Wildlife Response to Environmental Arctic Change

Home , Redpoll

U.S. Fish & Wildlife Service Wildlife Response to Environmental Arctic Change Predicting Future Habitats of Arctic Alaska Workshop Sponsors University of Alaska Fairbanks (UAF) International Arctic Research Center ABR Inc. UAF Institute of Arctic Biology Wildlife Conservation Society

Workshop and Report Coordination The Arctic Research Consortium of the U.S. (ARCUS) coordinated workshop planning and implementation, and report layout and editing.

rctic Res l A ea a r n c o h i t C a e n r n

e e

r r

n I

U s k n n iv a er rb sity ai of Alaska F

Workshop Speakers David Atkinson, Erik Beever, Eugenie Euskirchen, Brad Griffith, Geoffrey Haskett, Larry Hinzman, Torre Jorgenson, Doug Kane, Peter Larsen, Anna Liljedahl, Vladimir Romanovsky, Martha Shulski, Matthew Sturm, Amy Tidwell, and Mark Wipfli.

Report Reviewers Brian Lawhead, Anna Liljedahl, Wendy Loya, Neil Mochnacz, Larry Moulton, Stephen Murphy, Tom Paragi, Jim Reist, Vladimir Romanovsky, John Walsh, Matthew Whitman, and Steve Zack.

Photo Contributors Doug Canfield, Stephanie Clemens, Bryan Collver, Fred DeCicco, Elizabeth Eubanks, Richard Flanders, Larry Hinzman, Marie-Luce Hubert, Benjamin Jones, M. Torre Jorgenson, Jean-Louis Klein, Catherine Moitoret, Mitch Osborne, Leslie Pierce, Jake Schaas, Ted Swem, Ken Tape, Øivind Tøien, Ken Whitten, Chuck Young, Steve Zack, and James Zelenak.

Cover Image Ice floating in a lagoon near Collinson Point in the Arctic National Wildlife Refuge. Photo by Philip Martin.

Suggested Citation Martin, Philip D., Jennifer L. Jenkins, F. Jeffrey Adams, M. Torre Jorgenson, Angela C. Matz, David C. Payer, Patricia E. Reynolds, Amy C. Tidwell, and James R. Zelenak. 2009. Wildlife Response to Environmental Arctic Change: Predicting Future Habitats of Arctic Alaska. Report of the Wildlife Response to Environmental Arctic Change (WildREACH): Predicting Future Habitats of Arctic Alaska Workshop, 17-18 November 2008. Fairbanks, Alaska: U.S. Fish and Wildlife Service. 138 pages. Wildlife Response to Environmental Arctic Change Predicting Future Habitats of Arctic Alaska

Report from the Wildlife Response to Environmental Arctic Change (WildREACH): Predicting Future Habitats of Arctic Alaska Workshop 17–18 November 2008 Westmark Hotel Fairbanks, Alaska

Authors: Philip D. Martin, Jennifer L. Jenkins, F. Jeffrey Adams, M. Torre Jorgenson, Angela C. Matz, David C. Payer, Patricia E. Reynolds, Amy C. Tidwell, and James R. Zelenak

U.S. Fish & Wildlife Service 101 12th Avenue, Room 110 Fairbanks, Alaska 99701

Phone: 907-456-0325 http://alaska.fws.gov/

Table of Contents

Chapter 1: Executive Summary ...... 2 Chapter 2: Introduction ...... 8 Chapter 3: Arctic Alaska in a Changing Climate ...... 12 Chapter 4: Habitat Change ...... 38 Chapter 5: Climate Effects on Fish and Wildlife ...... 60 Chapter 6: Working Groups’ Summaries: Common Themes and Research Gaps ...... 94 Chapter 7: Conclusions and Recommendations ...... 110

Appendices Appendix 1: Methods for Climate Projections ...... 117 Appendix 2: Description and Distribution of Northern Alaska Ecotypes ...... 120 Appendix 3: WildREACH Workshop Agenda ...... 127 Appendix 4: WildREACH Workshop Participants ...... 129 Appendix 5: Working Group Charges and Members ...... 135 Chapter 1 Executive Summary

2 Climate is changing worldwide, In the Arctic, climate affects habitat but the Arctic is warming at a uniquely through the interdependen- rate almost twice the global aver- cies of permafrost, hydrology, and age. Changes already observed in vegetation. The deep, cold, continu- arctic terrestrial landscapes include ous permafrost of the North Slope rapidly eroding shorelines, melting represents a reservoir of resilience ground ice, and increased shrub for this landscape. Nevertheless, growth at high latitudes. Because enhanced seasonal melting of near- the Arctic will likely experience surface ice is already measurably early and disproportionately large altering habitats and hydrology. impacts of climate change, the U.S. Understanding how variation in Fish and Wildlife Service (Service) the type and quantity of ground ice has identified America’s Arctic as a influences a landscape’s susceptibil- priority region for developing man- ity to warming is fundamental to agement strategies to conserve fish, predicting the extent and magnitude wildlife, and their habitats. of habitat change.

The Service convened a Wildlife Hydrologic processes are a pivotal Response to Environmental Arctic determinant of climate-influenced Change (WildREACH) workshop on habitat change in arctic Alaska. 17–18 November 2008 in Fairbanks, Changes in overall water balance Alaska. Our goal was to identify the and in timing and magnitude of sea- priority research, modeling, and sonal water and energy fluxes will synthesis activities necessary to strongly affect habitat availability advance our understanding of the and quality for arctic-adapted spe- effects of climate change on birds, cies of fish and wildlife. The seasonal fish, and mammals of arctic Alaska, allocation of precipitation is key to focusing on terrestrial and freshwa- ecosystem response in an environ- ter systems. We used a conceptual ment where water remains frozen modeling approach to identify the most of the year. Despite the expec- potential changes that would most tation of higher annual precipitation, strongly influence habitat suitability models predict a generally drier for a broad suite of arctic species. In summer environment. Refining mod- doing so, we embarked on the first els to more confidently predict water essential step toward incorporating balance and the resultant water climate considerations into biological supply available to various habitat planning and conservation design types is one of our most important for the Arctic. The workshop was at- challenges. tended by over 100 participants representing federal and state agencies, academia, and commercial and non- profit organizations. WildREACH provided a forum for communication among specialists from multiple disciplines, a vital first step toward establishing effective partnerships. Summaries of each workshop report chapter are provided below.

Climate, Permafrost, Hydrology The average annual temperature of Alaska’s North Slope is projected to rise approximately 7°C by 2100. The magnitude of change is impre- cisely known, but Global Circulation Models identify northern Alaska as one of the fastest warming regions of the planet. Annual precipitation is also expected to increase, although there is less certainty surrounding this prediction.

The presence of ice-rich permafrost 3 soils makes arctic tundra uniquely vulnerable to the effects of warming. Photo from USFWS, Ikilyariak Creek, Arctic National Wildlife Refuge. Habitat Change • Floodplains are very dynamic Effects of climate change on North landscapes and could respond to Slope habitats will vary depend- climate change in a variety of ways. ing on the permafrost-influenced Floodplain processes are influ- geomorphic processes specific to enced more strongly by extreme particular ecosystems. It is useful to flood events than by average condi- consider the coastline, Coastal Plain, tions, and models of future flood Foothills, and floodplains separately. frequency and severity must be better developed in order to predict • In the coastal zone, rapid shoreline habitat change. erosion is occurring, associated with the retreat of summer sea Historically, tundra fires have been ice. Rising ocean temperatures, rare on the North Slope, but fire sea level rise, permafrost degrada- frequency will likely increase as the tion, increased storm surges, and climate warms. A positive feedback changes to river discharge and relationship exists whereby soils sediment transport will continue to tend toward a warmer and drier affect habitat availability and qual- condition after fire, which in turn ity in the coastal zone. promotes shrub growth and a more fire-prone landscape. Although wide- • The vast shallow wetlands of the spread conversion of North Slope Coastal Plain landscape are sensi- tundra to spruce forest is not ex- tive to changes in water balance pected within this century, increased that could lead to drying. Lakes shrub cover has been documented may enlarge through melting and in the Brooks Range and Foothills, erosion at their edges. Alterna- a trend that is expected to continue. tively, lakes may drain if surround- Changes in plant phenology (e.g., ing ice wedges degrade, resulting earlier green-up and senescence) in the formation of new drainage are certain to occur as spring melt networks. comes earlier.

• The hilly terrain of the Arctic Climate change may increase avail- Foothills is prone to thaw slumps ability and uptake of contaminants and gully formation. In the lower for fish, wildlife, and their habitats. Foothills, extremely ice-rich soils Contaminants currently contained are susceptible to ice wedge degra- within glacial ice, multi-year sea ice, dation, melting of massive ice, and and permafrost, including persistent formation or drainage of thermo- organic pollutants and mercury, karst lakes. will almost certainly be released to aquatic ecosystems as the temperature rises.

Climate Effects on Fish and Wildlife WildREACH workshop participants formed working groups for birds, fish, and mammals. Each working group developed conceptual models to illustrate hypotheses of likely pathways by which fish and wildlife populations of arctic Alaska may be affected by climate change. Hydro- logic process models for summer and winter provided linkages among climate variables, physical processes (hydrologic and permafrost), and habitat change. These processes were relevant to all species groups.

The bird working group developed conceptual models organized around four broad topics: abundance and distribution of surface water, vegeta- 4 tion community change, invertebrate community change, and coastal Spring melt is accompanied by a sharp peak in flow for rivers that arise in processes. the Brooks Range—changes in precipitation and warming temperatures may change flow regimes and sediment transport in arctic rivers. Photo from USFWS, Sadlerochit Mountains, Arctic National Wildlife Refuge. The fish working group developed a Basic biological studies, therefore, single conceptual model emphasiz- are also needed. Focal species should ing pathways related to the effects be chosen based on their predicted of increased water temperature and vulnerability to climate change and hydrologic changes related to soil potential to serve as indicators of moisture, glacial input, drainage hypothesized habitat changes. changes related to permafrost degradation, and changes in lake area. Conclusions and Recommendations WildREACH workshop discussions The mammal working group devel- revealed several specific information oped separate models for the sum- gaps within the four major thematic mer and winter seasons. Key factors areas previously listed (see Table in winter included changes in the on page 7). These gaps represent timing, amount, and nature of pre- the highest scientific priorities for cipitation (e.g., rain-on-snow events, scientific inquiry, which should be deeper snow). In summer, changes in pursued in an organized, multidisci- plant species composition, amount of plinary fashion. Specific recommen- forage, and seasonality were expect- dations include: ed to have the greatest potential for 1. Establishment of at least three affecting mammal populations. long-term observatories on the North Slope to collect integrated Common Themes and Research Gaps hydrologic, climate, and geophysi- Despite the uncertainty in project- cal data. The central mission of ing climate change impacts on arctic these observatories should be to species and habitats, workshop par- develop an understanding of the ticipants identified monitoring, re- response of permafrost (active search, and modeling priorities that layer dynamics), hydrologic, and will help improve our understanding ecological systems to changes in of future conditions. Specific infor- thermal regime. To ensure appli- mation gaps varied among species cability to fish and wildlife biology, groups, but most fell into four cross- water budgets should be estimated cutting themes: 1) changes in pre- for key ecotypes. cipitation and hydrology; 2) changes 2. Intensive observations at the in vegetation communities and phe- observatory sites should be supple- nology; 3) changes in abundance and mented by instrumentation (e.g., timing of invertebrate emergence; meteorology, radiation, stream dis- and 4) coastal dynamics. charge, soil moisture) at dispersed sites arrayed across important All working groups emphasized that environmental gradients. predictions regarding climate effects 3. Modeling that dynamically on fish and wildlife populations must couples soil thermal and hydrologic be tentative, given the uncertainty regimes, and biological systems at surrounding climate forecasts and appropriate spatial and temporal unavailability of models that couple scales. climate, geophysical, and ecological 4. Centralized data storage and in- processes at appropriate spatial and terpretation for the mutual benefit temporal scales. All working groups of multiple end-users. agreed that in order to more accurately predict climate change effects on species and habitats, multidisciplinary work is needed to better understand the underlying biological and physical processes that drive terrestrial and aquatic ecosystem function and the response of those systems to climate change. Hydro- logic processes, in particular, are pivotal determinants of climate-related habitat change, and enhanced data collection and modeling in this area will benefit multiple users.

All working groups emphasized that 5 information available on life history, habitat requirements, distribution, abundance, and demography is A flock of black brant migrate along the Beaufort Sea coast. Optimal timing inadequate for many arctic species. for bird migration could change under an altered climate regime. Photo by Philip Martin from Canning River delta, Arctic National Wildlife Refuge. We also recommend immediate expertise to implement conservation, attention to developing predictive research, and management at all models of habitat change, focusing scales. initially on processes that are occurring now and that act on short (e.g., The Service will improve commu- decadal) time scales. Priority topics nication and collaboration with the include: arctic research community to initiate 1. Coastal processes (e.g., erosion, building of wide-ranging partner- storm surge, deposition, vegetation ships. On a local, regional, and succession); national level the Service will: 2. Seasonality (e.g., plant phenology, 1. Work with the National Science animal migration, life stages of Foundation (NSF) to define climate aquatic invertebrates); research priorities relevant to re- 3. Shrub advance; source management agencies; 4. Fire regime (as a function of inter- 2. Increase collaboration with actions among climate, permafrost, academia and other researchers to and vegetation); and develop grant proposals that ad- 5. Thermokarst effects on surface dress priority questions; water storage, drainage systems, 3. Participate in planning and imple- and lakes. mentation of the interagency Study of Environmental Arctic Change The Service should engage the U.S. (SEARCH) Program to ensure Geological Survey (USGS) and oth- inclusion of research relevant to ers in a structured decision-making resource management agencies; process to refine the selection of 4. Work with arctic science program indicator species/parameters as managers in the research agencies components of a long-term climate (e.g., NSF, USGS, National Oceanic monitoring program. Upon reaching and Atmospheric Administration) consensus, management agencies to obtain funding for work that ad- should seek stable funding for moni- dresses priority questions; and toring these species/attributes. 5. Promote a collaborative approach to acquire, process, archive, and The Service recognizes that we disseminate essential satellite- must change the way we do busi- based remote sensing data prod- ness to succeed in managing fish, ucts (e.g., snow cover, green-up, wildlife, and their habitats in a and surface water) needed for rapidly changing climate. We can no regional-scale monitoring. longer manage for the status quo— we must manage for an uncertain Climate change presents an unprec- future. These challenges exceed the edented challenge to managers of capacity of any one agency, and we arctic natural resources. By initiat- must pool our collective resources. ing a collaborative process among By strategically targeting financial biologists, physical scientists, and resources, we can build Landscape managers, the WildREACH work- Conservation Cooperatives that shop successfully identified prior- increase capacity, eliminate redun- ity information gaps and activities dancy, and provide the technical needed to provide the basis for adaptive management of arctic fish and wildlife resources. Since the workshop, the Service has identified America’s Arctic as Alaska’s An undercut first Landscape Conservation bluff on the Region, which will be supported by Beaufort Sea the technical capacity housed in the coast, the result Northern Alaska Landscape Con- of a severe storm servation Cooperative. Adopting the in August 1980, WildREACH recommendations is illustrates the the next step in strengthening our susceptibility capacity to anticipate climate-related of ice-rich habitat change and to identify the coast to rapid most promising strategies to con- erosion. Photo serve fish and wildlife populations in 6 by Catherine America’s Arctic. Moitoret, Canning River delta, Arctic National Wildlife Refuge. Scientifc Priorities

Workshop participants identified important information gaps in our understanding of climate change effects on birds, mammals, and fish populations. The specific gaps varied among species groups, but most fell into four cross-cutting thematic areas and underlying research questions (see Chapter 6 for more details):

1. Precipitation, Water Balance, and Distribution of Surface Water a. How reliable are the projections for increasing precipitation and evapotranspiration? b. How will the annual precipitation input on the Coastal Plain and Foothills be allocated between winter (snow pack) and summer? c. How will changes in precipitation, evapotranspiration, and active layer depth alter summer surface water availability in shallow-water and mesic/wet tundra habitats? d. How will changing patterns of seasonal runoff affect stream flow? e. What is the contribution of groundwater in various systems, and is it sufficient to maintain year-round flow? f. Will drought conditions and changes in drainage patterns decrease water body connectivity? g. Which Coastal Plain lakes are susceptible to tapping (rapid drainage) and on what time scale? h. What are the expected changes in snowpack characteristics (depth, density, presence of ice layers) and how might these vary on a regional and local scale? i. How much change will occur in the timing of snow melt and snow onset? j. How will the frequency of rain-on-snow and severe winter storm events change?

2. Vegetation Community Composition and Phenology a. How will changes in the length and timing of the growing season influence plant phenology, including seasonal changes in nutritional quality? b. How will plant species composition shift in response to long-term climate change, and what are the implications for habitat structure and quality of the prevalent available forage (i.e., digestibility, nutrient content)? c. What is the time scale of expected shrub increase, and how will this vary by species/growth form (low vs. tall shrub) and ecoregion? d. What is the likelihood of widespread conversion from sedge and sedge-shrub meadow to bog meadow (paludification) and how would this affect herbivore and detritus-based trophic systems? e. How will changes in the seasonality of stream discharge and occurrence of flood events influence development of riparian vegetation communities?

3. Abundance and Phenology of Invertebrates a. How does earlier spring thaw affect timing of life cycle events and peak availability to predators? b. How does temperature affect growth and development of aquatic insects? c. What climate-related changes are likely in community composition of macroinvertebrates in stream, lake, and saturated soil environments? d. How will changes in the distribution and quality of surface waters and shifts from pelagic to benthic productivity in deep lakes affect availability of macroinvertebrates to fish and wildlife? e. How will warming and changing seasonality affect abundance and peak activity periods of biting insects and what are the bioenergetic consequences for caribou in particular? f. How will warming and changing seasonality affect the prevalence of parasites and disease vectors (e.g., nematode parasites of muskoxen and Dall’s sheep)?

4. Coastal Dynamics a. Will higher water temperatures, sea level rise, and retreat of summer sea ice cause degradation of the barrier island systems of the Beaufort and Chukchi seas? b. Will alluvial deltas continue to build or will rising sea levels outpace potential increases in sedimentation rates? c. How quickly will shoreline retreat result in newly breached lake basins? d. To what extent will coastal erosion, in combination with sea level rise, cause salinization of low-lying coastal areas? e. Will coastal wet sedge meadows establish at a rate equal to loss of this habitat through erosion and inundation? 7 f. Will increased fogginess/cloudiness exert a negative or positive feedback effect on air temperature in the coastal zone? What is the expected spatial extent of this effect? Chapter 2 Introduction

Climate warming is having profound effects on fish, wildlife, and their habitats, challenging resource agencies to manage ecosystems that are being fundamentally altered.

8 Climate is changing across the value wildlife habitat is by no means planet, but the Arctic is warming at restricted to these discrete loca- a rate almost twice the global aver- tions, however. Wetland and coastal age, and precipitation is projected habitats across the North Slope to increase significantly by the end provide nesting and rearing habitat of the century. Sea ice extent has for millions of birds that migrate decreased steadily over the past 40 north from four continents to exploit years, and the Arctic may be ice-free this seasonally productive habitat. in the summer within a decade. Arc- Four distinct caribou herds roam the tic terrestrial landscapes are also be- North Slope, from the Chukchi Sea ing altered by warming with rapidly to the Canadian border. This rich eroding shorelines, melting ground landscape also provides habitat for ice, and increased shrub growth three threatened species, a variety at high latitudes. The presence of of resident and anadromous fish spe- frozen ground creates an environ- cies, muskoxen, and furbearers. ment uniquely sensitive to warming, because permafrost holds water at Across this region, the Service the surface, creating vast wetlands manages significant trust resources and aquatic habitats in a region that including threatened species (Stell- would otherwise be classified as arid. er’s eider, spectacled eider, and polar If unabated, melting ground ice will bear) and migratory bird species of ultimately result in the drying of concern (Pacific black brant, yellow- northern Alaska with profound con- billed loons). Refuge-held lands sequences for fish and wildlife. within the Arctic are not only significant for their habitat values, as il- The mission of the U.S. Fish and lustrated in part by their importance Wildlife Service is, in part, to con- to caribou, fish, and migratory birds, serve fish, wildlife, and their habi- but also contain the only designated tats for future generations. To date, Wilderness in the Arctic. In addition, the Service’s management objectives the Service is responsible for provid- have generally been defined in terms ing opportunities to the residents of of maintaining the status quo or North Slope villages for continued restoring past conditions. Climate subsistence harvest. In the Arctic, change forces us to consider a future climate change is already impacting in which ecosystems and species Native cultures and natural resourc- assemblages will differ dramati- es. It is vital that stakeholders work cally from the historical condition. collaboratively to develop systematic Because the Arctic will likely experi- approaches for monitoring impacts, ence early and disproportionately predictive models, and strategies for large impacts of climate change, the management and conservation. Service has identified America’s Caribou (Rangifer tarandus) Arctic as a priority region for The Service convened a Wildlife in the Arctic National Wildlife developing management plans and Response to Environmental Arctic Refuge. Photo by Doug Canfield, adaptation strategies to conserve Change workshop (WildREACH) on USFWS. fish, wildlife, and their habitats.

The North Slope of Alaska en- compasses America’s terrestrial Arctic and is a large and relatively non-impacted landscape north of the Brooks Range. It is the only example of the arctic biome within the United States. Two well-known areas of high-value wildlife habitat are the portion of the Arctic Coastal Plain within the Arctic National Wildlife Refuge and the Teshekpuk Lake Special Area in the National Petroleum Reserve–Alaska. The Teshekpuk Lake region is a unique molting area for Pacific black brant and three other species of geese, and is used by many other water birds 9 for feeding, nesting, and molting. The refuge's Coastal Plain is best known as the calving ground of the Porcupine Caribou herd. High- 17–18 November 2008 in Fairbanks, design and implement conservation Alaska, to identify the priority strategies via LCCs. By identifying research, modeling, and synthesis areas of common interest among activities necessary to advance researchers and managers and our understanding of the effects of promoting collaborative efforts, we, climate change on birds, fish, and as a conservation community, will mammals dependent on the ter- improve the quality of information restrial and freshwater systems of available to support management arctic Alaska (see Appendix 3 for the decisions. WildREACH was the workshop agenda). Scientists from beginning of a collaborative process a variety of disciplines joined fish to develop conceptual models and to and wildlife managers to discuss the highlight areas of uncertainty and linkages between climate, hydro- information gaps, including aspects logic and geomorphic processes, and of both geophysical and biological ecosystem change (see Appendix 4 processes. Since the workshop, the for list of participants). A concep- Service has identified America’s tual modeling approach was used to Arctic as Alaska’s first Landscape identify potential changes that would Conservation Region, which will be most strongly influence habitat supported by the technical capacity suitability for a broad suite of arctic housed in the Northern Alaska LCC. species. The conceptual models are In order to provide better informa- considered the forerunners to a later tion and decision-making tools for generation of spatially explicit mod- managers, our next steps will be to els that will be designed to forecast begin to address information gaps changes in habitat availability and and development of habitat change quality. Landscape-change models models. The Service recognizes are critical to the Service and others that our success in conserving fish, working toward identifying those wildlife, and their habitats depends species and habitats most vulnerable on building shared capacity and to adverse climate-related impacts. establishing effective collaborative As models are refined and informed partnerships with specialists from by new research and monitoring multiple disciplines. data, we will improve our ability to design and implement manage- This report synthesizes the infor- ment strategies that are responsive mation presented at the workshop. to urgent conservation problems Background information on the and are logistically and financially region’s current and projected achievable. climate, permafrost, and hydrology is presented in Chapter 3; pro- The WildREACH workshop also cesses of geomorphic and habitat was a first step in establishing a change are outlined in Chapter 4; Landscape Conservation Coop- and conceptual models of fish and erative (LCC). Because the scope wildlife response to climate change and complexity of climate change are developed in Chapter 5. Priority problems surpass the abilities and research, monitoring, and modeling Tussock tundra covers vast areas budget of any one agency or organi- tasks are summarized in Chapter of the Arctic Foothills and Brooks zation, the Service envisions building 6, and final recommended actions Range. USFWS photo. shared capacity among scientists to are in Chapter 7. The priorities and conclusions are those of the Fish and Wildlife Service, as informed by input from the diverse group of workshop attendees. We recognize this as a first of many steps toward coordinating efforts with other resource agencies, researchers, and non-governmental organizations. At present, the scale of our knowledge does not match the scale of our management responsibilities. We look forward to continuing to develop this coordinated partnership to assess, monitor, and predict the impacts of 10 climate change on terrestrial and freshwater landscapes in America’s Arctic so that we may conserve this landscape and the animals who live there for future generations. Ice remnants in Lake Ikpikpuk in July 2005. Photo by Leslie Pierce, TREC 2005, courtesy of ARCUS.

11 Chapter 3 Arctic Alaska in a Changing Climate

This background chapter provides a brief introduction to the ecoregions of the North Slope of Alaska, an overview of both observed and projected climate change in the region, and some general information about permafrost and hydrologic processes in the Arctic.

12 Alaska’s North Slope covers ap- Range (Figure 3.2). Information for proximately 204,000 km2 of arctic each ecoregion is derived from Gal- North Slope lands, extending from the shores lant et al. (1995). of the Beaufort and Chukchi seas Ecoregions to the crest of the Brooks Range Arctic Coastal Plain mountains. Human population within The 50,000 km2 Arctic Coastal Plain the region is about 7,000 residents is the northernmost ecoregion in (2004 census), with activity and Alaska. The Coastal Plain is bound- infrastructure concentrated within ed on the west and north by the eight small communities and the Chukchi and Beaufort seas and ex- widespread industrial zone with a tends east nearly to the U.S.-Canada hub at Deadhorse/Prudhoe Bay. border. The region is underlain by Land management and ownership thick permafrost, up to 650 m deep is partitioned among seven major at Deadhorse/Prudhoe Bay. Perma- entities (Figure 3.1). The Bureau of frost-related surface features are Land Management (Department of common, including pingos, ice wedge the Interior [DOI]) is responsible polygons, peat ridges, and frost for management of a large portion boils. Most major streams originate of the region—nearly 45% of the from the other ecoregions to the total area or roughly 91,000 km2. The south. Lakes of both thermokarst State of Alaska is the second largest and non-thermokarst origin are landowner, responsible for 41,000 ubiquitous, constituting over 14% of km2 (20%). Fish and Wildlife Service the landscape. Streams west of the (DOI) lands account for approxi- Colville River are interconnected mately 35,000 km2 (17%), and Native with lakes and tend to be sluggish lands encompass about 26,000 km2 and meandering, while those east of (13%). The remaining lands are held the Colville River are braided and by the National Park Service (DOI), build deltas into the Arctic Ocean. private owners, and the Department Most of the smaller streams freeze of Defense. completely during winter.

The North Slope is divided into three ecoregions: Arctic Coastal Plain, Arctic Foothills, and Brooks

E A S B E A U F O R I Barrow T H S E C A K U Wainwright H C Atqasuk Kaktovik Point Lay Nuiqsut Deadhorse

Point Hope

Land ownership and infrastructure Anaktuvuk Pass Bureau of Land Management Fish and Wildlife Service National Park Service Native Corporation State of Alaska Existing infrastructure Kilometers 13 0 50 100 200 300 400 500

Figure 3.1. Land ownership and location of human infrastructure on Alaska’s North Slope. Map by USFWS from Alaska Department of Natural Resources data. Arctic Foothills age networks than the Coastal Plain. The Arctic Foothills, a 96,000 km2 Most streams tend to be swift, but band of rolling hills and plateaus, portions may be braided and smaller grades from the Coastal Plain to the streams dry or freeze during winter. Brooks Range. The Foothills stretch Flooding and channel shifting is from the Chukchi Sea in the west to common during breakup of river ice. the U.S-Canada border. The ecore- Lakes of glacial origin and oxbow gion is underlain by permafrost, and lakes, located along major streams, permafrost-related surface features are the predominant types in the are common on the landscape. The region. Foothills have better defined drain- Brooks Range The Brooks Range ecoregion is the Alaskan extension of the Rocky Mountains, of which 58,000 km2 lies north of the continental divide. The ecoregion covers most of the east- west extent of northern Alaska— from the U.S.-Canada border to within 100 km of the Chukchi Sea. In contrast to the Arctic Coastal Plain and Arctic Foothills, this ecoregion was extensively glaciated during the Pleistocene epoch, but the few remaining glaciers are limited to the eastern sector. The terrain is dominated by rugged mountain com- plexes, and continuous permafrost Ann underlies the region. The combina- ual tion of harsh climate, shallow soils, and highly erodible slopes result in sparse vegetation cover that is generally limited to valleys and lower hill slopes. Streams in the Brooks Range often have braided drainage patterns, with larger streams draining north and their tributaries flowing east or west. Lakes constitute less than 1% of the landscape and tend to occur in rock basins at the mouths of large glaciated valleys, in areas of ground and termi- nal moraines, and on floodplains of -8 -4 -2 -1 -0.5 0.2 0.5 1 2 4 8 major rivers. D M ec– ar–A Ja pr n– –M Fe a b y

Figure 3.2 (top). Ecoregions on the North Slope of Alaska. Map by Jun Sep –Ju t–O l– ct USFWS from USGS data (Gallant A –N ug o v et al. 1995). Figure 3.3 (bottom). Annual and seasonal temperature trends for the northern hemisphere, 1958–2008. 14 The scale indicates temperature change (°C) over the period, based on local linear trend. Data from Goddard Institute for Space Studies (http://data.giss.nasa.gov/gistemp), downloaded January 15, 2009. Climate Trends in the Arctic ing warmer summers and colder The arctic climate has warmed rap- winters in comparison to the coastal Climate idly over the past 50 years. Annual zone (Figures 3.6, 3.7, and 3.8 top average temperature has increased panels). Annual precipitation ranges nearly twice as fast as the rest of the from 150 mm (6 in) near the coast at world (ACIA 2004: 10). This polar Barrow to 600 mm (24 in) or more amplification of warming is attribut- in portions of the Brooks Range ed to: 1) positive feedback effects of (Figure 3.9 top panel). There are greater heat absorption associated only four stations in the region with with reduced snow and ice cover multi-decadal weather records, and on land and sea, 2) larger fraction only one (Barrow) with a continuous of energy going to warming rather record from the early 1900s until than evaporation relative to the trop- present (Shulski and Wendler 2007), ics, 3) shallower troposphere (lower resulting in a data-poor historical atmosphere) and frequent surface- record. Despite considerable an- based temperature inversions, and nual variation, however, the 50-year 4) atmospheric and oceanic circula- trend in mean annual temperature is tion (ACIA 2004: 20). Compared to positive (Figure 3.5), rising at a rate the rest of the circumpolar Arctic, of 0.45°C (0.81°F) per decade. Based terrestrial areas in northern Alaska, on the two best time series (Bar- western Canada, and central Russia row 1949–1996 and Barter Island have experienced the most rapid 1949–1988), precipitation on the warming (Figure 3.3). In northern Coastal Plain has declined in recent Alaska, the most pronounced warm- decades (Curtis et al. 1998). ing has occurred during winter and spring.

Observed Climate Change in Arctic Alaska Recent warming notwithstanding, the climate of arctic Alaska is cold and dry (Figure 3.4) with a mean annual air temperature (MAAT) of -12°C (10.4°F) and mean temperatures below 10°C (50°F) in every month of the year. Surface air temperature on land is strongly influenced by proximity to the coast, with inland areas experienc-

Figure 3.4 (top). Average monthly high (red line) and low (blue line) temperatures and precipitation (green bars) in the Arctic from 1971–2000. Data courtesy M. Shulski. 15 Figure 3.5 (bottom). 50-year trend in mean annual temperature at seven sites on the North Slope. Data courtesy M. Shulski. Table 3.1. Projected magnitude of increase from historic1 values for tempera- Projected Climate Change in Arctic ture and precipitation for mid-century (2051–2060) and end of century (2091– Alaska 2100) time series. Seasonal2 and annual data are summarized by ecoregion. Climate projections here are based on the composite of outputs from 5 Mid-century of the 15 General Circulation Models Temperature (increase °C) (GCMs) from the Intergovernmental Panel on Climate Change Fourth Ecoregion Winter Summer Annual Assessment Report (IPCC 2007), Arctic Coastal Plain 5.7 1.6 4.4 chosen for their superior perfor- Arctic Foothills 5.5 1.8 4.3 mance at high latitudes and available through the Scenarios Network for N. Brooks Range 5.2 2.0 4.1 Alaska Planning (SNAP, http://www. Precipitation (% increase) snap.uaf.edu; for details see Ap- Winter Summer Annual pendix 1). Based on these outputs, temperatures are projected to rise Arctic Coastal Plain 48 12 28 during the 21st century. Mean an- Arctic Foothills 33 10 21 nual air temperature for the decade 2091–2100 is expected to increase N. Brooks Range 13 12 13 6.5–7.5°C (Table 3.1, Figure 3.6) End of century compared to the baseline period Temperature (increase°C) 1961–1990, depending on location. Most of this warming will occur Ecoregion Winter Summer Annual during winter (October–May) and is Arctic Coastal Plain 9.5 3.0 7.3 expected to affect coastal areas more Arctic Foothills 9.1 3.3 7.2 than inland areas, most likely due to the influence of a longer marine N. Brooks Range 8.6 3.4 6.9 ice-free period (Table 3.1, Figure Precipitation (% increase) 3.7). Projected summer temperature Winter Summer Annual increases are of a lesser magnitude and more pronounced in inland areas Arctic Coastal Plain 77 27 50 (Table 3.1, Figure 3.8). Arctic Foothills 53 23 38 N. Brooks Range 36 19 22 By the end of the century, annual precipitation is projected to increase 1. Baseline temperature and precipitation values are based on the Parameter- by 20–60%, depending on location, Elevation Regression on Independent Slopes Model (PRISM) 1961–1990 dataset compared to the baseline period created by the PRISM Group. 1961–1990 (Table 3.1, Figure 3.9). 2. Summer is calculated as the average of June–September. Winter is calculated as Most of this increase is expected to the average of October–May. occur in winter, thereby contributing to a deeper snow pack. In both summer and winter, the relative precipitation increase in coastal areas is anticipated to be nearly twice Table 3.2. Projected magnitude of increase from historic1 values for frost-free that of the Brooks Range (Table season length for mid-century (2051) and end of century (2100) time series. 3.1, Figures 3.9 and 3.10). As noted Regression equations for calculation of frost-free season length, advance of above, the climate records from thaw, and delay in freeze provided by The Wilderness Society. Barrow and Barter Island indicate a decrease in precipitation over the Mid-century past half-century, unlike the increasing trend projected for the future by increase in the GCMs. frost-free Advance of Delay in freeze season length thaw Ecoregion (days) Along with warming temperatures and increased precipitation, the Arctic Coastal Plain 18 5 13 length of the frost-free season is Arctic Foothills 16 6 10 projected to increase across all three ecoregions (Table 3.2). By the end of N. Brooks Range 16 7 9 the century, frost-free season length End of century is projected to increase by 27–33 Arctic Coastal Plain 33 8 25 days over baseline values, depending on location. Projected change Arctic Foothills 28 10 17 in season length is due primarily to 16 N. Brooks Range 27 11 15 delayed onset of freezing in the fall, but also to an advance in first-thaw 1. Baseline temperature and precipitation values are based on the Parameter- date in spring (Table 3.2). Increases Elevation Regression on Independent Slopes Model (PRISM) 1961–1990 dataset created by the PRISM Group. continued page 21 Average Annual Air Temperatures (°C)

17 Figure 3.6. Average annual air temperatures (°C), baseline (top panel, 1961–1990) and projected (middle panel, 2051–2060, bottom panel, 2091–2100). Insets depict North Slope ecoregions (top panel) and projected change from baseline for periods 2051–2060 (middle panel) and 2091–2100 (bottom panel). Map created by The Nature Conservancy, Anchorage, Alaska. Winter Air Temperatures (°C)

18 Figure 3.7. Winter air temperatures (°C), baseline (top panel, 1961–1990) and projected (middle panel, 2051–2060, bottom panel, 2091–2100). Insets depict North Slope ecoregions (top panel) and projected change from baseline for periods 2051–2060 (middle panel) and 2091–2100 (bottom panel). Map created by The Nature Conservancy, Anchorage, Alaska. Summer Air Temperatures (°C)

19 Figure 3.8. Summer air temperatures (°C), baseline (top panel, 1961–1990) and projected (middle panel, 2051–2060, bottom panel, 2091–2100). Insets depict North Slope ecoregions (top panel) and projected change from baseline for periods 2051–2060 (middle panel) and 2091–2100 (bottom panel). Map created by The Nature Conservancy, Anchorage, Alaska. Average Annual Precipitation (mm)

20 Figure 3.9. Average annual precipitation (mm), baseline (top panel, 1961–1990) and projected (middle panel, 2051–2060, bottom panel, 2091–2100). Insets depict North Slope ecoregions (top panel) and projected change from baseline for periods 2051–2060 (middle panel) and 2091–2100 (bottom panel). Map created by The Nature Conservancy, Anchorage, Alaska. Continued from page 16 in summer temperature and duration of the frost-free period are expected to result in an increase in evapotranspiration, with drying projected across the three ecoregions (Figure 3.10). Near the middle of the century (2035–2044), the landscape may be 10–16% drier, and near the end of the century (2075–2084), the Mean June-September water balance for the Arctic Coastal Plain North Slope could be 23–37% drier (Figure 3.10). 500 P PET P-PET 400 Cloud cover, including fog, can 300 substantially alter the amount of 200 solar energy available at ground level by intercepting and reflect- 100 0 ing solar energy away from the mm earth’s surface and by emitting -100 1961-1990 2035-2044 2075-2084 longwave radiation (i.e., heat) back -200 towards the ground (e.g., Weller and -300 10% Drier Holmgren 1974). During summer, 23% Drier the balance between reflected and -400 emitted energy tends to be negative, -500 meaning that the net effect of cloud Time Period cover is a cooling of the surface

(Chapin et al. 2005). Historical data Mean May-September water balance for Arctic Foothills from Barrow shows that peak cloudi- P PET P-PET ness occurs in August and Septem- 600 ber, coinciding with the presence of 500 open water along the coast (Maykut 400 and Church 1973). Cloudiness in the 300 Arctic is expected to increase as sea 200 ice retreats, creating larger areas 100 mm of relatively warm open water for a 0 longer period (Walsh and Chapman -100 1961-1990 2035-2044 2075-2084 1998). While there is agreement that -200 changes in cloud cover will affect the -300 arctic landscape, there is substantial 12% Drier uncertainly regarding the magnitude -400 25% Drier -500 of those effects (ACIA 2005, Walsh Time Period and Chapman 1998).

Model Confdence Mean June-August water balance for Northern Brooks Range The point estimates for change 600 P PET P-PET in temperature and precipitation 500 should be considered in the context of model reliability. Figure 3.11 400 depicts the geographic pattern of 300 among-model variance in tempera- 200 ture and precipitation for the mid-

mm 100 century. The differences in model output can be attributed largely to 0 2075-2084 differences in the way each of the -100 1961-1990 2035-2044 five models treats the seasonal re- -200 treat of sea ice, which substantially -300 16% Drier influences projections for fall and 37% Drier winter temperatures (John Walsh, -400 International Arctic Research Cen- Time Period ter, personal communication). Sea Figure 3.10. Growing season water balance, baseline (1961–1990) and ice distribution is difficult to model projected (2036–2044 and 2075–2084) for the three North Slope ecoregions. and was one of several deficiencies Top panel: Arctic Coastal Plain, middle panel: Arctic Foothills, bottom 21 noted by Kattsov and Källén (2005) panel: Northern Brooks Range. P=precipitation; PET=potential in GCMs. Other identified areas evapotranspiration. When the difference between these two quantities of deficiency included representa- (P-PET) is negative, a drier condition is indicated. Figures courtesy Wendy tion of clouds, atmospheric bound- Loya and Brendan O’Brien, The Wilderness Society, Anchorage, Alaska. ary layers, freshwater discharge A retrospective analysis of the into marine waters, and vegetation performance of the five models feedbacks. (SNAP 2008) demonstrated that model output corresponded reason- Differences among models are not ably well to actual climate records the only potential source of uncer- from 1980–2000 at 32 stations across tainty in climate projections. An- Alaska. That analysis, however, other source of variation includes included only one arctic site (Bar- assumptions regarding atmospheric row). Models performed better for concentrations of greenhouse gases; temperature than precipitation due the results presented here are based to the inherent variability of pre- on just one of the many potential cipitation over time and space. At emissions scenarios assembled by Barrow, model outputs accounted for the IPCC Special Report on Emis- 85–91% of the variation in monthly sions Scenarios (Nakicenovic et al. mean temperature, but only 21–51% 2000). We chose to base our discus- of the variation in precipitation ag- sions on a “middle-of-the-road” gregated into four-month intervals. scenario (scenario A1B) but there On the circumpolar scale, Walsh is no assurance that this is the most (2005) also reported a large range of probable scenario. uncertainty in projections of future precipitation regimes. Given the im- Potential within-model variation portance of water balance to habitat is also not incorporated into this conditions, improvements in the reli- discussion; rather, results are based ability of projections of precipitation on a single run of each model with and evapotranspiration are urgently initial conditions set to the aver- needed. age over the period 1961–1990 (see Appendix 1). An “initial condition ensemble” approach would employ multiple runs of a single model with variable initial conditions. This type of analysis would permit the inclusion of natural variability through the statistical distribution of climatic conditions and the probability of extreme events (Kattsov and Källén 2005).

