficiencies.
      2. Update the plan every five years, or as needed, based on the findings of the yearly reviews.

Objective 4: Evaluate the information obtained and adjust management actions as necessary

Actions

1. Evaluate the success of the research project. Did it answer the research questions asked? (If not, it may be necessary to revise protocols and conduct additional research.)
2. If data gaps or additional questions were identified during the project, enter these into the list of data gaps presented in Section 4.1.4, and update list of research questions.
3. Use the information obtained from research projects to improve management of resources, as needed (adaptive management), or to justify current actions, if they are questioned.

by physical characteristics such as terrain, elevation, or other features unlikely to change rapidly. Elevation would be a good choice for Mauna Kea.

1. Use permanent plots for monitoring. This will allow managers to identify changes over time, and will preclude problems caused by naturally occurring variation between plots.
2. Optimize sample sizes to obtain the information desired. This requires an understanding of how much change needs to be detected and the variability of resources across space and time. The smaller the change needing detection or the greater the variability across space and time, the more samples that need be taken. Of course, sample size must be balanced with the availability of time and personnel. As a rough guide, NPS recommends a minimum sample size of six plots (per resource or community measured; e.g., māmane woodlands).
3. View the sample locations on a map using GPS and GIS technology, to ensure that adequate representation and coverage of natural resources is being achieved by the sampling protocol.
4. It is not necessary to visit all selected monitoring sites every year. Sampling protocols  can be designed to cover a greater area by using rotating sampling locations (see Section 4.1.2.2 for an example).

1. Attempt to sample the different resources in same sample areas (e.g., vegetation, birds, mammals and invertebrates, and abiotic factors), in the same locations or in closely spaced sites, so relationships between these resources can be investigated.
2. Use labor-saving techniques during fieldwork whenever possible and economically feasible. Two such techniques are remote sensing and aerial photography.
3. Consult a statistician before finalizing the sampling design to ensure that monitoring protocols are statistically sound.²¹
4. Doing a test run (or runs) before establishing permanent long-term monitoring or research protocols will help determine how many samples are needed and will allow the NRC to address any problems discovered in the sampling design. This will improve the overall quality of the data collected and ensure that flaws in the sampling design do not render data unusable. This may delay the start of the monitoring or research program but in the long term, it will improve the quality of the data collected.

Objective 1: Collect the right data and ensure data quality

Actions

1. When developing management actions, include a means to measure progress towards the management goal, such as monitoring site conditions regularly. The frequency of monitoring efforts will depend on the nature of the project and how rapidly the resource being managed is expected to respond.
   1. Examples of these are provided in the various component plans (e.g., Section 4.2, Threat Prevention and Control; Section 4.3, Natural Resources Preservation, Enhancement, and Restoration).
2. Each year, devise a list of questions to be answered by the monitoring and research activities covered during that time period.
   1. A running checklist of all monitoring requirements should be developed by the OMKM NRC. The list should include general trend-monitoring goals as well as a list of monitoring goals specific to projects, as described in Item 1, above.
3. Determine if monitoring and research methodologies, and the natural resources selected for monitoring, can provide the data needed to answer the questions posed by the monitoring and research programs.
   1. Consult with statisticians and local experts, if needed.
   2. If it is determined that the methods used, or the resources selected for monitoring, are not appropriate to answer questions, revise methodology, add additional monitoring activities, or begin monitoring of additional resources.
4. Develop and carry out inventory, monitoring, and research programs.
5. Follow protocols for proper data management, including Quality Assurance/Quality Control (QA/QC).
6. Ensure that natural resources staff has proper training in data collection, management, and analysis.

Objective 2: Analyze data to identify trends in natural resources status and to answer specific management questions

Actions

1. Analyze monitoring and research data annually and look for trends that indicate change in the conditions of natural resources. If evidence of degradation of resources is found, examine management activities and determine where management techniques and activities can be improved.
   1. Analyze data with the goal of answering the questions described in Objective 1, above.
2. Review management plan and monitoring and research programs yearly, to determine progress towards management goals.
3. Revise plans as needed, or every five years.
4. Produce “State of the Resources” reports (trend analysis reports) every five years (see Section 4.1.3.3).

Factors other than priority may also play into when a project is undertaken. For example, a project that is Low priority, but is inexpensive and easy to accomplish, may fit in nicely with a High or Medium priority project and provide additional useful information. When possible, projects that complement each other are cross-referenced so that consideration may be given to undertaking them simultaneously.

