Description of the Arctic LTER site and project The field site of the Arctic Long Term Ecological Research (LTER) project lies 170 km south of Prudhoe Bay in the foothills of Alaska’s North Slope near the Toolik Field Station (TFS) of the University of Alaska Fairbanks . The project goal is to describe the communities of organisms and their ecology, to measure changes that are occurring, and to predict the ecology of this region in the next century. Research at the Toolik Lake site began in the summer of 1975 when the completion of the gravel road alongside the Trans-Alaska Pipeline, now called the Dalton Highway, opened the road-less North Slope for research. This book synthesizes the research results from this site since 1975, as supported by various government agencies but mainly by NSF. This book focuses on the Arctic LTER project of NSF, funded since 1987. Toolik Lake is located in the foothills of the North Slope, the arctic region of northern Alaska, at 68o38’N and 149o36.4’W and an altitude of 719 m. The LTER research site is defined as the catchment south of the junction of two rivers: the Toolik Lake outlet and the headwaters of the Kuparuk River (Fig. 1.2). The view looking south from above Toolik Lake shows the Toolik Field Station, the inlet stream, numerous lakes on the inlet basin, and the Brooks Range (Fig. 1.3). The North Slope (about the size of Nebraska) encompasses the 207,000 km2 of Alaska that drain to the Arctic Ocean. There are three physiographic divisions: coastal plain (71,000 km2), foothills (62,400 km2), and mountains (73,600 km2). The North Slope is underlain by hundreds of meters of permafrost, frozen ground that includes soil, rock, and ice. Permafrost is a product of the low annual temperature (-8oC; it limits the plants and roots to the top meter of soil, called the active layer, which thaws during the three months of summer. Permafrost also holds the 300 mm of annual precipitation at the surface so the tundra is moist, rivers and streams plentiful, and wetlands, ponds, and lakes abundant. The largest disturbance affecting this site occurred when a series of glaciers advanced northward from the valleys of the Brooks Range. Today, the age of the surface soil across the research site is either >300,000 years, ~60,000 years, or ~10,000 years. The age affects the chemistry and topography with the result that there are characteristic drainage patterns, land forms, and even vegetation communities on each of these glacial tills. The dominant plants of the foothills are the sedges and grasses of tussock tundra, a vegetation type that covers some 80% of arctic Alaska. Low shrubs, including birches and willows, grow between the tussocks and abundantly along the streams. Nutrients, particularly nitrogen (N), are in short supply and primary productivity low. Low concentrations of N and phosphorus (P) are also found in streams and lakes; both are ultra-oligotrophic. During the nine months of winter the lakes are covered with ~1.5 m of ice and the streams are frozen solid. Advantages of the Site for LTER Research The LTER program includes more than 20 sites in most of the biomes of the U.S., surrounding oceans, and of the Polar regions. The only arctic site is at Toolik Lake. The choice of this site was primarily made on two criteria, gaps in fundamental knowledge about habitats in the Alaskan Arctic and logistics. The gaps in knowledge arose because most ecological research in the Alaskan Arctic was originally supported by the Office of Naval Research at the Naval Arctic Research Laboratory (NARL) on the flat coastal plain at Barrow. Most of the NARL research (1940’s through the 1960’s) took place in the immediate vicinity of Barrow, that is, in the wet sedge tundra and shallow ponds and lakes. These shallow lakes froze to the bottom each winter and therefore lacked fish. Because of the excellent laboratory and living quarters and the many papers and theses about the local ecology, Barrow and NARL became the obvious site for the International Biological Programme (IBP) of the early 1970’s, a comprehensive project funded by NSF to study cycling of carbon (C), N, and P in the tundra and ponds. Little was known about the foothills region of arctic Alaska. When the IBP ended, the aquatic scientists searched the North Slope for a new site that was both accessible and contained deep lakes with fish. At the same time the construction of the oil pipeline was underway and a gravel road was newly completed from Prudhoe Bay to beyond the Yukon River. The logistics advantage was obvious; a new research site could be located south of Prudhoe Bay along the road where inexpensive truck transport could replace the expensive air transport; Toolik Lake was the obvious choice as it was the only large and deep lake along the road, and limnological research began in 1975. The road also allowed sampling the entire suite of arctic vegetation found on the North Slope: wet sedge and alpine tundras, tussock tundra , hilltop heath, and riparian willows. The Toolik site is mostly surrounded by tussock tundra but it also includes small areas of heath, wet sedge, and riparian shrubs. There are several major streams and a number of large and small lakes in the LTER site. Ecologists at the site also make use of the differences in topography and soils caused by the age of the glacial tills deposited by the three distinct glacial advances. As a result of weathering, leaching, accumulation of organic matter, and vegetation preferences, the pH of soils ranges from 4 to 8. This glacial tills also result in differences in vegetation communities and in water