22 Figure 3.11. Model uncertainty for annual temperature (°C; upper panel) and precipitation (mm; lower panel) projections for 2051–2060. Insets depict projected temperature change relative to baseline (upper panel) and projected precipitation increase relative to baseline (lower panel). Map created by The Nature Conservancy, Anchorage, Alaska.

23 Permafrost is earth material (soil, erosion (Yoshikawa and Hinzman Permafrost rock, ice, and organic material) that 2003). Along the Beaufort Sea remains at or below 0°C for at least coastline, permafrost thickness is two consecutive years (van Everdin- generally in the range of 200–400 m gen 1988). Permafrost is unique to (650–1,300 ft) although a thickness of those areas characterized by low 650 m (2,130 ft) has been measured mean annual air and ground tem- at Prudhoe Bay (Gold and Lachen- peratures (Smith and Riseborough bruch 1973). 2002) and is believed to underlie approximately 13–18% of the exposed At the surface, above the perma- land in the northern hemisphere frost, soil thaws during the summer (Zhang et al. 2000). The areal dis- and freezes again in winter—this tribution of permafrost is divided zone is termed the “active layer.” into four zones based on the relative Soil texture and moisture play a proportions of the land underlain key role in determining active layer by permafrost (Brown et al. 1997): depth; gravelly soils tend to be well continuous (>90%), discontinuous drained with deep active layers, (50–90%), sporadic (10–50%), and whereas organic soils tend to be isolated (0–10%). These zones are poorly drained with shallow active loosely related to mean annual air layers. Active layer depth also varies temperature (MAAT), with continu- across the terrain: it can be as little ous permafrost generally beneath as 35 cm (14 in) in upland tussock areas with MAAT at or below -6° to tundra, with organic-rich soils to as -8°C (Smith and Riseborough 2002). deep as 130 cm (51 in) in riverine tall The North Slope falls within the willow shrub with sandy soils, and zone of continuous permafrost, and up to 4 m or more in coarse-grained thermokarst and thermal erosion sediments and in the Brooks Range features are found across most of the (Jorgenson et al. 2003). In addition region (Figure 3.12). Thawing of ice- to influencing active layer depth, to- rich permafrost causes the surface pography and texture of the predom- to subside, creating characteristic inant soil type determine the ground surface landforms termed thermo- ice content of permafrost (Box 3.1). karst features. Processes associated Within the permafrost layer, pockets with thermokarst include thaw, of perennially unfrozen ground, or ponding, surface and subsurface taliks, can form in locations where drainage, surface subsidence, and temperature is maintained above

Figure 3.12. Permafrost distribution and associated thermokarst and thermal erosion landforms on the North Slope of Alaska. Map by USFWS from Jorgenson et al. 2008. 0°C, such as under lakes, rivers, or greatest change in temperature in areas influenced by groundwater at the surface of the permafrost, a movement (van Everdingen 1988). warming of 3–4°C, was detected on the Arctic Coastal Plain. In Alaska, The thermal regime of permafrost the changes in permafrost tem- is mediated by topography, surface perature coincide with warming air water, soil moisture, groundwater temperatures in winter (Osterkamp movement, vegetation, and snow. 2005, 2007). Wet, near-surface soils increase heat transferred into the ground in Sensitivity of the landscape to cli- summer, while drier soils reduce that mate warming is greatly influenced heat transfer. Unfrozen water pro- by the quantity and nature of ground vides an important positive feedback ice (Box 3.1). On a local scale, the by enhancing permafrost degrada- stability of permafrost is affected by tion when impounded on the surface, terrain, surface water, groundwater when moving through the active movement, soil properties, vegeta- layer, or flowing through taliks. tion, and snow (Osterkamp 2007). If During the warm season, vegeta- permafrost does degrade, the type tion protects permafrost from thaw of thermokarst and the ecologi- by shading the ground surface from cal implications vary depending on incoming solar radiation and act- climate, landscape position, soils, ing as insulation (Streletskiy et al. hydrology, and amounts and types of 2008). Similarly, interception of snow ice (Jorgenson and Osterkamp 2005). by shrubs can act to insulate soils against extremely cold air temperatures in winter. Because permafrost is strongly influenced by ecological components such as vegetation structure and composition, its properties evolve along with successional patterns of ecosystem development (Jorgenson et al. 1998). In turn, the patterns of ice accumulation and degradation influence the patterns of vegetation and soil development. This co-evolution of permafrost and ecological characteristics at the ground surface is most evident after disturbance, such as river channel migration, lake drainage, and fire (Shur and Jorgenson 2007).

Over the past 2–3 decades, permafrost temperatures have generally increased 1–2°C at sites throughout the northern hemisphere, although some are stable (for review see Brown and Romanovsky 2008). Per- mafrost warmed about 2–3°C near Figure 3.13. Temperature records from deep wells along a north-south Prudhoe Bay from the mid-1980s to transect from Prudhoe Bay to the Brooks Range. Figure from Romanovsky 1997 (Romanovsky and Osterkamp et al. 2007. 1997). Long-term monitoring of deep wells along a north-south transect from Prudhoe Bay to the Brooks Range shows a general warming trend over the past 25 years, but with stable and cooling periods (Figure 3.13; Osterkamp 2005). The

25 The dependence of permafrost stability on both air temperatures and ecological factors adds complexity to the task of forecasting the biological consequences of climate change. Landscape-level models have been developed to project the distribution of permafrost under scenarios of climate change. For example, the GIPL model (http://www.gi.alaska.edu/sno- wice/Permafrost-lab/methods/modelling.html) incorporates the effects of surface geology, lithology, ground temperatures, soil properties, snow cover, vegetation, air temperature, and precipitation into a numeri- cal model of heat transfer through the atmosphere, snow, vegetation, and active layer. The model outputs predicted mean annual ground temperature at various depths, from which the persistence of permafrost may be inferred. When applied to long-term (decadal and longer time scale) averages, this approach shows an accuracy of +0.2–0.4ºC for mean annual ground temperature (Sazon- ova and Romanovsky, 2003). Using climate projections based on the SNAP data set under the A1B emissions scenario, permafrost distribution in arctic Alaska is expected to remain stable through the end of the century, as evidenced by projected mean annual soil temperature below 0°C (Figure 3.14).

Despite the relative stability projected for permafrost in arctic Alaska, recent observations suggest that warming temperatures can accelerate thermokarst processes at mean annual ground temperatures well below 0°C, as evidenced by an increase in both area and density of degrading ice wedges in a study area on the Arctic Coastal Plain (Jorgenson et al. 2006). In the near term, thermokarst processes, such as the degradation of ice wedges, that affect local drainage and vegetation are the likely agents of habitat change, rather than widespread deepening of the active layer or a shift to discontinuous permafrost.

Figure 3.14. Projected mean annual soil temperatures for Alaska at 1 m 26 depth, 2050–2059 (above) and 2090–2099 (below). Based on output from the GIPL 1.3 Permafrost Model using the SNAP climate data set. Figure courtesy of S. Marchenko and V. Romanovsky (Geophysical Institute Permafrost Laboratory, University of Alaska Fairbanks) and SNAP (University of Alaska Fairbanks). Box 3.1 Ground Ice Formation

Permafrost—frozen ground—varies in the amount and type of ice contained. Ground ice can take the form of: 1. massive ice (e.g., ice wedges and pingo ice), 2. ice contained within microscopic soil pores, or 3. discrete lenses of segregated ice that are highly variable in size and shape (van Everdingen 1988).

On the Arctic Coastal Plain, ice wedge polygons are the most common form of massive ice. The process of ice wedge formation was described by Lachen- bruch (1962). Ice wedges begin to form when frozen soil contracts and cracks in response to abrupt temperature drops during winter. In spring, water seeps down into the crack and freezes when it reaches the permafrost zone. The following winter, the growing ice wedge cracks in nearly the same location. In spring, the cracks are again filled with water that soon freezes after it enters the ice wedge. This iterative process of cracking and freezing forms a polygonally patterned network, which is, from above, similar in appearance to the cracks in drying mud, but with polygons measuring 5–40 m in diam- eter. Over a period of hundreds to thousands of years (Jorgenson et al. 1998), growing subterranean ice wedges displace soil upward, forming a polygonal-patterned surface with a microtopography consisting of troughs, rims, and basins (see diagram). Local hydrology and vegetation is strongly influenced by the density and form of ice wedge poly- Polygon rims gons. Trough

Basin Active layer

Permafrost Ice wedge

Above: Schematic illustrating formation of ice wedges by R. Mitchell/Inkworks from cited sources.

27 This section provides an overview of Hydrologic processes are a pivotal Hydrology the hydrology of the Alaskan Arctic determinant of climate-influenced and a number of potential future im- habitat change. In arctic Alaska, pacts of increasing air temperature continuous permafrost forms an and possible increasing precipita- impermeable layer below a shallow tion. There is a considerable amount seasonally frozen surface soil layer. of uncertainty associated with our In the Arctic Foothills, limited soil description of current conditions, storage capacity and topographic as well as projections of how arctic gradients result in high runoff ratios hydrology may differ in the future. in spring and summer. On the Coast- The uncertainty derives from many al Plain, the underlying permafrost sources, including difficulty mea- and low topographic gradient results suring hydrologic parameters in an in a landscape dominated by wet- extreme environment, limited obser- lands, ponds, and lakes (see photo vational data sets covering time pe- this page) in a region that would riods of several decades or less from otherwise be classified as arid on the only a few locations, and climate basis of annual precipitation. Deep, modeling errors. The arctic hydro- continuous permafrost is expected to logic cycle exhibits high natural persist for centuries in arctic Alaska, variability that can make it difficult even under scenarios of continued to clearly detect changes. Further- arctic warming. Nevertheless, more, the system is highly complex changes in overall water balance and involves many feedbacks that (Box 3.2) and timing and magnitude may accelerate some changes while of seasonal water and energy fluxes inhibiting and restoring stability in could significantly affect the aquatic reaction to other changes. and semi-aquatic habitats upon which many species depend.

The vast Arctic Coastal Plain is covered with lakes and ponds. Photo by Steve Zack, Wildlife Conservation Society. Box 3.2. Water Balance

The water budget of a basin can be described by the following equation:

(Ps + Pr) – (Es + ET) – R = Δ (Ssurface + Ssoil), where input occurs as either snow (Ps ) or rain (Pr) and output can take the form of either sublimation (Es), evapotranspiration (ET), or surface runoff (R). When this quantity is positive, there is a net increase in basin storage, partitioned into surface water storage (Ssurface) and soil moisture (Ssoil). When this quantity is negative, as occurs in a dry summer, a soil water storage deficit ensues. The larger the deficit at the end of the summer season, the greater proportion of next spring’s snow melt will be taken up to recharge ponds and lakes and absorbed into surface and near-surface soils.

In arctic Alaska, the best available water balance data are from the Ku- paruk watershed, including Imnavait Creek in the Kuparuk headwaters, and the adjacent Putuligayuk River (Bowling et al. 2003, Kane et al. 2000, 2008, Kane and Yang 2004; see map in Box 3.3, page 33).

Imnavait Creek and the Upper Kuparuk River are representative of systems in the Arctic Foothills, whereas the Putuligayuk River lies wholly within the Arctic Coastal Plain. Water input to the systems is represented by Snow Water Equivalent (SWE) and rainfall; water loss is through ET and runoff (snow and summer). As indicated in the figure, the hydrologic regime on the Coastal Plain (Putuligayuk River) is dominated by snowfall and snowmelt-generated runoff. Although snowfall and snowmelt are also very significant in the Foothills, it can be seen that summer rainfall is typically greater than end-of-winter SWE, and rainfall yields considerable runoff.

Average Water Balance Components

SWE Rain ET Snow Runoff Summer Runoff 800

700 179

600 57 500 100 140 400 80

34 300 109 74

Water Depth (mm) 262 200 60 163 85 100 115 69 95 0 Imnavait Upper Kuparuk Putuligayuk (1985-2003) (1996-2002) (1999-2007) 29

Basin Figure by Amy Tidwell for WildREACH from cited sources. Overview of Hydrologic Processes tion as snow (minimum 41% in 2002; by Season maximum 85% in 2007; Kane et al. Figure 3.15 is a generalized repre- 2004, 2008). sentation of the current hydrologic processes in arctic Alaska. Winter snow cover and winter temperatures strongly affect soil tem- Winter peratures. Snow acts as an insulator Winter generally lasts from October between the relatively warm soil and until mid- to late May. During the cold air temperatures throughout cold season, precipitation falls as the winter. Therefore, greater snow snow and is temporarily stored in depth and warmer air temperatures the snow pack at the ground surface. act to reduce winter heat loss from Snow accumulation is reduced to the soil, while more shallow snow some extent by sublimation (vapor- and lower air temperatures lead to ization directly from the solid phase). colder soil temperatures. Water stored in the snow pack at the end of winter is an important Spring Melt hydrologic quantity and is referred The spring season is very brief, to as the end-of-winter snow water with snow melt, spring flood, and equivalent or SWE (Figure 3.15: leaf-out occurring within a period of A). On average, SWE represents 2–3 weeks beginning in mid- to late approximately 40% of annual North May. During the spring, the snow Slope precipitation. However, this pack that developed during winter is can vary greatly from year to year lost through combined processes of and by location. For example, in the melting and evaporation (ablation) in upper Kuparuk River basin, SWE a short period of time, usually within represents approximately one-third 1–2 weeks (Figure 3.15: B). Once of total precipitation (minimum 14% ablation is complete, surface soil in 1999; maximum 42% in 2000), temperatures begin to increase and while the lower Kuparuk region soils begin to thaw. As surface soils on the Coastal Plain receives an thaw, they become much more per- average one-half of its precipita- meable, and water can more readily

30 Figure 3.15. Generalized representation of hydrologic processes in arctic Alaska in seasonal sequence. Specific features of the illustration are discussed in the text. Processes above the thick black line represent the primary mechanism of water movement from the surface to the atmosphere for each season. Processes below the black line represent movement of water from atmosphere to surface. Figure by Amy Tidwell for WildREACH from data in Kane et al. 2003, 2004, 2008. enter the soil column, provided that Fall Freeze it is not saturated (Figure 3.15: C). Freezing temperatures may occur Water released from the snow pack throughout the summer, but a hard during spring melt recharges lakes, freeze is likely any time between ponds, and emergent wetlands, and late August and mid-September. The drains off the surface into streams onset of persistent snow cover usu- (Figure 3.15: D). ally occurs in September. In early fall, surface soils begin to cool and Shortly following snow ablation, eventually reach the freezing point. streamflow rises sharply. Although Between mid- and late fall, the active much of the snowmelt drains into layer undergoes freeze-back, pri- the Arctic Ocean, a considerable marily at the surface but also at the portion remains on the surface in interface with the permafrost table the form of lakes, ponds, wetlands, (Figure 3.15: H). and soil moisture. Over the course of a few weeks, the inundated land surface area, and thus surface water storage, rapidly decreases as lakes and wetlands drain downslope until water surface elevations fall below their outlets (Figure 3.15: E). Surface water storage is further decreased as surface (lakes, ponds, etc.) and soil-stored water is lost to the atmosphere through evapotranspiration (the sum of evaporation and plant transpiration from the earth’s surface to the atmosphere).

Summer From early June through August and potentially into September, average daily temperatures remain above 0°C, and most precipitation occurs as rain and condensation. Early summer tends to have much lower precipitation than late summer. Dur- ing this period, evapotranspiration often exceeds precipitation (Bowling et al. 2003), and the landscape undergoes progressive drying (Figure 3.15: F). Rainfall typically increases throughout the season, while evapotranspiration declines due to reduced evaporation from the ground surface and plant transpiration. Con- sequently, late summer and early fall precipitation may reduce much of the surface drying that occurs in early summer (Figure 3.15: G).

Currently, the Coastal Plain exhibits little stream response to summer rainfall events. This is due to the high surface storage capacity, low topographic gradient, and progressive surface drying during summer. As rainfall arrives at the surface, it replenishes soil moisture and The Kuparuk River in winter. Photo by J. Benstead, University of surface storage and typically does Alabama. not result in significantly increased streamflow (Bowling et al. 2003). In contrast, significant rainfall events 31 do result in streamflow response in the Foothills (Box 3.3), where there is considerably less surface storage and steeper topographic gradients (Kane et al. 2003, 2008). Implications of Climate Change to Potential changes to winter precipi- Hydrologic Processes tation can directly impact the water A climate shift toward higher balance, spring floods, soil tempera- temperatures and/or increased tures, and wildlife foraging. Increas- precipitation will likely alter the ing winter precipitation would yield seasonal pattern described previ- greater end-of-winter SWE and ously. Potential changes as compared more effectively insulate soils from to current conditions are presented winter air temperatures. in the following descriptions. Spring Melt Winter Given a scenario of increased Increasing winter temperatures temperature and snowfall, effects may not directly impact the water on the timing of spring melt may balance on the North Slope but be offsetting to some extent—in- will alter depth and qualities of the creasing air temperatures facilitate snow pack, such as density. These earlier snowmelt, while increasing factors will affect several mammal snow depth retards it. Alternatively, species by altering access to forage if the winter snow accumulation does for grazers, quality of subnivean not increase due to a shorter winter habitats for small mammals, and season and/or if winter precipitation soil temperatures for hibernators. does not increase, then the aver- Warmer winter temperatures may age timing of spring melt would be be accompanied by an increased fre- earlier in spring. If earlier snow- quency of rain-on-snow events, mid- melt does develop, it would result winter snow melt, and ice formation in earlier surface storage recharge, within the snowpack (Kane 1997). peak streamflow, and onset of active As discussed previously, increasing layer thaw. Furthermore, because winter temperatures may also lead rainfall tends to be at a minimum to warmer soil temperatures, with early in the warm season, an earlier important implications for nutrient melt may result in a longer period cycling and vegetation (Sturm et al. of surface drying prior to increasing 2005). rainfall later in the summer season.

Many climate model projections Surface soils remain frozen during suggest that winter snowfall may snow ablation and the active layer also increase in the coming decades. plays little role in the spring streamflow peak. In the Arctic Foothills, where lakes and ponds are less prominent, spring peak flows will be determined largely by the end-of- winter SWE. On the Coastal Plain as well as Foothills areas with little topographic relief and significant surface storage (lakes and ponds), spring hydrologic response will depend on end-of-winter SWE as well as on the water content of the organic mat and soils. If the surface and near-surface water storage deficit is high due to dry conditions in the previous summer, then a larger volume of snowmelt the following spring will go directly to recharge surface water bodies and soils (see Box 3.2). Alternatively, if fall surface storage is high and vegetation and soil are wet, spring peak flow may be larger relative to available SWE. It is important to note that fall surface storage depends on many factors, such as the timing of precipitation, summer air temperatures, and veg- 32 etation changes. Summer Soil moisture is an important determinant of summer hydro- Polygonal ground and ponds on the Arctic Coastal Plain. USFWS photo. logic response. Runoff response to a Box 3.3 Inputs into Stream Ecosystems

The source and timing of freshwater inputs into stream systems of arctic Alaska varies by ecoregion. The Putuligayuk River is representative of a Coastal Plain watershed, with input dominated by snowmelt. As illustrated by the hydrograph in Panel 1, this system is characterized by a single large peak coinciding with the spring freshet, resulting from the rapid melting of the snowpack. After 80 the initial peak, discharge declines sharply and remains Putuligayuk River 2001 low throughout the mid-summer, with a slight rise at the 1 2002 end of summer corresponding to the period of increased 60 precipitation. The Upper Kuparuk River (Panel 2) in

the Foothills exemplifies a rainfall-dominated headwater stream. The spring snowmelt peak is still seen as 40

a significant hydrologic event, yet this regime is also (m3/s) Discharge characterized by multiple peaks associated with rainfall 20 throughout the summer. Relative to the Putuligayuk, the Upper Kuparuk is susceptible to flash floods, such as the 0 one in mid-August 2002. The Kuparuk River flows through 9-May 8-Jun 8-Jul 7-Aug 6-Sep all three ecoregions, and the Lower Kuparuk (Panel 3) has 160 a hydrologic regime with a mix of characteristics typical Upper Kuparuk River 2001 2002 of both Coastal Plain watersheds and rainfall-dominated 2 headwater systems. In the Lower Kuparuk, the spring 120 freshet produces a readily detectable peak in stream flow that quickly declines. Large precipitation events do pro- 80 duce additional increases in stream flow; these spikes in streamflow, however, are less pronounced than those seen (m3/s) Discharge in the upper reaches of the river. 40

Groundwater can also be a significant input into the annual 0 water budget of arctic stream systems. There are three 9-May 8-Jun 8-Jul 7-Aug 6-Sep general types of groundwater in the Arctic: supraperma- 1600 Lower Kuparuk River 2001 frost (within the active layer in summer), intra-permafrost 3 2002 (in taliks), and sub-permafrost (beneath the permafrost table). Although the thick, continuous permafrost of arctic 1200 Alaska restricts the input of sub-permafrost groundwater to surface waters, spring-fed stream systems are distrib- 800 uted widely in the eastern Brooks Range, most commonly in limestone areas at an elevation of about 600 m (Yoshi- (m3/s) Discharge kawa et al. 2007). Groundwater enters streams, becoming 400 baseflow, by migrating through the organic mat or unfrozen soil until it enters a drainage network or by flowing 0 directly into the system (i.e., spring flow). 9-May 8-Jun 8-Jul 7-Aug 6-Sep

Glacial discharge can also be a significant water source for systems originating in the Map (left) shows Brooks Range between the Sagavanirktok and Kongakut rivers. Glacier-fed systems location of Kuparuk are less sensitive to summer precipitation events, exhibit more diurnal variation, and and Putuligayuk less among-year variability in runoff (Hock et al. 2005). In the initial stages of warming, rivers. Numbers as melt rate increases, summer runoff will increase. Ulti- indicate the location mately, however, loss of glacial mass will reduce overall run- of gauging stations off and render these systems more susceptible to reduced used to generate flows in summer. the hydrographs (above). Map by Data sources: USFW. Hydrographs by Amy Tidwell for Upper Kuparuk and Putuligayuk Rivers: WildREACH from Kane DL and Hinzman LD. 2008. Climate data from the cited sources. North Slope Hydrology Research project. University of Alaska Fairbanks, Water and Environmental Research Cen- ter. URL: http://www.uaf.edu/water/projects/NorthSlope/. Fairbanks, Alaska, variously paged. November 2008.

Lower Kuparuk: U.S. Geological Survey. 2009. Surface Water data for Alaska: 33 USGS Surface-Water Daily Statistics. URL: http://water- data.usgs.gov/ak/nwis/dvstat? Accessed March 31, 2009. given storm can be vastly different drier surface in both wetlands and depending on soil conditions, such uplands even if the absolute volume as moisture content and active layer of stored water is higher. It is impor- depth. For example, when the soil tant to note, however, that this effect water deficit is high, there is greater may be offset by ground settle- capacity for soil to store water ment and vegetation changes. In temporarily for gradual release over the Foothills, a longer warm season a longer period, with the effect of may prolong the time available for reducing peak flow and prolonging soil moisture to move down gradient flow recession. and into drainage networks. This may lead to late summer and early The magnitude and direction of fall soil drying, while baseflows, the change in soil moisture under portion of streamflow derived from changing climate scenarios is not subsurface sources, may increase well understood. The state of soil due to additional shallow groundwa- moisture depends on total soil stor- ter flow. age capacity, recharge from rainfall, evapotranspiration, the ability of Increasing soil storage capacity and plants to access and exploit available more rapid moisture export through soil water, and topographic gradi- evapotranspiration may lead to ents. Although increasing tempera- decreased hydrologic response to tures will provide additional energy summer storms. As the active layer to drive evaporation and transpira- depth increases and surface stor- tion, actual evapotranspiration can age decreases, the Foothills will be water- or energy-limited. If early have muted hydrologic response to summer rainfall does not change storms but greater base flow due to considerably, then summer evapo- suprapermafrost groundwater flow transpiration can be expected to from soil moisture. On the Coastal become water-limited earlier in the Plain, reduced surface storage (e.g., season than is currently observed. lower lake levels) and drier soils will Fall evapotranspiration tends to be require greater rainfall recharge limited by available energy, as well before significant surface runoff to as the onset of plant senescence. streams can occur, absent changes Therefore, changes in late summer to drainage systems (see Coastal and early fall rainfall can potentially Plain, Chapter 4). Changes in have significant impacts on the state evapotranspiration and soil storage of surface water storage at the onset capacity can also lead to drier sur- of winter. face conditions, which then require a larger amount of the following year’s Warmer soil temperatures and spring snowmelt for recharge. This earlier onset of active layer thaw redistribution of water in spring may will enhance the potential depth of reduce peak streamflow to some seasonal thaw and, thus, soil water extent. storage capacity. Actual active layer thickness is complicated by soil Fall Freeze moisture-energy balance-vegetation Warmer soil temperatures and a feedbacks. Soil water content affects prolonged fall season will lead to the absorbtion of solar energy at the later and more gradual soil freez- surface—a wet surface is darker and ing. In extreme scenarios, warmer hence absorbs more heat than a dry fall soil temperatures and increased surface. Dry near-surface (upper 10 winter snowfall may lead to talik cm) organic soils act as an efficient formation if the active layer does not insulator, reducing heat transferred fully refreeze. Although this may into soil in summer (suppressing occur in some cases, it is not likely active layer depth) and heat flux out to occur on a large scale for a very of the soils in winter. In contrast, long time—on the order of centuries. wet soils in late summer retard the In the Foothills, prolonged thaw and freezing of the soil in fall due to the formation of taliks could mean con- large amounts of latent heat associ- tinued soil drainage throughout the ated with the phase change from winter and drier spring soil mois- water to ice. Similarly, ice-rich soils ture. This winter drainage may lead require more energy to thaw than to greater under-ice streamflow. Fall 34 ice-poor soils. and winter drainage may be less important on the Coastal Plain where If active layer thickness increases, soil moisture distribution is largely the distribution of stored water may determined by microtopography. shift downward. This redistribution of water in storage may produce a Literature Cited Gold LW, Lachenbruch AH. 1973. Jorgenson MT, Yoshikawa K, Thermal conditions in permafrost: Kanevskiy M, Shur Y, Romanovsky ACIA. 2004. Impacts of a Warming a review of North American litera- V, Marchenko S, Grosse G, Brown Arctic: Arctic Climate Impact As- ture. Permafrost: Second Interna- J, Jones B. 2008. Permafrost Char- sessment. Cambridge University tional Conference, North American acteristics of Alaska. In Kane DL, Press. Contribution, National Academy of Hinkel KM, eds., Ninth Interna- Sciences, Washington, DC. p. 3–23. tional Conference on Permafrost, ACIA. 2005. Arctic Climate Impact Extended Abstracts (p. 121-122). Assessment. New York: Cambridge Hock R, Jansson P, Braun L. 2005. Institute of Northern Engineering, University Press. 1042 p. http:// Modelling the reponse of mountain University of Alaska Fairbanks. www.acia.uaf.edu. glacier discharge to climate warming. In Huber UM, Burgmann Kane DL. 1997. The impact of Bowling LC, Kane DL, Gieck RE, HKM, Reasoner MA (eds) Global hydrologic perturbations on arctic Hinzman LD, Lettenmaier DR. change and mountain regions (a ecosystems induced by climate 2003. The role of surface storage state of knowledge overview), change. In Oechel WC, Callaghan in a low-gradient arctic watershed. Springer, Dordrecht., p. 243-252. T, Gilmanov T, Holten JI, Maxwell Water Resour. Res. 39(4), 1087, B, Molau U, Sveinbjornsson B doi:10.1029/2002WR001466. IPCC. 2007. Climate Change 2007: (eds). Global Change and Arctic Synthesis Report. Contribution of Terrestrial Systems (pp 63-81). Brown J, Ferrians OJ, Heginbot- Working Groups I, II, and III to Springer-Verlag, New York. tom, JA Jr., Melnikov ES. 1997. the Fourth Assessment Report of Circum-Arctic map of permafrost the Intergovernmental Panel on Kane DL, Gieck RE, Hinzman LD. and ground-ice conditions: U.S. Climate Change. Geneva, Switzer- 2008. Water balance for a low- Geological Survey Circum-Pacific land: IPCC. 104 p. gradient watershed in northern Map 45, 1 sheet. Alaska. In Kane DL, Hinkel KM Jorgenson MT, Osterkamp TE. 2005. (eds). Ninth International Confer- Brown J, Romanovsky VE. 2008. Response of boreal ecosystems ence on Permafrost (p. 883-888). Report from the International to varying modes of permafrost Institute of Northern Engineering, Permafrost Association: State of degradation: Can. J. For. Res. 35: University of Alaska Fairbanks (2 Permafrost in the First Decade of 2,100–2,111. Vols), 2140 pp. the 21st Century, Permafrost Peri- glac. Process. 19: 255-260. Jorgenson MT, Roth JE, Emers M, Kane DL, Hinzman LD, McNamara Schlentner S, Swanson DK, Pull- JP, Zhang Z, Benson CS. 2000. An Chapin FS, Sturm M, Serreze MC, man E, Mitchell J, Stickney AA. overview of a nested watershed McFadden JP, Key JR, Lloyd AH, 2003. An ecological land survey for study in Arctic Alaska. Nord. Hy- McGuire AD, Rupp TS, Lynch AH, the northeast planning area of the drol. 31(4/5): 245-266. Schimel JP, Beringer J, Chapman National Petroleum Reserve–Alas- WL, Epstein HE, Euskirchen ES, ka, 2002: Final Report prepared for Kane DL, McNamara JP, Yang D, Hinzman LD, Jia G, Ping C-L, ConocoPhillips, Anchorage, AK, by Olsson PQ, Gieck RE. 2003. An ex- Tape KD, Thompson CDC, Walker ABR, Inc., Fairbanks, AK, 128 p. treme rainfall/runoff event in arctic DA, Welker JM. 2005. Role of land- Alaska. J. Hydrometeorology, 4(6), surface changes in arctic summer Jorgenson MT, Shur Y, Walker 1220-1228. warming. Science 310: 657-660. HJ. 1998. Evolution of a perma- Published online 22 September frost dominated landscape on the Kane DL, Yang D. 2004. Overview 2005: 10.1126/science.1117368. Colville River Delta, northern of water balance determinations Alaska. In Lewkowicz, AG, Al- for high latitudes watersheds. In Curtis J, Wendler G, Stone R, Dut- lard M, eds., Proceedings of the Kane DL, Yang D (eds). Northern ton E. 1998. Precipitation decrease Seventh International Permafrost Research Basins Water Balance in the western Arctic, with special Conference: Sainte-Foy, Quebec, (p. 1-12). International Association emphasis on Barrow and Barter Universite Laval, Collection Nordi- of Hydrological Sciences. IAHS Island, Alaska. Int. J. Climatol. 18: cana, no. 57, p. 523–530. Publication 290. Wallingford, UK, 1687-1707. 271 pp. Jorgenson MT, Shur YL, Pullman Euskirchen ES, McGuire AD, ER. 2006. Abrupt increase in Kattsov VM, Källén E. 2005. Future Chapin FS. 2007. Energy feed- permafrost degradation in arctic climate change: modeling and backs of northern high-latitude Alaska. Geophys. Res. Lett. 33: scenarios for the Arctic. In ACIA. ecosystems to the climate system L02503, doi:10.1029/2005GL024960. 2005. Arctic Climate Impact As- due to reduced snow cover dur- sessment. New York: Cambridge ing 20th century warming. Global Jorgenson MT, Walker HJ, Brown University Press. (p. 99-150) http:// Change Biol 13: 2425-2438. J, Hinkel K, Shur Y, Osterkamp www.acia.uaf.edu. T, Ping C-L, Kanevskiy M, Eisner Gallant AL, Binnian EF, Omernik W, Rea C, Jensen, A. 2009. Coastal Lachenbruch AH. 1962. Mechanics JM, Shasby MB. 1995. Ecoregions Region of Northern Alaska: guide- of thermal contraction cracks and of Alaska. U.S. Geological Survey book to permafrost and related ice wedge polygons in permafrost, 35 Professional Paper 1567. 73 p., 2 features. Alaska Division of Geo- Spec. Pap. Geol. Soc. Am., 70. sheets, scale 1:5,000,000. physical and Geological Surveys; Guidebook 10. 156 p. Maykut GA, Church PE. 1973. Ra- Streletskiy DA, Shiklomanov NI, diation climate of Barrow, Alaska, Nelson FE, Klene AE. 2008. 1962-1966. J. Appl. Meterol. 12: Thirteen years of observations at 620-628. Alaskan CALM sites: Long-term active layer and ground surface Nakicenovic N, Swart N (eds). 2000. temperature trends. In Kane DL, Special Report on Emissions Sce- Hinkel KM, eds. Ninth Interna- narios: A Special Report of Work- tional Conference on Permafrost ing Group III of the Intergovern- (p. 1727-1732). Institute of North- mental Panel on Climate Change, ern Engineering, University of Cambridge University Press, Alaska Fairbanks (2Vols). Cambridge, U.K. http://www.grida. no/climate/ipcc/emission/index.htm. Sturm M, Schimel J, Michaelson G, Welker JM, Oberbauer SF, Liston Osterkamp TE. 2005. The recent GE, Fahnestock J, Romanovsky warming of permafrost in Alaska: VE. 2005. Winter biological pro- Global Planet. Change 49: 187–202. cesses could help convert arctic tundra to shrubland. Bioscience Osterkamp TE. 2007. Charac- 55(1): 17-26. teristics of the recent warming of permafrost in Alaska. van Everdingen R (ed). 1998 revised J. Geophys. Res. 112: F02S02, May 2005. Multi-language glossary doi:10.1029/2006JF000578. of permafrost and related ground- ice terms. Boulder, CO: National Romanovsky VE, Gruber S, In- Snow and Ice Data Center/World stanes A, Jin H, Marchenko SS, Data Center for Glaciology. http:// Smith SL, Trombotto D, Walter nsidc.org/fgdc/glossary/. KM. 2007. Frozen Ground. In: Global Outlook for Ice and Snow, Walsh J. 2005. Cryosphere and Earthprint, United Nations hydrology. In ACIA. 2005. Arctic Environment Programme/GRID, Climate Impact Assessment. New Arendal, Norway, pp. 181-200. York: Cambridge University Press. (p 183-242) http://www.acia.uaf.edu. Romanovsky VE, Osterkamp TE. 1997. Thawing of the active layer Walsh JE, Chapman WL. 1998. on the coastal plain of the Alaskan Arctic cloud-radiation-temperature Arctic. Permafrost Periglac. Pro- association in observational data cess. 8: 1–22. and atmospheric reanalyses. J. Clim. 11: 3030-3045. Sazonova TS , Romanovsky VE. 2003. A model for regional-scale Weller G, Holmgren B. 1974. The estimation of temporal and spatial microclimates of the Arctic Tundra. variability of active layer thickness J. Appl. Meteorol. 13:854-862. and mean annual ground temperatures. Permafrost Periglac. Pro- Yoshikawa K, Hinzman LD. 2003. cess. 14: 125-139. Shrinking thermokarst ponds and groundwater dynamics in discon- Shulski M, Wendler G. 2007. The tinuous permafrost near Council, Climate of Alaska. University of Alaska. Permafrost Periglac. Alaska Press. Fairbanks. Process. 14: 151-160. DOI:10.1002/ ppp.451. Smith M, Riseborough DW. 2002. Climate and the limits of perma- Zhang T, Heginbottom JA, Barry frost: a zonal analysis. Permafrost RG, Brown J. 2000. Further Periglac. Process. 13:1-15. statistics on the distribution of permafrost and ground ice in the SNAP. 2008. Validating SNAP Northern Hemisphere. Polar Geog. climate models. http://www.snap. 24: 126-131. uaf.edu/news/new-analysis-validity- snap-climate-models, accessed February 2009.

36 A male American tree sparrow (Spizella arborea) sings after arriving on his breeding grounds. Shrub-dependent songbird species will likely benefit from increased shrub cover on the North Slope. Photo by Ted Swem, USFWS.

Alder shrubs (Alnus crispa) colonizing the rims of polygons on the Colville River Delta. The colonizing seeds were dispersed by water movement along the floodplain corridor. This novel habitat may increase across the North Slope. Photo by M.T. Jorgenson. Chapter 4 Habitat Change

This chapter describes four North Slope landscapes influenced by the presence of permafrost—the coastline, coastal plain, foothills, and floodplains—and what we can project about how they may change under a warming climate. This chapter also outlines how processes such as fire, vegetation community change, and contaminant mobilization may affect future habitats in the region.

38 Coastline The Beaufort and Chukchi sea The Beaufort and Chukchi sea coasts are ice-free for 3–4 months Permafrost- coasts are characterized by lagoons per year, with the open-water period with sandy barrier islands, exposed occurring from July–September. infuenced coast with high peat bluffs, deltas, By early September, the sea ice and low-lying drained lake basins retreats to a distance of 300–500 that are occasionally flooded by km off Barrow, and 100–300 km off Geomorphic storm surges (Figure 4.1). These Kaktovik (Jorgenson and Brown dynamic environments are greatly 2005). In the past few decades, the Processes affected by sea ice, wind-driven length of the ice-free period in the waves and storm surges, surface Chukchi and western Beaufort seas water temperatures, coastal erosion has increased dramatically. From the and accretion, sedimentation by riv- late 1970s to 2006, the ice-free period ers and eroding coastal bluffs, and increased by an average of 50–95 long-shore currents. days, depending on the region, with

Current

Projected 39

Figure 4.1. Schematic of arctic coastline landscape, current (above) and projected (below). The projected landscape illustrates elements likely to change as a result of climate warming. Figure by R. Mitchell/Inkworks for WildREACH from cited sources. the greatest increase between Point Sea level rise measured at nine tide Lay and Barrow where the ice-free stations in Siberia averaged 2–3 mm season grew from approximately 30 per year from 1954–2007 (Richter- to 125 days (Rodrigues 2008, unpub- Menge et al. 2008). Global sea lished). Arctic sea ice extent reaches level rise in the period since 1993 is its annual minimum at summer’s end estimated to have occurred at a rate and has declined at a rate of 11% per of 3 mm per year, and projections decade since 1972 (Richter-Menge for cumulative rise by the end of the et al. 2008). Open-water conditions 21st century range from 0.18–0.59 increase the probability that strong m, but do not account for potential wind conditions will result in a storm changes in ice flow of the Greenland surge, because the presence of ice and Antarctic ice shelves (IPCC would otherwise inhibit wave forma- 2007). Some researchers believe that tion (Reimnitz and Maurer 1979). accelerated ice sheet melt will result in mean sea level rise exceeding 1 m by the end of this century (http:// climatecongress.ku.dk/newsroom/ rising_sealevels/, retrieved March 14, 2009).

Sea-surface temperature trends for the Arctic Ocean over the past century are characterized by a period of cooling from 1930–1965, followed by a period of warming, particularly pronounced since 1995 (Steele et al. 2008). Sea-surface temperatures for the Beaufort and Chukchi seas have been especially warm in this decade, e.g. 2–3°C higher in 2007 than the average for the previous 25 years (Richter-Menge et al. 2008).