The following information is provided as a resource for the NRC, and it is not the intention of this NRMP that all of the following inventory, monitoring and research activities be undertaken by OMKM. Due to staffing and funding constraints, it is recognized that only a small subset of these inventory, monitoring, and research activities will be undertaken.

4.1.4.1.2      Baseline Inventory

with those on subsequent images. For the large spatial area of the MKSR, this approach is cost-effective and it reduces potential impacts from ground-based surveying. The second technique is a ground-based method employing observations made by OMKM Rangers or staff familiar with features of concern that are within the high travel zones where the probability of resource impact from human activities are greater. Observations will be geospatially delineated using either a global positioning system (GPS), or by identifying the features or areas of interest on the high resolution ortho-rectified air photograph. This technique will include establishing photo-point monitoring of features. Delineated areas of concern from both techniques will be tagged with field notes and other information that describes or quantifies the feature of interest.

4.1.4.1.4      Research

Priority: Low²⁸

Although much is known regarding the geology and formation of Mauna Kea, new questions continually surface. Geologic research has the potential for testing hypotheses and answering questions that cannot be inferred through monitoring. However, much of the potential information that could be learned is not required to manage the resources. From a research perspective, identifying the lithology of the cinder cones in the summit region is somewhat of a prerequisite for understanding how water that infiltrates the cones moves beneath the surface. It would also assist geologists and hydrologists in their understanding  of the relationship between the hydrology of the summit region and the aquifers and streams of the larger Mauna Kea region. Further investigation of cinder cones may provide a more detailed understanding of how they were built, whether or not their lithology has any bearing on summit hydrogeology, and the limits of their structural integrity. Core samples are required for this type of investigation, to better understand lithology, hydraulic connectivity and flow paths.²⁹ In the event core sampling is not possible, ground penetrating radar, electrical sounding techniques, and acoustic methods could also be applied to provide useful, if incomplete data.

Questions:

1. What is the inner structure of individual cinder cones? What are the discrete layers composed of and what are their approximate layer depths?
2. Are there consistencies in how the material has been laid down between different cinder cones?
3. For each cinder cone, what are the physical characteristics of the different materials forming the bulk of the cone?

Research Projects:

1. Substrate analysis at varying depths in construction areas
2. Geological analysis at varying depths in construction areas

28 This type of investigation may warrant higher priority if new projects are proposed for the summit region.

29 A potential limitation to this project or similar projects that result in disturbance to subsurface resources is the sensitivity required when conducting these types of activities. It is possible that this type of research would not be permitted due to cultural considerations or concerns about disturbance to natural resources resulting from the study. Relevant data could be gathered during any permitted construction activities.

Techniques: Conduct an erosion assessment to identify locations and causes of accelerated erosion, and make recommendations to correct problems.³³ The assessment should also summarize the inventoried sites and prioritize them for treatments. All sites should be catalogued into GIS and tagged with supporting documentation.

4.1.4.2.4      Research

Priority: Medium

The focus of surficial feature and soils research will be to investigate the effectiveness of treatments and management actions to remediate accelerated erosion implemented as part of the Threat Prevention and Control Program (see Section 4.2.3.4). Identifying which methodologies work best in the high-elevation, arid conditions of the Mauna Kea summit region will facilitate continued implementation of successful erosion control activities. In addition, research can be conducted on identifying the impacts of some of the less obvious contributors to erosion (e.g., invasive plants, feral ungulates) in order to provide additional justification for control of these threats.

Questions:

1. What erosion control methodologies are most effective to reduce various types of accelerated erosion?
2. Are invasive plant species or feral ungulates contributing significantly to accelerated erosion?

Research Projects:

1. Determine which erosion control methodologies are most effective for different types of erosion. This will be achieved through trial of a range of methodologies and long-term monitoring of their success or failure.
2. Determine the impact of invasive plant species and feral ungulates on substrate and potential contributions to erosion. Coordinate with other research on impacts of these threats on vegetation resources (see Sections 4.1.4.6.4.2 and 4.1.4.9.4).

 

that changes between subsequent samples can be detected. All monitoring data should be compiled into GIS.

4.1.4.3.4      Research

Priority: High

Due to a variety of reasons, including no regulatory requirements to do so, the hydro-geology of Mauna Kea’s summit region has not been investigated. Simple one dimensional modeling for very limited areas has been conducted. In order to understand how land uses in the summit area affect or do not affect water resources beneath and downslope of the MKSR, it is recommended that a robust multi-dimensional hydro-geologic investigation be conducted.