chemistry in lakes and streams. For many ecological questions, the Toolik site also is a pristine end-member of a gradient. For example, the fish communities and populations in many of the lakes and streams have not been modified by sportsman transplants of game fish. There is virtually no nitrogen deposition from acid rain and the western Arctic location of the site minimizes the cross-pole deposition of pollutants (Ford et al. 1995). This contrasts with the Canadian Arctic where deposition of pollutants from Europe can be high. The integrity of the site as a long-term monitoring and experimental location is assured by the US Bureau of Land Management (BLM) control of the land, by the BLM permits made to the University of Alaska, and by the BLM designation of the LTER site as a Research Natural Area. There do not seem to be any oil or gas deposits on the site. Finally, the site has become the location for a large number of research projects and the resulting accumulation of a large quantity of ecological research. It is quite probably the best known arctic research site. This is due to the combination of logistics provided by NSF support which includes helicopter transportation, the availability of the LTER database with all the information since 1975, and availability of climate information from the LTER and the TFS. The LTER data set provides researchers with climate data, soil chemistry information, vegetation distribution information, and other ecological information needed on any project. Another factor that attracts many non-LTER scientists is the availability of the long-term terrestrial experiments of warming the tundra, fertilizing the tundra and streams, reducing the available solar radiation, and excluding grazing. Scientists are allowed to sample and measure the effects of more than 20 years of experimental treatments.
The Natural Setting
The Kuparuk Basin
The LTER site lies in the Kuparuk River Basin, a 9,000 km2 catchment stretching from the edge of the Brooks Range on the south to the Arctic Ocean on the north. The elevation varies from sea level at the north to 720 m at Toolik Lake to several thousand meters at the southern crest of the basin on the slopes of the Brooks Range. The combination of distance from the ice-covered Arctic Ocean in the north and the high elevations of the Brooks Range in the south results in a summer air temperature range of 5oC; it is coldest closest to the ocean and in the mountains at the south end of the basin and warmest mid-basin. Irradiance is closely correlated with temperatures. Zhang et al. (1996) point out, however, that in the winter a temperature inversion is often set up with a maximum height of around 500 m. The result is that the coastal region, lying beneath the inversion, is often colder than the southern half of the drainage basin where the elevation is above 500 m. Tundra vegetation covers the Kuparuk River basin but there are differences in tundra communities and in the amounts of plant material. The map of Walker (www.ArcticAtlas.org) for vegetation distribution (Fig. 1.6) shows that wet tundra (e.g., sedges) dominates the northern part or coastal plain part of the basin. There are areas of moist non-acidic tundra in the northern half of the basin as well as small areas near Toolik Lake in the south. Much of the basin is moist acidic tundra dominated by cotton grass (Eriophorum vaginatum) also called tussock tundra. Large shrubs (e.g., willow spp.), found along the rivers south of the coastal plain, reach their maximum development in the central part of the watershed. The satellite view shows that the maximum “greenness” (NDVI) occurs in the mid-part of the basin. The NDVI can be converted to leaf area index (LAI) which is the area of leaf per area of land surface; the LAI is also highest in the middle parts of the river basin and, reasonably enough, so is the primary productivity as modeled by Williams et al. (2001) at the 1 km2 scale. The Kuparuk River flows north from the edge of the Brooks Range to the Arctic Ocean at Prudhoe Bay. Flow begins in late May or early June with the meltwater peak. Typically there are a few increases in flow during the summer caused by rainstorms and then a cessation of flow in late September. The chemistry of the river is described in McNamara et al. (2008) and Kling et al. (1992). Shallow lakes and ponds formed by thermokarst activity, which is the collapse and settling of soil caused by the melting of buried ice, are abundant on the coastal plain of the Kuparuk basin even though the region was never glaciated. Ponds with a maximum depth of 50 cm often form in patterned ground created by ice wedges formed in the permafrost. These ponds, which freeze solid every winter, were extensively studied at Barrow during the IBP (see Hobbie 1980). Lakes with a maximum depth of around 2.5 m form on the coastal plain when the ice within the soil melts. The biology and chemistry of some of these ponds and lakes is reported in Kling et al. (1992) The LTER Site The site encompasses the upper headwaters of the Kuparuk River including the drainage basin of Toolik Lake and the Kuparuk River basin above the confluence of the outlet stream from Toolik Lake (Fig. 1.2). In practical terms, the research is concentrated in the drainage basins of Toolik Lake, Imnavait Creek, and of the Kuparuk River above the Dalton Highway bridge. The Sagavanirktok River and its tributary, Oksrukuyik Creek, are also sites of research but are not officially an LTER site. The U.S. Bureau of Land Management (BLM) has designated nearly the entire LTER research site a Research Natural Area.