The combined effect of sea level rise, increased frequency of storm surges, and increased water temperature has already resulted in a substantial increase in erosion rates on the Beaufort Sea coast (Jorgenson and Brown 2005, Jones et al. 2009). Lunar tides along the Beaufort Sea coast are only on the order of 20 cm, but water levels are strongly affected by wind direction, with west winds associated with higher tidal stage (Jorgenson 2009). Strong winds can raise water levels as much as 2 m (Reimnitz and Maurer 1979), and these storm surges can result in very rapid coastline erosion, particularly in sectors characterized by exposed ice-rich peat bluffs. Erosion of ice-rich bluffs involves formation of a thermoerosional niche (Figure 4.2) and subsequent collapse of bluff materials (Reimnitz and Maurer 1979). Unlike coastlines in temperate areas, erosion is affected by thaw- Figure 4.2 (above). The effects of a major storm with strong west winds on ing as well as mechanical processes. Barter Island in August 2008. The undercutting of thermoerosional niches Erosional rates of 1–2 m per year and the collapse of large blocks helps fragment and accelerate the rate of are typical for many sectors of the coastal erosion along ice-rich portions of the Beaufort Sea coast. Below the Beaufort coast, excluding river human figure, a cross section of the thin active layer and an ice wedge is 40 deltas, where accretion is occurring. clearly visible. Photo by Mitch Osborne, USFWS. The 60-km section north of Tes- Figure 4.3 (below): Salt-killed tundra is prevalent along low-lying portions hepuk Lake is retreating much more of the coast that have been inundated by storm surges. After inundation by rapidly, however, with mean annual salt water, iron staining and a white salt crust at the surface are common. erosion rates having increased from Photo by M.T. Jorgenson. an average of 6.8 m per year (1955– 1979) to 13.6 m per year (2002–2007). discharge in arctic rivers is much For most shoreline types the rate less than the global average, due was even higher—18 m per year—in to watershed characteristics that the most recent period (Jones et al. include a thin weathering crust, low 2009). If this rate were to remain precipitation, extensive permafrost, constant over the next century, a low temperatures, wetland-dominat- coastal retreat of 1.8 km would be ed landscape, and low level of human expected, representing a loss of 8% activity (Gordeev 2006). Models of the Teshekpuk Lake Goose Molt- predict a 30% increase in sediment ing Area (see box 5.1 on page 72). load for every 2°C of warming in the drainage basin, and a 10% increase By comparison, the coastal reach in sediment transport for a 20% between the Colville and Saga- increase in discharge (Syvitski 2002, vanirktok rivers (within which Morehead et al. 2003). Projected most of the North Slope oil and gas increases in temperature and pre- infrastructure is located) retreated cipitation in arctic Alaska suggest at a historic rate of less than 2 m per a trend toward increased rates of year (Jorgenson and Brown 2005). sedimentation. The increased supply At a hypothetical higher rate of 4 m of sediment to river deltas may com- per year held constant over the next pensate for, or outpace, the effect of century, the resultant 200 m retreat sea level rise. would represent a loss of less than 1% of the Arctic Coastal Plain ter- The barrier island-lagoon system restrial habitats in this sector. provides important summer feeding habitat and migration corridors In addition to accelerating shoreline for waterfowl and anadromous fish. erosion, increased storm heights ac- The barrier islands are believed to companied by the projected sea level be erosional remnants of ancient rise will cause additional flooding mainland shores and composed of and salinization of low-lying ter- materials not generally supplied rain. Inundation with salt water may by modern depositional processes kill the existing vegetation, result- (Hopkins and Hartz 1978, Morack ing in relatively barren patches of and Rogers 1981). Some retain salt-killed tundra (Figure 4.3). A remnant vegetated tundra underlain distinctive coastal wet sedge tundra by permafrost, but most are con- Erosion along a segment of the community characterized by salt structional islands largely devoid of Beaufort Sea coastline based tolerant species (e.g., Puccinellia vegetation and in a continual state upon vertical aerial photography phryganodes and Carex subspatha- of migration (westward and towards captured in 1955, 1979, and 2002. cea) may develop in wetter areas the mainland) and morphological Image from B. Jones, USGS. (Flint et al. 2008). Coastal wet sedge tundra is currently rare, and its future abundance will reflect the equilibrium between loss from erosion and inundation and increase due to colonization in salinized terrain.

The persistence of coastal delta environments depends on the degree 1979 2002 to which sedimentation keeps pace with sea level rise. Information on the baseline rates of sedimentation for North Slope rivers is very limit- 1979 2002 ed. On the Colville River Delta, a large breakup or precipitation event can result in accumulation of 10 cm or more, and an 41 average nearshore deposition rate of 1 kg/m/year was estimated (Brown and Jorgenson 2005). In general, sediment change. These islands are typically period of open water and increased less than 1 m in elevation and no occurrence of larger waves is at least higher than 3 m (Hopkins and Hartz partially responsible for this accel- 1978) and, therefore, are subject eration. Ice-push events require the to overwash during storm events. coincidence of strong onshore winds Storm-associated erosion commonly and a high density of broken ice, and breaches the islands and splits them this may occur less frequently as sea into smaller pieces, which may later ice retreats farther offshore in sum- coalesce due to longshore transport mer. Warming ocean temperatures and redeposition of sediments. Ice- also may play a role, however, as push events, during which gravel is even the constructional islands may dredged and deposited by wind-driv- be partially composed of ice-bonded en ice floes, may play an important sediments (Morack and Rogers role in maintaining the islands by 1981), which inhibit longshore sedi- periodically concentrating materials ment transport (Thomas Ravens and in a discrete zone. William Lee, University of Alaska Anchorage, personal communica- Preliminary evidence suggests that tion). These trends suggest that the Beaufort Sea barrier island the deterioration or disappearance system may be disintegrating. For of the existing system of barrier example, Narwhal Island, which is islands is possible over a relatively east of Prudhoe Bay, has greatly short period. diminished in surface area since Figure 4.4. Conceptual model 1955, and the migration rate from The coastline is a dynamic environ- illustrating linkages between 1990–2007 (24 m/yr) greatly exceeds ment subject to continual change. physical processes and coastal that from 1955–1990 (5 m/yr). Total Climate change may affect the habitat. Grey boxes identify climate surface area of barrier islands in the equilibrium among various coastal drivers, blue boxes indicate physical central Beaufort Sea (Colville River processes, however, and result in a processes, and white boxes indicate to Point Thomson) has decreased net change in habitat availability. landscape responses. The light-blue approximately 4% from the 1940s The linkages among coastal process- box contains habitats for which we to the 2000s, and the rate of change es, as affected by temperature and are less certain of the direction of is greater during the period since precipitation changes, are illustrated the habitat change. 1980 (Gibbs et al. 2008). A longer in a conceptual model (Figure 4.4).

Temperature, precipitation and coastal processes

! TEMPERATURE ! PRECIPITATION

↑ Degradation of ↑ Ocean ↑ Storm Surge ↑ Sea-level ↑ River discharge coastal permafrost temperature (↓ Sea Ice) and sediment transport

↑ Nearshore deposition of nutrients and deltaic ↑ Coastal erosion sediments

! Barrier islands Δ Salt-killed tundra

Δ Saline mud flats and coastal wet sedge tundra ↓ Protected ! Terrestrial ↑ Breached lagoons habitats along lakes coastline ∆ Delta mudflats (accretion or erosion) 42 Opposing influences create uncertainty regarding direction of change for mudflats, coastal wet sedge tundra, and salt-killed tundra Arctic Coastal Plain In low-lying basins, the water of Regional differences in the distribu- deep lakes and shallow ponds will be tion of late-Pleistocene surficial de- fully recharged by spring snow melt posits (Figure 4.5) strongly influence with excess water running off, as- the presence and characteristics of suming constant or increased winter lake basins. Ice-rich deposits, such precipitation. Under a warming as delta and glacio-marine deposits, scenario where summer precipita- are associated with abundant thaw tion is insufficient to compensate lakes and drained basins. Lakes for increased evapotranspiration, in ice-poor deposits, such as eolian deeper midsummer draw-down may sand and slightly pebbly till, formed be expected in lakes and ponds, but in low-lying swales during the early water levels should increase again Holocene and expanded through ero- with fall rains and spring melt in sion of fine-grained sediments (Jor- the following year. Shallow ponds genson and Shur 2007) because the and wet sedge meadows, which are volume of ground ice was insufficient dependent on direct precipitation for to allow thaw lake development. water input, would be most sensitive to a negative shift in water balance; The development and expansion if drier summer conditions become of a drainage network during the prevalent, desiccation of these wet- Holocene has resulted in drainage lands may occur (Smol and Douglas of many large lakes, particularly 2007). over the past 5,000 years (Hinkel et al. 2003, Jorgenson and Shur 2007). Effects of ice wedge thawing are Lake drainage may be triggered by already evident on the Arctic a variety of processes, such as ice Coastal Plain. A large increase in wedge erosion, headward stream both area and density of degrading erosion, tapping, bank overflow, ice wedges has been observed in or coastal erosion. Once lakes are study areas near the Colville River drained, the relatively warm water Delta (Jorgenson et al. 2006). The in deep waterbodies that promotes accelerated thermokarst degrada- talik formation is lost, and the tion, probably associated with record exposed sediments are once again subject to permafrost aggradation. The sandy margins tend to aggrade little ice, while ice segregation and ice wedge development in the organic-rich silty centers is prevalent. This differential ice accumulation typically causes the drained centers to dome up and shift water to the lower lying sandy margins. Small ponds created by this hydrologic shift are abundant around the margins. These typically fill in with limnic sediments in the center and sedge peat around the margins. Stabilization of a drained lake basin and re-establishment of polygonal terrain is a process that may take centuries to millennia (Jorgenson and Shur 2007).

The response of Coastal Plain deposits to a warmer and wetter environment is expected to have a dramatic effect on the stability of ice wedges accompanied by redistribution of Figure 4.5. Regional differences in the distribution of surficial deposits, water on the landscape. Because ice all of late Pleistocene origin, are a reflection of various past processes. wedges are formed just below the The coastal areas are characterized by glaciomarine pebbly silt, glacial active layer and are in close equilib- deposition of slightly pebbly silty sand from a northern ice sheet, and rium with the existing climate, an deposition of marine sand by marine transgression. The western portion of the Arctic Coastal Plain is covered by an eolian sand. The lower portion of abrupt warming will cause nearly all 43 ice wedges to degrade (Box 4.1, Fig- the Arctic Foothills is blanketed by thick loess (silt). The upper Foothills are ure 4.6). The redistribution of water, underlain by bedrock near the surface; ridges typically have rocky residual however, will depend on whether the soils, slopes are mantled with organic-rich colluvium (accumulations of site is a water-shedding slope or a sediment transported downslope by gravity), and basins are filled with water-gathering basin. organic-rich fine-grained colluvium. From Jorgenson et al. 2009. warm temperatures from 1989–1998, stable to increasing (Smith et al. has resulted in a deepening network 2005, Riordan et al. 2006, Mars and of water-filled thermokarst troughs Houseknecht 2007) or too sensitive and pits and drying polygon centers. to short-term variability in precipita- Based on the general distribution of tion regime to allow detection of a ice-rich soils, surface changes of this long-term trend (Plug et al. 2008). type could potentially affect 10–30% Large, deep lakes are expected to of arctic lowland landscapes. continue to expand in surface area, due to shoreline erosion by wind- Thermokarst lakes within the driven waves and lake expansion lower latitude zone of discontinuous from thermokarst. Erosion rates permafrost may drain as a result of averaged 0.08 m/yr for large (>20 a deepening talik that completely ha), deep (>1.5 m) lakes at three penetrates the permafrost, as study sites on the Coastal Plain west Figure 4.6. Schematic of Arctic documented on Alaska’s Seward of the Colville River Delta (Jorgen- Coastal Plain landscape, current Peninsula (Yoshikawa and Hinzman son et al. 2003), while shorelines (above) and projected (below). The 2003). On the North Slope, however, of small or shallow (≤1.5 m) lakes projected landscape illustrates continuous deep permafrost would eroded at rates averaging <0.04 m/ elements likely to change as a result preclude this mechanism. In con- yr. Based on extrapolation from Jor- of climate warming. Figure by R. tinuous permafrost environments, genson et al. (2003), the increment Mitchell/Inkworks for WildREACH studies have concluded that the in lake area due to expansion may from cited sources. surface area of large lakes is either comprise ~1–3% of the landscape by the end of the century, although this underestimates the effect if rates of lake expansion increase significantly due to a longer ice-free season and warmer water.

Most drained lake basins formed during a period of extensive drainage during the mid-Holocene (Jor- genson and Shur 2007) and contem- porary lake drainage is uncommon. With expected warming, however, degrading ice wedges may pro- gressively integrate into drainage channels with a lower base elevation resulting in increased frequency of lake-tapping (sudden drainage) events. Where ice-rich soils and a topographic gradient exist (e.g., Current adjacent to a stream or the coast), sequential lake tapping could occur (McGraw 2008).

Projected Box 4.1. Permafrost Degradation

The susceptibility of arctic landscapes to thermokarst processes depends on ground ice content, and this varies according to the surficial geology and geomorphic setting. Ground ice may occur as segregated ice (smaller stuctures interspersed within the permafrost) or massive bodies such as ice wedges, and the relative abundance of each type affects how the surface will respond to warming, at both the landscape scale and over distances of a few meters. On the Arctic Coastal Plain, mean segregated ice volume in the upper 2 m ranged from 45% in inactive sand dunes to 71% in alluvial-marine deposits (Jorgenson et al. 1998, 2003). The mean volume of ice wedges in the upper 2 m varied from less than 1% in ice-poor margins of lakes to a high of 33% in ice-rich lake centers (Jorgenson and Shur 2007). Understanding how the type and amount of ground ice varies across the landscape is essential to forecasting how habitats may change in response to warming temperatures. Above: This polygonal terrain is Ice wedges, with top surfaces located just 30-40 cm beneath the surface, located in the Niguanak Lowlands are particularly sensitive to warming air temperatures (Jorgenson et al. on the coastal plain of the Arctic 2006). Melting of the tundra surface results in varying amounts of ground National Wildlife Refuge. The settlement, with ice wedges melting faster than the intervening polygon bright green polygon troughs are centers. The subsiding surface overlying the melting ice wedges forms associated with the reticulate- water-filled troughs with deeper pools at the intersections (see photos). patterned network of subterranean As the trough system develops and deepens, a steeper drainage gradient ice wedges. The wet troughs support redistributes water from broad polygon centers into narrow, deep troughs. lush sedge vegetation, in contrast to The landscape may shift from a more uniform mesic state to one of alter- the drier polygon centers. Photo by nating patches of dry polygon centers and flooded troughs. In settings M.T. Jorgenson. with sufficient topographic gradient, which occur commonly in the transitions from upland ridges to lake basins and floodplains, deepening troughs can develop into a drainage network that promotes runoff, lowers the water table, and results in more widespread drying. In contrast to the upland setting, ice wedge degradation in topographic basins will create more capacity for water storage but should have little effect on water levels.

Left: Aerial view of polygonal pools formed by ice wedge degradation west of the Colville River Delta. Below: Ground level view of the same location. Both photos by M.T. Jorgenson.

45 Arctic Foothills have little effect on surface hydrol- The upland ecosystems of the Arctic ogy. Under drying conditions, little Foothills have three major distinc- change is expected. tive geomorphic environments: 1) rocky residual soils on ridges, 2) The colluvium-mantled hillsides are gentle slopes mantled with ice- and likely to be very sensitive to climate organic-rich colluvium (material warming. The soils on colluvium, deposited at slope-bottom by grav- typical of mid- to lower slopes, tend ity) over bedrock, and 3) Pleistocene to be highly organic, saturated, loess on the lower Foothills (Figure and have abundant ice wedges and 4.7). Response of each Foothills ter- segregated ice near the permafrost rain condition is discussed sepa- table. Because the active layer is un- rately below. derlain by ice-rich permafrost, thaw slumps are likely to become abun- The rocky residual soils are well- dant on the sloping surfaces (Gooseff drained, have deep active layers, and et al. 2009). The slumping will create are relatively thaw stable. Rocky new thaw lakes, expose new soil Figure 4.7. Schematic of Arctic ridges probably will undergo little to plant colonization, and increase Foothills landscape, current (above) geomorphic change under a warmer sediment transport in runoff (Figure and projected (below). The projected and wetter climate. The soils will 4.7). Gullies are likely to become landscape illustrates elements likely remain well drained and thaw stable. common where water flow through to change as a result of climate The active layer will likely increase ice wedge networks causes the warming. Figure by R. Mitchell/ and taliks may develop on some ground surface to collapse. The gul- Inkworks for WildREACH from south-facing slopes, but this will lies then contribute to channelization cited sources. of flow and drying of lakes and intervening ridges. In some areas, water tracks may deepen without deeper gully formation but still serve to channel suprapermafrost groundwater flow. On very gentle slopes, ice wedges are likely to degrade without prominent gully formation.

Landscapes characterized by the presence of extremely ice-rich loess (yedoma) of late Pleistocene origin are highly sensitive to warming and have the potential for drastic change.

Current

46 Projected Yedoma is abundant across the lower Floodplains foothills and may occupy roughly Floodplains are active geomorphic 20% of the overall foothills landscape environments because of sedimen- (Carter 1988). In this type of terrain, tation caused by flooding, erosion masses of ice can extend to depths of caused by channel migration, and up to 30 m (Carter 1988). Along the deposition of wind-blown sand from lower Colville River, the yedoma has barren river bars (Figure 4.8). The only 0.5–1 m of soil covering 10–25 varying flooding regime creates a m of ice. Because of the massive ice, sequence of deposits that include deep thermokarst lakes are com- massive or crossbedded sand in the mon on this terrain. The degraded active channel; rippled interbed- yedoma landscape of the Seward ded sands and fine-grained mate- Peninsula may provide a good analog rial (fines) with distinctive detrital of how the arctic landscape could be organics that are the remains of peat altered by widespread ice degrada- banks eroded upstream; layered tion. First, thermokarst develops fines caused by vertical accretion Figure 4.8. Schematic of arctic in the network of shallow Holocene of silts during overbank flooding; floodplain landscape, current ice wedges near the surface of the layered organics and silts created by (above) and projected (below). The permafrost. Next, the thermokarst the accumulation of organic matter projected landscape illustrates troughs and pits expand into ther- between infrequent flooding events; elements likely to change as a result mokarst ponds and lakes that thaw and massive organics that accumu- of climate warming. Figure by R. into the underlying massive ice. Fi- late on higher floodplains that are Mitchell/Inkworks for WildREACH nally, expansion of the thermokarst rarely flooded (Shur and Jorgenson from cited sources. lakes and drainage networks causes the lakes to drain, forming a thaw lake plain—this process, however, is likely to take hundreds to thousands of years.

Current

47 Projected 1998). During floodplain evolution, In response to warming, permafrost the deposits are modified by the ag- aggradation will be retarded in the gradation of segregated and wedge barren portions of active floodplains, ice, which deforms the surface, and degradation will be acceler- affects water runoff, and increases ated on the inactive and abandoned the susceptibility of older terrain to floodplains. Ice wedges formed in thermokarst. These deposits support the later stages of floodplain devel- a successional sequence of ecotypes opment will degrade, thus lower- from river to riverine barrens, tall ing the water table and instigating shrub, low shrub, dwarf shrub, formation of drainage networks that and wet sedge meadows. River- will accelerate drainage of riverine ine ecosystems are not abundant, lakes. The consequences of altered comprising ~8% of the Coastal Plain precipitation (amount, seasonality, and ~4% of the Foothills. They are and frequency of extreme events), highly productive ecosystems, how- discharge, flooding, sedimentation, ever, serving as conduits of water, and erosion are more uncertain. sediments, and nutrients, and are Many of these processes are more used by a wide range of species. sensitive to extreme events rather than to average conditions. A scenar- Riverine ecosystems could respond io of increased precipitation will be to climate change in a variety of accompanied by increased flooding, ways depending on the amount of sedimentation, and erosion. This, in warming and the balance between turn, should favor more productive evapotranspiration and precipitation. early successional ecosystems. In contrast, decreased runoff associated with drying during midsummer may lead to increased channel stability and increased shrub growth on the stabilized active floodplain.

Floodplain lakes include oxbow and thaw lakes. These lakes are important to fish and bird species and play an important role in trapping sediments and the biogeochemical processing of water before it is discharged to the ocean. Fish access, as well as nutrient status, is related to the degree of connectivity with river systems, the duration of which can vary from constant (all summer) to a transitory connection limited to high-water events (especially spring flood). For example, when flow from upstream snow melt begins before the onset of melt downstream, ice dams may spread water over vast areas in arctic river deltas, recon- necting and recharging lakes. In the MacKenzie River Delta, the degree to which lakes are connected depends on their distance from the nearest active channel and elevation. During low-water years, associated with decreased ice-dam flooding, lakes greater than 4 m above sea level are not recharged or connected to riverine source waters (Lesack and Marsh 2007). If changes in freeze-thaw cycles cause years with little ice-damming to become more frequent, these lakes may dry, sedi- 48 Above: A male Wilson’s warbler (Wilsonia pusilla) singing during the ment transport to river deltas may breeding season. Songbird species may move into expanded shrub habitats increase, and biogeochemistry of if active arctic floodplains stabilize in drier summers. Photo by Ted Swem, river water discharged to the ocean USFWS. Below: The floodplain of the Katakturuk River in the Arctic may be altered. National Wildlife Refuge shows the effects of past flooding and erosion events on the vegetation community. USFWS photo. Historically, tundra fires have been Fire frequency on the Seward rare on the North Slope, with only Peninsula is greater than that on Fire Regime 10 known occurrences north of 69°N the North Slope (Racine and Jandt from 1956 to 2007 (Racine and Jandt 2008). The two regions are similar, 2008). In the fall of 2007, however, however, with respect to frequency coinciding with an anomalously of lightning strikes (http://geology. warm and dry season, a 1,000 km2 com/articles/lightning-map.shtml, area burned between the Itkillik and retrieved April 2009) and vegetation. Nanushuk rivers in the central Arc- Differing soil moisture, therefore, tic Foothills; this is the largest North most likely accounts for the differ- Slope fire on record (Racine and ence in fire frequency. In Barrow, Jandt 2008). This event underscores under near-average summer condi- the potential for more frequent, tions, polygon rims and troughs larger tundra fires under a climate remained >80% saturated after late scenario of warmer, drier summers. June (Liljedahl et al. 2009). The wa- A severe fire reduces surface albedo ter table on tussock tundra hillsides and combusts the insulating surface in the upper Kuparuk River water- organic layer, increasing soil ther- shed typically remains within 10 mal conductivity and heat flow into cm of the ground surface (Hinzman the ground (Yoshikawa et al. 2002). et. al. 1993), whereas soils at 10 cm Post-fire soil moisture increases in depth at a Seward Peninsula tus- the short term (Liljedahl et al. 2007), sock site declined to ~40% saturated which also increases soil thermal after the spring peak (Liljedahl conductivity and promotes warming. et al. 2007). Predicting the future Soil warming is accompanied by at North Slope fire regime, therefore, least a temporary thickening of the is closely linked with projection of active layer and, if the soils are ice- summer water balance. rich, can stimulate thermokarst. In some settings, long-term change in active layer thickness and a change in vegetation community (decreased moss cover and increased shrub cover) ensues, reinforcing a shift toward warmer, drier near-surface soils (Yoshikawa et al. 2002). On the Seward Peninsula, active layer thickness diminished to pre-fire levels within a few decades on flatter terrain, but a thickened active layer persisted on steeper slopes (Racine et al. 2004). The post-fire warming and drying on slopes promoted a shift to willow shrub vegetation on the steeper slopes, but tussock sedge recovered well on the poorly drained sites. Sphagnum moss and lichen failed to recover after 24 years, except for Sphagnum in a wet meadow site (Racine et al. 2004).

The Anaktuvuk Fire ignited on July 16, 2007, and burned until early October. This MODIS satellite image was obtained on September 25, 2007, and the red dots indicate areas still burning on that date (NASA, http://earthobservatory.nasa.gov/NaturalHazards/view. php?id=19139&oldid=14550).

49 Current Vegetation Communities gional differences in vegetation re- Vegetation The North Slope of Alaska com- flect differences in climate, topogra- prises less than 3% of the circum- phy, and parent material (Table 4.1, polar arctic region mapped by the Figure 4.9, 4.10, Appendix 2). CAVM Team (2003) but contains 26% of the area classified as sedge, The Coastal Plain landscape is moss, dwarf-shrub wetland, primar- a mosaic of low-lying lacustrine ily on the Coastal Plain, and 24% of (lake-related) basins separated by the area classified as tussock-sedge, intervening higher terrain. Lacus- dwarf-shrub, moss tundra, mostly trine ecosystems include deep (>1.5 in the Foothills (Figure 4.9). Ecore- m) lakes, shallow (<1.5 m) lakes and ponds, lacustrine grass marsh (dominated by Arctophila fulva in water >0.3 m deep), lacustrine sedge marsh (Carex aquatilis and Eriophorum angustifolium in water 0.1–0.3 m deep), and lacustrine wet sedge tundra (C. aquatilis, E. angustifolium, forbs and mosses in water <0.1 m deep). The older, higher terrain between basins also supports lowland wet sedge tundra in swales, lowland moist sedge-shrub tundra on lower slopes, and upland tussock tundra on upper slopes and gentle ridges. Prostrate willows and Dryas (a small flowering shrub) occur on moist sedge-shrub tundra, but typically comprise <35% cover.

Figure 4.9 (left). Vegetation of arctic Alaska based on the Circumpolar Arctic Vegetation Map (CAVM 2003). Map by FWS, based on CAVM data. Available online at http://data.arcticatlas.org/geodata/ circumpolar/cavm/.

Arctic Coastal Plain Arctic Foothills Brooks Range 15% 10% 17% 23% Other 28% Other 35% Other 10% Alpine Low Moist 7 % Upland Moist Noncarbonate Sedge-shrub Low Moist Upland Shrubby Sedge-shrub Dwarf Shrub Sedge-shrub Tussock 12% Coastal Water Upland Moist Alpine 8 % Noncarbonate Sedge-shrub 14% Low Wet Barrens Sedge-shrub Upland Low Upland Tussock Upland 20% Tussock Shrub Birch- Upland Shrubby Upland Low Shrub willow Tussock Birch-willow Lowland Lake 10% 14% 20% 19% 14% 25%

Alpine Noncarbonate Dwarf Shrub Tundra Upland Tussock Tundra Lowland Lake

Alpine Noncarbonate Barrens Upland Moist Sedge-shrub Tundra Coastal Water 50 Upland Shrubby Tussock Tundra Lowland Moist Sedge-shrub Tundra Other

Upland Low Shrub Birch-willow Tundra Lowland Wet Sedge-shrub Tundra

Figure 4.10. Distribution of ecotypes among ecoregions of the North Slope, after Jorgenson and Heiner 2003. Uplands in the Foothills are charac- (C. bigelowii, D. integrifolia) on terized by shrub-dominated vegeta- circum-alkaline soils. Less common tion. Dominant ecotypes include are lowland wet sedge tundra and upland Dryas and ericaceous dwarf lowland low birch-willow shrub in shrub on dry rocky ridges, upland swales, toe-slopes, and basins. low birch-willow shrub (Betula nana, Salix. planifolia, S. glauca) Upland tall alder shrub occurs on on better drained soils, upland mid- to upper slopes of the Brooks shrubby tussock tundra (B. nana, S. Range but is <1% of the landscape. planifolia, E. vaginatum) and up- Tall alder-willow shrub occurs in land tussock tundra (E. vaginatum) riverine settings in both the Foothills on saturated organic-rich soils, and and the Brooks Range, but is <0.1% upland moist sedge-shrub tundra of the landscape in either ecoregion.

Table 4.1 Distribution of ecotypes1 among ecoregions of the North Slope. More abundant (>5% of area) ecotypes are highlighted in bright yellow. Brooks North Slope Ecotype Coastal Plain Foothills Range (total area) Area % Area % Area % Area % (km2) (km2) (km2) (km2) Coastal Barrens 507.7 0.9 9.5 0.0 0.0 0.0 517.2 0.2 Coastal Wet Sedge Tundra 1253.8 2.3 40.9 0.0 0.0 0.0 1294.7 0.6 Coastal Water 6741.6 12.1 301.4 0.3 0.0 0.0 7043.0 3.4 Coastal Grass and Dwarf Shrub Tundra 1320.0 2.4 79.2 0.1 0.0 0.0 1399.2 0.7 Riverine Barrens 548.4 1.0 921.1 1.0 228.3 0.4 1697.8 0.8 Riverine Willow Shrub Tundra 75.9 0.1 920.9 1.0 146.9 0.3 1143.7 0.5 Riverine Moist Sedge-Shrub Tundra 1374.4 2.5 2948.2 3.1 579.5 1.0 4902.1 2.3 Riverine Wet Sedge Tundra 951.3 1.7 914.4 1.0 135.3 0.2 2000.9 1.0 Riverine Waters 554.2 1.0 610.9 0.6 132.1 0.2 1297.2 0.6 Riverine Dryas Dwarf Shrub Tundra 0.0 0.0 4.8 0.0 270.2 0.5 275.0 0.1 Riverine Spruce Forest 0.0 0.0 0.0 0.0 1.8 0.0 1.8 0.0 Lowland Wet Sedge Tundra 11239.0 20.2 3511.4 3.7 643.6 1.1 15394.0 7.4 Lowland Lake 7983.2 14.4 1665.3 1.7 246.3 0.4 9894.9 4.7 Lowland Moist Sedge-Shrub Tundra 12671.0 22.8 6923.7 7.2 0.0 0.0 19595.5 9.4 Lowland Low Birch-Willow Shrub 683.2 1.2 1293.7 1.4 297.6 0.5 2274.6 1.1 Lowland Spruce Forest 0.0 0.0 0.0 0.0 11.5 0.0 11.5 0.0 Upland Tussock Tundra 7480.6 13.5 9682.7 10.1 0.0 0.0 17163.7 8.2 Upland Dryas Dwarf Shrub Tundra 891.5 1.6 551.7 0.6 1089.5 1.9 2532.9 1.2 Upland Shrubby Tussock Tundra 17.6 0.0 33469.3 34.9 10448.6 18.2 43938.2 21.0 Upland Low Shrub Birch-Willow Tundra 1199.5 2.2 23831.6 24.9 11357.8 19.8 36390.4 17.4 Upland Moist Sedge-Shrub Tundra 18.1 0.0 7869.5 8.2 5621.5 9.8 13510.1 6.5 Upland Tall Alder Shrub 0.0 0.0 0.0 0.0 211.8 0.4 211.8 0.1 Upland Spruce Forest 0.0 0.0 0.4 0.0 41.2 0.1 41.6 0.0 Alpine Glaciers 0.0 0.0 0.0 0.0 198.6 0.3 198.6 0.1 Alpine Non-carbonate Barrens 0.0 0.0 62.7 0.1 8133.0 14.2 8195.8 3.9 Alpine Carbonate Barrens 0.0 0.0 0.0 0.0 103.2 0.2 103.2 0.0 Alpine Mafic Barrens 0.0 0.0 0.0 0.0 103.1 0.2 103.1 0.0 Alpine Non-carbonate Dwarf Shrub Tundra 0.0 0.0 51.5 0.1 16060.3 28.0 16112.0 7.7 Alpine Carbonate Dwarf Shrub Tundra 0.0 0.0 0.0 0.0 129.5 0.2 129.5 0.1 Alpine Mafic Dwarf Shrub Tundra 0.0 0.0 0.0 0.0 185.4 0.3 185.4 0.1 Riverine Alder-Willow Shrub 0.0 0.0 0.0 0.0 2.0 0.0 2.0 0.0 51 Cloud, Snow, and Ice 5.4 0.0 123.7 0.1 1071.7 1.9 1200.8 0.6 Total 55516.4 26.6 95788.7 45.9 57450.1 27.5 208762.5 100 1. Source: Jorgenson and Heiner (2003). Projected Changes in Vegetation Patches of balsam poplar forest oc- Communities cur in scattered locations across the Treeline northern Brooks Range and Foot- The Arctic Climate Impact Assess- hills, in floodplain settings with year- ment (ACIA 2005, p. 332) predicted round groundwater flow (Bockheim very likely replacement of tundra et al. 2003). They also occur in nu- by forest, and some continent-scale merous small patches on floodplains biome shift models predict that a and hillsides in northwestern Alaska taiga environment will replace North where mean annual air tempera- Slope tundra by the year 2100 (Gon- tures are -6 to -8 °C (Jorgenson et al. zalez et al. 2005). Although north- 2004). Because these trees release ward shift of the spruce treeline has highly mobile wind-dispersed seeds been observed (Lloyd et al. 2002), and are adapted to growing on well- modeling studies focused on the drained, early successional habitats, Alaska forest-tundra ecotone predict balsam poplar should be able to that a shift from upland tundra rapidly advance down floodplains to spruce forest (5% tree cover) across arctic Alaska in response to would take >150 years (Chapin and warming temperatures. Starfield 1997, Rupp et al. 2000a). Furthermore, the position of the Shrub Expansion Brooks Range coincides with the Shrubs can have dramatic effects on latitudinal treeline and is expected local microclimates and may provide to pose a significant topographic a positive feedback that contributes barrier to forest establishment on to global warming (Sturm et al. the Alaska North Slope (Rupp et al. 2001, Pomeroy et al. 2006). Shrubs 2000b). Conversion of North Slope trap snow and can increase winter tundra to spruce forest is therefore soil temperatures, and the snow considered an unlikely scenario cover change can affect the timing within the century time frame. and magnitude of all surface energy balance components during the melt and increase runoff late in the snow- 6 July 1950 melt period (Liston et al. 2002). Within the Brooks Range and Foot- hills, an expansion of shrub tundra into areas formerly dominated by sedge is evident over the time scale of a half century. Increased abundance of alder, willow, and dwarf birch is detectable via comparison of historical oblique aerial photos 27 July 2002 (1940s) with modern photography (Sturm et al. 2001, Stow et al. 2004, Tape et al. 2006). The study sites are located in the Brooks Range and Arctic Foothills, primarily in the Colville River watershed. Expansion of alder is most conspicuous in the photographs, but field observations confirm that willow and shrub birch cover has increased as well. These results pertain mainly to hill slopes and valley bottoms.

Vegetation on the broad interfluve areas of the Foothills is primarily tussock-shrub tundra, with the tussock-forming sedge Eriopho- rum vaginatum codominant with low shrubs, principally dwarf birch A systematic comparison of photos of Alaska’s North Slope taken in Betula nana and willow Salix pul- the mid-20th century with recent photos of the same locations reveals an chra. Plot experiments conducted 52 increase in shrub vegetation over time (Sturm et al. 2001). In 2002, alder at Toolik Field Station suggest that shrubs cover more of the landscape along the Nimiuktuk River, a tributary relative cover and canopy height of the western Noatak River, compared to conditions in 1950. The vegetation of shrubs (principally B. nana) will of the western Brooks Range and North Slope appears to be responding increase under conditions of warmer to changes in climate, with implications for surface energy exchange and summer temperatures and deeper carbon flux. Photos courtesy of Ken Tape. snow (Wahren et al. 2005). These Paludifcation results are consistent with increased Paludification refers to the process shrub and graminoid height and of establishment of peat-forming cover after artificial warming at 10 plant communities on both dry lands other international experimental and in bodies of water (bog forma- tundra biome sites (Walker et al. tion from water-body infilling may 2006). Increased birch cover could be considered a distinct process, promote fire spread, because shrub terrestrialization, but here we refer birch is relatively flammable com- to all bog formation as “paludifica- pared with other shrubs and grami- tion”). Klinger (1990) proposed that noids (Higuera et al. 2008). mosses such as Sphagnum play a key role in driving this successional Remote sensing methods, most process. Initially, according to this notably the normalized difference hypothesis, organic acids released vegetation index (NDVI) have also by moss cause mortality of fine roots been used to assess vegetation in vascular plants, setting in motion trends in the Arctic. NDVI values a shift in plant species composition, are a measure of “greenness” and with an increasingly acidic and moss- correlate well with above-ground dominated bog as the stable climax plant biomass. Retrospective studies community. In the boreal and arctic show an increase in NDVI in arctic environment, conditions created Alaska over the periods 1981–2001 by Sphagnum normally produce (Jia et al. 2003) and 1990–1999 (Stow a positive feedback that enhances et al. 2004). Increasing NDVI values paludification by providing cooler, could reflect increased shrub cover increasingly acidic soils leading to and stature, but this interpretation increased peat formation, a thin- is not straightforward. A 22-year ner active layer, and permafrost record analyzed by Verbyla (2008) aggradation (Auerbach et al. 1997). demonstrated a significant positive Extrinsic disturbance that results in trend in the annual maximum NDVI drier soils and/or higher alkalinity on the Arctic Coastal Plain and leads to conditions less favorable for Arctic Foothills, with no significant Sphagnum growth (Spatt and Miller trend in the Brooks Range. Increas- 1981). ing biomass on the Arctic Coastal Plain is not likely to be associated Present-day acidic tussock tundra with shrub increase, as shrubs that dominates the Arctic Foothills constitute a very low proportion of may have evolved from arid grass- the cover. The only available study forb communities that were preva- from the Coastal Plain based on lent during the late Pleistocene, field measurements (Jorgenson and through paludification in the warm- Buchholtz 2003) reported a small but er, moist Holocene environment significant decline in non-vascular (Walker et al. 2001). Tussock tundra plant cover, but no significant change communities are characterized by Figure 4.11. Thermokarst pits on in vascular plant cover or shrub plants that are relatively unpalatable the Hulahula Lowlands, Arctic height in plot studies covering the to herbivores, with high concentra- National Wildlife Refuge. Orange period 1984–2002. On the wet and tions of toxic secondary compounds, coloration on trough margins are lowland moist sedge tundra of the and thus offer less favorable habi- patches of Sphagnum moss. Photo by Arctic Coastal Plain, therefore, the tat for herbivores. Conversion of M.T. Jorgenson. increasing NDVI values reported by Verbyla (2008) might be indicative of more robust graminoid growth. An- ecdotal evidence of shrub invasion in this ecoregion includes the establishment of alders (Alnus spp.) on polygon rims in the Colville River Delta (M.T. Jorgenson, ABR Inc., personal communication; see image page 37). There is no strong evidence, however, for widespread shrub expansion on the Arctic Coastal Plain.

53 shallow water and sedge tundra on Phenology the Arctic Coastal Plain to acidic Growth and photosynthetic activ- bog habitats would have profound ity commences almost immediately ecological implications, given that after snowmelt exposes plants in acidification impedes nutrient avail- spring and the ground thaws. Data ability, lowers productivity, and collected as part of the International creates favorable conditions for Tundra Experiment (ITEX) sug- slower-growing sedges and heath gest that experimental warming of shrubs (Szumigalski and Bayley tundra can produce “small to moder- 1997, Thormann and Bayley 1997). ate” advance in green-up (Arft et al. 1999). Similarly, NDVI data from In the northern hemisphere, Sphag- northeast Alaska spanning 12 years num reaches peak abundance in the (Figure 4.12) shows that in warmer climate zone featuring 630–1,300 mm years (1990s), green-up is advanced annual precipitation and -2°C to -6°C approximately two weeks on the MAAT (Gajewski et al. 2001), both Coastal Plain. Peak greenness oc- warmer and wetter than current curs in mid-July regardless of mean conditions in arctic Alaska. In con- temperature, but biomass averages trast to the Arctic, wet drained lake higher in the warmer years. Absent basins in the warmer environment of other influences, warming air tem- the Seward Peninsula (MAAT of -6 peratures are projected to advance °C) are characterized by vegetation spring thaw an average of 8–11 days co-dominated by sedges and Sphag- by the end of the century, and as num moss (Jorgenson et al. 2004). If indicted by the ITEX results, may the Seward Peninsula coastal climate advance the onset of annual plant is a reasonable analog for the future growth and development. Deepening climate of the Coastal Plain, we of the snowpack and greater cloud may expect increased abundance of cover in spring, however, could offset Sphagnum in the Arctic and poten- the effect of increased air tempera- tial conversion of wet sedge tundra ture, slowing snow ablation and pos- to bog meadow (M.T. Jorgenson, sibly delaying onset of green-up. ABR Inc., personal communication). The rate at which peatlands Senescence in plants is a highly might develop on the Coastal Plain is regulated, species-specific process unknown. Holocene peatland forma- controlled by both endogenous (e.g., tion in northern Canada developed nutrient drain, plant hormones) and over a period >1,000 years (Kuhry exogenous (e.g., photoperiod, tem- et al. 1992), but Sphagnum has been perature) factors (Kelly et al. 1988). observed to fill thermokarst pits The potential impacts of increased (Figure 4.11) on the Arctic Coastal temperature on plant senescence Plain over the past half century are not well understood. Indeed, (M.T. Jorgenson, ABR Inc., personal experimental warming of tundra has communication). produced conflicting results. The ITEX experiment reported little effect on timing of senescence, indicat- ing that photoperiod might play a more important role in late-season phenology (Arft et al. 1999). In contrast, an experiment in northeast Greenland using an alternate warming technique showed that warmer temperatures postponed the onset of senescence (Marchand et al. 2004). In further contrast, NDVI data from northeast Alaska indicated that the decline in senescence began earlier in the warmer years (Figure 4.12). Given the conflicting evidence and the complexity surrounding onset and control of plant senescence, it cannot be assumed that warmer fall temperatures will extend the period 54 of greenness. Figure 4.12. Median normalized difference vegetation index (NDVI) within the late July concentrated use area of satellite-collared adult females from the Porcupine caribou herd in the 1980s (1985–1989) compared to the 1990s (1990–2001). Source: Brad Griffith, USGS, Alaska Cooperative Wildlife Research Unit, unpublished data. Contaminants generated through- Weidinmeyer and Friedli (2007) out the globe are deposited across estimated mercury emissions from Contaminants the arctic landscape through at- wildfires in Alaska and the lower 48 mospheric deposition and ocean states to average 44 metric tons per currents, in addition to local sources. year from 2002–2006. This amount Climate change may increase both is roughly equivalent to 30% of the transport and mobilization of these total mercury emissions permitted contaminants to fish, wildlife, and by the U.S. Environmental Protec- their habitats. For example, climate tion Agency in 2002. While emis- change may increase contaminant sions due to fire were quite variable mobilization from previously con- in Alaska, it is notable that in 2004 tained sources. Coastal erosion and fire-related mercury emissions in sea level rise may breach or erode Alaska (32 metric tons) exceeded sewage lagoons, landfills, dumps, the combined total amount from the tailing ponds, drilling mud pits, and lower 48 states (23 metric tons). In oil-contaminated soils (AMAP 2003), 2005, mercury emissions from fires releasing contaminants into the in Alaska (22 metric tons) were only nearshore environment. Shoreline slightly lower than the total for the fuel delivery, storage, and pipe- other states (28 metric tons). line systems may also be damaged or destroyed; in 2007, Newtok, Mercury concentrations in fish have Alaska, lost its barge landing and been shown to increase following a 1,000-gallon fuel tank in a storm fires in Alberta, Canada, both by (CCAAG 2008). Increased discharge increasing mercury inputs and by from large river systems and the restructuring food webs; top preda- associated increases in sediment tors switched from invertebrate prey transport also will bring additional to smaller fish (Kelly et al. 2006). contaminant mobilization. For ex- Collectively, these studies suggest ample, mercury concentrations were that these fire-related mercury positively correlated with river flow emissions will continue, assuming and were greatest during spring fire incidence remains constant or freshets and high flow events in the increases with a warming climate. If Mackenzie River Delta (Leitch et al. wetland area increases as predicted 2007); increased mercury concentra- by some climate models, the area tions may in part reflect increased available for methylation of mercury surface inundation and bank erosion. will also increase. Increased mercu- Contaminants currently contained ry mobilization and methylation will within glacial ice, multi-year sea ice, cause mercury levels in all trophic and permafrost, including persistent levels to increase; already, mercury organic pollutants and mercury, will is a major contaminant of concern almost certainly be mobilized to for arctic biota. aquatic ecosystems as these systems melt. Additionally, long-range atmospheric transport will continue to be a source of contaminants to the Arctic.