Questions:

1. How much water that reaches the aquifers and streams is from rain and snow that falls across the MKSR?
2. What is the fate of potential contaminants contained in runoff and waste effluent discharged across the MKSR?

Research Projects:

1. Hydro-geologic investigation³⁹

Table 4.1-6. Climate and weather data gaps and actions to fill them

4.1.4.4.2      Baseline Inventory

 

equivalent) should be deployed across the summit area of the MKSR at the beginning of each snow season and removed upon snow melt. Data should be statistically summarized annually.

4.1.4.4.4      Research

Priority: Medium

While it is generally agreed that anthropogenic impacts are contributing to global climate increases, the effects to the climatic regime of the Hawaiian Islands can only be postulated. Mauna Kea provides a unique platform for testing hypotheses and assessing meso-scale changes to atmospheric attributes. Research into how shifts in global weather patterns may impact the persistence and elevation of trade wind inversion are of specific interest due in part to controls the inversion places on precipitation frequency and its magnitude across the MKSR.

It has been suggested that wind patterns play a significant role in maintaining arthropod communities at the summit of Mauna Kea (Howarth and Stone 1982; Howarth et al. 1999; Englund et al. 2002; Porter and Englund 2006). Global climate change has the potential to impact the ecology of arthropod communities through changes in local wind vectors on MKSR. It is known that the summit area aeolian ecosystem is a function of wind, and changes to the wind, in both its magnitude and direction can potentially have an adverse impact on native arthropods. Changes to wind vectors could be induced by large scale atmospheric forcing induced either by climate change or locally by obstructions such as buildings. How these potential changes alter the transport and deposition of wind transported food sources across the summit area is unknown. Research into these potential changes may provide biologists with information on how to better manage the summit arthropod communities.

Questions:

1. Where does the aeolian drift (food sources for the summit arthropod communities) come from?
   1. How does the wind flow up Mauna Kea?
   2. How do observatories and buildings alter wind patterns at the summit?
   3. Will shifts in global scale weather patterns alter the persistence and elevation of the tradewind inversion? How will these changes affect other meteorological variables across the MKSR?

Research Projects:

1. Conduct wind flow studies and models to determine sources of aeolian drift at summit.
2. Conduct airflow analysis. Conant et al. (2004) recommend field studies of impacts of buildings on aeolian drift and moisture through deployment of an array of moisture sensors, sediment collectors, and seed traps (to capture aeolian drift).
3. Model windflow patterns under various climate change regimes to determine impact of climate change on aeolian ecosystems.
4. Develop (or support development) of climate change models of high elevation areas in the Hawaiian Islands. Collect and analyze climate data for both MKSR and Hale Pōhaku.
5. Model climate change and vegetation response for subalpine and alpine regions.

4.1.4.5.2      Baseline Inventory

4.1.4.5.4      Research

Priority: Medium

At this time no noise-related research is being recommended. The focus of air quality research will be to investigate the effectiveness of treatments and management actions to remediate generation of fugitive dust implemented as part of the Threat Prevention and Control Program (see Section 4.2.3.2). Identifying which methodologies work best in the high-elevation, arid conditions of the Mauna Kea summit region will facilitate continued implementation of successful dust abatement control activities. In addition, research can be conducted on identifying the impacts of some of the less significant contributors to dust (e.g., trails) in order to provide additional justification for control of this threat. Further research can include studies to determine if there are human health risks associated with prolonged dust exposure, if dust fallout is adversely impacting biological resources, and where particulates are generated.

Questions:

1. What types of particulates are in the air at MKSR and Hale Pōhaku?
2. Where are the particulates coming from?
3. Are the levels of dust-sized air-borne particulates currently present within the air shed potentially harmful to human health?
4. Are dust-sized particulates negatively impacting the health of the local biological communities?
5. Are there chemical constituents of the dust that may be adding to the nutrient load of Mauna Kea’s upper elevation soils?

Research Projects:

1. Conduct literature search to determine impacts of certain air quality parameters on human health.
2. Conduct literature search to determine impacts of certain air quality parameters on lichens and mosses.
3. Investigate potential linkages between dust-sized, air-borne particulates and local biotic health (e.g., covering leaves and reduction of photosynthesis, clogging of cinder substrate, reduction of available wēkiu bug habitat).

4.1.4.6     Plants

Locations and Resources Included: All vegetation communities found at Hale Pōhaku and MKSR, with special emphasis on māmane woodlands and alpine stone desert lichen and moss communities.