The climate at Toolik Lake is characterized as low Arctic (Hobbie et al. 2003). The average annual temperature is -8oC but during the summer months of June, July, and August the average temperature may climb above 10oC (Fig. 1.7). Winter temperatures average -20oC in the coldest months. While snow may fall on any day of the year, a persistent snow cover begins in mid-to-late September and lasts until late May. Maximum snow depth is around 30 cm but drifts form behind every obstruction and in every hollow. The sun is above the horizon continuously for more than two months in the summer and there is no sun for two months in the winter. During June, July, and August of the same period in Fig. 1.7, the 3 month rainfall averaged 188 mm while total annual precipitation ranged from 250-350 mm. Plant photosynthesis begins as soon as the snow melts but by this time, late May, nearly half of the total solar radiation for the year is past. The low annual temperature means that permafrost, or permanently frozen ground, is present to a depth of about 200 m throughout the LTER site. Each summer the top layer of soil thaws to a depth of 29-46 cm (range from 1990 to 2000) depending upon how warm and wet the summer happens to be (Hobbie et al. 2003). Note that in Fig. 1.7 the soil freezing is not complete for some months after the insulating snow cover develops. Not only does permafrost restrict the rooting zone of plants to the active layer, it also acts like bedrock and seals the soils to water penetration. The result is that water from snowmelt and rain is held in the active layer, especially in the organic matter-rich upper 10-20 cm, and the soils are usually moist despite the low precipitation. When there is enough precipitation to saturate the soil, the resulting runoff is “flashy”. That is, there is a quick peak of flow in the streams but there is little water storage and the recession from the peak is rapid. The distribution of vegetation is dependent upon topography (dry ridge tops, moist hill slopes, lowland areas of water saturated soils) and upon the soil chemistry as determined by the age of the soils (Hamilton 2003, see Chapter 3). As shown in Fig. 1.8, the oldest soils developed on glacial till from the Sagavanirktok glacial advance (>~300,000 years ago), the next oldest soils on till from the Itkillik I advance (~60,000 years ago), and the youngest soils developed on till from the Itkillik II advance (~10,000 years ago). As a result of interactions of topography with age of soils, there are four dominant terrestrial ecosystems: moist acidic tussock, moist non-acidic tussock, heath, and wet sedge (see Chapter 5). Most of the LTER site is covered with acidic tussock tundra (Fig. 1.6). Here, a number of shrubs (willow, dwarf birch) and forbs are always present but the sedge Eriophorum dominates. Nitrogen availability limits primary productivity and nitrogen is rapidly recycled. Net ecosystem productivity is around 140 g C m-2 yr-1, while grazing by large and small herbivores is minimal. Non-acidic tussock vegetation grows on the youngest soils and tends to lack dwarf birch. Large caribou herds pass through the site every 5 to 10 years, and grizzly bear and wolves are the chief predators. The streams in the LTER site are small, 1-10 m across, and relatively shallow. Most streams have bottoms covered with rocks, but some streams have peaty bottoms. Because nutrients are tightly held in the plants and soils, only small amounts move into the stream water and the resulting inorganic N and P concentrations are exceedingly low. Most of the N and P is bound in organic forms as dissolved organic matter (DOM). Only a little of the nutrients in the DOM is available to microbes, perhaps none to algae. As a consequence, stream primary productivity is exceedingly low; most photosynthesis occurs in algae (diatoms) attached either to the stream bottom or to tubes of insect larvae. The food web of the stream is based on the algae, and insect larvae consume the algae and are themselves consumed by other insect larvae or fish. Blackfly larvae filter particles from the water; most of their assimilated food is dislodged algae. There is only one species of fish, the arctic grayling, living in the streams. Because the streams freeze completely each fall, the fish must migrate some tens of kilometers to deep lakes where they survive the winter beneath the ice cover. The only large lake in the LTER site is Toolik Lake, a 26 m deep lake with an area of 1.5 km2. A number of smaller lakes near Toolik Lake, such as those shown in Fig. 1.3, have been used for experiments manipulating nutrients and fish species. All the lakes are ultra oligotrophic with a primary productivity around 10 g C m-2 y-1. Both N and P have very low concentrations, so much so that either can be limiting primary production at any given time. The primary productivity of the water column in the lakes comes mainly from the photosynthesis of nanoplankton, which are mostly small flagellates. Because the algal production and biomass are so low, the zooplankton grazers are also in low