Increased number or intensity of fires could result in additional mercury inputs to the arctic tundra. Turetsky et al. (2006) evaluated climate-related changes in fire incidence and associated mercury emissions. They found that mercury formerly sequestered in cold, wet peat soils is released to the environment during fires in Canadian boreal forests and present a growing threat to sensitive aquatic habitats and northern food chains as the climate warms. Estimates of circumboreal mercury emissions presented in Top: The Anaktuvuk Fire of 2007 was by far the largest fire ever recorded this study are 15-fold greater than 55 estimates that did not account for north of the Brooks Range. Photo courtesy Arctic Long Term Ecological mercury stored in peat soils. Research program. Bottom: Debris from a dump at the abandoned military site at Nuvagapak Point in the Arctic National Wildlife Refuge. Coastal erosion is exposing buried industrial and household waste. USFWS photo. Literature Cited Chapin FS, Starfield AM. 1997. IPCC. 2007. Climate Change 2007: Time lags and novel ecosystems Synthesis Report. Contribution of AMAP. 2003. AMAP Assessment in response to transient climatic Working Groups I, II and III to the 2002: The Influence of Global change in arctic Alaska. Clim. Fourth Assessment Report of the Change on contaminant pathways Change 35: 449-461. Intergovernmental Panel on Cli- to, within, and from the Arctic. mate Change. Geneva, Switzerland: Arctic Monitoring and Assessment Flint PL, Mallek MJ, King RJ, IPCC. 104 p. Programme (AMAP), Oslo, Nor- Schmutz JA , Bollinger KS , way. xi+65 pp. Derksen DV. 2008. Changes in the Jia G, Epstein HE, Walker abundance and spatial distribution DA. 2003. Greening of arc- Arft AM, Walker MD, Gurevitch J, of geese molting near Teshekpuk tic Alaska, 1981–2001. Geo- Alatalo JM, Bret-Harte MS, Dale Lake, Alaska: interspecific compe- physical Research Letters, 30, M, Diemer M, F. Gugerli MF, Hen- tition or ecological change. Polar 2067,doi:10.1029/2003GL018268. ry GHR, Jones MH, Hollister RD, Biol. 31: 549-556. IJnsdttir LS, Laine K, Lévesque Jones BM, Arp CD, Jorgenson MT, E, Marion GM, Molau U, Mlgaard Gajewski K, Viau A , Atkinson M, Hinkel KM, Schmutz JA , Flint PL. P, Nordenhäll U, Raszhivin V, Rob- Sawada D, Wilson S. 2001. Sphag- 2009. Increase in the rate and uni- inson CH, Starr G, Strenstrm A, num peatland distribution in formity of coastline erosion in Arc- Totland Ø., Turner PL, Walker LJ, northern America and Eurasia dur- tic Alaska. Geophys. Res. Lett. 36, Webber PJ, Welker JM, Wookey ing the past 21,000 years. Global L03503, doi:101029/2008GL036205. PA. 1999. Responses of tundra Biogeochem. Cycles 15: 297–310. plants to experimental warming: Jorgenson J, Buchholtz C. 2003. meta-analysis of the International Gibbs AE, Richmond BM, Erikson Eighteen years of vegetation Tundra Experiment. Ecol. Monogr. L. 2008. Regional shoreline change monitoring in the Arctic National 69: 491-511. along the North Slope of Alaska. Wildlife Refuge. Alaska. Unpub- In EOS Transactions, American lished Abstract, SEARCH Open Auerbach NA. Walker MD, Walker Geophysical Union Fall Meeting, Science Meeting. http://siempre. DA. 1997. Effects of roadside dis- Abstract C11C-0521. arcus.org/4DACTION/wi_pos_dis- turbance on substrate and vegeta- playAbstract/7/600, accessed April tion properties in arctic tundra. Gonzalez P, Neilson RP, Drapek RJ. 2009. Ecol. Appl. 7: 218-235. 2005. Climate change vegetation shifts across global ecoregions. Jorgenson MT, Brown J. 2005. Clas- Bockheim, JG, O’Brien JD, Munroe Ecological Society of America An- sification of the Alaskan Beau- JS, Hinkel KM. 2003. Factors af- nual Meeting Abstracts 90: 228. fort Sea Coast and estimation of fecting the distribution of Populus sediment and carbon inputs from balsamifera on the North Slope of Gooseff MN, Balser A, Bowden WB, coastal erosion. Geo. Mar.Lett Alaska, USA . Arct. Antarct. Alp. Jones JB. 2009. Effects of hillslope 25:69-80. DOI 10.1007/s00367-004- Res. 35:331-340. thermokarst in northern Alaska. 0188-8. EOS, Trans. Am. Geophys. Vol. 90, Callaghan, T. 2005. Arctic tundra doi:10.1029/2009EO040001, 2009, Jorgenson MT, Heiner M. 2003. Eco- and polar desert ecosystems. In published online January 4, 2009. systems of Northern Alaska. Un- ACIA. 2005. Arctic Climate Impact published 1:2.5 million-scale map Assessment. New York: Cambridge Gordeev VV. 2006. Fluvial sediment produced by ABR, Inc., Fairbanks, University Press. (p 244-352) flux to the Arctic Ocean. Geom- AK and The Nature Conservancy, http://www.acia.uaf.edu. prhology 80: 94-104. Anchorage, AK. (Available at www.nature.org/wherewework/ Carter LD. 1988. Loess and deep Higuera PE, Brubaker LB, An- northamerica/states/alaska/files/ thermokarst basins in arctic Alas- derson PM, Brown TA, Kennedy aya2.pdf). ka. In Senneset K, ed., Proceedings AT, Feng SH. 2008. Frequent fires of the Fifth International Confer- in ancient shrub tundra: implica- Jorgenson MT, Pullman ER, Shur ence on Permafrost, Trondheim, tions of paleorecords for arctic YL. 2003. Geomorphology of the Norway: Norway, Tapir Publishers, environmental change. PLoS ONE Northeast Planning Area of the v. 1, p. 706–711. 3(3): e0001744. doi:10.1371/journal. National Petroleum Reserve- pone.0001744. Alaska, 2002, second annual report. CAVM Team. 2003. Circumpolar Phillips Alaska, Inc., Anchorage. Arctic Vegetation Map. Scale Hinzman LD, Kane DL, Everett 67 p. 1:7,500,000. Conservation of Arctic KE. 1993. Hillslope hydrology in Flora and Fauna (CAFF) Map No. an Arctic setting. In Proceedings Jorgenson MT, Roth JE, Emers M, 1. U.S. Fish and Wildlife Service, of Sixth International Conference Davis W, Schlentner SF, Macander Anchorage, Alaska. on Permafrost, July 5-9, Beijing MJ. 2004. Landcover mapping China, South China Technology for Bering Land Bridge National CCAAG. 2008. Climate Change Press, p. 267-271. Preserve and Cape Krusenstern Adaptation Advisory Group. Alaska National Monument, Northwestern 56 Climate Change Strategy: Public Hopkins DM, Hartz RW. 1978. Alaska. Final Report prepared for Infrastructure. Materials prepared Coastal morphology, coastal ero- National Park Service, Anchorage, for CCAAG meeting, May 16, 2008. sion, and barrier islands of the AK, by ABR, Inc., Fairbanks, AK. http://www.akclimatechange.us/ Beaufort Sea, Alaska. USGS Open 129 p. ewebeditpro/items/O97F17336.pdf. File Report 78-1063. 54 p. Jorgenson MT, Roth JE, Emers M, Kuhry P, Nicholson BJ, Gignac LD, McGraw M. 2008. The degradation Schlentner S, Swanson DK, Pull- Vitt DH, Bayley SE. 1993. Devel- of ice wedges in the Colville River man E, Mitchell J, Stickney AA. opment of Sphagnum-dominated Delta and their role in pond drain- 2003. An ecological land survey for peatlands in boreal continental age. In Kane DL, Hinkel KM, eds. the northeast planning area of the Canada. Can. J. Bot. 71:10-22. Ninth International Conference on National Petroleum Reserve–Alas- Permafrost (p. 1161-1166). Institute ka, 2002: Final Report prepared for Leitch DR, Carrie J, Lean D, of Northern Engineering, Univer- ConocoPhillips, Anchorage, AK, by Macdonald RW, Stern GA, Wang sity of Alaska Fairbanks (2Vols). ABR, Inc., Fairbanks, AK, 128 p. F. 2007. The delivery of mercury to the Beaufort Sea of the Arctic Morack JL, Rogers JC. 1981. Seis- Jorgenson MT, Shur Y. 2007. Ocean by the Mackenzie River. Sci. mic evidence of shallow permafrost Evolution of lakes and basins Total Environ. 373: 178-195. beneath islands in the Beaufort in northern Alaska and discus- Sea, Alaska. Arctic 34: 169-174. sion of the thaw lake cycle. J. Lesack LFW, Marsh P. 2007. Geophys. Res 112. FO2S17, Lengthening plus shortening of Morehead MD, Syviitski JP, Hutton doi:10.1029/2006JF000531. river-to-lake connection times in EW, Peckham SD. 2003. Modeling the Mackenzie River Delta re- the temporal variability in the flux Jorgenson, MT, Shur YL, Pull- spectively via two global change of sediment from ungauged river man ER. 2006. Abrupt increase in mechanisms along the arctic coast. basins. Glob. Planet. Change 39: permafrost degradation in Arctic Geophys. Res. Lett. 34: L23404, 95-110. Alaska. Geophys. Res. Lett. 33: doi:10.1029/200GL031656. L02503, doi:10.1029/2005GL024960. Plug LJ, Walls C, Scott BM. 2008. Liljedahl AK, Hinzman LD, Busey Tundra lake changes from 1978 Jorgenson MT, Shur Y, Walker RC, Yoshikawa K. 2007. Physical to 2001 on the Tuktoyaktuk Pen- HJ. 1998. Evolution of a perma- short-term changes after a tussock insula, western Canadian Arctic. frost dominated landscape on the tundra fire,Seward Peninsula, Alas- Geophys. Res. Lett: 35, L03502, Colville River Delta, northern ka. J Geophys. Res, 112, F02S07, doi:10.1029/2007GL032303. Alaska. In Lewkowicz, AG, Al- doi:10.1029/2006JF000554. lard M, eds., Proceedings of the Pomeroy JW, Bewley DS, Essery Seventh International Permafrost Liljedahl AK, Hinzman LD, Tweedie RLH, Hedstrom NR. Link T, Conference: Sainte-Foy, Quebec, CE , and Busey RC. 2009. An Granger RJ, Sicart JE, Ellis CR, Universite Laval, Collection Nordi- empirical attempt of watershed Janowicz JR . 2006. Shrub tundra cana, no. 57, p. 523–530. mapping in micro-scale resolution, snowmelt. Hydrol. Process. 20: POSTER, Lessons from Continuity 923-941. Jorgenson MT, Walker HJ, Brown and Change in the Fourth Inter- J, Hinkel K, Shur Y, Osterkamp national Polar Year Symposium, Racine C, Jandt R. 2008. The 2007 T, Ping C-L, Kanevskiy M, Eisner March 4-7, Fairbanks, Alaska. “Anaktuvuk River” tundra fire on W, Rea C, Jensen A. 2009. Coastal the Arctic Slope of Alaska: a new Region of Northern Alaska: guide- Liston G, Mcfadden J, Sturm M, phenomenon? In Kane DL, Hinkel book to permafrost and related Pielke R. 2002. Modelled changes KM, eds., Ninth International Con- features. Alaska Division of Geo- in arctic tundra snow, energy and ference on Permafrost, Extended physical and Geological Surveys; moisture fluxes due to increased Abstracts (p. 247-248). Institute of Guidebook 10. 156 p. shrubs. Global Change Biol. 8:17- Northern Engineering, University 32. of Alaska Fairbanks. Kelly EN, Schindler DW, St. Louis VL, Donald DB, Vladicka KE. Lloyd AH, Rupp TS, Fastie CL, Racine C, Jandt R, Meyers C, Den- 2006. Forest fire increases mercury Starfield AM. Patterns and dy- nis J. 2004. Tundra fire and vegeta- accumulation by fishes via food namics of treeline advance on tion change along a hillslope on the web restructuring and increased the Seward Peninsula, Alaska. Seward Peninsula, Alaska, USA. mercury inputs. Proceedings of the J. Geophys. Res. 108 D2, Arct, Antarct, Alp. Res. 36:1-10. National Academy of Sciences 103: doi:10.1029/2001JD000852, 2003. 19380–19385. Reimnitz E, Maurer DK. 1979. Ef- Marchand FL, Nijs I, Heuer M, fects of storm surges on the Beau- Kelly MO, Davies PJ, Woolhouse Mertens S , Kockelbergh F, Pon- fort Sea coast, northern Alaska. HW. 1988. The control of whole tailler J, Impens I, Beyens L. 2004. Arctic 32: 329-344. plant senescence. Crit. Rev. Plant Climate warming postpones senes- Sci. 7: 139-173. cence in high arctic tundra. Arct. Antarct. Alp. Res. 36: 390-394. Klinger LF. 1990. Global patterns in community succession: 1. Bryo- Mars JC, Houseknecht DW. 2007. phytes and forest decline. Mem- Quantitative remote sensing study oirs of the Torrey Botanical Club indicates doubling of coastal ero- 24:1-50. sion rate in past 50 yr along a segment of the Arctic coast of Alaska. 57 Geology 35: 583-586. Richter-Menge J, Overland J, Smol JP, Douglas MSV. 2007. Cross- Turetsky MR, Harden JW, Friedli Svoboda M, Box J, Loonen MJJE, ing the final ecological threshold HR, Flannigan M, Payne N, Proshutinsky A, Romanovsky V, in high Arctic ponds. Proc. Natl. Crock J, Radke L. 2006. Wildlifes Russell D, Sawatzk CD, Simpkins Acad. Sci. USA 104: 12395–97. threaten mercury stocks in north- M, Armstrong R, Ashik I, Bai L.-S, ern soils. Geophys. Res. Lett. 33: Bromwich D, Cappelen J, Carmack Spatt PD, Miller MC. 1981. Growth L16043.1-L16043. E, Comiso J, Ebbinge B, Frolov I, conditions and vitality of Sphag- Gascard JC, Itoh M, Jia GJ, Krish- num in a tundra community along Verbyla D. 2008. The greening and field R, McLaughlin F, Meier W, the Alaska Pipeline Haul Road. browning of Alaska based on 1982- Mikkelsen N, Morison J, Mote T, Arctic 34:48-54. 2003 satellite data. Global Ecol. Nghiem S, Perovich D, Polyakov I, Biogeogr. 17: 547-555. Reist J, Rudels B, Schauer U, Shi- Steele M, Ermold W, Zhang J. 2008. klomanov A, Shimada K, Sokolov V, Arctic Ocean surface warming Wahren C-HA, Walker MD, Bret- Steele M, Timmermans M-L, Toole trends over the past 100 years. Harte MS. (2005) Vegetation J, Veenhuis B, Walker D, Walsh Geophys. Res. Lett., 35, L02614, responses in Alaskan arctic tundra J, Wang M, Weidick A, Zckler C. doi:10.1029/2007GL031651. after eight years of a summer 2008. Arctic Report Card 2008, warming and winter snow manipu- NOAA, Silver Springs, Md. (Avail- Stow DA, Hope A, McGuire D, lation experiment. Global Change able at http://www.arctic.noaa.gov/ Verbyla D, Gamon J, Huemmrich Biology 11: 537-552. reportcard). F, Houston S, Racine C, Sturm M, Tape K, Hinzman L, Yoshikawa K, Walker DA, Bockheim JG, Chapin Riordan B, Verbyla D, McGuire Tweedie C, Noyle B, Silapsawan FS, Eugster W, Nelson FE, Ping AD. 2006. Shrinking ponds in C, Douglas D, Griffith B, Jia G, CL. 2001. Calcium-rich tundra, subarctic Alaska based on 1950- Epstein H, Walker D, Daeschner wildlife, and the “mammoth 2002 remotely sensed images. S, Petersen A, Zhou L, Myneni R. steppe.” Quaternary Sci. Rev. J. Geophys. Res. 111, G04002, 2004. Remote sensing of vegetation 20:149163. doi:10.1029/2005JG000150. and land-cover change in arctic tundra ecosystems. Remote Sens. Walker MD, Wahren CH, Hollister Rodrigues J. 2008. The increase in Environ. 89: 281–308. RD, Henry GHR, Ahlquist LE, the length of the ice-free season as Alatalo JM, Bret-Harte MS, Calef further indication of the rapid de- Sturm M, McFadden JP, Liston GE, MP, Callaghan TV, Carroll AB, cline of the Arctic sea ice. Abstract Chapin FS, Racine CH, Holmgren Epstein HE, Jonsdottir IS, Klein #C51A-0542, American Geophysi- J. 2001. Shrub-snow interactions JA, Magnusson B, Molau U, Ober- cal Union, Fall Meeting. in arctic tundra: a hypothesis with bauer SF, Rewa SP, Robinson CH, climatic implications. J. Climate 14: Shaver GR, Suding KN, Thompson Rodrigues J. Unpublished. The in- 336-344. CC, Tolvanen A, Totland O, Turner crease in the length of the ice-free PL, Tweedie CE, Webber PJ, season in the Arctic. Sturm M, Racine C, Tape K. 2001. Wookey PA. 2006. Plant community Increasing shrub abundance in the responses to experimental warm- Rupp TS, Chapin FS, Starfield AM. Arctic. Nature 411: 546–547. ing across the tundra biome. Proc. 2000a. Response of subarctic veg- Natl. Acad. Sci. USA 103: 1342- etation to transient climatic change Syvitski, JPM. 2002. Sediment 1346. on the Seward Peninsula in north- transport in Arctic rivers; implica- west Alaska. Global Change Biol. 6: tions for a warmer future. Polar Wiedinmeyer C, Friedli H. 2007. 541–555. Res. 21: 323-330. Mercury emission estimates from fires: An initial inventory for the Rupp TS, Starfield AM, Chapin FS. Szumigalski AR, Bayley SE. 1997. United States. Environmental Sci- 2000b. A frame-based spatially Net aboveground primary pro- ence and Technology 41: 8092-8098. explicit model of subarctic vegeta- duction along a peatland gradient tion response to climatic change: in central Alberta in relation to Yoshikawa K, Bolton WR, Ro- comparison with a point model. environmental factors. Ecoscience manovsky VE, Fukuda M, Landscape Ecol. 15: 383–400. 4:385-393. Hinzman LD. 2002. Impacts of wildfire on the permafrost in the Shur YL, Jorgenson MT. 1998. Tape K, Sturm M, Racine C. 2006. borealforests of Interior Alaska, Cryostructure development on the The evidence for shrub expansion J. Geophys. Res., 107, 8148, floodplain of the Colville River Del- in northern Alaska and the pan- doi:10.1029/2001JD000438. [printed ta, northern Alaska. In Lewkowicz, Arctic. Global Change Biol. 12: 108(D1), 2003]. AG, Allard M, eds., Proceedings of 686-702. the Seventh International Perma- Yoshikawa K, Hinzman L. 2003. frost Conference: Sainte-Foy, Que- Thormann,MN, Bayley SE. 1997. Shrinking thermokarst ponds and bec, Universite Laval, Collection Aboveground net primary produc- groundwater dynamics in discon- Nordicana, no. 57, p. 993-1000. tion along a bog-fen-marsh gradi- tinuous permafrost near Council, ent in southern boreal Alberta, Alaska. Permafrost Periglac. Pro- 58 Smith LC, Sheng Y, MacDonald Canada. Ecoscience 4:374-384. cess 14: 151-160. GM, Hinzman L. 2005. Science 308: 1429. Caribou (Rangifer tarandus) foraging near Toolik Field Station; the animals’ ability to access vegetation under snow is key to their overwinter survival. Photo by Øivind Tien.

59 Chapter 5 Climate Effects on Fish and Wildlife

In the previous chapter, important physical and biological agents that will alter arctic habitats in response to climate change were described. In this chapter, we explore likely pathways by which fish and wildlife populations will be affected.

60 The WildREACH workshop partici- is likely to result in novel species pants worked in breakout sessions assemblages accompanied by altered Introduction (see Appendices 3–5) to consider competitive interactions, predator- the mechanisms by which climate- prey relationships, and parasite-host related changes in habitat would relationships. Predicting the nature affect fish and wildlife. Workshop of these changes was considered too participants acknowledged the im- speculative at this stage. mense uncertainty associated with climate projections and the manner Hydrologic Process Models in which physical and ecological Hydrologic processes are found to systems will respond to an altered be a critical link between climate thermal regime. Against this back- change and habitat suitability. Issues ground, participants were cautious of water balance, precipitation, and to avoid poorly supported predic- surface water distribution were of tions regarding fish and wildlife paramount concern for all species response. For each taxonomic group groups. Two simple conceptual mod- of wildlife (i.e., birds, fish, and els are presented (summer [Figure mammals) graphic and narrative 5.1] and winter [Figure 5.2]), outlin- conceptual models were developed ing pathways by which changes in that represent hypotheses linking air temperature and precipitation climate change, habitat change, and regime will affect aspects of surface biological effects. It is important hydrology most relevant to fish to recognize that these models are and wildlife habitat. These models intended as hypotheses to be tested provide the foundation for the taxon- through observation and experi- specific models that follow later in ment, rather than evidence-based this section. Summer and winter conclusions. seasons are considered separately since winter constitutes most of The models are not intended as an the year in the Arctic, yet the short exhaustive catalog of all possible summer growing season is critical to effects; rather they represent a reproduction and body condition for selection based on our judgment of most species. both plausibility and magnitude of effect. We focus attention on changes Summer Processes associated with physical processes 1. Evapotranspiration is the pri- and vegetation change. Trophic mary means by which surface and interactions are considered at the soil water loss occurs and thus is community and system level, but an a key factor in controlling arctic assessment of other multi-species hydrology (Kane 1997, Bowling et interactions is largely deferred. We al. 2003). There is evidence that acknowledge that climate change portions of the Coastal Plain are

Summer: Temperature, precipitation, and surface hydrology

! PRECIPITATION ! TEMPERATURE 1 2 5

! Evapotranspiration ! Active layer depth ! Thermokarst

3 4 !"soil moisture and ! water temperature !"glacial input surface storage

Local redistribution of water and new New drainage networks and drying ! Lake area drainage networks (Arctic Coastal interfluves (Arctic Foothills) 61 Plain); lake drainage in some regions

Figure 5.1. Conceptual model illustrating the effects of increased summer air temperature and precipitation on aspects of surface hydrology relevant to habitat change. Gray text boxes identify climate drivers, blue boxes identify physical processes, and white boxes represent habitat effects. already experiencing increasingly temperature. There is a paucity drier conditions as precipitation of long-term temperature records fails to compensate for evapotrans- for arctic lakes and streams, but piration (Hinzman et al. 2005). what data are available suggest While increasing temperatures will a warming trend. Toolik Lake, promote evapotranspiration, it is located within the Foothills, shows important to recognize that the ac- a 3°C warming of the epilimnion tual rate of evapotranspiration can (upper layer of lake) from 1975 to be either water- or energy-limited. 1992 (O’Brien et al. 1997). Warming If projected increases in summer of water in the Arctic is not limited precipitation do not compensate to lakes, as large rivers, such as the for increased evapotranspiration, Lena River in Siberia, also show a then summer drying will occur warming trend (Yang et al. 2002). earlier and to a greater degree Milner et al. (1997) reported that than currently. Fall evapotranspira- daily maximum temperatures in tion tends to be limited by available a tundra stream reached as high energy, as well as by the onset of as 21°C, although the maximum plant senescence. Soil satura- monthly temperature was 13.5°C. tion and surface water storage at If these data are representative, it the onset of winter is sensitive to suggests that water temperature in changes in rainfall, which tends to arctic Alaska lakes and streams is occur in late summer and fall. highly sensitive to climate change.

2. Warmer soil temperatures and 4. As the rate of ice loss continues to earlier onset of active layer thaw increase for Brooks Range glaciers will increase the depth of seasonal (Nolan et al. 2006), these glaciers thaw. As active layer thickness are expected to disappear in the increases, the soil storage capacity next 100–200 years (Weller et al. will increase and water may perco- 2007). Initially the magnitude and late downward, resulting in drying temporal extent of glacial runoff at the surface. Streamflow records will increase, leading to increased from gauges in the Canadian sediment delivery to deltaic mud Northwest Territories and portions flats and the ocean and greater of the Yukon River show increases connectivity to floodplain lakes. in baseflow during winter, which When ice loss reaches the point at has been attributed to perma- which meltwater outflow decreases frost degradation (St. Jacques or ceases, glacial-fed systems will and Sauchyn 2009, Walvoord and show reduced flow rates, lower Striegl 2007). In the Foothills, a turbidity, and altered water quality. longer warm season may prolong These changes will be particularly the time available for soil moisture important in the eastern portion to move down gradient and into of the North Slope where glaciers drainage networks. This may lead exert strong controls over flow re- to drying of shallow soils in late gime, habitat suitability, and forma- summer and early fall, while base- tion of coastal deltas. flow increases due to additional shallow groundwater flow. Fall and 5. Recent observations in arctic winter drainage may be less impor- Alaska suggest that permafrost is tant on the Coastal Plain because warming (Romanovsky and Oster- soil moisture distribution is largely kamp 1997, Osterkamp and Jor- determined by microtopography. genson 2006, Clow 2008) and that Deepening of the active layer and warming temperatures can acceler- subsequent redistribution of water, ate thermokarst processes at mean however, should reduce surface annual ground temperatures well storage (e.g., lower lake levels) and below 0°C (Jorgenson et al. 2006). produce drier near-surface soils. Permafrost temperatures are likely to continue to increase under the 3. A warming climate combined with expected scenario of increasing increased snow cover is expected air temperature and precipitation. to have a significant impact on the This in turn will likely acceler- annual heat budget of arctic lakes ate the formation of thermokarst (for review see Schindler and Smol features. Types of thermokarst 62 2006). Increased snow cover will features and ecological implications insulate lakes and result in thinner of permafrost degradation will vary ice. Thinner lake ice will melt faster depending on location, landscape in spring, leading to earlier ice-out position, soils, hydrology, and and earlier seasonal rise in water amounts and types of ice (Jorgen- son and Osterkamp 2005). In areas Taras et al. (2002) estimated that with little topographic relief, such increasing snow depth by 15 cm as portions of the Coastal Plain, ice would increase temperatures at wedge degradation can produce the soil/snow interface by 0.5–3°C. water-filled pits and troughs that It follows that warmer winter air drain water from flooded polygon temperatures in combination with centers (Jorgenson et al. 2006). increased snow depth will likely Where topographic gradients are produce warmer soil and subnivean large enough to allow for rapid (air layer between the snow and drainage, suprapermafrost ground- the soil surface) temperatures. water flow can deepen the active layer and produce features known 2. Increased snowpack depth will as water tracks (Jorgenson et al. likely produce higher snow water 2008). Similarly, thermokarst gul- equivalent (SWE) values. lies can form when surface water becomes channelized, deepens the 3. Control of timing and duration of active layer, and thaws ice-rich snowmelt is complex. In general, permafrost (Jorgenson et al. 2008). snowmelt is advanced in warm, Thermokarst lakes, common across wet springs and delayed under cool the western portion of the Coastal and dry conditions (Zhang et al. Plain, will also respond to a warm- 2001). Therefore, under a scenario ing climate. Lakes can undergo of warming air temperatures it is lateral expansion by the process possible that spring melt will occur of thermal and mechanical erosion earlier in the year. Snow cover (Jorgenson and Shur 2007, Jor- records from Barrow suggest that genson et al. 2008). As permafrost the snowmelt date is becoming in- warms and begins to degrade, lake creasing variable and has advanced area will likely increase as shore- by 10 days since 1941 (Hinzman et lines become more susceptible to al. 2005), suggesting that a shift to erosion. an earlier onset of spring is already occurring. Winter Processes 1. Snow acts as an insulator between 4. On the North Slope, mid-winter relatively warm soil and cold air freeze-thaw cycles and rain-on- throughout the winter. Tempera- snow events can occur but are rare ture records from the Kuparuk (Kane et al. 2000). Warmer temper- River basin show that snow cover atures during winter could result in attenuates average winter air tem- more frequent thaw-freeze cycles perature by 34–47% and diurnal and rain-on-snow events. If these air temperatures by 40–96% (Taras icing events become increasingly et al. 2002). Where snow is deep, common, the snowpack will become soil temperatures are warmer, and, denser and grain size will increase. in extreme cases, deep snowdrifts Such a change would likely have can prevent soils from freezing profound consequences for a num- (Sturm et al. 2005). Using a model ber of mammals. developed for the Kuparuk basin,

Winter: Temperature and precipitation

↑ PRECIPITATION ↑ TEMPERATURE

↑ End- of- winter soil and 1 subnivian temperatures

↑ Rain on snow events 4

63 ↑ End-of-winter SWE Advance in first day 4 ↑ Thaw– freeze cycles Changes in snow 2 3 above freezing 4 structure Figure 5.2. Conceptual model illustrating the effects of increased winter air temperature and precipitation on aspects of surface hydrology relevant to habitat change. Gray text boxes identify climate drivers and blue boxes identify physical processes. Over 200 species of birds have been used preferentially by many brood- Birds observed on the North Slope, but rearing and post-breeding waterfowl many are of casual occurrence. Over and shorebirds, particularly brant 90 species (Table 5.1) regularly use and snow geese (Johnson 2000, terrestrial and freshwater habitats Sedinger and Stickney 2000). River- for nesting, brood rearing, and fall delta mudflats provide important staging. Nearly all are migratory, pre-migration habitat for shorebirds occupying the region for some por- (Connors et al. 1979, Andres 1994, tion of the summer season (May to Alaska Shorebird Group 2008). Lake September). A few species, including basins breached by coastal erosion rock and willow ptarmigan, common may provide especially favorable raven, snowy owl, and American staging habitat for post-breeding dipper (Johnson and Herter 1989), shorebirds (Audrey Taylor, Univer- are known to occur on a year-round sity of Alaska, unpublished data). basis. Several waterfowl species perform The avifauna of arctic Alaska is a post-breeding molt migration to dominated numerically by waterfowl specific North Slope sites at which and shorebirds that are dependent they complete wing molt, remaining on a variety of wetland habitat flightless for several weeks. Notably, types distributed broadly across the tens of thousands of geese of four Coastal Plain. All of the waterfowl species converge on the large lake (21 species), loons (4 species), grebes area north of Teshekpuk Lake each (1 species), and gulls/terns (4 spe- year from breeding areas in north- cies) that regularly occur in arctic ern and western Alaska, arctic Rus- Alaska are dependent on open-water sia, and Canada (King and Hodges habitats for foraging, brood rearing, 1979, Derksen et al. 1979). Similarly, and/or molting (Bergman et al. 1977, large numbers of long-tailed ducks USFWS 1993). Shorebirds (25 spe- use nearshore waters of Beaufort cies) vary in their breeding habitat Sea coastal lagoons for molting preference, with some species using (Johnson and Richardson 1981, moist upland tundra and others de- Johnson 1985). pendent on aquatic or semi-aquatic habitats (Cotter and Andres 2000, Diets vary among species groups, Troy 2000, Brown et al. 2007). While but many arctic birds rely heav- wetland-dependent species are most ily on invertebrate prey. Of the 94 prevalent on the Coastal Plain, the species listed in Table 5.1, 37 spe- diversity of perching birds is great- cies (39%) are entirely dependent est in the Brooks Range and Foot- on invertebrate prey, and 72 species hills, where shrub-associated species (77%) rely at least to some degree on with taiga affinities reach the north- invertebrates. Insects are a particu- ern limits of their ranges (Johnson larly important component of the and Herter 1989, Poole 2005). diet for juveniles of widely divergent taxonomic groups, ranging from Many species of waterfowl and dabbling ducks to sparrows. Her- shorebirds use coastal habitats bivorous species are few but include (mainland beaches, river deltas, geese and ptarmigans, which are protected lagoons and their shore- of special management significance lines, and barrier islands) for brood because of their importance to sub- rearing, molt, and staging prior to sistence and sport hunters. Piscivory fall migration (Connors et al. 1979, is also uncommon, with loons and Connors et al. 1981, Johnson and terns the most prominent examples. Richardson 1982, Connors 1984, Twelve species rely significantly on Gill et al. 1985, Johnson et al. 1993, bird and mammal prey, including Fischer and Larned 2003). Main- hawks, falcons, eagles, owls, jaegers, land and barrier-island beaches are and gulls. used by high densities of staging shorebirds (e.g., red and red-necked Predation is a major cause of nest phalaropes), likely attracted by failure among arctic-breeding water- marine invertebrates concentrated fowl and shorebirds (Day 1998, Troy by wind-driven water currents (Con- 2000, Sovada et al. 2001, Meltofte et nors et al. 1979). Barrier islands are al. 2007), likely influencing produc- 64 used by nesting common eiders to tivity and population sizes for some the virtual exclusion of other habi- species, particularly colonial nesters. tats (Johnson and Herter 1989, Dau Arctic foxes are the most important and Larned 2007). Coastal wet sedge predator of bird eggs and young (also known as arctic salt marsh) is (Day 1998, Burgess 2000, Troy 2000, Bety et al. 2001), and they also take microtine rodents (lemmings and adults of some species. Other im- voles), and predation on tundra- portant predators include glaucous nesting birds appears to be inversely gulls, jaegers, common ravens, and related to the abundance of mamma- brown bears (Day 1998, Troy 2000, lian prey (Day 1998, Bety et al. 2001, Bety et al. 2002, NRC 2003). Foxes, 2002, ACIA 2005: 301, Meltofte et al. gulls, and jaegers rely heavily on 2007).

Table 5.1. Bird species occurring regularly in terrestrial Habitat Associations2 and freshwater habitats of arctic Alaska (Johnson and Herter 1989) classifed by diet (Poole 2005) and habitat Wetlands Shrubs Coastal association (The Nature Conservancy [TNC] and ABR, Inc., unpublished data). ater all 1 hrubS Common Name Scientific Name Diet Lake Emergent Marsh Sedge Wet Moist Sedge- T Low Barrens Sedge Wet W Waterfowl Tundra Swan Cygnus columbianus H 3 3 3 3 2 Greater White-fronted Anser albifrons H 3 3 3 2 2 Goose Snow Goose Chen caerulescens H 2 2 2 Brant Branta bernicla H 3 3 3 3 3 2 Canada Goose Branta canadensis H 2 2 2 3 American Wigeon Anas americana H, I 2 2 2 2 Mallard Anas platyrhynchos H, I 2 2 Northern Shoveler Anas clypeata H, I 2 2 2 2 Northern Pintail Anas acuta H, I 2 3 3 3 2 2 Green-winged Teal Anas crecca H, I 2 2 2 2 Greater Scaup Aythya marila H, I 2 2 2 2 Lesser Scaup Aythya affinis I 2 Steller’s Eider Polysticta stelleri I 2 2 3 Spectacled Eider Somateria fischeri H, I 3 3 3 2 2 King Eider Somateria spectabilis H, I 2 3 3 2 2 2 Common Eider Somateria mollisima I 2 3 2 Harlequin Duck Histrionicus histrionicus I 2 2 Surf Scoter Melanitta perspicillata I 2 White-winged Scoter Melanitta fusca I 2 3 Long-tailed Duck Clangula hyemalis H, I 3 3 3 2 2 2 3 Red-breasted Merganser3 Mergus serrator I, P Grouse Willow Ptarmigan Lagopus lagopus H 3 3 2 2 Rock Ptarmigan Lagopus muta H 3 3 2 Loons Red-throated Loon Gavia stellata P 3 3 3 2 2 Pacific Loon Gavia pacifica P, I 3 3 3 2 2 2 Common Loon Gavia immer P 2 Yellow-billed Loon Gavia adamsii P 3 2 2 Grebes Red-necked Grebe3 Podiceps grisegena P, I Continued next page 65 Table 5.1. Continued Habitat Associations2 Wetlands Shrubs Coastal tEmergen ater 1 all Common Name Scientific Name Diet Lake Marsh Sedge Wet Moist Sedge- Shrub T Low Barrens Sedge Wet W Hawks, Eagles, Falcons Northern Harrier3 Circus cyaneus C Rough-legged Hawk3 Buteo lagopus C Golden Eagle3 Aquila chrysaetos C Merlin4 Falco columbarius C Gyrfalcon3 Falco rusticolus C Peregrine Falcon3 Falco peregrinus C Cranes Sandhill Crane3 Grus canadensis H, I, C Shorebirds Black-bellied Plover Pluvialis squatarola I 3 3 2 American Golden-Plover Pluvialsi dominica I 3 3 2 Semipalmated Plover Charadrius semipalmatus I 2 Spotted Sandpiper4 Actitis macularius I Wandering Tattler3 Tringa incana I Upland Sandpiper3 Bartramia longicauda I Whimbrel Numenius phaeopus I, F 2 Bar-tailed Godwit Limosa lapponica I, F 3 2 3 Ruddy Turnstone Arenaria interpres I 2 2 2 3 2 Surfbird3 Aphriza virgata I Red Knot Calidris canutus I, H 2 2 3 Sanderling Calidris alba I 2 3 Semipalmated Sandpiper Calidris pusilla I 3 3 3 2 Western Sandpiper Calidris mauri I 3 3 Least Sandpiper Calidris minutilla I 2 2 2 2 White-rumped Sandpiper Calidris fuscicollis I 3 3 2 Baird’s Sandpiper Calidris bairdii I 2 3 2 3 Pectoral Sandpiper Calidris melanotos I 2 3 3 2 2 Dunlin Calidris alpina I 3 3 2 2 Stilt Sandpiper Calidris himantopus I 3 3 2 Buff-breasted Sandpiper Tryngites subruficollis I 2 3 3 2 Long-billed Dowitcher Limnodromus scolopaceus I 2 3 3 2 2 Wilson’s Snipe Gallinago delicata H, I 2 3 2 2

1. Diet categories: H=herbivore (shoots, leaves), F=herbivore (fruit), S=herbivore (seeds), I=invertebrates, C=carnivore (mammals/birds), P=piscivore, A=anthropogenic. 2. Habitat associations: 2=medium, 3=high, based on review of habitat use by 89 species (TNC/ ABR). Only habitat use indices of 2 and 3 are reported, regardless of data quality. Alpine, Dwarf Shrub, Forested, Bluff and Riverine Water habitats are not presented here because few species exhibited medium or high use of these habitats. Some TNC/ABR habitats are combined here: Emergent Marsh includes both Lacustrine Marsh and Riverine Marsh categories presented in TNC/ABR; Wet Sedge includes Lowland Wet Sedge Tundra and Riverine Wet Sedge Tundra categories; Moist Sedge-Shrub includes Upland Moist Sedge-Shrub Tundra, Lowland Moist 66 Sedge-Shrub Tundra, and Riverine Moist Sedge-Shrub Tundra categories; Tall Shrubs includes Upland Tall Alder Shrub and Riverine Tall Alder-Willow Shrub categories; and Low Shrubs includes Upland Low Birch-Willow Shrub Tundra, Upland Shrubby Tussock Tundra, Lowland Low Birch-Willow Shrub, and Riverine Low Willow-Shrub Tundra categories. 3. Species reviewed in TNC/ABR, but none have index ≥ 2 for habitats presented in this table. 4. Species not reviewed in TNC/ABR. Table 5.1. Continued Habitat Associations2 Wetlands Shrubs Coastal ater 1 Common Name Scientific Name Diet Lake Emergent Marsh Sedge Wet Moist Sedge- Shrub Tall Low Barrens Sedge Wet W Shorebirds, continued Red-necked Phalarope Phalaropus lobatus I 3 2 3 3 2 Red Phalarope Phalaropus fulicarius I 2 3 3 2 Gulls and Terns Glaucous Gull Larus hyperboreus C, P, A 3 2 2 2 3 3 Mew Gull4 Larus canus C, P, A, I Sabine’s Gull Xema sabini H, I, P 3 3 3 2 3 3 2 Arctic Tern Sterna paradisaea I, P 2 2 2 3 3 2 Jaegers Pomarine Jaeger Stercorarius pomarinus C 2 2 Parasitic Jaeger Stercorarius parasiticus C 3 2 Long-tailed Jaeger Stercorarius longicaudus C, I 3 3 Owls Snowy Owl3 Bubo scandiacus C Short-eared Owl Asio flammeus C 2 Perching Birds Say’s Phoebe4 Sayornis saya I Horned Lark4 Eremophila alpestris I, S Cliff Swallow4 Petrochelidon pyrrhonota I Common Raven3 Corvus corax H, C, P, A Gray-headed Chickadee4 Poecile cincta I, S American Dipper3 Cinclus mexicanus I, P Arctic Warbler4 Phylloscopus borealis I Bluethroat Luscinia svecica I 2 3 Northern Wheatear3 Oenanthe oenanthe H, I Gray-cheeked Thrush Catharus minimus I, F 3 3 American Robin4 Turdus migratorius I, F Eastern Yellow Wagtail Motacilla tschutschensis I 2 3 American Pipit3 Anthus rubescens I Northern Shrike Lanius excubitor I, C 3 3 Yellow Warbler4 Dendroica petechia I Wilson’s Warbler4 Wilsonia pusilla I American Tree Sparrow Spizella arborea I 3 3 Savannah Sparrow Passerculus sandwichensis I, S 3 Fox Sparrow Passerella iliaca I 3 3 White-crowned Sparrow Zonotrichia leucophrys I, S 3 2 Lapland Longspur Calcarius lapponicus I, S 3 3 2 2 Smith’s Longspur Calcarius pictus I, S 3 2 3 Snow Bunting Plectrophenax nivalis I, S 3 3 Common Redpoll Carduelis flammea I, S 2 3 Hoary Redpoll Carduelis hornemanni I, S 2 3 67 Effects of Climate Change on Birds of dependent species across the Arctic Arctic Alaska Coastal Plain. Shrub expansion Many bird species characteristic of on the North Slope, if it results in high latitudes have broad circum- displacement of open moist and polar distributions. Within Alaska, wet sedge tundra habitats, would the breeding range of some tundra- similarly reduce habitat quality and associated species extends along the availability. Chukchi Sea coast and as far south as the Yukon-Kuskokwim Delta (e.g., Increased temperatures and longer tundra swan, common eider, black- open-water and growing seasons are bellied plover, bar-tailed godwit, and likely to increase primary and sec- western sandpiper). Other species ondary productivity in aquatic and are widely distributed across Alaska, semi-aquatic systems, thus increas- including alpine areas and wetlands ing food abundance for invertebrate- in the boreal forest zone (e.g., Pacific dependent species and herbivores. loon, long-tailed duck, American Changes in seasonal patterns of food golden plover, arctic tern, and availability and quality, however, Lapland longspur). Considering only could be detrimental if birds can- the Alaska portion of their range, not adjust migration, breeding, and populations of several species are molting schedules to optimize exploi- concentrated in the Arctic. These in- tation of food resources. This could clude: yellow-billed loon, snow goose, result in a “trophic mismatch,” in king eider, spectacled eider, Steller’s which the seasonal timing of life cy- eider, red phalarope, stilt sand- cle events is out of phase with that of piper, ruddy turnstone, red knot, critical food resources (Coppack and white-rumped sandpiper, pectoral Both 2002). At present, the timing of sandpiper, buff-breasted sandpiper, breeding activities for many species long-billed dowitcher, glaucous gull, of arctic birds appears closely linked pomarine jaeger, snowy owl, and to peak insect emergence (Hurd and Smith’s longspur (Poole 2005). While Pitelka 1954, Holmes 1966, MacLean arctic Alaska may serve as a refu- 1980). If birds are unable to respond gium for species with more southerly to climate-mediated changes in these distributions, these arctic specialists peak events, breeding success and are likely the most vulnerable to population size will likely be affected climate warming. (Both et al. 2006, Moller et al. 2008, Tulp and Schekkerman 2008). Simi- The potential for warmer summers larly, changes in plant phenology and and delayed freeze-up would likely increased plant growth may reduce improve reproductive success for the availability of high-quality forage some bird species. For example, for geese. This could reduce gosling there is evidence that shorebird growth and survival (Dickey et al. chick growth and survival is con- 2008) and restrict nutrient uptake strained by cold weather conditions among molting geese (Schmutz et al. (Soloviev et al. 2006), thus a warm- unpublished). ing climate could increase productivity in these species. A longer open- The presence or absence of fish in water season should also improve arctic lakes heavily influences the in- fledging success for species like vertebrate community (Stross et al. red-throated loons, for which early 1980) and, therefore, prey availabil- freezing temperatures are a sig- ity for aquatic birds. Changes in flow nificant source of juvenile mortality regimes that prevent fish from en- (Dickson 1993). tering lakes would be detrimental to piscivores, but reduced competition If warmer summers result in dry- for invertebrate prey will likely ben- ing of wetlands, however, species efit other bird species. Additionally, that rely on shallow water and wet in northern Alaska, species diversity meadow habitats could be pro- of soil invertebrates increases sub- foundly affected. A long-term drying stantially along a climatic gradient trend would likely lead to changes in away from the colder coastal zone vegetation community composition (MacLean 1975), suggesting that and productivity of invertebrates, longer summer seasons will result affecting herbivorous species as in range shifts and changes in the 68 well as those dependent on arthro- composition of the soil invertebrate pods. Widespread paludification fauna. The potential effects of such (expansion of bogs and peat-forming changes on birds are unclear and environments, see page 53) could would depend on their ability to reduce habitat quality for wetland- exploit these new resources. Changes in coastal habitats will Finally, there is anecdotal evidence affect availability of bird habitat. that red foxes may be increas- Coastal erosion, accompanied by ing in number on the North Slope lake-breaching and salinization and documentation of competition of adjacent low-lying areas, may between the arctic fox and the larger result in changes in vegetation that red fox, with the latter dominating influence habitat suitability differ- (Pamperin et al. 2006). The effect entially for birds (Flint et al. 2008). on breeding birds of increasing red Increased stream sediment loads fox numbers and their potential may result from increasing drainage exclusion of arctic foxes is unclear. basin surface temperatures (Syvitski Red foxes would likely occur at 2002) and thermokarst-associated lower densities during the breeding bank erosion (Walsh et al. 2005); the season, but they are more capable of persistence of deltaic mud-flat habi- killing or chasing off large adult wa- tat is dependent on the balance of terfowl that often succeed at defend- deposition rate and inundation from ing nests and deterring predation sea level rise. attempts by the smaller arctic fox.

Climate warming may affect the relationship between arctic birds and their predators in several ways. Pop- ulation cycles of lemmings and other small rodents affect productivity of waterfowl and shorebirds because these bird species (especially at egg and chick stages) serve as alternate prey when rodent populations are low (Bety et al. 2001, 2002, ACIA 2005: 301). The inverse relationship between lemming abundance and nest success is less apparent on the North Slope than the Eurasian Arctic (Day 1998). Nevertheless, climate-related changes in rodent population cycles (Ims and Fuglei 2005, Kausrud et al. 2008) could indirectly influence bird productivity and abundance.

Warming may lead to further increases in human economic activity (e.g., increased shipping and offshore oil development), and this activity may further influence abundance or distribution of predators. Arctic foxes, glaucous gulls, common ravens, and brown bears tend to concentrate in areas of human activity on the North Slope (Day 1998, NRC 2003). Increased development could increase localized densities of predators, although population- level effects on birds are unknown and could likely be at least partially mitigated (Liebezeit et al. in press). A red fox (Vulpes vulpes) carrying a spectacled eider (Somateria Further increases in human eco- fischeri) near Prudhoe Bay, Alaska. Climate warming may allow red fox nomic activity may also result in populations to expand in arctic Alaska. Photo by Bryan Collver, BryLyn increased industrial contamination, Collver Art. including oil spills.

69 Conceptual Models: Potential Climate- mediated Impacts to Arctic Birds Four conceptual models were developed describing potential climate- mediated habitat changes on North Slope birds. Narratives describing elements of each model are presented below.

Abundance and Distribution of Surface Water (Figure 5.3) 1. Aquatic and semi-aquatic habitats of the North Slope support large numbers of birds, both herbivores and invertebrate-dependent species (Bergman et al. 1977, Derksen et al. 1981, Martin and Moitoret 1981, Meehan and Jennings 1988, USFWS 1993, TERA 1994). Persistently dry summer conditions could result in conversion of aquatic marsh and wet sedge meadow to moist sedge meadow, and moist sedge meadow to tussock tundra. Widespread loss of aquatic and semi-aquatic habitat across the Arctic Coastal Plain would likely decrease invertebrate productivity and availability. This alteration would represent a loss of habitat quantity and quality, leading to changes in bird distributions as well as decreases in productivity and abundance, with population- level effects possible for some species.

70 Above: A male Steller’s eider (Polysticta stelleri) near Barrow; the species is listed as threatened under the Endangered Species Act. Photo by Ted Swem, USFWS. Below: Long-billed dowitchers (Limnodromus scolopaceus) at Prudhoe Bay, Alaska, flocking in August prior to migration. Photo by Bryan Collver, BryLyn Collver Art. 2. Permafrost degradation in polygo- with stands of water sedge Carex nal terrain will continue to drive aquatilus and emergent pendant local redistribution of surface wa- grass Arctophila fulva. Emergent ter from shallow polygon centers Carex and Arctophila stands have to deeper polygon troughs. Loss high primary productivity (Alex- of productivity in drying polygon ander et al. 1980) and support high centers would likely be offset by densities of aquatic invertebrates increased primary and secondary (Bergman et al. 1977). Arctophila productivity of expanding thermo- wetlands are preferred habitat karst pits, including higher biomass for Pacific and red-throated loons, of aquatic insects (MacLean 1980, tundra swans, black brant, north- Martin 1983). Habitat heterogene- ern pintails, long-tailed ducks, ity characteristic of thermokarst white-winged scoters, king eiders, terrain has been linked to higher and both red and red-necked phala- local-scale nest density for species ropes (Derksen et al. 1981, Troy such as American golden plover, 1988), and are also important to semipalmated sandpiper, and red- threatened spectacled and Steller’s necked phalarope (Troy 2000) and eiders (Quakenbush et al. 1995, higher general use by some water- USFWS 1996). fowl such as greater white-fronted goose (Troy 1991). It is unclear, 4. In the Foothills ecoregion, devel- however, whether landscape-wide opment of new drainage networks acceleration of thermokarst would and drying of intervening ridges be a net benefit to these or other may result in transition from species. tussock tundra to dwarf birch vegetation. The effects on birds of 3. The formation of new drainage such a transition are uncertain and networks could cause increased perhaps minor. lake drainage in some landscapes, resulting in a loss of open-water 5. Large, deep lakes unaffected by habitats. Depending on the extent new drainage networks are expect- Figure 5.3. Conceptual model of to which this occurs, it could affect ed to continue expanding due to how changes in abundance and the distribution, productivity, and thermokarst and shoreline erosion distribution of surface water could abundance of species such as loons, by wind-driven waves. This will impact birds. Blue text boxes terns, and some diving ducks. Lake likely affect only a small portion identify physical drivers, green drainage can also result in the of the Coastal Plain landscape and boxes identify habitat effects, and development of productive drained will have only minor effects on bird white boxes summarize biological lake-basin complex wetlands distribution and abundance. implications of habitat change.

Abundance and distribution of surface water: BIRDS

Δ soil moisture and Local redistribution of water New drainage networks, ↑ Lake area surface storage and new drainage networks drying inferfluves (Arctic (Arctic Coastal Plain). Foothills) 1 2 3 4 5 Minor effect

↑ Drying of saturated soils and ↑ Drying of polygon ↑ Water in thermokarst Lake drainage ↓ Water table, drying shallow wetlands centers pits and troughs of intervening ridges

↓ Invertebrate productivity ↑ 1 ° and 2 ° (invertebrate) ↓ Large open - ↑ Drained -lake Minor effect and availability productivity in thermokarst water bodies basin -complex pits wetlands

71 ↓ Productivity and abundance; ↑ Productivity and abundance; ↓ Productivity and abundance; ↑ Productivity and abundance; ∆ in distribution of some ∆ in distribution of some ∆ in distribution of loons, terns ∆ in distribution of phalaropes, shorebirds (e.g. red phalarope, shorebirds (e.g. red-necked and some diving ducks. some shorebirds and some pectoral sandpiper) and phalarope), geese and dabbling waterfowl (e.g. geese, dabbling waterfowl. ducks. ducks) Box 5.1 Avian Population Response to Ecological Change Along the Arctic Coastal Plain

The Teshekpuk Lake Special Area (TLSA), northeast of Teshekpuk Lake in the National Petroleum Reserve–Alaska, represents one of the most important goose molting habitats in the circumpolar arctic (Flint et al. 2008). To understand how variability in the physical environment may manifest biological change, the U.S. Geological Survey (USGS) investigated the extent to which goose distribution has changed over time, whether such change was a consequence of habitat change, and what physical and biological processes were responsible. Managers need this understanding to better predict future distributions of molting geese, thereby improving land management decisions.

The USGS analyzed survey data collected by the U.S. Fish and Wildlife Service since 1976 and discovered that goose distribution has changed. In particular, black brant shifted from inland lakes to coastal habitats while greater white-fronted geese concurrently increased in numbers. The change in distribution could either have been related to habitat change or Alaska interspecific competition could have forced brant to use suboptimal habitats. Examination of a time-series of aerial imagery confirmed that high rates of coastal erosion combined with periodic storm surges have led to saltwater intrusion into freshwater habitats, and the associated changes in salinity may have altered foraging conditions for geese in the coastal zone (Jones et al. 2009).

The results from multiple lines of inquiry suggested that the change in distribution has not been detrimental to brant. There was no difference in body condition between birds molting in traditional inland habitats compared to those molting along the coast. Moreover, body condition appears to have improved in recent years compared with birds measured in the 1980s. Analyzing forage plant species across a range of habitats within the TLSA revealed that salt- tolerant plant species, most commonly found along the coast, tended to have higher nitrogen content than species sampled along inland lakes, an indicator of higher nutritional quality. Interestingly, in the warmer year (2006) plant Above: Location of the Teshekpuk productivity was higher in terms of net biomass, but nutrient quality Lake Special Area (orange outline) was lower in terms of percent nitrogen and the carbon-to-nitrogen and area used by concentrations of ratio. molting geese (green outline). Map by USFWS from Bureau of Land These results demonstrate the difficulty in predicting how climate Management data. change may impact habitat and wildlife. On the surface, a change in Below: Brant (Branta bernicla) are distribution from historical molting grounds seems troubling. How- exclusively high arctic breeders ever, changes in goose distribution at TLSA that are correlated with that depend on particular coastal climate change suggest that geese have shifted habitats in response vegetation communities to feed to improving conditions along the coast. This redistribution of birds young goslings. Following breeding, from inland lakes likely reduced interspecific competition, resulting in 72 brant congregate at specific areas overall improvements in body condition across all habitats. We antici- such as Teshekpuk Lake where pate further shifts in wildlife distributions and changes in plant forage they molt in large flocks before fall species productivity and nutrient composition as warming patterns migration. Photo by Stephanie continue to alter the landscape. Clemens, USFWS. Vegetation Community Changes 2. The extent to which paludification (Figure 5.4) may impact bird populations and 1. Increased shrub abundance could distribution depends on the rate at have variable effects, depending on which this process may occur and the landscape setting. Increased the degree to which diminished abundance of tall shrubs in riparian primary productivity influences the or hillslope settings would permit availability of invertebrate prey. range expansion and increased Acidification may reduce diversity abundance of a variety of songbirds and biomass of aquatic inverte- with affinities for tall shrub. At brates (Hendrey et al. 1976). present, tall shrub habitat is ex- ceedingly scarce (<1% of the land- 3. Only a few species of birds are scape), and thus an increase would herbivorous during the summer have the net effect of increasing season. For herbivorous geese, regional biodiversity. Conversion of forage quality rather than quantity tussock tundra to shrubby tussock is thought to be limiting, and for- tundra in the Foothills and Coastal age quality affects gosling growth Plain, particularly in the form of rates and subsequent survival increased stature and density of (Lindholm et al. 1994, Sedinger et dwarf birch (Betula spp.), would al. 1995). Preliminary studies near have uncertain consequences for Teshekpuk Lake (Schmutz et al. bird distribution and abundance. unpublished) suggest that warmer Broad-scale expansion of shrub temperatures during the growing communities at the expense of wet season result in greater plant bio- sedge tundra and moist sedge- mass but diminished forage quality. shrub tundra on the Coastal Plain There is evidence suggesting that would reduce habitat availabil- reduced gosling size in years of ity for a broad suite of wetland- higher spring temperatures may be adapted species (Table 5.1). The attributed to a mismatch between threshold at which shorebird and hatching dates of goslings and waterfowl production would be re- timing of the peak forage quality duced, however, and the likelihood (Dickey et al. 2008). of reaching that threshold are not well understood.

Vegetation community changes: BIRDS

Increased shrub abundance ↑ Paludification Changes in plant phenology; ↑ 1° productivity and plant biomass

1 2 3

↑ Habitat for ↓ Habitat for ↓ Habitat for ↓ 1° and 2° ↑ Forage quantity ↓ Forage quality Timing mismatch between emergence of shrub -associated water -dependent most species (invertebrate) high -quality forage and species. species productivity in herbivore life history shallow wetlands Minor effect

↓ Egg and young production,

↓ Productivity and abundance, ∆ in fledging success, fledgling survival; distribution of phalaropes, some Body condition during molt, post - shorebirds and some waterfowl ↓ ↑ Productivity and abundance; molt and pre -migration among ∆ in distribution of some herbivores passerines, ptarmigan.

Figure 5.4. Conceptual model of how vegetation community changes could impact birds. Blue text boxes identify physical drivers, green boxes identify habitat effects, and white boxes summarize biological implications of habitat change. 73 Invertebrate Community Changes decrease in availability of benthic (Figure 5.5) invertebrate larvae could affect 1. The timing of insect emergence brood-rearing waterfowl and pre- is closely related to the timing migration body condition of shore- of snowmelt, and might advance birds, perhaps resulting in reduced with warmer spring temperatures juvenile survival among both, with (Hye and Forchhammer 2008, subsequent declines in recruitment Tulp and Schekkerman 2008). If and, ultimately, population size. warming alters the timing and patterns of insect emergence and peak 2. Other things equal, warmer abundance, and birds are unable temperatures and longer growing to compensate for these changes, seasons should result in increased “trophic mismatches” may result. productivity of soil and aquatic If birds cannot alter their mi- invertebrates. Increases in pro- gration and breeding schedules duction associated with increased accordingly, nest success, fledgling temperatures, permafrost degrada- survival, and pre-migration body tion, and longer open-water and condition will likely decline. growing seasons may initially lead to increased primary and second- Beyond short-term phenologi- ary productivity in both pelagic and cal response to changing climate, benthic habitats. If birds are able seasonality may have profound to exploit these increases in prey long-term consequences for the abundance, the result should be im- distribution and abundance of proved habitat quality that would arctic arthropods. Multi-year life translate into increased productiv- cycles occur in many arctic inver- ity and abundance, and perhaps tebrates (MacLean 1980, Chernov higher densities and expanded 1985) and the prevalence of multi- distributions among shorebirds and year life cycles results in a large invertebrate-dependent waterfowl standing biomass of larval inver- and passerines. Increased nutrient tebrates available to predators input into lakes, however, may lead Figure 5.5. Conceptual model of how throughout the summer season. A to eutrophication which would de- invertebrate community changes shift to shorter, even annual, life crease light penetration and could may impact birds. Blue text boxes cycles could substantially influ- ultimately lead to a decline in ben- identify physical drivers, green ence the larval biomass available to thic primary production (Vadebon- boxes identify habitat effects, and birds during portions of the breed- coeur et al. 2003). This decline may white boxes summarize biological ing season, although the relation- translate into decreased benthic implications of habitat change. ship is complex (MacLean 1980). A invertebrate biomass/production

Invertebrate community changes: BIRDS

Local redistribution of water and new Δ soil moisture and Growing season and ↑ drainage networks (Arctic Coastal Plain surface storage warmer water temperatures and Arctic Foothills).

1 3 ↑ Water in thermokarst ↑ Drying of saturated soils, pits/ troughs shallow wetlands and polygon centers 2

Trophic mismatch for invertebrate - ↑ Invertebrate productivity dependent species

↓ Multi -year invertebrate life cycles ↑ Habitat some invertebrate - ↓ Invertebrate productivity and dependent birds availability – meadows and polygons

↓ Biomass, end -of -season ↓ Habitat for some invertebrate - benthic invertebrate larva dependent species

↓ Fledging success, fledgling ↓ Body condition during molt, post- ↑ Productivity and abundance: ∆ ↓ Productivity and abundance: survival; ↓ Body condition molt and pre-migration among in distribution of phalaropes, ∆ in distribution of phalaropes, some shorebirds and some during molt, post-molt and pre- invertebrate-dependent species some shorebirds and some waterfowl migration waterfowl and/or a change in benthic inverte- Loss or major reduction of barrier brate community composition that islands would likely cause changes may negatively impact both birds in nearshore temperature and and fish. salinity regimes, perhaps reducing invertebrate abundance and, there- 3. If warming results in broad- fore, the quality of foraging habitat scale drying of moist and shallow for species such as long-tailed wet habitats, a decline in soil and ducks. Loss of barrier islands aquatic invertebrates would be would also reduce shoreline forag- expected across the North Slope. ing habitat for shorebirds such as This would reduce habitat quality phalaropes because their plankton- and availability for all shorebirds ic prey frequently becomes concen- and most waterfowl, with associ- trated along island water lines. ated decreases in productivity and abundance as well as changes in 2. The increase in coastal erosion distributions for many species. projected over the next century is not expected to result in a sig- Coastal Processes and Habitats nificant loss of terrestrial habitats (Figure 5.6) along much of the coast; however, 1. A warming Arctic may cause the potential loss of 8% or more a number of changes in barrier of the goose molting area north of islands of the Beaufort and Chuk- Teshekpuk Lake could have signifi- chi seas. Potential changes include cant consequences for the tens of reduced size, increased rates of mi- thousands of geese that molt there gration and erosion, and a greater annually. likelihood of overwashing and breaching by storm surges. Most 3. Climate-induced changes in of the 2,000–3,000 common eiders coastal processes could poten- that breed along the Beaufort and tially affect other coastal habitats, Chukchi sea coasts (Johnson and including coastal wet sedge tundra, Herter 1989, Dau and Larned 2007) salt-killed tundra, and delta mud nest on coastal barrier islands, flats. Coastal wet sedge and deltaic and loss of this specialized habitat mud flats currently are sparsely would likely lead to substantial distributed but important habitats, reduction in this population. More and any loss would be considered frequent or severe overwash events detrimental to waterfowl and would reduce nest success through shorebirds. Salt-killed tundra is direct egg mortality, as well as of low value as nesting habitat diminished nesting habitat suit- because of limited plant cover so an ability, due to removal of sheltering increase of this type at the expense driftwood, detritus, and vegetation of more productive habitat would clumps (Dau and Larned 2007). be detrimental. The total area affected, however, is expected to be relatively minor on a regional scale.

Coastal processes and habitats: BIRDS

Opposing influences create uncertainty regarding direction of ↓ Barrier islands and protected ↓ Terrestrial habitats change for mudflats, coastal wet sedge tundra, and salt-killed lagoons along coastline tundra

Δ Saline mud flats and Δ Salt- killed ∆ Delta mudflats 1 2 3 coastal wet sedge tundra tundra

↓ Nesting habitat for common eiders. ↓ Foraging, nesting, brood- Δ Habitat quality and quantity Δ Minor effect ∆ Shorebird pre- rearing and staging habitats for brood-rearing brant, post- migration staging ↓ Foraging and/or molting habitat for for some waterfowl and breeding waterfowl and habitat 75 waterfowl, loons, gulls, terns, and some shorebirds. shorebirds. shorebirds.

Figure 5.6. Conceptual model of how changes in coastal processes and habitats could impact birds. Blue text boxes identify physical drivers and white boxes summarize biological implications of habitat changes. Aquatic Habitats of the North Slope of Coastal Plain lakes are abundant, Fish Alaska occupying approximately 14% of the Aquatic habitats on the North Slope Coastal Plain ecoregion (Jorgenson are characterized by low average and Heiner 2003). The area they temperatures, low prey densities, encompass is considered to be the short open water periods each year, second largest lake district in Alaska and limited overwintering areas. (Arp and Jones 2008). Many Coastal Many fish populations are reliant on Plain lakes are derived from ther- seasonal movements among habitats mokarst processes (Hobbie 1973) (Craig 1989). North Slope streams and are relatively shallow (<9 m), can be categorized by ecoregion, including Teshekpuk Lake, the larg- size, and origin (Craig and McCart est lake on the North Slope (NRC 1975, Craig 1989). Mountain streams 2003). Smaller and shallower (<2 m) originate in the Brooks Range and lakes typically freeze to the bottom Arctic Foothills, are prevalent east and only contain liquid water during of the Colville River, and are char- the warm season. Coastal Plain lakes acterized by glacial and clear water are dependent on surface runoff for systems of variable size, with flows recharge and are subject to substan- derived mostly from springs and tial evaporative loss during sum- surface runoff (Craig and McCart mer (Miller et al. 1980). Although 1975). These streams are typically landlocked lakes are present on the cooler and have higher discharges Coastal Plain, many lake systems than other waterbodies in the region are seasonally linked by shallow and typically freeze to the bottom streams, forming complex networks in winter, with only spring areas of connected habitats. and deep pools containing liquid water in the winter (Craig 1989). As Marine and estuarine environments the largest mountain stream and on the North Slope are no less the largest drainage on the North extreme than freshwater habitats. Slope, the Colville River intercepts Many areas of the North Slope coast smaller mountain streams across the are characterized by barrier islands western Brooks Range. West and that form shallow lagoon systems north of the Colville, most streams and by large river deltas, both of originate on the Coastal Plain. These which provide important fish habitat Coastal Plain streams are typically (NRC 2003). During summer and smaller, warmer, and slower-moving, depending on prevailing winds, and many do not contain liquid topography, and nearshore currents, water in winter (Craig 1989); the freshwater flows from North Slope larger of these drainages, however, rivers mix with coastal waters to contain liquid water in deep pools produce a narrow nearshore band and springs throughout the year of relatively warm, brackish water (Morris et al. 2006). Many Coastal (Craig 1984). During winter when Plain streams meander through freshwater input is minimal, near- lake networks with their discharges shore water may become hyper- dependent on surface runoff. These saline due to concentrated salts from systems also include the smaller tun- ice formation, and the temperature dra streams that typically occur as of under-ice water can reach -2°C or tributaries across the Coastal Plain. colder (Griffiths et al. 1977). These cold waters are not suitable habitat Lakes across the North Slope can for anadromous or freshwater fish. also be categorized as mountain Tidal fluctuations are on the order of lakes or Coastal Plain lakes, based 20 cm and do not play a major role in on ecoregion, size, and origin mixing and exchange between brack- (Hobbie 1973). Mountain lakes are ish and marine waters (Griffiths et uncommon, occupying only 1.7% of al. 1977). the Foothills and 0.4% of the Brooks Range ecoregions (Jorgenson and Heiner 2003) and are typically of glacial origin, formed by moraines, scour, or deposited ice blocks (kettle lakes; Hobbie 1973). Many of these lakes are large and deep with 76 recharge associated with surface runoff and groundwater sources. Because of their depth, most of these lakes do not freeze to the bottom. Freshwater Resident and Anadromous waters. Alaska blackfish (Dallia pec- Fish of the North Slope of Alaska toralis), northern pike (Esox lucius), Species in the family Salmonidae slimy sculpin (Cottus cognatus), and are perhaps the most diverse group longnose sucker (Catostomus catos- of fishes that use North Slope tomus) are also found in freshwater freshwater habitats. They include habitats of the western North Slope. lake trout (Salvelinus namaycush) Habitat use for a subset of fish spe- and Arctic char (S. alpinus), which cies is summarized in Table 5.2. live exclusively in freshwater lake systems (Morrow 1980, Reist et al. 1997). Arctic grayling (Thymallus arcticus) and round whitefish (Pro- Table 5.2. Generalized life Habitat Use2 sopium cylindraceum) live in lakes history and seasonal habitat and rivers and are rarely encoun- use by common North Slope Open-water Season Ice-cover Season tered in coastal waters. Most Dolly freshwater and anadromous Streams Lakes Coastal Streams Lakes Varden (S. malma) populations fsh. Water are anadromous, while others are resident and live entirely in freshwa- Deep Deep

ter. Similarly, both anadromous and Small Small Large Large Life Ocean freshwater resident populations of Shallow History

least cisco (Coregonus sardinella) Near-shore Species Strategy1 occur on the North Slope (Seigle 2003, Moulton et al. 1997). Broad A X X X X X Least Cisco whitefish (C. nasus) and humpback R X X whitefish (C. pidschian) populations Arctic Cisco A X X X X are typically anadromous, although freshwater populations may exist in Round Whitefish R X X X X X X certain lakes or in upstream reaches Broad Whitefish A X X X X X of the Colville River (Craig 1989). Humpback Arctic cisco (C. autumnalis) encoun- A X X X X X X Whitefish tered on the North Slope of Alaska are entirely anadromous and return Lake Trout R X X to the Mackenzie River in northern Arctic Char R X X Canada to spawn (Fechhelm et al. A X X X X X X 2007). Chum salmon (Oncorhynchus Dolly Varden keta), pink salmon (O. gorbuscha), R X X X X and other Pacific salmon species Pink Salmon A X X X X occur in low numbers in nearshore coastal waters of the Beaufort Sea Chum Salmon A X X X X each summer (Stephenson 2005). Arctic Grayling R X X X X X X X Small numbers of spawning chum Ninespine A X X X X X X X X and pink salmon are regularly ob- Stickleback served in the Colville River and oc- R X X X X X X X X casionally observed in other streams as well. Aside from chum salmon in the Colville River, it is not clear 1. Life history strategies: A=Anadromous, R=Resident. Least cisco, Dolly whether Pacific salmon observed in Varden, and ninespine stickleback have populations in both categories. North Slope rivers are members of 2. Habitat Definitions: Large streams have sufficient flow to allow instream self-sustaining populations or strays (springs and deep areas) and estuarine (river delta) overwintering from populations in drainages of the habitat. Small streams have insufficient flow to develop estuarine Chukchi Sea or farther south. habitats, but some of these streams may provide instream overwintering habitat in the form of springs and deep pools. Deep lakes do not freeze to A number of non-salmonid fishes of the bottom allowing year-round use by fish. Shallow lakes do not provide several families have also adapted to overwintering habitat but may be used by fish during the open water aquatic habitats on the North Slope. season if there is access. Coastal water (near-shore) is marine water that Ninespine stickleback (Pungitius is somewhat warmer and of lower salinity than the ocean in summer, due pungitius) populations can be either to freshwater inflow during the open water period; these habitats take on resident or anadromous (Morrow fully marine characteristics of salinity and temperature during winter, 1980). They are found in freshwater which precludes their use by freshwater or anadromous species during and nearshore habitats across the that season. Coastal water (ocean) is fully marine habitat that does North Slope, and they play a critical not allow overwintering of any salmonid species because of low water role in the food webs of piscivorous temperature. birds (Poole 2005) and fish. Burbot 77 (Lota lota) are common in stream and lake habitats of the western North Slope (Morris 2003), but they are rarely captured in coastal Effects of Climate Change on Fish of Higher than optimum water tem- Arctic Alaska peratures could result in decreased The Intergovernmental Panel on Cli- biomass and yield, leading to altered mate Change (IPCC) and the Arctic rates and locations of colonization, Climate Impact Assessment (ACIA) extinction, competition, and produc- both identified the Arctic as an area tivity (Tonn 1990). Increasing tem- where climate change effects will be peratures will have a direct effect on observed most readily (IPCC 2007, available habitats, including popula- ACIA 2005). Furthermore, aquatic tions reliant on colder water below systems within the region are ex- the thermocline in lakes (Reist et al. pected to act as key indicators of the 2006a). Increased air temperature timing, rate, intensity, and effects of and duration of the ice-free season the change. Fish response to climate may promote thermal stratification change is difficult to quantify due to in lakes that currently remain un- the lack of basic biological informa- stratifed due to wind-mixing, short tion and the incomplete understand- ice-free season, and low air tem- ing of climate effects on ecological peratures. Over the past 40 years, processes of freshwater systems presence and strength of late sum- (ACIA 2005). There is little doubt, mer thermal stratification in sev- however, that a changing climate has eral large deep lakes in the Brooks the potential for widespread conse- Range has increased dramatically quences for the physical, chemical, (G. Burkart, National Park Service, and biological processes that are key personal communication). Juvenile components of fish habitat in arctic fish constrained to the warm epilim- Alaska (Reist et al. 2006a). Reduced nion (surface layer) may experience flow and increased air temperatures lower growth rates because greater that cause higher water tempera- food availability may not compen- tures may also have direct effects sate for higher metabolic rate, while on individual fitness and population adult fish seeking thermal refuge in resilience (Svenning and Gullestad the hypolimnion (deep layer) may 2002). experience anoxic conditions due to the increae in pelagic productivity in A warming climate is likely to the warmer surface waters. increase ecosystem productivity, resulting in increased biomass and Longer term changes may also lead yields of many species (Reist et al. to a disturbance of synchronized 2006b). The magnitude of change in environmental cues, such as pho- ecosystem productivity and biomass toperiod and water temperature, will depend on local conditions and which may drive major life history population tolerances. Freshwater actions, including gonadal matura- resident fish in lakes may show in- tion and fertilization success (Reist creased production in comparison to et al. 2006a). Changes in groundwa- those populations in flowing water. ter flows may affect the type and Increased productivity in nearshore amount of instream sediment and areas could boost returns of anadro- substrate, alter chemical composi- mous fish. However, increased pro- tion, and change the temperature ductivity in freshwater and estuarine of the water, leading to increased systems could be offset if water physiological stress for populations temperatures rise past optima. not adapted to these new conditions. Changes to the physical and chemical properties of water may reduce incubation success and the availability of overwintering habitat. In addition, groundwater can alter the timing, extent, and duration of ice cover, which may lead to changes in habitat structure. Changes in both groundwater and precipitation runoff may affect the flow regimes of rivers and streams and disrupt the migration patterns of freshwater and anadromous fish (Prowse et al. 78 2006). An increase in sea level and coastal erosion may also disrupt traditional migration patterns or make Dolly Varden char (Salvelinus malma malma) in arctic Alaska can be found current habitats unavailable (ACIA in both anadromous and resident populations. Photo by Fred DeCicco, 2005). Alaska Department of Fish and Game. Access to Spawning and and Peterson (1992) described a case Overwintering Habitats in the summer of 1990 where water Periods of low stream flow and low levels in the upper Kuparuk River water levels in lakes could have were low, many riffles were nearly significant effects on migratory pat- dry, and many of the deeper pools terns of many fish populations (Reist were isolated. A similar instance et al. 2006a, 2006b). Sufficient flows was noted in the upper John River during particular seasons are critical near Anaktuvuk Pass in the Brooks to fish movements and also reduce Range (G. Burkart, National Park stranding events (Power et al. 1999). Service, personal communication). Arctic grayling in several areas of The degree to which changes in wa- the stream became stranded, and ter balance impact fish will depend only after a heavy rain event in in part on the type of watershed September were water levels high they occupy and the time of year. enough to allow these fish to migrate Anadromous and many freshwater to overwintering areas. If similar resident fish populations on the circumstances become more com- North Slope are highly migratory mon across the North Slope, fish and are reliant on corridors with suf- may be delayed in their migrations, ficient discharges to complete their and they may not be able to access annual life cycle. While anadromous spawning or overwintering areas populations move between freshwa- prior to freeze-up, with direct mor- ter and marine environments (Craig tality likely. 1989), freshwater residents move among individual lakes and streams, Increased Water Temperatures as well as within lake and stream Periods of low stream flow and low systems (Morris 2003). During the water levels in lakes could also have spring freshet (water flow resulting significant effects on the tempera- from sudden rain or melting snow), ture regimes of the habitats required anadromous populations move from by many fish populations (Reist et al. overwintering areas in freshwater 2006a). Fish seek out habitats with to summer feeding areas in ma- water temperatures that fall within rine water. These populations then their thermal preference (Coutant leave marine waters in late summer 1987). These habitat areas can be and ascend rivers to spawn and dispersed, and the availability of overwinter. Freshwater residents water at the preferred temperatures that spawn in the spring are also limit seasonal and spatial fish distri- dependent on the freshet when they bution. As such, water temperatures move from overwintering areas to influence migration patterns (Sven- spawning areas. After spawning, ning and Gullestad 2002), can pose a they move to summer feeding areas, barrier to movement, and influence and in late summer and fall return the availability of pathways to access to overwintering areas. Freshwater preferred habitats (Coutant 1987). residents that spawn in the fall move Water temperature also affects directly from overwintering to feed- physiological processes, including ing areas in the spring, and in late growth, gonadal maturation, and summer and fall return to spawning egg incubation, as well as population and overwintering areas. Because productivity and survival (Reist et the freshet is driven by snowmelt, al. 2006a). Warmer waters may also sufficient flows during spring are increase the prevalence of diseases expected to be available for fish to and parasites (Reist et al. 2006b) and move to preferred habitats, regard- increase the uptake and toxicity of less of their life history strategy. contaminants (Schiedek et al. 2007).

Key impacts could occur in late If streams summer and early fall, however, on the North when warmer air temperatures and Slope become increased evapotranspiration reduce shallower, fish the amount of water stored at the may be unable to surface, causing a drying effect that access seasonal may limit access to spawning and habitats such as overwintering habitats. Even in overwintering those areas where stream baseflow areas. Photo 79 is likely to increase or be maintained by Larry (Kane 1997), hot, dry summers could Hinzman. result in at least a temporary loss of connectivity. For example, Deegan Conceptual Model: Effects of and Arctic char. In settings where Changed Surface Hydrology and fish move between deeper overwin- Water Temperature tering lakes and productive shallow We developed a conceptual model lakes for summer foraging, how- describing potential effects of ever, a loss of connectivity could changed surface hydrology and wa- result in reduced food availability ter temperature on North Slope fish or stranding. (Figure 5.7). Narratives describing elements of the model are presented A shift from surface storage to below. soil storage of water could lower water levels in coastal streams 1. As soil moisture and surface stor- and lakes and reduce connectivity age change, baseflow in mountain among these waterbodies. Species streams during summer and fall that rely on these connections to will likely increase and may be suf- access critical habitats, such as ficient to maintain the physical con- broad whitefish, least cisco, Arctic nections among required habitats grayling, and ninespine stickleback within a watershed; if increased (Moulton 2007), may be forced into precipitation is insufficient to com- less preferred habitats (Wrona et pensate for increased evapotrans- al. 2006). A lack of connectivity piration, however, loss of connectiv- between streams and their del- ity is probable. Moreover, surface tas may also affect the amount of flow may also be controlled by thaw overwintering habitat available and depth beneath the channel, and a the timing of annual migrations for deeper active layer may allow more resident species. For anadromous underground flow and drying at species, the loss of connectivity the surface. Water levels in moun- may restrict access to spawning tain lakes may decrease, but should and overwintering areas as fish remain sufficient in most cases to return from summer feeding in support populations of lake trout estuarine and marine waters.

Temperature and surface hydrology: FISH

Δ soil moisture and Δ water Δ glacial input Local redistribution New drainage ↑ Lake area surface storage temperature of water and new networks and drainage networks drying interfluves 1 (Arctic Coastal Plain) (Arctic Foothills) 2 3 4 5

Pits and troughs intercept Sequential lake drainage where ↓ surface flow ↑ baseflow for suprapermafrost water topographic gradients are great for streams in streams in the and reduce baseflow enough to allow for drainage the Arctic Arctic Foothills Coastal Plain (i.e., Mountain steams) Loss of connectivity; Loss of summer foraging drying of streams. habitat resulting in reduced impedes migration and ↑ growth rates and productivity. stranding.

Little impact to fish if When ice loss reaches the point where Little direct baseflow maintains input to streams decreases, this will impact to fish. waterbody connectivity change flow regimes and water quality (see text). of glacial streams.