numbers and often have little grazing control over their prey. Another major source of primary productivity is benthic algae attached to rocks and sediments in the shallow regions of the lake. These algae are grazed by snails which, in turn, are consumed by lake trout or burbot, the top of the food web. Yet another lake food web, the microbial food web, has its origins in the primary production of lake algae and terrestrial plants (O’Brien et al 1997). In Toolik Lake, however, the numbers and production of bacteria and the numbers of their predator, small colorless flagellates, is much higher than expected for an ultra oligotrophic lake. The extra organic matter, which may be the major factor affecting bacteria, comes from the dissolved organic matter transported by streams from land into the lake (Crump et al. 2003, Chapter 8)
Vegetation and Human History
The most recent glacial advance, the Itkillik II, retreated from the site around 13,000 years before present (Fig. 1.8). Pollen records and radiocarbon dates show that the initial herb tundra of sedge and grass was replaced by a birch-dominated shrub tundra about 12,000 yr B.P, and from ~ 9,500 to 7,000 B.P. alder became established on the North Slope (Brown and Kreig 1983, and see Chapter 4). Radiocarbon dates from the Sagavanirktok River valley and elsewhere in the north-central Brooks Range also indicate that peat-forming plants and woody shrubs became relatively widespread during this interval. Bixby (1993) reports evidence for a warming peak ~7,000 yr B.P. and an unchanged vegetation since that time. Thus the vegetation has been stable at Toolik Lake for the past 7,000 years. The LTER site is close to two of the oldest sites in North America for human habitation. One, the Gallagher Flint Station (Brown and Krieg 1983), is some 40 km to the northeast where charcoal dating gives a value of 10,400 yr B.P. The location is atop a small kame that provides a wide view of the Sagavanirktok River valley. Another site some 200 km to the west, the Mesa Site (Kunz and Mann 1997), has a similar aspect on top of a small hill where dating of charcoal gives a value of 11,700 yr B.P. Both of these sites are evidently places where hunters of the Paleoindian Tradition would sit and scan the valleys for game while reworking flint projectile points and other stone tools. At this time the large herbivores were likely horse, bison, antelope, and mammoth (Kunz and Mann 1997). Archeological evidence, especially the abundant tent rings, indicates that for the past 400 years Nunamiut (“mountain”) Iñupiat hunters and fishermen camped on the glacial moraines and other dry locations in the Toolik basin (Huryn and Hobbie in press). Oil was discovered at Prudhoe Bay in 1968 and in 1970 a construction camp to build a supply road to Prudhoe was set up at Toolik Lake. This 400-person camp was initially set up by truck transport along a winter road over frozen tundra, the Hickel Highway. In 1974 the gravel haul road, now the Dalton Highway, was completed from Prudhoe Bay to the Yukon River and the construction camp closed soon after the oil pipeline was completed in 1976. The scientific camp at Toolik was set up in the summer of 1975 and has evolved to become the Toolik Field Station. In recent years the road has been opened to the public. Caribou hunters arrive in late summer for a bow-hunting and rifle season. Tourist buses (one per day) pass Toolik Lake and there are a number of private vehicles as well. The BLM regulations do not allow camping or off-road vehicles in the Toolik Research Natural Area. Disturbance Ecologists now recognize how disturbances shape landscapes (Turner et al. 2003). In fact, ecological systems are seldom in equilibrium; rather, they are often in some phase of recovery from a prior disturbance. Some of the disturbances discussed by Turner are fire, flood or drought, hurricane, insect pests, exotic species, and land use. The landscape of the Arctic LTER site is unusual in that the major disturbance, the advance of a glacier, occurred some 13,000 years ago. As noted above, there has been little change in the vegetation for the past 7,000 years. Fire does affect some tundra systems, and is the key disturbance that has shaped ecosystems of the Bonanza Creek LTER, the other Alaska LTER project near Fairbanks in the boreal forest. For arctic Alaska, Walker (1996) stated that tundra fires are rare and that there was no record of a major fire in arctic tundra: the general belief was the tundra was too moist for a fire to spread. But in 2007 a combination of a dry summer and many lightning strikes on the North Slope created a fire that burned close to 1,000 km2 of tussock tundra northwest of Toolik Lake. This “natural experiment” is under current study to determine the changes in terrestrial carbon lost to the atmosphere and to the streams and lakes (Rocha et al. 2011). The natural changes that take place in a landscape after disturbance by glaciers include the geomorphic softening of landscape, the leaching and development of soils, the growth of plants, the thickening of soil organic