Loss of connectivity ↓ availability of thermal refugia during summer (e.g., deeper between Coastal and thermoclines in lakes and reduced areas of cold water in streams and lakes streams) and ↑ strength and prevalence of thermoclines. impedes migration and ↑ stranding. Physiological impacts to fish including increased metabolism, change in productivity (short term ↑ followed by long-term ↓), and direct mortality, incidence of parasites/disease. 80

Figure 5.7. Conceptual model describing potential effects on changes in surface hydrology and water temperature on North Slope fish. Blue text boxes identify physical drivers, green boxes identify habitat effects, and white boxes summarize biological implications of habitat change. 2. Increased water temperatures in 4. As air temperatures increase, late summer may prevent fish from the distribution of surface water reaching spawning and overwinter- on the Coastal Plain is likely to ing areas after summer feeding. change due to both local redistri- The magnitude of temperature bution of water and formation of change and loss of thermal refugia new drainage networks (McGraw will differ among stream systems, 2008). Newly formed pits and with those systems fed by springs troughs can intercept subsurface or glaciers (see element 3) likely flows, decreasing baseflow into having cooler water that will al- streams and reducing connectivity low normal fish movements. Fish among streams and lakes. Where inhabiting mountain lakes, such as topographic gradients are large lake trout and Arctic char, may be enough, new drainage networks confined to deeper waters as water may form, resulting in the loss of temperatures increase (Reist et connectivity among previously used al. 2006c), and may be restricted corridors, potentially disrupting from moving into shallow areas migration. and summer feeding areas along the shoreline. Migratory fish that 5. Increased thermokarst along use Coastal Plain streams and lakeshores will cause soils to slump lakes may be the most affected by into the lakes (Prowse et al. 2006), increases in water temperatures. expanding lake surface area and Although migratory fish may be increasing the amount of shoreline able to find suitable temperatures habitat. Provided that lake expan- in deep portions of these waterbod- sion does not result in breaching ies, warmer temperatures at the of the lake margins, this will result surface may interfere with emigra- in a slight increase in fish habitat tion. Also, warmer temperatures in availability. streams may constrain movements and result in fish becoming stranded in isolated pockets of cooler water (Power 1999). Increased water temperatures could lead to higher metabolism and higher productivity, but will likely result in long term decreases in yield (Tonn 1990). Consistently high temperatures could also lead to direct mortality.

3. Migratory fish rely on sufficient late summer discharge to reach spawning and overwintering habitats. Glacial stream systems and the fish that use them are likely to be significantly impacted by a warming climate. As the rate of ice loss continues to increase for Brooks Range glaciers (No- lan et al. 2006), these glaciers are expected to disappear in the next 100–200 years (Weller et al. 2007). When ice loss reaches the point at which meltwater input decreases, sufficient discharge may not be available to allow Dolly Varden and other species that use glacial systems to complete their migrations from summer feeding areas to spawning and overwintering habitats. The decrease in amount of meltwater may also change turbidity and other water quality charac- 81 teristics to which these populations have adapted. Ublutuoch River during spring breakup (above) and mid-summer (below). Strong seasonal variation in flow regime is a characteristic of Coastal Plain streams that could be accentuated by climate change. Photo by Richard Kemnitz, Bureau of Land Management. Twenty-eight species of terrestrial using only one type of tundra and/ Terrestrial mammals live in arctic Alaska from or a very specific diet; others have the crest of the Brooks Range north broad niches, using a wide variety Mammals to the Beaufort Sea (Bee and Hall of habitats and/or foods. Species 1956, Wilson and Ruff 1999, Table vary in their geographic distribu- 5.3). Polar bears are usually clas- tions from those that are endemic sified as marine mammals, but, to northern Alaska, to those found because they use terrestrial habitats worldwide throughout the northern for birthing dens and scavenging hemisphere (Table 5.3). For many carcasses, they are included here. broadly distributed species, arctic Some arctic species are restricted Alaska includes the northern limit of to very narrow ecological niches, their geographic range. Table 5.3. Species, habitats, and geographic range of mammals found in arctic Alaska, north of the Brooks Range mountains. Common name Scientific name Habitats (niche breadth) Range Red-backed vole Myodes rutilus Forest + tundra (3) 6 Singing vole Microtus miurus Tundra dry shrub (2) 3 N Tundra vole Microtus oeconomus Tundra taiga moist graminods (3) 6 Collared lemming Dicrostonyx groenlandicus Arctic tundra shrub (1) 2 Brown lemming Lemmus trimucronatus Arctic+ subarctic graminoids (2) 4 Snowshoe hare1 Lepus americanus Forest + tundra (3) 5 N Alaskan hare1 Lepus othus Tundra (1) 1 N Porcupine Erethizon dorsatum Forest + tundra (3) 5 N Arctic ground squirrel Spermophilus parryii Forest, tundra, meadows (3) 3 N Alaska marmot Marmota broweri Tundra alpine rocks (1) 1 N Dall’s sheep Ovis dalli Arctic+subarctic alpine rocks (1) 3 N Caribou Rangifer tarandus Tundra + open woodlands (3) 6 Muskox Ovibos moschatus Tundra graminoid + shrub (2) 2 Moose Alces americanus Forest + tundra shrub (3) 6 Barren-ground shrew Sorex ugyunak Tundra graminoid+shrub (2) 2 Tundra shrew Sorex tundrensis Forest+tundra graminoid shrub (3) 5 Dusky shrew Sorex monticolus Forest+tundra wet areas (3) 5 N Alaska tiny shrew Sorex yukonicus Forest, riparian shrub (3) 2 Least weasel Mustela nivalis Forest, prairies, tundra (3) 6 Ermine Mustela erminea Forest, meadows, tundra (3) 6 Mink Neovison vison Streams, lakes, marshes (2) 5 N Arctic fox Vulpes lagopus Arctic tundra and coasts (2) 4 Red fox Vulpes vulpes Forests, deserts, tundra (3) 7 River otter Lontra canadensis Rivers + lakes: fish (2) 5 N Wolverine Gulo gulo Forests, taiga, tundra (3) 6 Wolf Canis lupus Forests, tundra: ungulates (2) 7 Grizzly bear Ursus arctos Forests, taiga, tundra, deserts (3) 6 Polar bear2 Ursus maritimus Arctic sea ice: ice seals (1) 4 Niche breadth (number of habitats/types of food): 1=few; 2=moderate; 3=many. Geographic range: 1=endemic, arctic/west Alaska; 2=arctic Alaska/Canada; 3=Alaska/ northern Canada; 4=circumpolar arctic; 5=Alaska through west or south U.S.; 6=circumpolar-boreal north to mid-latitudes; 7=most of northern hemisphere; N=northern extent of range. Species are grouped as in Tables 5.4, 5.5. Scientific names and ecological information from Wilson and Ruff (1999) and Wilson and Reeder (2005). Geographic distribution and other ecological information summarized from MacDonald and Cook (2008) 82 and the IUCN Red List 2008 (www.iucnredlist.org). 1. Both species have been recorded on the North Slope. The range of L. othus evidently has contracted from the North Slope in the past century; L. americanus is known to occur today along the Colville River. Despite some taxomonic uncertainty, L. othus appears to be a distinct species (Waltari and Cook 2005). 2. Polar bears use terrestrial habitats for natal dens and scavenging carcasses. Arctic mammals can be grouped Reproductive strategies play an by diet, body size, winter survival important role in the fitness of arctic strategies, and reproductive strate- mammals (Table 5.4). Most arctic gies (Table 5.4). Arctic mammals eat mammals are blind, naked, and plants (herbivores), other animals helpless (altricial) at birth and need (carnivores), or a mixed diet of shelter and parental care that can plants and animals (omnivores). Ter- last from a few weeks to ≥2 years. restrial mammals range in size from In contrast, some medium-sized her- the 10-g tundra shrew to the 600-kg bivores (e.g., hares and porcupines) polar bear. Arctic mammals can be and large ungulates (e.g., caribou, classified as small (<0.1 kg), medium Dall sheep, muskoxen, and moose) (0.5–10 kg), large (50–90 kg), or very produce more mature (precocial) large (>150 kg), based on the aver- offspring. All young mammals age mass of adult females (Wilson obtain their initial nutrition from and Ruff 1999). milk produced by their mothers, and

Table 5.4. Arctic mammals grouped by body size, diet, and life history strategies. Group characteristics Group Common name Body size and diet Winter strategy Reproductive strategy Red backed vole Altricial offspring Singing vole Small herbivores Subnivian (active Short gestation, lactation, H1 Tundra vole beneath snow) parental care <0.1 kg Store food Multiple medium to Collared lemming large litters Brown lemming Snowshoe hare Precocial offspring Medium-sized Short to long gestation, H2a herbivores Active lactation, parental care Alaskan hare 0.5–10 kg Single & multiple small to Porcupine large litters

Arctic ground squirrel Medium-sized Dormant in den Altricial offspring (hibernation) Short to medium gestation, H2b herbivores Fatten before denning short lactation, social groups Alaska marmot 0.5–10 kg Use body reserves Single small to medium litters Dall sheep Large/very large Active Precocial offspring H3 Caribou herbivores Adapted to cold & Long gestation, lactation, (ungulates) Muskox less food parental care 50 200 kg Use body reserves Single birth of 1–2 offspring Moose – Barren-ground shrew Altricial offspring Short gestation & lactation; Tundra shrew short to medium care Small carnivores Active and/or C1 Dusky shrew Single/multiple medium to subnivean <0.1 kg large litters Least weasel Delayed implantation in Ermine mustelids (weasels) Mink Arctic fox Red fox Medium/large Altricial offspring Active Short to medium gestation & C2 River otter carnivores Adapted to cold lactation; short to long care Wolverine 0.5–45 kg Single small to large litters Lynx Wolf

Grizzly bear Altricial offspring Very large Dormant in den Delayed implantation, medium C3 carnivores/omnivores Fatten before denning gestation, very long lactation Polar bear >200 kg Use body reserves & care Single small litters 83 Ecological information summarized from Wilson and Ruff (1999) and various authors from IUCN Red List 2008 (www.iucnredlist.org). lactation is a major energetic cost Even species that typically migrate, for reproductive females (Vaughan such as caribou, move only short dis- et al. 2000, Persson 2005). The length tances relative to most birds and are of gestation, lactation, and parental therefore exposed to severe winter care ranges from a few weeks to ≥2 conditions, requiring specific adapta- years (Wilson and Ruff 1999). Most tions to cold and resource shortages. arctic mammals reproduce once per year, and litter sizes vary from 1 to The previously identified mammal >20. Some small and medium-sized groups (Table 5.4) also vary with mammals (e.g., shrews, lemmings, respect to their habitat and physio- voles, snowshoe hares, and least logical requirements for survival and weasels) can have >2 litters per reproduction (Table 5.5). Identifica- year. In contrast, some very large tion of these essential elements is a herbivores and carnivores, e.g., mus- useful starting point for developing koxen and grizzly bears, have few hypotheses regarding species’ vul- offspring at intervals of 2 or more nerability to climate-related habitat years (Wilson and Ruff 1999). change. During the short growing season, mammals must have access Unlike most birds that are present to adequate food to replace body in the Arctic only during the brief reserves lost the previous winter summer season, most mammals re- and depleted during pregnancy and main year-round. Therefore, winter lactation. They must also accumu- survival strategies are critical (Table late body fat to survive the coming 5.4). Arctic mammals survive the winter. In winter, mammals must long winter by: conserve energy, access stored food 1. adapting to cold temperatures or fat reserves, and find appropriate and low food availability and living shelter. Females that are pregnant in the arctic landscape year-round and lactate during the winter season (surface active); require additional body reserves and 2. storing food and remaining active natal nests or dens to shelter help- in winter beneath insulating snow less offspring. cover (subnivean); or 3. putting on large fat reserves and spending the winter in dens in true hibernation or a torpid (dormant) state (Blix 2005).

Table 5.5. Seasonal requirements of arctic mammals. Group Mammals Winter season needs Growing season needs Access to stored food Abundant green forage Microtines (lemmings Hoar frost layer Food to store for winter H1 and voles) Snow for insulation Natal and post natal nests Natal nests Runways + escape cover Winter forage Abundant green forage H2a Hares and porcupines Shelter for offspring Shelter for offspring Escape cover Escape cover Abundant green forage Ground squirrels and Winter dens H2b Burrows for escape/sleep marmots Snow for insulation Large pre-winter body reserves Large ungulates (sheep, Adequate winter forage Abundant green forage H3 caribou, muskoxen, Escape terrain (sheep) Escape terrain (sheep, moose) moose) Soft shallow snow Pre-winter body reserves Invertebrate/microtine prey Shrews and small Invertebrate/microtine prey C1 Natal nests mustelids (weasels) Nests Snow for insulation Medium/large mustelids, Winter prey or carcasses Summer prey or carcasses C2 foxes, lynx, wolves Natal and/or permanent dens Dens/shelter Access to food Natal and/or winter dens C3 Bears Large pre-winter body Adequate snow for insulation 84 reserves Effects of Climate Change on Mammals Conceptual Models: Changes in Winter of Arctic Alaska and Growing-season Conditions Climate change will have different Two conceptual models were devel- effects on arctic mammals in differ- oped summarizing potential effects ent seasons. Winter is the dominant of changing conditions on arctic season in the Arctic, lasting for mammals during winter and during 8–9 months of the year. Although the short growing season (Figures temperatures below 0°C and snow 5.8, 5.9). We assumed a scenario of can occur any time of the year, these increasing temperatures and pre- conditions generally exist from cipitation, primarily in winter (Walsh September through May. As a result et. al 2008, IPCC 2007). Numbered of climate change, temperatures and elements within the model are de- precipitation are expected to in- scribed more fully below. crease, primarily in winter. Warmer winter temperatures will likely Winter affect amounts of precipitation, the 1. Warmer temperatures in winter density and hardness of snow, and will likely result in an increase in will likely increase the frequency of thaw-freeze cycles and the num- rain-on-snow or “icing” events. ber of rain-on-snow (icing) events. Denser, harder snow and the for- The growing season in the Arctic oc- mation of ice layers within the snow curs from early to mid-June through pack or at ground level will reduce early to mid-August. Although this is accessibility to forage and increase a relatively short period in the annu- energetic costs for ungulates and al cycle, the growing season is key to other mammals active throughout survival. Herbivores and omnivores the winter. A severe icing event in depend on the high digestibility and October 2003 resulted in the deaths nutritional value of green plants of thousands of muskoxen on to replace body reserves used up Banks Island (Grenfell and Put- during the long winter, to meet the konen 2008). Ice layers and chang- demands of pregnancy and lactation, es in snow structure may cause the and to fatten before the next long loss of nests and runways and could Figure 5.8. Conceptual model winter season. Several species, like decrease access to stored food used illustrating possible effects of muskoxen, do not breed until body by subnivean herbivores. The ef- changing conditions during the mass reaches a minimum threshold fect of ice layers on the transporta- winter season (September–May) on (White et. al 1997). tion of oxygen and carbon dioxide arctic mammals. Blue text boxes through snow is unknown, but this identify physical drivers, green may be important for denning and boxes identify habitat effects, and subnivean mammals. white boxes summarize biological implications of habitat change.

85 2. Deeper, harder snow will be 4. Warmer temperatures will result energetically costly to ungulates in a shorter snow season and and possibly some predators who reduced snow extent in late winter. must travel through snow or dig Earlier breakup and flooding may through snow to find food. Deeper affect subnivean mammals and snow could benefit some species, mammals in winter dens. Warmer however. For example, deeper snow temperatures will likely result in would provide additional insulation early emergence from winter dens for denning and subnivean mam- and the subsequent death of altri- mals and might create additional cial neonatal offspring like bear denning habitat for polar bears. cubs, canid (wolves, foxes) pups, The insulative effect of deeper and young rodents. snow will also increase winter soil temperatures, potentially enhanc- Growing Season ing microbial activity and nutrient 1. Improved nutrient availability, availability for plants (see Growing as a result of warmer winter soil Season below). temperatures, could increase overall primary productivity/for age 3. Warmer winter temperatures availability, but also may contribute and a shorter snow season would to a shift toward a more shrubby result in the early emergence of environment (Sturm et al. 2005), some plants. This would provide with divergent effects on mammal early green forage and reduced populations (see below). energetic costs for some ungulates and possibly other species, but this 2. With a longer, warmer growing shift could also disrupt the seasonal season, plant biomass will likely synchrony between herbivores and increase, providing more forage for the plants upon which they feed. large and small herbivores, includ- Emerging plant species have high- ing lemmings, voles, squirrels, and er forage quality than do plants at large ungulates. In experiments, later phenological stages (Jorgen- cool overcast summers were ac- Figure 5.9. Conceptual model son et al. 2002). Herbivores such companied by lower productivity illustrating possible effects of as caribou migrate to calving areas of grasses and forbs (Lenart et al. changing conditions during the and give birth when tussock sedges 2002, Rachlow and Bowyer 1998), growing season (June–August) are emerging. Earlier emergence but a long-term trend toward on arctic mammals. Blue text box may result in forage of lesser qual- warmer and drier conditions may identifies physical drivers, green ity being available during calving result in a shift to increased shrub boxes identify habitat effects, and and peak lactation when nutritional cover. Shrubby tundra and boreal white boxes summarize biological demands reach a yearly maximum forests are expanding northward implications of habitat change. (Griffith et al. 2002).

86 and up mountain slopes (Sturm et et al. 2006). Increases in biting al. 2001, Danby and Hik 2007). If insects like bot and warble flies will wet graminoid tundra is replaced likely cause stress and declines in by drier shrubbier tundra, singing body condition in caribou and other voles, collared lemmings, snowshoe species (Thomas and Kilaan 1990). hares, porcupines, barren ground If drier summer conditions prevail, shrews, and moose could benefit, soil invertebrate communities will and brown lemmings, tundra voles, change; shrews would likely benefit and tundra and dusky shrews may if there were an overall increase in not. The diversity of arctic-adapted invertebrate abundance. species will likely decline if some arctic habitats disappear, but overall diversity may increase as additional species shift their ranges northward.

3. Warmer temperatures and a longer growing season will likely increase the abundance of invertebrates, parasites, and disease organisms that could potentially negatively affect all arctic mammals. For example, life cycles of protostrongylid nematodes (Uming- makstrongylus pallikuukensis in muskoxen and Paralaphostrongy- lus odocoilei and P. stilesi in Dall’s sheep) shorten as temperatures warm and the frost-free season lengthens, increasing the incidence of disease (Kutz et al. 2004, Jenkins

Above: A group of muskoxen (Ovibos moschatus) on the North Slope. Although the species was extirpated from Alaska in the late 19th century, it was successfully reintroduced in the 20th century. Photo by Richard Flanders. Below: A Dall’s sheep ram (Ovis dalli) in the Brooks Range. Photo by Ken Whitten. Literature Cited Bowling LC, Kane DL, Gieck RE, Cotter P, Andres B. 2000. Nest den- Hinzman LD, Lettenmaier DP. sity of shorebirds inland from the ACIA. 2005. Arctic Climate Impact 2003. The role of surface storage Beaufort Sea coast, Alaska. Can. Assessment. New York: Cambridge in a low-gradient arctic water- Field. Nat. 114: 287-291. University Press. 1042 p. http:// shed. Water Resour. Res. 39: 1087. www.acia.uaf.edu. doi:10.1029/2002WR001466, 2003. Coutant CC. 1987. Thermal preference: when does an asset become Alaska Shorebird Group. 2008. Brown S, Bart J, Lanctot RB, John- a liability? Environ. Biol. Fish. 18: Alaska Shorebird Conservation son JA, Kendall S, Payer D, John- 161-172. Plan. Version II. Alaska Shorebird son J. 2007. Shorebird abundance Group, Anchorage, AK. Available and distribution on the coastal Craig PC. 1984. Fish use of coastal online at http://alaska.fws.gov/ plain of the Arctic National Wildlife waters of the Alaska Beaufort Sea: mbsp/mbm/shorebirds/plans.htm Refuge. Condor 109: 1-14. a review. Trans. Am. Fish. Soc. 113: 265-282. Alexander V, Stanley DW, Daley Burgess RM. 2000. Arctic Fox. RJ, McRoy CP. 1980. Primary In Truett JC, Johnson SR. The Craig PC. 1989. An introduction to producers. In Hobbie JE, ed, natural history of an Arctic oilfield: anadromous fishes in the Alaskan Limnology of Tundra Ponds (p 179- development and the biota (p 159- Arctic. In Norton D, ed. Biological 250). Stroudsburg, PA.: Dowden, 178). Academic Press. Papers of the University of Alaska, Hutchinson, and Ross, Inc, Research Advances on Anadro- Chernov YI. 1985. The Living mous Fish in Arctic Alaska and Andres BA. 1994. Coastal zone use Tundra. Cambridge: Cambridge Canada (p 27-54) . University of by postbreeding shorebirds in University Press. Alaska-Fairbanks, Fairbanks, AK. northern Alaska. J. Wildl. Manage. 58: 206-213. Clow GD. 2008. Continued per- Craig PC, McCart PJ. 1975. Classifi- mafrost warming in northwest cation of stream types in Beaufort Arp CD, Jones BM. 2008. Geogra- Alaska as detected by the DOI/ Sea drainages between Prudhoe phy of Alaska Lake Districts: Iden- GTN-P borehole array, Alaska. In Bay, Alaska, and the Mackenzie tification, Description, and Analysis Kane DL, Hinkel KM. eds., Ninth Delta, N.W.T., Canada. Arctic Al- of Lake-rich Regions of a Diverse International Conference on Per- pine Res. 7: 183-198. and Dynamic State. U.S. Geologi- mafrost Extended Abstracts (p 47). cal Survey Scientific Investigations Institute of Northern Engineering, Danby RK, Hik DS. 2007. Variability, Report 2008-5215, Anchorage, AK. University of Alaska Fairbanks. contingency and rapid change in recent subarctic alpine tree line Bee JW, Hall ER. 1956. Mammals Connors PG. 1984. Ecology of dynamics. J. Ecol. 95:352-363. of northern Alaska on the Arctic shorebirds in the Alaskan Beaufort Slope. Univ. Kansas Mus. Nat. littoral zone. In: Barnes R, Schell, Dau CP, Larned WM. 2007. Aerial Hist. Misc. Publ. No. 8, 309 pp. DM, Ramirez E, eds. The Alaskan population survey of common Beaufort Sea: Ecosystems and en- eiders and other waterbirds in near Bergman, RD, Howard RL, Abra- vironments (p. 403– 416). Academic shore waters and along barrier ham KF, Weller MW. 1977. Water Press. islands of the Arctic Coastal Plain birds and their wetland resources of Alaska, 22-24 June 2007. Unpubl. in relation to oil development at Connors PG, Connors CS, Smith Report, USFWS, Office of Migra- Storkersen Point, Alaska. U.S. Fish KG. 1981. Shorebird littoral zone tory Bird Management, Anchor- Wildl. Serv., Resour. Publ. 129. ecology of the Alaska Beaufort age, AK. Washington, D.C. Coast. Final Report of Principal Investigators, Outer Continental Day RH. 1998. Predator popula- Bety J, Gauthier G, Giroux J-F, Shelf Environment Assessment tions and predation intensity on Korpimaki E. 2001. Are goose nest Program, National Oceanic and tundra-nesting birds in relation to success and lemming cycles linked? Atmospheric Administration human development. ABR, Inc., Interplay between nest density and 23:295-396. Fairbanks, AK. iv + 106 pp. predators. Oikos 93: 388-400. Connors PG, Myers JP, Pitelka FA. Deegan LA, Peterson BJ. 1992. Bety J, Gauthier G, Korpimaki E, 1979. Seasonal habitat use by Arc- Whole-river fertilization stimulates Giroux J-F. 2002. Shared predators tic Alaskan shorebirds. Studies in fish production in an arctic tundra and indirect trophic interactions: Avian Biology 2: 100-111. river. Can. J. Fish. Aquat. Sci. 49: lemming cycles and arctic-nesting 1890-1901. geese. J. Anim. Ecol. 71: 88-98. Coppack T, Both C. 2002. Predicting life-cycle adaption of migratory Derksen DV, Rothe TC, Eldridge Blix AS. 2005. Arctic animals and birds to global climate change. In WD. 1981. Use of wetland habitats their adaptations to life on the Both C, Piersma T, eds. The avian by birds in the National Petroleum edge. Tapir Academic Press. 296 p. calendar: exploring biological Reserve-Alaska. U.S. Dept. of the hurdles in the annual cycle. Proc. Interior, Fish and Wildlife Service 88 Both C, Bouwhuis S, Lessells CM, 3rd Conf. European Orn. Union, Resource Publication 141. Wash- Visser ME. 2006. Climate change Grnigen, August 2001. Ardea 90: ington, D.C. and population declines in a long- 369-378. distance migratory bird. Nature 441: 81–83. Derksen DV, Weller MW, Eldridge Griffith B, Douglas DC, Walsh NW, Hye TT, Forchhammer MC. 2008. WD. 1979. Distributional ecology Young DD, McCabe TR, Russell Phenology of high-arctic arthro- of geese molting near Teshekpuk DE, White RG, Cameron RE, pods: effects of climate on spatial, Lake, National Petroleum Reserve- Whitten KR. 2002. The Porcupine seasonal, and inter-annual varia- Alaska. In Jarvis RL, Bartonek JC, caribou herd. In Douglas DC, tion. In Meltofte H, Christensen eds, Management and Biology of Reynolds PE, Rhode EB, eds. TR, Elberling B, Forchhammer Pacific Flyway Geese (p 189-207). Arctic Refuge coastal plain terres- MC, Rasch M, eds., Advances in Oregon State University Book- trial wildlife research summaries Ecological Research 40. High arctic stores, Corvallis, OR. (p 8-37). U.S. Geological Survey, ecosystem dynamics in a changing Biological Resources Division, Bio- climate: ten years of monitoring Dickey M-H, Gauthier G, Cadieux logical Science Report USGS/BRD/ and research at Zackenberg Re- MC. 2008. Climatic effects on the BSR-2002-0001. search Station, Northeast Green- breeding phenology and reproduc- land (p 300-322). London: Academic tive success of an arctic-nesting Griffiths W, DenBeste J, Craig P. Press. goose species. Glob. Change Biol. 1977. Fisheries investigations in 14: 1973-1985. a coastal region of the Beaufort Hurd PD, Pitelka FA. 1954. The role Sea (Kaktovik Lagoon, Alaska). of insects in the economy of certain Dickson DL. 1993. Breeding biology Arctic Gas Biological Report Series Arctic Alaskan birds. In Proceed- of red-throated loons in the Cana- Volume 40, Fisheries Investigations ings of the 3rd Alaskan Scientific dian Beaufort Sea region. Arctic along the North Slope from Prud- Conference: Science in Alaska 1952 46: 1–7. hoe Bay, Alaska to the Mackenzie (p. 136-137) 2-27. College, AK. Divi- Delta, N.W.T. sion of the American Association Fechhelm RG, Streever B, Gallaway for the Advancement of Science. BJ. 2007. The Arctic cisco (Core- Hendrey G, Baalsrud K, Traaen T, gonus autumnalis) subsistence Laake M, Raddum G. 1976. Acid Ims RA, Fuglei E. 2005. Trophic and commercial fisheries, Colville precipitation: some hydrological interaction cycles in tundra eco- River, Alaska: a conceptual model. changes. Ambio 5: 224-227. systems and the impact of climate Arctic 60: 421-429. change. Bioscience 55: 311-322. Hinzman LD, Bettez ND, Bolton Fischer JB Larned WM. 2003. Sum- WR, Chapin FS, Dyurgerov MB, IPCC. 2007. Climate Change 2007: mer distribution of marine birds in Fastie CL, Griffith B, Hollister Synthesis Report. Contribution of the western Beaufort Sea. Arctic RD, Hope A, Huntington HP, Working Groups I, II and III to the 57: 143-159. Jensen AM, Jia GJ, Jorgenson T, Fourth Assessment Report of the Kane DL, Klein DR, Kofinas G, Intergovernmental Panel on Cli- Flint PL, Mallek EJ, King RJ, Sch- Lynch AH, Lloyd AH, McGuire mate Change. Geneva, Switzerland: mutz JA, Bollinger KS, Derksen AD, Nelson FE, Nolan M, Oechel IPCC. 104 pp. DV. 2008. Changes in abundance WC, Osterkamp TE, Racine CH, and spatial distribution of geese Romanovsky VE, Stone RS, IUCN 2008. 2008 IUCN Red List molting near Teshekpuk Lake, Stow DA, Sturm M, Tweedie CE, of Threatened Species. Available Alaska: interspecific competition or Vourlitis GL, Walker MD, Walker online at www.iucnredlist.org. ecological change? Polar Biol. 31: DA, Webber PJ, Welker J, Winker 549-545. KS, Yoshikawa K. 2005. Evidence Jenkins EJ, Veitch AM, Kutz SJ, and implications of recent climate Hobert EP, Polley L. 2006. Climate Gill RE, Handel CH, Connors PC. change in northern Alaska and change and the epidemiology of 1985. Bird utilization of Peard Bay other arctic regions. Climatic protostrongylid nematodes in and vicinity. In: Kinney, PJ, ed., Change 72:251-298. northern ecosystems: Parela- Environmental characterization phostrongylus odocoilei and and biological utilization of Peard Hobbie JE. 1973. Arctic Limnol- Protostrongylus stilesi in Dall’s Bay. Anchorage: Bureau of Land ogy: A Review. In Britton ME, ed . sheep (Ovis d. dalli). Parasitology Management/National Oceanic Alaskan arctic tundra. Proceedings 132:387- 401. and Atmospheric Administration, of the 25th Anniversary Celebra- Outer Continental Shelf Environ- tion of the Naval Arctic Research Johnson SR. 1985. Adaptations of mental Assessment Program. Final Laboratory (p 127-168). the long-tailed duck (Clangula hye- Reports of Principal Investigators malis L.) during the period of molt 35:244-323. Holmes RT. 1966. Breeding ecology in Arctic Alaska. Proc. Internat. and annual cycle adaptations of the Ornithol. Congress 18: 530-540. Grenfell TC, Putkonen J. 2008. A red-backed sandpiper (Calidris method for the detection of the se- alpina) in northern Alaska. Condor Johnson SR. 2000. Lesser Snow vere rain-on-snow event on Banks 68: 3-46. Goose. In Truett JC, Johnson SR. Island, October 2003, using passive The natural history of an Arctic microwave remote sensing. Water oilfield: development and the biota Resour. Res. 44:3425-3434. (p 233-257). Academic Press.

Johnson SR, Herter DR. 1989. The 89 birds of the Beaufort Sea. British Petroleum Exploration (Alaska), Anchorage, AK. 372 pp. Johnson SR, Richardson WJ. 1981. Jorgenson MT, Shur Y, Osterkamp Lindholm A, Gauthier G, Desrochers Beaufort Sea barrier island lagoon TE. 2008. Thermokarst in Alaska. A. 1994. Effects of hatch date and ecological process studies, Part 3: In Kane DL, Hinkel KM, eds. food supply on gosling growth in Birds. Environmental assessment Ninth International Conference arctic-nesting greater snow geese. of the Alaskan Continental Shelf, on Permafrost (p. 1-8). Institute of Condor 96:898-908. Final reports of principal investi- Northern Engineering, University gators. Vol. 7, Biological studies of Alaska Fairbanks (2Vols). Macdonald SO, Cook JA. 2009. Re- prepared for U.S. Depart. Comm., cent mammals of Alaska. Univer- Nat. Oceanic and Atmospheric Jorgenson MT, Shur YL, Pullman sity of Alaska Press, Fairbanks. Admin., Office of Mar. Poll. Assess. ER. 2006. Abrupt increase in and U.S. Depart. Int., Bur. Land permafrost degradation in Arctic MacLean SF. 1975. Ecology of Manage. Alaska. Geophys. Res. Lett. 33: tundra invertebrates at Prudhoe L02503, doi:10.1029/2005GL024960. Bay, Alaska. In Brown J, ed., Johnson SR, Richardson WJ. 1982. Ecological Investigations of the Waterbird migration near the Kane DL. 1997. The impact of Tundra Biome in the Prudhoe Bay Yukon and Alaskan coast of the hydrologic perturbations on arctic Region, Alaska. Biological Papers Beaufort Sea: II. Moult migration ecosystems induced by climate of the University of Alaska, Special of seaducks in summer. Arctic 35: change. In Oechel WC, Callaghan Report 2 (p 114-123). University of 291–301. T, Gilmanov T, Holten JI, Maxwell Alaska, Fairbanks, AK. B, Molau U, Sveinbjornsson B, eds. Johnson SR, Wiggins DA, Wain- Global Change and Arctic Terres- MacLean SF. 1980. The detritus- wright PF. 1993. Late summer trial Ecosystems (p 63-81). New based trophic system. In Brown J. abundance and distribution of York: Springer Verlag Inc. Miller MC, Tieszen LL, Bunnell, marine birds in Kasegaluk Lagoon, FL, eds., An Arctic Ecosystem: Chukchi Sea, Alaska. Arctic 46: Kane DL, Hinzman LD, McNamara The coastal tundra at Barrow, 212-227. JP, Zhang Z, Benson CS. 2000. An Alaska (p. 411-457). Stroudsburg, overview of a nested watershed PA: Dowden, Hutchinson, and Jones BM, Arp CD, Jorgenson MT, study in Arctic Alaska. Nord. Hy- Ross, Inc. Hinkel KM, Schmutz JA , Flint PL. drol. 31: 245-266. 2009. Increase in the rate and uni- Martin PD. 1983. Bird use of arctic formity of coastline erosion in Arc- Kausrud, KL, Mysterud A, Steen tundra habitats at Canning River tic Alaska. Geophys. Res. Lett. 36, H, Vik JO, Østbye E, Cazelles B, Delta, Alaska. M.S. Thesis, Univ. of L03503, doi:101029/2008GL036205. Framstad E, Eikeset AM, Myster- Alaska, Fairbanks, AK. ud I, Solhy T, Stenseth NC. 2008. Jorgenson J, Udevitz MS, Felix NA. Linking climate change to lemming Martin PD, Moitoret CS. 1981. Bird 2002. Forage quantity and qual- cycles. Nature 456: 93-98. populations and habitat use, Can- ity. In Douglas DC, Reynolds PE, ning River Delta, Alaska. Unpubl. Rhode EB, eds. Arctic Refuge King JG, Hodges JL. 1979. A pre- report. U.S. Fish Wildl. Serv., coastal plain terrestrial wildlife liminary analysis of goose banding Arctic National Wildlife Refuge, research summaries (p 45-60). on Alaska’s Arctic Slope. In Jarvis Fairbanks, AK. U.S. Geological Survey, Biologi- RL, Bartonek JC, eds, Manage- cal Resources Division, Biological ment and Biology of Pacific Flyway McGraw M. 2008. The degradation Science Report USGS/BRD/BSR- Geese (Pages 176-188). Oregon of ice wedges in the Colville River 2002-0001. State University Bookstores, Cor- Delta and their role in pond drain- vallis, OR. age. In Kane DL, Hinkel KM, eds. Jorgenson MT, Heiner M. 2003. Eco- Ninth International Conference on systems of Northern Alaska. Un- Kutz SJ, Hoberg EP, Nagy J, Polley Permafrost (p 1161-1166). Institute published 1:2.5 million-scale map L, Elkins B. 2004. Emerging para- of Northern Engineering, Univer- produced by ABR, Inc., Fairbanks, sitic infections in arctic ungulates. sity of Alaska Fairbanks (2Vols). AK and The Nature Conservancy, Integr. Comp. Biol. 44:109-118. Anchorage, AK. (Available at Meehan R, Jennings TW. 1988. www.nature.org/wherewework/ Lenart EA, Bowyer RT, Ver Hoef J, Characterization and value ranking northamerica/states/alaska/files/ Ruess RW. 2002. Climate change of waterbird habitat on the Colville aya2.pdf). and caribou: effects of summer River Delta, Alaska. U.S. Fish and weather on forage. Can. J. Zool. Wildlife Service, Anchorage, AK. Jorgenson MT, Osterkamp TE. 2005. 80:664-678. Response of boreal ecosystems Meltofte H, Piersma T, Boyd H, to varying modes of permafrost Liebezeit JR, Kendall SJ, Brown S, McCaffery B, Ganter B, Golovnyuk degradation. Can. J. Forest Res. Johnson CB, Martin P, McDonald VV, Graham K, Gratto-Trevor CL, 35: 2100–2111. TL, Payer DC, Rae CL, Streever Morrison RIG, Nol E, Rsner B, Wildman AM, Zack S. 2009. H-U, Schamel D, Schekkerman H, Jorgenson MT, Shur Y. 2007. Influence of human development Soloviev MY, Tomkovich PS, Tracy Evolution of lakes and basins on nest survival of tundra birds, DM, Tulp I, Wennerberg L. 2007. 90 in northern Alaska and discus- Arctic Coastal Plain, Alaska. Eco- Effects of climate variation on the sion of the thaw lake cycle. J. logical Applications (in press). breeding ecology of Arctic shore- Geophys. Res 112. FO2S17, birds. Meddelelser om Grnland doi:10.1029/2006JF000531. Bioscience 59. Copenhagen, Danish Polar Center. Miller MC, Prentki RT, R.J. Bars- Nolan M, Arendt A, Rabus B. 2006. Rachlow JL, Bowyer RT. 1998. date. 1980. Physics. In Hobbie JE, Volume change of McCall Glacier, Habitat selection by Dall’s sheep: ed, Limnology of Tundra Ponds (p Arctic Alaska, from 1956 to 2003. material trade-offs. J. Zool. 245: 51-75). Stroudsburg, PA: Dowden, Ann. Glaciol. 42: 409-416. 457-465. Hutchinson, and Ross, Inc. NRC. 2003. Cumulative environmen- Reist JD, Johnson JD, Carmichael Milner AM, Irons JG, Oswood MW. tal effects of oil and gas activities TJ. 1997. Variation and specific 1997. The Alaskan landscape: an on Alaska’s North Slope. National identity of char from northwest- introduction for limnologists. In Research Council of the National ern Arctic Canada and Alaska. In Milner AM, Oswood MW, eds, Academies. The National Acad- Reynolds J, ed. Fish ecology in Freshwaters of Alaska: ecological emies Press, Washington, D.C. Arctic North America. American syntheses (p 1-44). Ecological Stud- Fisheries Society Symposium 19 (p ies 119. Springer, New York. O’Brien WJ, Bahr M, Hershey AE, 250-261). Bethesda, MD. Hobbie JE, Kipphut GW, Kling Moller AP, Rubolini D, Lehikoinen GW, Kling H, McDonald M, Miller Reist JD, Wrona FJ, Prowse TD, E. 2008. Populations of migratory MC, Rublee P, Vestal JR. 1997. The Power M, Dempson JB, Beamish bird species that did not show a limnology of Toolik Lake. In Milner RJ, King JR, Carmichael TJ, phenological response to climate AM, Oswood MW, eds, Freshwaters Sawatzky CD. 2006a. General ef- change are declining. Proc. Natl. of Alaska: ecological syntheses (p fects of climate change on Arctic Acad. Sci. 105: 16195-16200. 61-106). Ecological Studies 119. fishes and fish populations. Ambio Springer, New York. 35: 370-380. Morris W. 2003. Seasonal movements and habitat use of Arctic grayling Osterkamp TE, Jorgenson MT. 2006. Reist JD, Wrona FJ, Prowse TD, (Thymallus arcticus), burbot (Lota Warming of permafrost in the Arc- Power M, Dempson JB, Power M, lota), and broad whitefish (Corego- tic National Wildlife Refuge, Alas- Kock G, Carmichael TJ, Sawatzky nus nasus) within the Fish Creek ka. Permafrost Periglac.17:65-69. CD, Lehtonen H, Tallman RF. drainage of the National Petroleum 2006b. Effects of climate change Reserve-Alaska, 2001–2002. Alaska Pamperin NJ, Follmann EH, and UV radiation on fisheries for Department of Natural Resources, Petersen B. 2006. Interspecific arctic freshwater and anadromous Technical Report No. 03-02, Fair- killing of an arctic fox by a red fox species. Ambio 35: 402-410. banks. at Prudhoe Bay, Alaska. Arctic 59: 361-364. Reist JD, Wrona FJ, Prowse TD, Morris WA, Moulton LL, Bacon Power M, Dempson JB, King JR, J, Rose JR, Whitman M. 2006. Persson J. 2005. Female wolverine Beamish RJ. 2006c. An overview Seasonal movements and habitat (Gulo gulo) reproduction: repro- of effects of climate change on se- use by broad whitefish (Coregonus ductive costs and winter food avail- lected arctic freshwater and anad- nasus) in the Teshekpuk Lake ability. Can. J. Zool. 83: 1453-1459. romous fishes. Ambio 35: 381-387. region of the National Petroleum Reserve-Alaska, 2003–2005. Poole A. (Editor). 2005. The Birds of Romanovsky VE, Osterkamp T. Technical Report No. 06-04. Alaska North America Online: http://bna. 1997. Thawing of the active layer Department of Natural Resources, birds.cornell.edu/BNA/. Cornell on the coastal plain of the Alaskan Office of Habitat Management and Laboratory of Ornithology, Ithaca, arctic. Permafrost. Periglac. 8: Permitting: Juneau, Alaska. 98 pp. NY. 1-22.

Morrow JE. 1980. The Freshwater Power G, Brown RS, Imhof JG. 1999. Schiedek D, Sundelin B, Readman Fishes of Alaska. Alaska North- Groundwater and fish—insights JW, Macdonald RW. 2007. Inter- west Publishing Company, Anchor- from northern North America. actions between climate change age, AK. Hydrol. Process. 13: 401-422. and contaminants. Mar. Poll. Bull. (54):1845–1856. Moulton L. 2007. Distribution and Prowse TD, Wrona FJ, Reist JD, quantity of potential fish habitat in Gibson JJ, Hobbie JE, Levesque Schindler DW, Smol JP. 2006. Cu- the northeast National Petroleum LMJ, Vincent WF. 2006. Climate mulative effects of climate warm- Reserve-Alaska Planning Area. change effects on hydroecology ing and other human activities on MJM Research, Lopez Island, WA. of arctic freshwater ecosystems. freshwaters of arctic and subarctic Ambio 35: 347-358. North America. Ambio 35: 160-168. Moulton LL, Philo LM, George JC. 1997. Some reproductive character- Quakenbush LT, Suydam RS, Flu- Schmutz JA, Reed JA, Flint PL. Un- istics of least ciscoes and humpback etsch KM, Donaldson CL. 1995. publ. Growth and nutrient content whitefish in Dease Inlet, Alaska. Breeding biology of Steller’s eiders of Arctic tundra vegetation and the In Reynolds R, ed. Fish ecology in nesting near Barrow, Alaska, potential consequences to herbi- Arctic North America. American 1991–1994. U.S. Fish and Wildl. vores in a warming climate. USGS Fisheries Society Symposium 19 (p Serv. Fairbanks, AK. Alaska Science Center, Anchorage, 119-126). Bethesda MD. AK. 91 Sedinger JS, Flint PL, Lindberg Sturm M, Schimel J, Michaelson G, Tulp I, Schekkerman H. 2008. Has MS. 1995. Environmental influ- Welker JM, Oberbauer SF, Liston prey availability for arctic birds ence on life-history traits: growth, GE, Fahnestock J, Romanovsky advanced with climate change? survival and fecundity in black VE. 2005. Winter biological pro- Hind¬casting the abundance of brant (Branta bernicla). Ecology cesses could help convert Arctic tundra arthropods using weather 76: 2404-2414. tundra to shrubland. BioScience and seasonal variation. Arctic 61: 55: 17-26. 48-60. Sedinger JS, Stickney AA. 2000. Black Brant. In Truett JC, John- Svenning MA, Gullestad N. 2002. U.S. Fish and Wildlife Service. son SR. The natural history of an Adaptations to stochastic envi- 1993. Bird-habitat associations of Arctic oilfield: development and the ronmental variations: the effects the coastal plain of north central biota (p 221-232). Academic Press. of seasonal temperatures on the Alaska. Anchorage, Alaska. 300 pp. migratory window of Svalbard + App. Seigle JC. 2003. Determination of Arctic charr. Environ. Biol. Fish. marine migratory behavior and its 64: 165-174. U.S. Fish and Wildlife Service. 1996. relationship to selected physical Spectacled Eider Recovery Plan. traits for least cisco (Coregonus Syvitski JPM. 2002. Sediment dis- Anchorage, AK. 157 pp. sardinella) of the Western Arctic charge variability in Arctic rivers: Coastal Plain, Alaska. Master’s implications for a warmer future. Vadeboncoueur Y, Jeppeson E, Thesis, Oregon State University, Polar Res. 21: 323-330. Vander Zanden MJ, Schierup HH, Corvallis. Christoffersen K, Lodge DM. Taras B, Sturm M, Liston GE. 2002. 2003. From Greenland to green Soloviev MY, Minton CDT, Tom- Snow—ground interface tempera- lakes: Cultural eutrophication and kovich PS. 2006. Breeding per- tures in the Kuparuk River Basin, the loss of benthic pathways in formance of tundra waders in arctic Alaska: measurements and lakes. Limnol. Oceanogr. 48: 1408- response to rodent abundance and model. J. Hydrometeorol. 3:377- 1418. weather from Taimyr to Chukotka, 394. Siberia. In Boere GC, Galbraith Vaughan T, Ryan J, Czaplewski N. CA, Stroud DA, eds. Waterbirds TERA. 1994. Bird use of coastal 2000. Mammalogy, 4th Edition. around the world (p 121-137). The tundra at Prudhoe Bay, Alaska: Toronto: Brooks Cole. Stationery Office, Edinburgh, UK. 1991-1992. Report prepared by Troy Ecological Research Associ- Walsh JE, Anisimov O, Hagen JOM, Sovada MA, Anthony RM, Batt ates, Anchorage, Alaska, for BP Jakobsson T, Oerlemans J, Prowse BDJ. 2001. Predation on waterfowl Exploration (Alaska), Inc., Anchor- TD, Romanovsky V, Savelieva in arctic tundra and prairie breed- age, AK. N, Serreze M, Shiklomanov I, ing areas: a review. Wildl. Soc. Bull. Solomon S. 2005. Cryosphere and 29:6-15. Thomas DC, Kilaan HPL. 1990. hydrology. In ACIA. 2005 Arctic Warble infestations in some Cana- Climate Impact Assessment (p Stephenson SA. 2005. The distribu- dian caribou and their significance. 183-242). New York: Cambridge tion of Pacific salmon (Oncorhyn- Rangifer Special Issue No. 3: 409- University Press. 1042 p. http:// chus spp.) in the Canadian west- 417. www.acia.uaf.edu ern arctic. Canadian Manuscript Report of Fisheries and Aquatic Tonn WM. 1990. Climate change Walsh JE, Chapman WL, Ro- Sciences 2737, Winnipeg. and fish communities: a conceptual manovsky V. 2008. Global climate framework. Trans. Am. Fish. Soc. model performance over Alaska St. Jacques J, Sauchyn DJ. 2009. 119: 337-352. and Greenland. J. Climate 21:6156- Increasing winter baseflow and 6174. mean annual streamflow from Troy DM. 1988. Bird use of the possible permafrost thawing in the Prudhoe Bay oil field during the Waltari E, Cook JA. 2005. Hares on Northwest Territories, Canada. 1986 nesting season. LGL Alaska ice: phylogeography and historical Geophys. Res. Lett. 36:L01401, Research Associates, Inc. Anchor- demographics of Lepus arcticus, L. doi:10.1029/2008GL035822. age, AK. othus, and L. timidus (Mammalia: Lagomorpha). Molecular Ecology Stross RG, Miller MC, Daley RJ. Troy DM. 1991. Bird use of dis- 14:3005-3016. 1980. Zooplankton. In Hobbie JE, turbed tundra at Prudhoe Bay, ed, Limnology of Tundra Ponds Alaska: bird and nest abundance Walvoord MA, Striegl RG. 2007. (p 251-296). Stroudsburg, PA.: along the abandoned peat roads, Increased groundwater to stream Dowden, Hutchinson, and Ross, 1988-1989. Report prepared by discharge from permafrost thaw- Inc. Troy Ecological Research Associ- ing in the Yukon River basin: ates, Anchorage, Alaska, for BP potential impacts on lateral ex- Sturm M, Racine C, Tape K. 2001. Exploration (Alaska), Inc., Anchor- port of carbon and nitrogen. Increasing shrub abundance in the age, AK. Geophys. Res. Lett. 34:L12402, Arctic. Nature 411: 546-547. doi:10.1029/2007GL030216. 92 Troy DM. 2000. Shorebirds. In Tru- ett JC, Johnson SR. The natural history of an Arctic oilfield: development and the biota (p 277-303). Academic Press. Weller G, Nolan M, Wendler G, Wrona FJ, Prowse TD, Reist JD, Benson C, Echelmeyer K, Un- Hobbie JE, Levesque LMJ, tersteiner N. 2007. Fifty Years of Vincent WF. 2006. Key findings, McCall Glacier Research: From science gaps, and policy recommen- the International Geophysical Year dations. Ambio 35: 411-415. 1957–58 to the International Polar Year 2007–08. Arctic 60: 1-10. Yang D, Kane DL, Hinzman LD, Zhang X, Zhang T, Ye H. 2002. White RO, Rowell JE, Hauer WE. Siberian Lena River hydrologic 1997. The role of nutrition, body regime and recent change. J. of condition and lactation on calving Geophys. Res. 107(D23): 4694. success in muskoxen. J. Zool. Lond. doi:10.1029/2002JD002542, 2002. 243: 13-20. Zhang T, Stamnes K, Bowling SA. Wilson DE, Reeder, DM. Eds. 2005. 2001. Impact of the atmospheric Mammal species of the world: A thickness on the atmospheric taxonomic and geographic refer- downwelling longwave radiation ence. Third Edition. Johns Hopkins and snowmelt under clear-sky con- University Press, Baltimore. ditions in the Arctic and Subarctic. J. Climate 14: 920-939. Wilson DE, Ruff S (eds). 1999. The Smithsonian Book of North American Mammals. Smithsonian Institution Press. Washington, D.C. 750 pp.

Hatching snowy owls (Bubo scandiacus). Snowy owls breed across the circumpolar Arctic. Photo by Elizabeth Eubanks (PolarTREC 2008), courtesy of ARCUS. Chapter 6 Working Groups’ Summaries: Common Themes and Research Gaps

This chapter synthesizes the discussions of the WildREACH working groups for birds, fish, and mammals.

94 The working groups were multidisciplinary groups of scientists and multi-organization groups of managers (see Appendix 5). The working groups’ findings are organized into four sections:

Integrated Research and Monitor- ing. All working groups emphasized the importance of interdisciplinary studies and recognized that collaboration between geophysical and biological specialists is paramount.

Common Themes Across Species Groups. Specific information gaps varied among species groups, but most fell into four crosscutting themes: 1) changes in precipitation and hydrology; 2) changes in vegetation communities and phenology; 3) changes in abundance and timing of invertebrate emergence; and 4) coastal dynamics.

Fish and Wildlife Ecology Data. All working groups emphasized that information available on life history, habitat requirements, distribution, abundance, and demography is inadequate for many arctic species. More complete information on basic life history is needed to refine hypotheses and improve our predictions regarding species’ response to environmental change.

Focal Species in the Context of Cli- mate Change. All working groups developed a preliminary list of focal species perceived to be sensitive indicators of climate change.

Above: Image of the Arctic National Wildlife Refuge and the Beaufort Sea acquired by the Multi-angle Imaging SpectroRadiometer’s nadir (vertical-viewing) camera on August 16, 2000, during Terra orbit 3532. The area represented by the image is approximately 380 km x 540 km. Image Credit: NASA/GSFC/LaRC/JPL, MISR Team. Below: Arctic grayling (Thymallus arcticus) have highly variable life history strategies depending on location. Some populations are highly migratory while others are relatively sedentary. Wikimedia Commons 95 photo. We cannot accurately predict the ef- Satellite-based remote sensing ap- Need for fects of climate change on organisms proaches are essential for assessing without sufficient information about attributes such as snow cover, green- Integrated the underlying biological and physi- up, and surface water on a regional cal processes that drive terrestrial scale. Processing and archiving and aquatic ecosystem functions these data sets in a readily usable Research and (Wrona et al. 2006). Acquiring this form will require multi-agency co- understanding will require addi- ordination. At an intermediate scale Monitoring tional effort in three complementary (e.g., watershed), high-resolution areas: satellite or aircraft-mounted sensors 1. Long-term monitoring of physical can provide elevation data for digital and biological parameters; elevation models or detailed stream 2. Field observations and focused and landcover mapping. These re- process studies to clarify the mech- mote sensing efforts combined with anisms by which environmental spatial and retrospective analyses variables influence fish and wildlife will better inform planning efforts populations; and for costly long-term monitoring and 3. Interdisciplinary synthesis model- process studies conducted on the ing. ground. Cost considerations will likely limit ground-based efforts to This approach is consistent with that relatively few sites, thus careful site proposed by Vrsmarty et al. (2000) selection will be important. in their assessment of the needs for developing a synthetic understand- Multidisciplinary monitoring pro- ing of the arctic hydrologic cycle. grams, integrated with research, The strong linkages and feedbacks will provide the best opportunity between climate, permafrost, hydrol- for developing a mechanistic under- ogy, and habitat conditions highlight standing of climate-related habitat the critical need for comprehensive, change. In this framework, intensive cross-disciplinary monitoring and in situ research and monitoring research programs. Integrated long- would be scaled up to the landscape term data sets from representative and regional level through modeling locations are needed to test hypoth- informed by remote sensing tech- eses regarding the mechanisms by nologies. Monitoring and research which climate affects habitat condi- should include replication across tions and to parameterize models of dominant environmental or eco- landscape change. Similarly, moving logical gradients using a distributed toward more specific predictions network of sites that are sampled at regarding specific fish and wildlife a lower intensity. responses to habitat change will require both long-term monitoring and There are some existing long-term focused research approaches. monitoring efforts on the North Slope of Alaska, which provide a Long-term data sets are particularly limited foundation and baseline data scarce in Alaska, where organized for an integrative approach to long- scientific programs are recent, and term monitoring of fish, wildlife, and access to remote areas is difficult key environmental variables. For and expensive. Initial investment example, existing facilities at Toolik should be placed in “rescue” and as- Lake and Barrow should serve as sembly of legacy data to be used for “flagship observatories” (Shaver et retrospective analyses. These data al. 2004) around which a research sets may be limited by resolution and monitoring program could be and spatial scale; however, they may built. An essential element of an provide valuable insights into: observatory is a sustained observa- 1. the extent, spatial scale, and inten- tional time series consisting of mea- sity of monitoring efforts necessary surements of a core set of environ- to develop efficient and relevant mental variables. The Arctic Long approaches for long-term data col- Term Ecological Research (LTER) lection and modeling needs, site at Toolik Lake provides the best 2. mechanisms of change and their example of integrated monitoring relative importance to fish and in northern Alaska (http://ecosys- wildlife populations, and tems.mbl.edu/ARC/Datatable.html, 96 3. additional gaps in data not ad- retrieved April 10, 2009), but data dressed during this workshop. pertaining to higher trophic-level organisms are largely absent from the Future approaches should include existing program, with the exception establishment of integrated long- of freshwater fish. Suitable models term data sets at multiple scales. for integration of fish and wildlife monitoring with environmental data meteorological stations with coor- exist, however, at sites such as Zack- dinated data storage and analysis. enburg Research Station, northeast In considering climate effects, we Greenland (Meltofte et al. 2008). are constrained to some degree by The National Park Service’s Arctic the ways in which we aggregate Network Inventory and Monitor- data. Projections of future climate ing Program provides an example conditions tend to focus on metrics that emphasizes the use of fish and describing annual or seasonal aver- wildlife as “vital sign” indicators ages. While practical and often use- of ecosystem function (http://sci- ful, evaluation of potential climate ence.nature.nps.gov/im/units/arcn/, change effects based solely on aver- retrieved 10 April 2009). These age conditions overlooks the impor- existing programs provide useful tance of extreme events on fish and models, but data collection at the wildlife populations. For instance, temporal and spatial scale necessary events such as rain-on-snow in to effectively address climate change winter (which have killed thousands issues will require implementation of muskoxen and caribou) and major of monitoring at multiple sites, and floods in summer (which have killed strong collaboration among federal, dozens of muskoxen, and likely cari- state and local agencies, academic bou and moose) have a potential to institutions, conservation organiza- devastate populations if they occur tions, and private industry. with increased frequency. Greater attention must be paid to modeling Much of the uncertainty in project- variability in climate conditions. ing future habitat conditions stems Similarly, conditions during the sea- from an insufficient understand- sonal transition periods (spring thaw, ing of arctic hydrologic systems. freeze-up) can be more critical for In the Arctic, the thermal and fish and wildlife populations than the hydrologic regimes are uniquely relatively stable mid-winter or sum- linked through changes in the active mer periods. Greater consideration layer (the ground layer that experi- of the key issues related to seasonal ences seasonal freeze and thaw), so transitions is warranted. coordinated study of the coupling between permafrost dynamics, soil moisture, and ecological systems is fundamental to predicting the extent and magnitude of habitat change. High priority must be placed on Box 6.1. Examples of the coordinated set of measurements establishing a distributed network that might be made at an arctic hydrologic observatory study of watershed-based study sites to site. After Vrsmarty et al. (2000). collect data essential to improved synthesis modeling of environmental Hydrologic and other geophysical measurements change. A proposed set of measure- • Precipitation amount (year round) ments for hydrological study sites • Evapotranspiration and sublimation • Solar flux and surface energy measurements (modified from Vrsmarty et al. • Snow pack 2000) is provided in Box 6.1. Some of • Snow redistribution these parameters have been mea- • Snow melt sured in the Kuparuk River water- • Soil thermal properties and their variation shed, beginning as early as 1985. We - Temperature profiles recommend establishing a long-term - Active layer depth hydrological observatory within the - Permafrost temperature Kuparuk River watershed and at - Thermal conductivity least two additional watersheds to • Infiltration on frozen and unfrozen soils better represent variability across • Soil moisture • Runoff flow paths the North Slope. Data obtained at • Stream and large river discharge hydrologic observatories will be • High-resolution and accurate digital elevation models instrumental in creating predictive • Distribution of surface waters, including seasonal flooding and models that will help land manag- connectivity between lakes and rivers ers better understand and manage • Timing of freeze-thaw cycles fish and wildlife habitat in a rapidly changing climate. Biological and biogeochemical measurements • Precipitation chemistry Because climate modelling forms • Vegetation surveys (ecotype maps and community composition) 97 the basis for projections of habitat • Soil mapping, including ice content • Monitoring of vegetation, soil, and groundwater chemistry change, continued effort to improve • Physical, chemical, and biological parameters in stream, river, the reliability of down-scaled climate and lake ecosystems models is crucial, and would likely • Isotope and other tracers for discharge entering the Arctic Ocean benefit from an expanded network of Although climate change effects will jected wetter and warmer winter Common differ among species and among environment will have important taxonomic groups, a common set of implications for resident mam- Themes Across environmental drivers and processes mals (see below) but minimal direct will affect habitat suitability for a consequences to birds because broad suite of species. Four major most are migratory and absent in Species Groups themes emerge as relevant to a winter. For arctic fish populations, diversity of fish and wildlife species overwintering habitat availability groups: is currently limiting, and a warmer, 1. Hydrologic processes, including wetter winter environment would precipitation, water balance, and most likely relax that constraint. distribution of surface water; The consequences of a hypothetical 2. Vegetation, including community summer drying regime, however, are composition and phenology; perceived as a greater threat to fish 3. Invertebrates (primarily aquatic and birds. High priority is assigned or semi-aquatic forms), including to addressing the following ques- productivity and phenology; and tions: 4. Coastal dynamics, including the • How reliable are the projections interacting effects of erosion, for increasing precipitation and sedimentation, stream discharge, evapotranspirtion? inundation, and plant successional • How will the annual precipitation processes. input on the Coastal Plain and Foothills be allocated between win- Within these four general areas, ter (snow pack) and summer? there are specific needs for improv- • How will changes in precipitation, ing understanding of the physical evapotranspiration, and active processes that drive habitat change layer depth alter summer surface and the associated ecological re- water availability in shallow-water sponses, summarized in Table 6.1 and mesic/wet tundra habitats? and further elaborated below. • How will changing patterns of seasonal runoff affect stream flow? Hydrologic Processes: Precipitation, • What is the contribution of Water Balance, and Distribution of groundwater in various systems, Surface Water and is it sufficient to maintain year- Understanding water balance and round flow? partitioning of water among habitat • Will drought conditions and chang- units is a key issue, with important es in drainage patterns decrease direct and indirect implications water body connectivity? across all fish and wildlife taxa. The changes associated with the pro- Lakes may be susceptible to draining as a result of newly developed thermokarst drainage networks, thus: • Which Coastal Plain lakes are susceptible to draining and on what time scale?

Increasing temperatures and precipitation in winter will affect food acquisition and energy balance of non-denning and subnivean mammals. For mammals, the quantity and nature of winter precipitation is of paramount interest, with a particular focus on the following questions: • What are expected changes in snowpack characteristics (depth, density, presence of ice layers) and how might these vary on a regional and local scale? • How much change will occur in 98 the timing of snow melt and snow onset? A collared lemming (Dicrostonyx groenlandicus) in a subnivean runway. • How will the frequency of rain- Lemmings spend the majority of the year in subnivean spaces, and the on-snow and severe winter storm characteristics of the snowpack are likely to be critical to their survival and events change? reproduction. Photo by Jean-Louis Klein and Marie-Luce Hubert. Vegetation Community Change port of terrestrial productivity into Changes in vegetation have impor- the aquatic system in the form of tant implications for several ecosys- leaves and insects. This input could tem processes. Vegetation exerts an increase with increased floodplain important influence on soil moisture shrub growth, with potential impli- and temperature, active layer thick- cations for invertebrate community ness, and permafrost stability, and composition and productivity. Shore- vice versa. Changes in both plant line stability and water temperature phenology and the relative abun- is also influenced by the vegetation dance of forage species will affect adjacent to water bodies. For birds the palatability, nutritional quality, and mammals, the influence of and quantity of available food for changing vegetation communities is herbivores (primarily mammals in more direct, with the following is- arctic Alaska). Changes in primary sues considered highest priority: productivity will directly affect the • How will changes in temperature carrying capacity of habitat for and precipitation affect plant aquatic and terrestrial herbivores, community composition, includ- and indirectly affect higher trophic ing changes in habitat structure levels. Vegetation is a primary com- and nutritional quality of available ponent of habitat structure, which forage (i.e., digestibility, nutrient may affect suitability for some spe- content)? cies independent of trophic effects. • What is the expected rate of shrub increase, and how will this vary by Vegetation change will affect fish species/growth form (low vs. tall indirectly, primarily through ex- shrub) and ecoregion?

Table 6.1. Pathways by which climate-infuenced habitat change could infuence arctic fsh and wildlife, categorized by major themes. Affected Species Theme Habitat Change Pathway Groups Lake drainage due to development of new drainage networks Birds, Fish Drying of shallow-water and mesid/wet tundra habitats Birds Changes in flow regimes, as influenced by water source Fish Water: Precipitation, water Drought-related loss of connectivity between water bodies Fish balance, and distribution of surface water Change in snowpack characteristics, e.g., depth and density Mammals Change in the length and timing of the snow season Fish, Mammals Change in the frequency and timing of extreme events, i.e., rain- Mammals on-snow events and major winter storms Changes in aquatic trophic systems, including potential shift Birds, Fish from benthic to pelagic production in lakes Change in plant phenology, accompanied by change in the timing Birds, Mammals of forage nutritional quality Change in plant community composition: −Change in forage quantity and quality, at community level Birds, Mammals −Shrub expansion Birds, Mammals Vegetation: Seasonality and community composition −Paludification and consequences for both herbivore and Birds detritus-based trophic systems Increased primary productivity of terrestrial vegetation and Fish export of nutrients into aquatic systems Change in riparian vegetation, as influenced by flow regime and Birds, Fish, Mammals frequency/severity of flooding Change in productivity, abundance, and seasonality of life cycles Birds, Fish Invertebrates: Phenology and for aquatic and semi-aquatic invertebrates abundance Change in timing of emergence and abundance of biting insects, Mammals parasites, and disease vectors Degradation of barrier island systems and alterations to physi- Birds, Fish cal/chemical environment of lagoons Increased fog/clouds and subsequent effects on evapotranspira- Coastal Dynamics: Interactions Birds, Mammals 99 of coastal erosion/deposition, tion, snow melt, and plant phenology river discharge, sedimentation, Change in availability of coastal habitats, specifically coastal wet Birds, Mammals inundation, succession sedge and deltaic mud flats Change in availability of deep-channel habitat in major river Fish deltas • How will changes in the length parasites and vectors of disease, and timing of the growing season rather than prey. Related priority influence plant phenology, includ- questions include: ing seasonal changes in nutritional • How will warming and chang- quality? ing seasonality affect abundance • What is the likelihood of wide- and peak activity periods of biting spread conversion from sedge insects, and what are the bioener- and sedge-shrub meadow to bog getic consequences for caribou? meadow (paludification), and how • How will warming and changing would this affect herbivore and seasonality affect the prevalence of detritus-based trophic systems? parasites and disease vectors (e.g., • How will changes in the seasonal- nematode parasites of muskoxen ity of stream discharge and oc- and Dall’s sheep)? currence of flood events influence development of riparian vegetation Coastal Dynamics communities? Loss of summer sea ice, increased coastal erosion, rising sea level, Invertebrate Populations: Temperature warmer ocean waters, changing and Seasonality Infuences seasonality of river discharge, and Most species of birds and fish in altered sedimentation rates interact arctic Alaska are predators of to affect coastal habitats. Coastal aquatic and semi-aquatic macroin- lagoons of the Beaufort and Chukchi vertebrates (e.g., insects, crusta- seas are important summer feed- ceans, gastropods) and are therefore ing habitat and migration corridors vulnerable to changes in productiv- for birds and fish, while deltas and ity, abundance, and seasonality of life coastal wet sedge tundra are impor- cycle stages. This is an exceptionally tant foraging habitats for shorebirds poorly studied aspect of arctic ecol- and waterfowl. Although coastal ogy, and experimental research and habitats are dynamic by nature, cli- monitoring efforts to address the mate change may cause directional following areas are a high priority: change that alters the availability • How does earlier spring thaw of some habitats. Related priority affect timing of invertebrate life questions include: cycle events and peak availability • Will higher water temperatures, to predators? sea level rise, and retreat of sum- • How does temperature affect mer sea ice cause degradation of growth and development of aquatic the barrier island systems of the insects? Beaufort and Chukchi seas? The USFWS has determined • What climate-related changes are • Will alluvial deltas continue to that listing the yellow-billed loon likely in community composition of build or will rising sea levels out- (Gavia adamsii) as a threatened or macroinvertebrates in stream, lake, pace potential increases in sedi- endangered species is warranted and saturated soil environments? mentation rates? under the Endangered Species • How will changes in the distribu- • How quickly will shoreline retreat Act, but that listing is precluded tion and quality of surface waters result in newly breached lake by other higher priority species. and shifts from pelagic to benthic basins? The “warranted but precluded” productivity in deep lakes affect • To what extent will coastal ero- finding was published in the Federal availability of macroinvertebrates sion, in combination with sea level Register on March 25, 2009. The to fish and wildlife? rise, cause salinization of low-lying yellow-billed loon is now designated coastal areas? as a candidate species. Photo by Ted For most mammals, invertebrates • Will coastal wet sedge meadows Swem, USFWS. are more important in their role as establish at a rate equal to loss of this habitat through erosion and inundation?

A longer ice-free season will strongly influence climate in the coastal zone. A longer open water season may increase the prevalence of fog, which will have a cooling effect in spring while a warming effect in fall. • Will increased fogginess/cloudiness exert a negative or positive 100 feedback effect on air temperature in the coastal zone? What would be the expected spatial extent of this effect? With the exception of a few well- • Spawning and overwintering areas studied species, such as caribou, our are known for some populations of Fish and knowledge of life history, habitat Dolly Varden in the eastern North requirements, distribution, abun- Slope and broad whitefish in the Wildlife dance, and demography is incom- Teshekpuk Lake region, but the plete or poor for most arctic species; extent of inter-drainage movement additional basic biological data are and the habitats used at different Ecology needed for all but the most general life stages for these and other spe- forecasts regarding species response cies are poorly known. to a warming climate. Because the • Little is known about the factors region is remote, basic knowledge of that limit Pacific salmon popula- status and distribution is often inad- tions on the North Slope and the equate, as illustrated in the following potential for range expansion into examples: arctic waters. • Little is known about the distri- • Knowledge of bird distribution is bution and natural history of the heavily biased toward the coastal Alaska marmot, an alpine-adapted areas, with very sparse data from species that has a limited and dis- the Foothills and northern Brooks junct population. Range, particularly west of the • The Alaska Tiny Shrew is a newly Dalton Highway. described species (discovered • Habitat models that reliably in 1997) with only a few dozen predict bird species occurrence specimens collected to date, two of and/or abundance on the basis of which are from the North Slope. measurable habitat parameters are lacking.

Alaska marmots (Marmota broweri) are endemic to mountains north of the Yukon River (Gunderson et al. 2009); evidence suggests that these highly social animals hibernate in groups. Photo by Jake Schaas, Toolik Field Station.

101 Cost constraints and practicality A breakdown in the cyclicity of argue for conducting biological and population highs could influence life history investigations on focal nutrient cycling, plant community species or related species groups, composition, avian and mammalian chosen on the basis of criteria that predator population dynamics, includes consideration of climate and productivity of shorebirds and impacts; this topic is explored below. waterfowl. It is not known whether lemming population dynamics in Because species respond individu- arctic Alaska are similar to those alistically to climate drivers, it is reported from Eurasia. expected that novel assemblages of • The Fennoscandian population species will arise. Given the difficulty of arctic fox is considered endan- inherent in predicting the composi- gered (Angerbjrn et al. 2008). tion of new species assemblages, Competition with red fox, along consideration of interspecific inter- with the influence of warming on actions is largely deferred, except prey (lemming) availability, has at the broad level of trophic dynam- been suggested as contributing to ics. In specific instances, however, the non-recovery of this arctic fox observed changes in other arctic population (Frafjord 2003, Kil- regions suggest promising avenues lengreen et al. 2006). Anecdotal for investigation: evidence from northern Alaska • In most portions of the Arctic, documents interspecific competi- lemming populations exhibit tion (Pamperin et al. 2006) but is strong cyclicity, but observa- insufficient to demonstrate broad- tions in Eurasia over the past few scale displacement of the arctic decades indicate that lemming fox. The relative abundance of the cycles are becoming less regular two species may be a useful early and of lesser amplitude (Ims and indicator of a climate related shift Fuglei 2005, Kausrud et al. 2008). in predator-prey dynamics.

102 The arctic fox (Vulpes lagopus), found across the circumpolar Arctic, is an important predator of small mammals and birds in summer and ranges out onto the sea ice in winter. Photo by Ken Whitten. Developing a better understanding of the physical processes that Choosing Focal influence ecosystem function is an essential foundation for managing Species in the fish and wildlife populations affected by climate change. Basic biological information, however, such as Context of seasonal habitat requirements, is deficient for many species. Un- Climate Change derstanding the response of arctic fish and wildlife to climate change requires improved understanding of physical and ecological processes, as well as enhanced knowledge of natural history and population ecology of arctic species.

Practical considerations dictate that resource management agencies choose a subset of species to actively study and manage—traditionally, those species whose numbers have or will decline without management intervention or which have significant public recreational or commercial value. Choosing these “focal species” is a challenging task, best undertaken in a structured fashion based on objective criteria. In the context of the Fish and Wildlife Service’s “Strategic Habitat Con- servation” approach, it has been suggested that these criteria reflect the importance of the species relative to its ecological significance, management significance, legal mandates, and feasibility of implementing long-term, landscape-based adaptive management.

103 Above: Both species of phalarope found in northern Alaska —red phalaropes (Phalaropus fulicarius) and red- necked phalaropes (Phalaropus lobatus)—may be good indicator species for climate change effects. Photo by Stephanie Clemens, USFWS. Below: This polar bear (Ursus maritimus) ranged about 125 miles inland along the Dalton Highway in October 2002 before returning to the coast. Photo by Richard Flanders. Climate change adds complexity to The uncertainties surrounding sce- the process of selecting focal species. narios of climate-associated habitat While existing lists (e.g., endan- change exacerbate the difficulties in gered species listing, NatureServe choosing appropriate focal species heritage ranking, International for conservation planning. Our ap- Union for Conservation of Nature proach has been to develop hypoth- [IUCN] Red List) reflect population eses regarding projected habitat response to past events, consider- change in the form of conceptual ation of climate change forces us models (Chapter 5). For each model, to consider a future in which over we have proposed reasonably fore- 40% of species are at risk of extinc- seeable fish and wildlife responses. tion globally (IPCC 2007), and the Here, we propose monitoring priori- arctic tundra region may experi- ties based on selection of species/ ence up to 74% turnover in species parameters believed to be sensi- assemblage (Lawler et al. 2009). tive indicators of the hypothesized Intrinsic life history traits that may habitat changes (Table 6.2). The be associated with greater extinction resulting list includes both common risk in a changing climate may be and rare species—some that would incorporated into a ranking process benefit from a warming climate and (Foden et al. 2008). An example of an some for which negative effects are application of this approach to arctic postulated. In this framework, focal terrestrial mammals is provided in species are selected, at least in part, Box 6.2. While a valuable first step, on the basis of their perceived value this approach does not incorporate as tests of specific models of habitat specific projections of habitat change change. This list should be consid- on a local or regional scale. ered preliminary and should be refined through a further structured decision-making process. A second additional criterion to consider when refining selection of focal species is the timescale on which different processes of change occur. Some processes will alter habitats relatively quickly, while others operate on much longer timescales (Table 6.3). As the priority list develops, it may be necessary to add or re- move species. The process must be undertaken in full awareness that initial projections of future habitat condition may be inaccurate (thus understanding of species’ vulnerability limited), but this should not deter us from initiating biological monitoring. As data are collected and synthesized, conceptual models can be adjusted to reflect changes in our understanding or changes in processes and species/environmental responses.

A common eider (Somateria mollisima) sits on her nest on a barrier island. Common eiders breed colonially, and females commonly return to their natal islands. Current evidence suggests that the existing system of barrier islands in northern Alaska may deteriorate or diminish in a relatively short time frame. Photo by James Zelenak, USFWS.

104 105

Less resilient More resilient 7 7 8 9 9 9 9 12 12 13 13 13 14 14 14 14 15 15 15 15 11 11 11 12 12 12 12 12 Index of potential resilience Polar bear Polar Alaska marmot Species Dall sheep Barren-ground shrew Collared lemming Alaskan hare Porcupine Porcupine Moose shrew Tundra Dusky shrew Caribou Least weasel Ermine Wolverine Snowshoe hare Red fox Wolf Red backed vole vole Tundra Muskox Arctic fox Singing vole Arctic ground squirrel Mink River otter Lynx Grizzly bear Brown lemming C3 C1 C1 C1 C2 C2 C2 C2 C2 C2 C2 C2 C2 C3 H3 H1 H3 H3 H1 H1 H3 H1 H1 H2a H2a H2a H2b H2b Group In contrast, species like moose that are In contrast, species like moose that are broadly distributed and have recently Arctic expanded their distribution into the likely have an advantage as temperatures estimated relative resilience warm. We in scores for representative species living the Alaskan arctic based on geographic range, niche breadth, population status, Our re- and life-history strategies (Table). sults suggest that polar bears and Alaska marmots are likely to be most vulnerable to expected changes in climate. Arctic mammals sorted by their relative vulnerability to climate change. An index of relative resilience was calculated from world-wide geographic range niche 5.3), breadth of ecological (Table Species Act and IUCN Red List 5.3), status under the Endangered (Table (1=threatened, 2=of least concern) and life history strategies (1=low reproducing carnivores [ungulates [bears], 2=low reproducing herbivores and porcupines], 3=moderate reproducing carnivores [shrews, mustelids, canids, felids] and herbivores [squirrels], and 4=high reproducing bears were the only species herbivores [microtines and hares]). Polar considered threatened; wolverines are listed as “near threatened” under IUCN criteria, but Alaska populations are not threatened. Effects of climate change on terrestrial change of climate Effects depend of the Arctic mammals change or resistance to vulnerability species’ relative on the individual niche distribution and function of geographic Resilience is a (resilience). a broad Mammals that have breadth. general- range, are habitat geographic foods are more eat a variety of ists, and/or likely to survive changing environmental that are highly spe- conditions than those currentthe for cialized environment arctic in the circumpolar and are found only example, north or in northern Alaska. For like Alaska marmotsan endemic species northernthat occurs only in and central to be adversely af- Alaska is more likely than red-backed fected by climate change Alaska and voles, which live throughout northernCanada. Arctic-adapted species and arctic like polar bears, muskoxen, to warmingfoxes are more vulnerable temperatures than species like grizzly bears and red foxes that are found throughout the northern hemisphere in a 5.3). Mon- variety of habitats (see Table tane species (e.g., Alaska marmots, Dall’s to sheep) may be particularly vulnerable occurringchanges gradi- elevational along ents that essentially shrink their available habitat. Box 6.2. Ranking arctic terrestrial mammals by resilience to climate change climate to by resilience mammals terrestrial arctic Ranking Box 6.2. Table 6.2. Proposed indicators of climate change effects on fsh and wildlife.

Positive Species or Parameter to Projected Change in or Rationale for Response to Projected Species Group Measure Habitat Negative Habitat Change Effect Birds Changes in fish Warming lakes could increase Distribution, Yellow-billed loon availability, lake productivity, but loss of connectivity fledging success + or - drainage could inhibit fish migration Warming could improve availability Increased productivity of macroinvertebrates fed to chicks; Pacific loon Fledging success in warmer lakes + longer ice-free season allows more time to fledge Red phalarope, Abundance, Drying of wet Loss of preferred foraging habitat and/ pectoral distribution sedge meadow - or decreased food availability sandpiper Loss of preferred foraging habitat in Drying of wet sedge Red-necked Abundance, lowlands (drying of wet sedge mead- meadow, increased phalarope distribution + or - ow), but increased thermokarst could thermorkarst allow expansion into upland areas Geese (black Change in plant Growth rates are sensitive to forage brant, greater Gosling growth rates phenology - quality during the fledging period white-fronted) Timing of arrival and Potential trophic mismatch if timing of Change in timing of nesting; shorebird migration and nesting does Shorebirds aquatic insect life cycle chick growth and - not match temperature-regulated tim- stages survival ing of insect abundance Perching birds Shrub-associated species will expand Abundance, (sparrows. war- Increased shrubbiness their range and local abundance as distribution + blers, etc.) shrubs increase Reduced availability of gravel islands Abundance, Loss of barrier islands, limits nesting habitat and/or increased Common eider nest success increased storms - frequency of storm overwash increases nest loss Abundance and Loss of barrier islands reduces avail- Change in lagoon Long-tailed duck distribution during ability of habitat for resting and systems - molt stage disrupts trophic system of lagoons Fish + (until Growth rate, Increased water upper productivity, age at temperature Arctic grayling lethal Sensitive, ubiquitous maturity, within- (associated with tempera- drainage distribution availability of food) ture is reached) Growth rate, produc- Increased water tem- Fish passage will depend on tivity, age of matu- Broad whitefish perature; loss of wa- connectivity between lakes, small rity, within-drainage - terbody connectivity streams, and other habitats distribution Increased water High site fidelity; specific habitat Dolly Varden Population estimates temperature; habitat - requirements fragmentation Increased water Assume that expansion of range would All salmon Regional distribution temperature + have a positive effect Changes in water Arctic char and Perhaps narrow range of temperature Regional distribution quality and increasing lake trout - tolerance temperatures Changes in water quality; changes in pH Rapid changes in response to Aquatic insects/ Species abundance and 106 (resulting from acidi- environmental changes; easily invertebrates composition + or - fication of terrestrial sampled habitats)

Continued next page Table 6.2. Continued

Positive Species or Parameter to Projected Change in or Rationale for Response to Projected Species Group Measure Habitat Negative Habitat Change Effect Mammals Shift in distribution, decline in Use of onshore Loss of summer sea ice Polar bear abundance are likely to occur as sea habitats/denning and principle prey - ice disappears Limited distribution; little is known Distribution, Alaska marmot Loss of alpine habitats about this endemic species that has a life history - limited and disjunct population Loss of alpine habitats; Higher energetic costs, more parasites Trends in abundance, more rain-on-snow Dall’s sheep and diseases, changes in plant phenol- distribution events, deeper snow, - ogy and communities warmer summers Arctic-adapted species lives in arctic More rain-on-snow Alaska year-round. Less access to Trends in abundance, events, deeper snow, winter forage, higher energetic costs, Muskox distribution warmer summers, - more diseases and parasites may more shrubs offset positive aspects of increasing summer biomass Less access to winter forage, loss of More rain-on-snow lichens from increased fire or competi- events, deeper snow, tion with other vegetation, timing of Trends in abundance, warmer summers, migration uncoupled from optimal Caribou distribution fewer lichens, changes - foraging, more insect harassment, in plant phenology and parasites, and diseases, and increased community structure energetic costs may offset positive aspects of increases summer biomass Arctic-adapted species likely to be af- Lemmings and fected by changes in food and shelter; Distribution and Changes in distribution barren ground important in food webs; little known relative trend and population cycles + or - shrews about barren ground shrews (may be difficult to study) Arctic-adapted carnivore; possible competition with red foxes and disappear- Distribution and Changes in abundance Arctic fox ance of sea ice may affect distribution relative trend and distribution - and abundance; important predator of birds.