horizons, the erosion of stream channels and formation of water tracks, and the formation of ice wedges. Because the LTER site includes soils of three different geologic ages, these processes have produced different results across the landscape. For example, the youngest soils have a pH of 6-7 while the soils of the older surfaces are quite acidic because of leaching of carbonates and accumulation of dissolved organic matter. Sphagnum moss is only found on the low pH surface. One natural disturbance that appears to be increasing as the climate warms is called thermokarst. When this melting of ice happens in the permafrost, whole hillsides may shift and new streams may form. Not only is the vegetation disturbed or destroyed but water erosion may transport soil and nutrients into lakes and rivers, affecting the growth of algae and the survival of freshwater communities. An interesting interaction between fire and thermokarst that was found in the large 2007 fire described above, is that when fire disturbs the insulating vegetation on the soil surface, deeper soils warm, ice melts, and thermokarst disturbances of the land surface become very common, especially around streams and lakes. Walker (1996) discussed in detail the possible anthropogenic disturbances in the Arctic, many of which are only found in industrial sites such as the Prudhoe Bay oil fields. Those disturbances of possible importance at Toolik Lake include: trash and solid waste, diesel or gasoline spills, thermokarst and thermal erosion, snow drifts from roads and buildings, impoundments, fire, off-road vehicle trails, roads and road dust, gravel borrow pits, acid rain or increased sulfates, air-borne pollutants, and climate change. Because of large-scale patterns of air movement, this region of the Arctic receives very low pollutant loads (Ford et al. 1995) but even in this isolated site PCBs and organochlorine pesticides are present in fish (Wilson et al. 1995). Although there are no major disturbances affecting the landscape around Toolik Lake, there are minor disturbances that must be kept in mind when interpreting biological and chemical records. This region of the Arctic is experiencing warming and has so for the past 150 years. At Barrow, there has been an increase in the annual average air temperature by nearly 2oC in the past 30 years. Details of the record, projections of climate into the future, and the responses of the Arctic LTER ecosystems to this warming are discussed later in the book. At Toolik Lake there is evidence that some species of plants now have more biomass and are more abundant, and that the alkalinity of streams and lakes has increased in the last 30 years. This change in alkalinity is very likely caused by increased weathering of previously frozen glacial till but the exact process remains unknown. Another disturbance was the research (1975) and construction camps (1970) at Toolik Lake. The construction camp was sited at the north end of the lake and any drainage of waste water entered rivers downstream from the lake. The scientific camp (now the Toolik Field Station) always had a strict policy that all wastes are removed from the site and taken for disposal to Prudhoe Bay. The field station is located on a former gravel borrow pit. The building of the road was, however, a measurable disturbance (Walker 1996). One reason was that seven gravel borrow pits were excavated within the Toolik Lake drainage basin. In a typical borrow pit, the upper several meters of a glacial kame, a small hill of water-sorted sand and gravel, was removed for road construction. The resulting surface has no organic soil and is very dry. As a result, there is little revegetation; in some cases grasses were seeded and the sites fertilized. More important, the underlying permafrost thaws over the years and exposes fresh glacial till and gravels to erosion and weathering. Hobbie et al. (1999) describe the stream flowing at the edge of one of these pits in which the stream supplies 5% of the water entering Toolik Lake but 35% of the phosphate. A second disturbance from the road is road dust. Since the road opened in 1974, each truck that passes creates a dust plume, especially in the summer. The dust is deposited downwind in a log-decay relationship, where the amount of dust deposited decreases logarithmically with the log of distance from the road. Everett (1980) measured the dust load near Toolik over 96 days to be 200 g m-2 at 10 m and 1.5 g m-2at 1000 m from the road. One effect is an earlier snow melt by days and weeks along the road corridor caused by a lowered albedo of the snow. Another effect is a change in reflectance of the road corridor that is visible in summer satellite pictures. While some environmental changes near the road have occurred (see Leadley et al. 1996), the overall importance of dust deposition is difficult to determine; however, there is no evidence so far of any widespread major changes in vegetation or water chemistry. 
- . The biogeochemistry and zoogeography of lakes and rivers in arctic Alaska. Hydrobiologia. 1992 ;240:1-14.