Table 6.3. Relative timeline of climate-driven processes of change. Timeline Process Erosion, construction, and migration of barrier islands Ecosystem phenology and productivity Annual to decadal Expansion of lake shorelines (due to erosion) Thermokarst, pond formation, and gully formation Change in fire regime Change in species composition Change in length of ice-free season Decadal to century Change in sea-surface temperatures Lake tapping Shrub advance Formation of a thaw lake plain Loss of continuous permafrost Century to millennial Stabilization of drained lake basin 107 Re-establishment of polygonal terrain Unknown Paludification (decadal to millennial) Literature Cited Kausrud K, Mysterud A, Steen Wrona FJ, Prowse TD, Reist JD, H, Vik JO, Østbye E, Cazelles B, Hobbie JE, Levesque LMJ, Angerbjrn A, Hersteinsson P, Tan- Framstad E, Eikeset AM, Myster- Vincent WF. 2006. Key findings, nerfeldt M. 2008. Alopex lagopus. ud I, Solhy T, Stenseth NC. 2008. science gaps, and policy recommen- In: IUCN 2008. 2008 IUCN Red Linking climate change to lemming dations. Ambio 35. List of Threatened Species. www. cycles. Nature 456: 93-98. iucnredlist.org. Downloaded on April 25, 2009. Killengreen ST, Ims RA, Ypoccoz NG, Bråtten KA, Henden JA, Foden W, Mace G, Vié J-C, Angulo andSchott T. 2006. Structural char- A, Butchart S, DeVantier L, Dublin acteristics of a low Arctic tundra H, Gutsche A, Stuart S, Turak ecosystem and the retreat of the E. 2008. Species susceptibility to Arctic fox. Biol. Cons. 135: 459-472. climate change impacts. In J-C Vié, C Hilton-Taylor, SN Stuart, eds. Kutz SJ, Hoberg EP, Nagy J, Polley The 2008 Review of The IUCN Red L, Elkins B. 2004. “Emerging” List of Threatened Species. IUCN parasitic infections in arctic ungu- Gland, Switzerland. lates. Integrative and Comparative Biology 44:109-119. Frafjord K. 2003. Ecology and use of arctic fox Alopex lagopus dens Lawler JJ, Shafer SL, White D, in Norway: tradition overtaken Kareiva P, Maurer EP, Blaustein by interspecific competition? Biol. AR, Bartlein PJ. 2009. Projected Cons. 111:445-453. climate-induced faunal change in the western hemisphere. Ecology Gunderson AM, Jacobsen BK, 90: 588-597. http://esapubs.org/Ar- Olson LE. 2009. Revised distribu- chive/ecol/E090/041/default.htm. tion of the Alaska marmot, Mar- mota broweri, and confirmation Meltofte H, Christensen TR, Eber- of parapatry with hoary marmots. Journal of Mammalogy. 90:859-869. ling B, Forchhammer MC, Rasch M, (eds). 2008. High-Arctic eco- Ims RA, Fuglei E. 2005. Trophic system dynamics in a changing cli- interaction cycles in tundra eco- mate: ten years of monitoring and systems and the impact of climate research at Zackenberg Research change. Bioscience 55: 311–322. Station, Northeast Greenland. Advances in Ecological Research: IPCC. 2007. Climate change 2007: 40. Elsevier, London, U.K. synthesis report. Contribution of Working Groups I, II and III to Pamperin NJ, Follmann EH, the Fourth Assessment Report of Petersen B. 2006. Interspecific the Intergovernmental Panel on killing of an arctic fox by a red fox Climate Change. RK Pachauri, A at Prudhoe Bay, Alaska. Arctic 59: Reisinger, L Bernstein, P Bosch, O 361-364. Canziani, Z Chen, R Christ, O Da- vidson, W Hare, S Huq, D Karoly, Shaver GR, Callaghan TV, Tweedie V Kattsov, Z Kundzewicz, J Liu, U CE, Webber PJ. 2004. Flagship Lohmann, M Manning, T Matsuno, observatories for arctic environ- B Menne, B Metz, M Mirza, N mental research and monitoring. Nicholls, L Nurse, R Pachauri, J Report of a workshop held No- Palutikof, M Parry, D Qin, N Ra- vember 18-20, 2004. Report to US vindranath, A Reisinger, J Ren, K National Science Foundation Office Riahi, C Rosenzweig, M Rusticucci, of Polar Programs and the Interna- S Schneider, Y Sokona, S Solomon, tional Arctic Science Committee. P Stott, R Stouffer, T Sugiyama, R Vrsmarty CJ, Hinzman LD, Pe- Swart, D Tirpak, C Vogel, G Yohe, terson BJ, Bromwich DH, Hamil- eds. Cambridge University Press, ton LC, Morison J, Romanovsky Cambridge, U.K. VE, Sturm M, Webb RS. 2001. Jenkins EJ, Veitch AM, Kutz SJ, The hydrologic cycle and its role Hobert EP, Polley L. 2006. Climate in Arctic and global environmental change and the epidemiology of change: a rationale and strategy for protogtrongylid nematodes in synthesis study. Fairbanks, Alaska: Arctic Research Consortium of the 108 northern ecosystems: Parela- phostrongylus odocoilei and U.S. 84 pp. Protostrongylus stilesi in Dall’s sheep (Ovis d. dalli). Parasitology 132:3:387-401. An arctic ground squirrel (Spermophilus parryii) emerges from his burrow in spring after hibernating for more than six months underground. Photo by Øivind Tien.

109 Chapter 7 Conclusions and Recommendations

Land and resource managers know that climate change will strongly affect the future landscape of America’s Arctic. Under a mid-range carbon emissions scenario, mean annual air temperature for the North Slope is projected to rise ~7°C by the end of the century, and precipitation is projected to increase by up to 50% in coastal areas. Despite the expected increase in precipitation, warmer summers are expected to result in significant drying.

110 Arctic aquatic and terrestrial The WildREACH workshop and habitats are shaped by the complex related discussions have identified interactions among climate, soils, important scientific priorities and permafrost, hydrology, and vegeta- activities required to improve our tion. Arctic habitats are among the ability to forecast—and thus most sensitive to warming because respond to—climate change effects of the pivotal role played by perma- on arctic fish and wildlife. frost, which can only persist at low temperature. Some arctic-adapted species will respond to habitat change through range shifts, but other species may be vulnerable to extirpation from the North Slope if suitable refugia are not available.

Climate change requires that managers of arctic species and landscapes anticipate the future condition of habitats and populations, but they cannot do this at present because too little is known about the linkages between climate, habitat, and organisms. Global Circulation Models, down-scaled to the North Slope region, are beginning to paint a clearer picture of the magnitude of climate change that may be expected within the century; these projections, however, are subject to large uncertainty. Our ability to translate climate change projections into specific forecasts of habitat condition is currently rudimentary. Indeed, it is not yet possible to link climate scenarios to fish and wildlife population status with the level of confidence A male spectacled eider (Somateria fischeri) near Barrow; the species is needed to support development and listed as threatened under the Endangered Species Act. Photo by Ted Swem, implementation of conservation USFWS. plans.

111

The Porcupine caribou herd (Rangifer tarandus) crosses the 1002 area of the Arctic National Wildlife Refuge. Photo by Chuck Young, USFWS. Workshop participants identified 1. Precipitation, Water Balance, and Scientifc important information gaps in our Distribution of Surface Water understanding of climate change a. How reliable are the projections Priorities effects on birds, mammals, and fish for increasing precipitation and populations. The specific gaps varied evapotranspiration? among species groups, but most b. How will the annual precipita- fell into four cross-cutting the- tion input on the Coastal Plain matic areas and underlying research and Foothills be allocated between questions (see Chapter 6 for more winter (snow pack) and summer? details): c. How will changes in precipitation, evapotranspiration, and active layer depth alter summer surface water availability in shallow-water and mesic/wet tundra habitats? d. How will changing patterns of seasonal runoff affect stream flow? e. What is the contribution of groundwater in various systems, and is it sufficient to maintain year- round flow? f. Will drought conditions and changes in drainage patterns decrease water body connectivity? g. Which Coastal Plain lakes are susceptible to tapping (rapid drainage) and on what time scale? h. What are the expected changes in snowpack characteristics (depth, density, presence of ice layers), and how might these vary on a regional and local scale? i. How much change will occur in the timing of snow melt and snow onset? j. How will the frequency of rain- on-snow and severe winter storm events change?

A pectoral sandpiper (Calidris melanotos) near Barrow. Photo by Ted Swem, USFWS.

112 2. Vegetation Community Composition 4. Coastal Dynamics and Phenology a. Will higher water temperatures, a. How will changes in the length sea level rise, and retreat of sum- and timing of the growing season mer sea ice cause degradation of influence plant phenology, includ- the barrier island systems of the ing seasonal changes in nutritional Beaufort and Chukchi seas? quality? b. Will alluvial deltas continue to b. How will plant species composi- build or will rising sea levels out- tion shift in response to long-term pace potential increases in sedi- climate change, and what are the mentation rates? implications for habitat structure c. How quickly will shoreline retreat and quality of the prevalent avail- result in newly breached lake able forage (i.e., digestibility, nutri- basins? ent content)? d. To what extent will coastal ero- c. What is the time scale of expected sion, in combination with sea level shrub increase, and how will this rise, cause salinization of low-lying vary by species/growth form (low coastal areas? vs. tall shrub) and ecoregion? e. Will coastal wet sedge meadows d. What is the likelihood of wide- establish at a rate equal to loss of spread conversion from sedge this habitat through erosion and and sedge-shrub meadow to bog inundation? meadow (paludification) and how f. Will increased fogginess/cloudi- would this affect herbivore and ness exert a negative or positive detritus-based trophic systems? feedback effect on air temperature e. How will changes in the seasonal- in the coastal zone? What is the ex- ity of stream discharge and oc- pected spatial extent of this effect? currence of flood events influence development of riparian vegetation Relevant to all of the four major communities? themes above, the study and modeling of hydrologic processes emerged 3. Abundance and Phenology of as critical to understanding habitat Invertebrates change. Furthermore, hydrologic a. How does earlier spring thaw af- data and models are needed by a fect timing of life cycle events and broad constituency of user groups, peak availability to predators? including engineers and infrastruc- b. How does temperature affect ture planners. growth and development of aquatic insects? c. What climate-related changes are likely in community composition of macroinvertebrates in stream, lake, and saturated soil environments? d. How will changes in the distribution and quality of surface waters and shifts from pelagic to benthic productivity in deep lakes affect availability of macroinvertebrates to fish and wildlife? e. How will warming and changing seasonality affect abundance and peak activity periods of biting insects, and what are the bioenergetic consequences for caribou in particular? f. How will warming and changing seasonality affect the prevalence of parasites and disease vectors (e.g., nematode parasites of muskoxen and Dall’s sheep)?

113 Ptarmigan (Lagopus sp.) are nomadic in winter, moving in flocks from one sheltered slope or patch of food to another from November to March. Photo by Øivind Tien. The scientific priorities previously 3. Modeling that dynamically Recommended listed should be addressed by activi- couples soil thermal and hydrologic ties including: monitoring of physi- regimes, and biological systems at Research cal and biological parameters; field appropriate spatial and temporal observations and focused research scales. activities designed to clarify the 4. Centralized data storage and in- Activities mechanisms by which environmental terpretation for the mutual benefit variables influence fish and wildlife of multiple end-users. populations; and development and refinement of arctic process models Development of Habitat Change that integrate physical and biological Models parameters. More specific research We recommend immediate attention activity recommendations are de- to developing habitat change models, tailed below. focusing initially on processes that are ongoing and occurring on a short Hydrologic Research (decadal) time scale. Priority topics Hydrologic data are sparse for the include: North Slope, principally due to the •Coastal processes (e.g., erosion, cost of acquisition. Recognizing that storm surge, deposition, succes- cost sharing among agencies and sion); others is necessary, we recommend •Seasonality (e.g., plant phenology, four initiatives in this area: animal migration, life stages of 1. Establishment of at least three aquatic invertebrates); long-term observatories on the •Shrub advance; North Slope to collect integrated •Fire regime (as a function of inter- hydrologic, climate, and geophysi- actions among climate, permafrost, cal data. The central mission of and vegetation); and these observatories should be to •Thermokarst effects on surface develop an understanding of the water and lake drainage. response of permafrost (active layer dynamics), hydrologic, and Selection of Indicator Species and ecological systems to changes in Parameters thermal regime. To ensure appli- Deficiencies in our knowledge of cability to fish and wildlife biology, basic natural history and ecology of water budgets should be estimated arctic species currently hinder our for key ecotypes. ability to assess responses to climate 2. Intensive observations at the change. We suggest choosing focal observatory sites should be supple- species that reflect sensitivity to mented by instrumentation (e.g., hypothesized climate change effects. meteorology, radiation, stream dis- The WildREACH working groups charge, soil moisture) at dispersed made progress toward this goal, but sites arrayed across important we recommend additional work in environmental gradients. this area. Specifically, the Service should engage the U.S. Geologi- cal Survey (USGS) and others in a structured decision-making process to refine the selection of indicator species/parameters as components of a long-term climate monitoring program. Upon reaching consensus, management agencies should seek stable funding for monitoring these species/attributes.

114

A flock of long-tailed ducks (Clangula hyemalis) migrating along the coast near Barrow. Photo by Ted Swem, USFWS. Climate change is transformative. quire financial commitment from a The challenge of conserving fish coalition of organizations and the Agency and wildlife populations in the face formation of teams of specialists of climate change already exceeds representing multiple disciplines. Coordination the capacity of any single entity. The Toward those ends, the U.S. Fish scale and scope of climate change and Wildlife Service is committed to demands change in the way that improving communication and col- and resource agencies do business—we laboration with the arctic research must pool already constrained hu- community. The Service will: Collaboration man and financial resources. The 1. Work with the National Science future of conservation lies in land- Foundation (NSF) to define priori- scape approaches that are developed ties for basic and applied research and implemented with other federal, programs targeted toward climate state, and non-governmental organi- research relevant to resource man- zation (NGO) partners. We recognize agement agencies; that there is no overarching business 2. Seek opportunities to collabo- model that fits all agencies’ man- rate with academic institutions to dates, but there is ample common frame grant proposals in ways that ground for agencies and NGOs to address the priorities of fish and engage in collaborative conserva- wildlife managers; tion. Conservation partnerships will 3. Participate in planning and imple- provide efficiencies through strate- mentation of the interagency Study gic targeting of financial resources of Environmental Arctic Change and provide an opportunity to co- (SEARCH) Program to ensure develop an adaptive, landscape-level inclusion of research relevant to response. resource management agencies; 4. Work with arctic science program Since the WildREACH workshop, managers in the research agencies the Service has promoted conserva- (e.g., National Science Foundation, tion partnerships across landscapes USGS, National Oceanic and Atmo- that align with Bird Conservation spheric Administration) to identify Regions (BCRs). These Landscape funding sources to address priority Conservation Regions (LCRs) will questions; and be supported by Landscape Con- 5. Promote a collaborative approach servation Cooperatives (LCCs) to acquire, process, archive, and that house the technical expertise disseminate essential satellite- from multiple partners to address based remote sensing data prod- the science needs of each LCR. The ucts (e.g., snow cover, green-up, Service has designated America’s and surface water) needed for Arctic as Alaska’s first LCR, affirm- regional-scale monitoring. ing the high priority to be placed on developing partnerships and The Service is currently working research plans and implementing with other federal and state man- conservation for the Arctic LCR. agers to develop mechanisms to The Arctic LCR will be supported by facilitate coordinated interagency the Northern Alaska LCC located planning, budget implementation, in Fairbanks. The Northern Alaska and data-sharing on a statewide LCC will work with USGS, through basis via the Climate Change its National Climate Change and Roundtable. The State of Alaska’s Wildlife Science Center and/or Governor’s subcabinet (http://www. Regional Climate Science Hub, climatechange.alaska.gov/) is also and other partners to establish the working to identify priority tasks technical and organizational capacity to address mitigation and adapta- to address the science needs of the tion within the context of social and Arctic LCR. economic factors. As we designate LCRs and build the LCCs that sup- Engaging in the broad scope of port them, resource management scientific inquiry recommended by agencies, research institutions, and workshop participants will re- NGOs must implement interagency

115 agreements that commit resources and obligation, however, to lay the to coordinated long-term research foundation of information needed to and monitoring, biological planning, inform future management deci- and conservation design and deliv- sions. The WildREACH workshop ery. was successful in identifying priority information gaps and activities Climate change presents an un- needed to provide the basis for precedented challenge to managers adaptive management of arctic fish of arctic natural resources. The and wildlife resources. Adopting complexity with which interacting these recommendations will greatly systems may respond to chang- strengthen our capacity to form ing climate forces us to accept an anticipatory responses to climate- uncomfortable level of uncertainty, related habitat change, and will and the time scales of change are help us identify the most promising long relative to the normal manage- strategies to protect arctic fish and ment cycle. We have the opportunity wildlife populations.

116

Candle ice (small, icicle-shaped pieces of lake ice) is rapidly blown to shore on a lake on the Ikpikpuk Delta in July, crumpling the soft lake margin. It took only 4–5 minutes for the ice to come ashore. Photo by Leslie Pierce (TREC 2005), courtesy of ARCUS. Climate Data Sets analysis to only the best performing Temperature and precipitation models is one approach to narrow- Appendix 1: projections were based on the ing the uncertainty of the projected downscaled General Circulation climate change (Walsh et al. 2008). Methods for Model (GCM) output distributed by the University of Alaska Sce- Each model run is based on the narios Network for Alaska Planning Intergovernmental Panel on Cli- Climate (SNAP; http://www.snap.uaf.edu/ mate Change (IPCC) A1B green- about), based on the work of Walsh house gas emission scenario (IPCC Projections et al. (2008). Output from each model 2007), which assumes fast economic contains spatially explicit mean growth, an eventual decline in global temperature (°C) and total precipita- population, development of more tion (mm) grids for every month of efficient technology, and the contin- every year for 120 years (1980–2099) ued use of fossil fuels for some, but for the entire state of Alaska. These not all, energy sources. The A1B data were downloaded to the De- scenario is considered a “middle- partment of Atmospheric Sciences of-the-road” scenario among those at the University of Illinois from presented by the IPCC. The A1B, the World Climate Research Pro- B1 (slow greenhouse gas increase), gramme (WCRP ) Coupled Model and A2 (more rapid greenhouse gas Intercomparison Project phase 3 increase) scenarios differ little by (CMIP3) multi-model dataset. The mid-century but diverge by 2100. five models used were: The projected climate changes • ECHAM5/MPI from the Max are generally proportional to the Planck Institute for Meteorology in respective increases of greenhouse Germany; gas concentrations, so the B1 and • GFDL-CM2.1 produced by the A2 changes of climate can be scaled National Oceanic and Atmospheric to the A1B changes. The projec- Administration (NOAA) Geophysi- tions used in this report are derived cal Fluid Dynamics Laboratory, a from a composite of the five model unit of the Department of Com- outputs. merce, in the U.S.; • MIROC3.2MedRes produced Baseline temperature and precipi- by the Center for Climate Sys- tation data were derived from the tem Research at the University Parameter-Elevation Regression on of Tokyo, the National Institute Independent Slopes Model (PRISM) for Environmental Studies, and dataset created by the PRISM the Frontier Research Center for Group (Oregon State University, Global Change, a unit of the Japan www.prism.oregonstate.edu). These Agency for Marine-Earth Science data consist of 12 gridded mean and Technology (JAMSTEC), in maximum temperature, mean Japan; minimum temperature, and total • UKMO-HadCM3 produced by the precipitation files at 2-km resolution, Hadley Centre for Climate Predic- one for each month averaged over tion and Research and the Met 1961–1990 for the state of Alaska. Office in the U.K.; and This dataset was created using ob- • CGCM3.1 produced by the Cana- servation data from weather stations dian Centre for Climate Modelling across Alaska and spatially interpo- and Analysis in Canada. lated over intervening areas using weighted regression incorporating More documentation on each model elevation and terrain effects on cli- is available at the CMIP3 website mate (Daly et al. 2002, Simpson et al. (http://www-pcmdi.llnl.gov/ipcc/mod- 2005). Walsh et al. (2008) calculated el_documentation/ipcc_model_docu- GCM baseline values for each of mentation.php). Walsh et al. (2008) the five selected models using mean objectively evaluated 15 models over monthly outputs for 1961–1990. They Alaska—the above five models were then calculated differences between ranked highest and provided output projected GCM values and baseline that most closely matched actual GCM values for each year and cre- Alaska climate data for the years ated “anomaly grids” representing 1958–2000 (SNAP; http://www.snap. these differences. The anomaly grids uaf.edu/about). The best perform- were added to the PRISM baseline ing models in that analysis tend to to create fine-scale (2-km) grids for 117 be the models that project the most monthly mean temperature and pre- change in climate. Limiting the cipitation projections out to 2099. Temperature and Precipitation Freeze-Up, Thaw, and the Frost-Free Projections Season Projections The Nature Conservancy calcu- The Wilderness Society generated lated the projected decadal mean data for onset of freeze-up, thaw, for each month of the year, for and length of frost-free season. Data both temperature and precipita- were analyzed using the open source tion. Monthly values were averaged application R (R Development Core (for temperature) or summed (for Team 2005) and ArcGIS 9.3 (ESRI, precipitation) across months to yield Redlands, California). Climate data seasonal and annual values. Mid- were monthly means and totals, but century and end-of-century intervals freeze-up and spring thaw dates are were selected to represent projected often variable within days; therefore, change. Mid-century (2051–2060) a ramp function was used to linearly and end-of-century (2091–2100) interpolate the day of freeze-up and grids were generated by calculating thaw for the 1980–2099 composite the differences between projected (Euskirchen et al. 2007). A 12 x 120 values from the PRISM baseline matrix (12 months, 120 years) of value (temperature) or the percent spatial mean values was developed change from the PRISM baseline and assumed to represent the value value (precipitation). A spatial mean for the 15th day of that month. The was then calculated to produce the slope between these values was cal- summaries by ecoregion. For this culated and added to each progres- document, summer is defined as sive day beyond the 15th until the 15th June through September and winter of the next month was reached. This as October through May. procedure was repeated for each month. For the purposes of these analyses, freeze-up was defined as the first day of the year when the interpolated value fell below 0°C, and thaw was defined as the first day above 0°C. The number of days between these two dates was defined as the frost-free season, calculated for all 120 years and plotted over time. Rate of change over time was calculated using a linear regression. This analysis is not indicative of the actual dates of freeze-up and thaw or the natural variability surrounding them (e.g., several freezing and unfreezing events may occur), however, it does give an indication of how much these actual dates may be delayed or advanced under each scenario of climate change.

118 Potential Evapotranspiration and Water Sparse cloud cover data made it Availability Projections necessary to interpolate regional The Wildness Society calculated values. Cloud cover data for the potential evapotranspiration (PET) Coastal Plain are averages of ACRC for each of the three North Slope weather station data from Barrow ecoregions using a modified ver- and Barter Island (the only stations sion of the Priestly-Taylor method available for the region). Values for adapted to a monthly time step using the Brooks Range are assumed to energy balance equations from Allen be the same as those measured in et al. (1998). The two time slices used Bettles (the closest high-altitude for PET projections are different station available). Because none of than those used for temperature the weather stations located within and precipitation and are centered the Arctic Foothills contained cloud around 2035–2044 and 2075–2084. cover measurements, cloud cover Climate variable inputs included data for this ecoregion is repre-

mean monthly temperature (Tmean), sented by an average of value from mean monthly minimum tempera- Barrow, Barter Island, and Bettles.

ture, (Tmin), mean monthly maximum temperature (Tmax), and percent Monthly estimates of PET were sub- cloud cover. Spatial averages for tracted from total monthly precipita- each ecoregion were derived from tion (P) values for each time period the baseline PRISM dataset for each to estimate changes in the mean ecoregion. Mean monthly minimum growing season water balance (P- and maximum temperature grids PET) for each ecoregion over time. were created by subtracting the These analyses were restricted to corresponding PRISM Tmean grids months with average temperatures from the projected Tmean grids to exceeding 0°C and did not include obtain monthly anomalies from the the additional days expected to con- baseline PRISM climatology. These tribute to a longer growing season as anomaly grids were then added to determined by the separate frost-

the PRISM Tmax temperature grids free season regression analyses. and subtracted from the PRISM Tmin grids to create 12 projected Tmax and Literature Cited 12 projected Tmin temperature grids for the two time periods. For months Allen RG, Perieira LS, Raes D, with mean temperatures below 0°C, Smith M. 1998. Crop evaptranspi- PET is assumed to be zero. Mean ration: Guidelines for computing cloudiness, as percent sky cover, for crop water requirements. FAO each ecoregion was estimated using Irrigation and Drainage Paper 56, station data available at the Alaska Rome, Italy. Climate Research Center (ACRC) website (http://climate.gi.alaska.edu/ Daly, C, Gibson WP, Taylor GH, Climate/index.html) and assumed Johnson GL, Pasteris P. 2002. A to remain constant throughout knowledge-based approach to the future projections. Additional inputs statistical mapping of climate. Clim included ordinal date and latitude. Res 22: 99-113. The 15th of January served as an initial ordinal date and progressed Euskirchen ES, McGuire AD, through the year at a 30-day time Chapin FS. 2007. Energy feed- step. For example, the ordinal date backs of northern high-latitude used for the month of February ecosystems to the climate system was the 45th, and for August it was due to reduced snow cover dur- the 225th. The latitude coordinates ing 20th century warming. Global used for the Arctic Coastal Plain, Change Biol 13: 2425-2438. Arctic Foothills, and Brooks Range were 70.37°N, 69.42°N, and 68.80°N, Simpson JJ, Hufford GL, Daly respectively. C, Berg JS, Fleming MD. 2005. Comparing maps of mean monthly surface temperature and precipita- Growing season months (Tmean >0°C) used for the Coastal Plain and Foot- tion for Alaska and adjacent areas hills analysis remained consistent of Canada produced by two differ- throughout future projections, span- ent methods. Arctic 58: 137-161. ning the months of June through 119 September. The growing season in Walsh JE, Chapman WL, Ro- the Brooks Range expanded, howev- manovsky V. 2008. Global climate er, from the historical June–August model performance over Alaska to include September by 2035–2044 and Greenland. J. Climate 21:6156- and May by 2075–2084. 6174. Appendix 2: Description and Distribution of Northern Alaska Ecotypes

Terrestrial habitat types adapted from the ecotypes described by Jorgenson and Heiner (2003) are described in Table A2.1. The distribution of ecotypes is depicted in Figure A2.1.

Table A2.1. Description of terrestrial habitat types found in northern Alaska. From Jorgenson and Heiner (2003).

Class Description ALPINE ECOTYPES Perennially frozen snow and ice at high elevations in the Brooks Range, typically on north- Alpine Glaciers facing slopes. Barren (<5% plant cover) to partially vegetated (5–30%) areas on non-carbonate bedrock and talus slopes above treeline in the Brooks Range. Bedrock includes felsic intrusive (e.g., granite, granodiorite), non-carbonate metamorphic (e.g., slate, schist), and non-carbonated sedimentary (e.g., conglomerate, sandstone, shale) rocks that generally have low calcium and sodium and Alpine Non-carbonate high aluminum concentrations that lead to acidic soils. Soils are rocky, excessively drained, lack- Barrens ing in surface organic accumulations, and strongly acidic (pH <5.5). At high elevations, common species include Geum glaciale, Saxifraga bronchialis, S. flagellaris, S. nivalis, S. eschscholtzii, and crustose and fruticose lichens. Lower elevations have species similar to Alpine Non-carbonate Dwarf Shrub Tundra. Barren (<5% plant cover) to partially vegetated (5–30%) areas on carbonate bedrock and talus slopes above treeline in the Brooks Range. Bedrock includes both sedimentary (limestone, do- Alpine Carbonate lostone) and metamorphic (marble) carbonate rocks. Soils are rocky, excessively drained, lacking Barrens in surface organics, and alkaline (pH >7.3). Common pioneering plants include Dryas integrifolia, D. octopetala, Saxifraga oppositifolia, Potentilla uniflora, Oxytropis nigrescens, O. arctica, and Carex rupestris. Barren areas on intermediate, mafic, and ultramafic plutonic rocks above treeline in the Brooks Range that typically have dark-colored mineral assemblages with abundant iron and magne- Alpine Mafic Barrens sium. Soils are rocky, excessively drained, lacking in surface organic accumulations, and are neutral to alkaline. Some areas have high levels of trace metals. Areas usually are devoid of vegetation. Areas on noncarbonate bedrock and talus slopes above treeline in the Brooks Range with dwarf shrub vegetation. Soils are rocky, excessively drained, have very thin surface organic accumulations, and are strongly acidic. Vegetation is dominated by dwarf shrubs including Dryas Alpine Non-carbonate octopetala (mostly south slopes), Salix phlebophylla, S. arctica, Loiseleuria procumbens, Diap- Dwarf Shrub Tundra ensia lapponica, Arctostaphylos alpina, Empetrum nigrum, Vaccinium uliginosum, and Cas- siope tetragona (north slopes). Other species include Carex podocarpa, C. bigelowii, Hierochloe alpina, Cladina mitis, C. rangiferina, and Rhizocarpon geographicum. Areas on carbonate bedrock and talus slopes above treeline in the Brooks Range with dwarf shrub vegetation. Soils are rocky, excessively drained, rich in humus, and alkaline. Vegetation is dominated by dwarf shrubs including Dryas integrifolia (mostly south slopes), D. octopetala, Alpine Carbonate Cassiope tetragona (north slopes), Salix arctica, and Arctostaphylos alpina. Other species Dwarf Shrub Tundra include Carex rupestris, C. bigelowii, Saxifraga oppositifolia, Potentilla uniflora, Oxytropis nigrescens, O. arctica, Nephroma arcticum, Rhytidium rugosum, Flavocetraria cucullata, and Thamnolia vermicularis. Areas on intermediate, mafic, and ultramafic plutonic rocks above treeline in the Brooks Range with dwarf shrub vegetation. Rocks have dark-colored mineral assemblages with abundant iron Alpine Mafic and magnesium. Soils are rocky, excessively drained, lacking in surface organic accumulations, Dwarf Shrub Tundra and are neutral to alkaline. Some areas have high levels of trace metals. Vegetation is poorly described for this type but it probably is similar to that described for Alpine Non-carbonate Dwarf Shrub Tundra. 120 Continued page 122 121 Figure A2.1. Distribution of ecotypes in northern Alaska (Jorgenson and Heiner 2003). Table A2.1. Continued Class Description UPLAND ECOTYPES Upland areas on mid- to upper slopes on weathered bedrock, colluvium, and glacial till with vegetation dominated by needleleaf trees. Soils are loamy to rocky, well-drained, have moderately thick organic horizons, are acidic, and may or may not have permafrost. This late-successional forest is dominated by an open to closed canopy of Picea glauca, but can include minor amounts Upland Spruce Forest of Betula papyrifera and P. mariana. Understory plants include Alnus crispa, Vaccinium vitis-idaea, Ledum groenlandicum, Empetrum nigrum, Rosa acicularis, Cornus canadensis, Shepherdia canadensis, Spiraea beauverdiana, Linnaea borealis, Calamagrostis canadensis, Hylocomium splendens, and Pleurozium schreberi. Upland areas on mid- to upper slopes on weathered bedrock, colluvium, and glacial till with vegetation co-dominated by broadleaf and needleleaf trees. Soils are well-drained, have thin organic horizons, are moderately acidic, and usually lack permafrost. This mid-successional mixed forest Upland Birch-Aspen- is dominated by an open to closed canopy of Betula papyrifera, Populus tremuloides, and Picea Spruce Forest glauca. Understory plants include Alnus crispa, Salix glauca, Vaccinium vitis-idaea, Ledum groenlandicum, Rosa acicularis, Cornus canadensis, Shepherdia canadensis, Linnaea borealis, Calamagrostis canadensis, and feathermosses. Upland areas on mid- to upper slopes on weathered bedrock, colluvium, and glacial till with vegetation dominated by broadleaf deciduous trees. Soils are loamy to rocky, well-drained, have thin organic horizons, are acidic, and usually lack permafrost. The mid-successional forest is Upland Birch-Aspen dominated by an open to closed canopy of Betula papyrifera and Populus tremuloides. Under- Forest story plants include Alnus crispa, Salix glauca, Vaccinium vitis-idaea, Ledum groenlandicum, Rosa acicularis, Cornus canadensis, Shepherdia canadensis, Spiraea beauverdiana, Linnaea borealis, Calamagrostis canadensis, and feathermosses. Upland areas on mid- to upper slopes on weathered bedrock, colluvium, and glacial till with vegetation dominated by tall shrubs. Soils are loamy to rocky, well-drained, have thin organic horizons, are acidic, and usually lack permafrost. Vegetation is dominated by an open to closed canopy of Alnus crispa, although Salix pulchra, Salix glauca, and Betula glandulosa occasion- Upland Tall Alder Shrub ally are abundant. Understory species include Vaccinium uliginosum, Vaccinium vitis-idaea, Betula nana, B. glandulosa, Ledum groenlandicum, Empetrum nigrum, Equisetum arvense, Spiraea beauverdiana, Calamagrostis canadensis, and Petasites frigidus. Mosses include Sphagnum spp., Hylocomium splendens, and Dicranum spp. Upland areas on mid- to upper slopes on weathered bedrock, colluvium, and glacial till with vegetation dominated by low shrubs. Soils are loamy to rocky, well-drained, have moderately thick organic horizons, are acidic, and usually have permafrost. Vegetation has an open to closed Upland Low Birch- canopy of Betula nana and/or Salix pulchra. Other species include Salix glauca, Vaccinium Willow Shrub Tundra uliginosum, V. vitis-idaea, Ledum decumbens, Empetrum nigrum, Arctostaphylos alpina, Dryas octopetala, D. integrifolia, Salix reticulata, Equisetum arvense, Carex bigelowii, and the mosses and lichens Hylocomium splendens, Tomentypnum nitens, Sphagnum spp., Aulacom- nium palustre, Dicranum spp., Cladina rangiferina, and Flavocetraria cucullata. Upland windswept ridges and upper slopes on weathered bedrock, colluvium, inactive sand dunes, and coastal plain deposits with vegetation dominated by dwarf shrubs. Soils are well- drained, loamy to rocky, have thin organic horizons, and are circumneutral to acidic. Common dwarf shrubs include Dryas octopetala (mostly south slopes), D. integrifolia, Salix phlebo- Upland Dryas Dwarf phylla, S. arctica, S. reticulata, Loiseleuria procumbens, Diapensia lapponica, Arctostaphylos Shrub Tundra alpina, Empetrum nigrum, Vaccinium uliginosum, Ledum decumbens, and Cassiope tetragona (north slopes). Other common species include Carex bigelowii, C. scirpoidea, Arctagrostis latifolia, Equisetum variegatum, Tomentypnum nitens, Hylocomium splendens, and Cladina stellaris. Gently sloping uplands and ridges on loess and colluvium over bedrock and glacial till, primarily within the Brooks Foothills (>120 m elevation), with vegetation co-dominated by tussock- forming sedges and low shrubs. Soils are somewhat poorly drained, loamy, have moderately thick surface organics, are acidic, and are underlain by ice-rich permafrost. The open low shrub Upland Shrubby canopy of Betula nana and Salix pulchra usually overtop the Eriophorum vaginatum tussocks. Tussock Tundra Other dominant plants include E. angustifolium, Carex bigelowii, Ledum decumbens, Vac- cinium vitis-idaea, V. uliginosum, Rubus chamaemorus, Hylocomium splendens, Sphagnum spp., Aulacomnium palustre, Cladina rangiferina, C. arbuscula, C. mitis, and Flavocetraria cucullata. Gently sloping uplands and ridges on loess, colluvium, and coastal plain deposits, primarily within the Beaufort Coastal Plain (<120 m elevation), with vegetation dominated by tussock-forming sedges. Soils are moist, somewhat poorly drained, loamy, and have moderately thick surface organics, are circumneutral to acidic, and are underlain by ice-rich permafrost. Vegetation is 2 122 Upland Tussock Tundra dominated by Eriophorum vaginatum. On circumneutral soils, Carex bigelowii, Dryas integrifolia, Salix pulchra, Cassiope tetragona, S. reticulata, Tomentypnum nitens, and Hylocomium splendens are common. On acidic soils, dominant plants include E. angustifolium, Betula nana, Salix pulchra, Ledum decumbens, Vaccinium vitis-idaea, Rubus chamaemorus, Hylocomium splendens, Sphagnum spp., Aulacomnium palustre, and Cladina rangiferina. Class Description Upland ridges and upper slopes on weathered bedrock, loess-mantled bedrock, colluvium, and glacial till, with vegetation co-dominated by sedges and low and dwarf shrubs. Soils are loamy to rocky, somewhat poorly drained, have moderately thick surface organics, and are alkaline to acidic depending on substratum. On acidic soils more common in the upper foothills and moun- Upland Moist Sedge- tains, dominant plants include Betula nana, Salix pulchra, Carex aquatilis, Eriophorum an- Shrub Tundra gustifolium, and Sphagnum spp. On circumneutral to alkaline soils more common on the coastal plain and lower foothills, dominant plants include Salix lanata richardsonii, Dryas integrifolia, S. reticulata, Arctostaphylos rubra, Rhododendron lapponicum, Equisetum arvense, Carex bigelowii, Tomentypnum nitens, and Thamnolia vermicularis. LOWLAND ECOTYPES Low-lying flats and gentle slopes on colluvium and abandoned floodplains with vegetation dominated by needleleaf forests. Soils are wet, somewhat poorly drained, have moderately thick surface organics, are acidic, and usually are underlain by permafrost. The open tree canopy (usually 5–10 m high) is dominated by Picea mariana, although P. glauca, Larix laricina, and Lowland Spruce Betula papyrifera occasionally can be present in small amounts. In the wettest areas the trees Forest can be very stunted. Common understory plants include Salix pulchra, Betula nana, Vaccinium uliginosum, Ledum groenlandicum, Potentilla fruticosa, Rubus chamaemorus, Equisetum arvense, and Carex bigelowii. Mosses and lichens include Sphagnum spp., Hylocomium splendens, Pleurozium schreberi, Cladonia spp., Nephroma spp., Cetraria spp., and Peltigera spp. Low-lying flats and lower slopes on drained lake basins, abandoned floodplains, colluvium, and coastal plain deposits with vegetation dominated by low shrubs. Soils typically are poorly drained, loamy, have moderately thick surface organics, are acidic, and are underlain by permafrost. The open to closed low shrub canopy is dominated by Salix pulchra and Betula nana. Lowland Low Birch- On acidic soils other common species include Ledum decumbens, Vaccinium uliginosum, V. Willow Shrub Tundra vitis-idaea, Empetrum nigrum, Petasites frigidus, Rubus chamaemorus, Eriophorum angustifolium, Carex aquatilis, Calamagrostis canadensis, and Sphagnum spp. On circumneutral to alkaline soils, Salix lanata richardsonii, S. reticulata, Dryas integrifolia, Arctostaphylos rubra, Equisetum arvense, Eriophorum angustifolium, and Carex aquatilis are common. Low-lying flats and gentle slopes on drained lake basins, abandoned floodplains, colluvium, and coastal plain deposits, particularly on the Beaufort Coastal Plain, with vegetation co-dominated by sedges and low or dwarf shrubs. Soils are saturated at intermediate depths (>15 cm), loamy with moderately thick surface organics, are circumneutral to alkaline, and are underlain by ice- Lowland Moist Sedge- rich permafrost. Sites generally are free of surface water during summer. Vegetation is dominat- Shrub Tundra ed by Carex aquatilis, C. bigelowii, Eriophorum angustifolium, and Dryas integrifolia. Other common species include Salix lanata richardsonii, S. pulchra, S. reticulata, Tomentypnum nitens, and Hylocomium splendens. Acidic vegetation could not be adequately differentiated from non-acidic vegetation on the Beaufort Coastal Plain. Low-lying flats and drainages on drained lake basins, abandoned floodplains, colluvium, and coastal plain deposits, particularly on the Beaufort Coastal Plain, with vegetation dominated by sedges. Soils are poorly drained, have moderately thick to thick (10–50 cm) surface organics over silt loam, usually circumneutral, and are underlain by ice-rich permafrost. Ice wedge development in older landscapes creates distinctive low-centered polygons. The surface gener- Lowland Wet Sedge ally is flooded during early summer (depth <0.3 m) and drains later, but soils remain saturated Tundra ≥15 cm from the surface throughout the growing season. Vegetation is dominated by Carex aquatilis and Eriophorum angustifolium, while willows, including Salix lanata richardsonii and S. pulchra, often are present but usually not co-dominant. Other common species include Dryas integrifolia, S. reticulata, C. bigelowii, and Equisetum scirpoides on higher microsites and polygon rims. Shallow (<1.5 m) ponds and deep (≥1.5 m) lakes resulting from thawing of ice-rich permafrost, primarily on the coastal plain and distal portions of abandoned floodplains. In shallow ponds, water freezes to the bottom during winter, thaws by early to mid-June, and is warmer than Lowland Lake water in deep lakes. In deep lakes, water does not freeze to the bottom during winter in deeper portions of the lake. Sediments are loamy to sandy. These lakes lack riverine influences (flooding), but they may have distinct outlets or connections to rivers.

Continued next page

123 Table A2.1. Continued

Class Description LACUSTRINE ECOTYPES Barren or partially vegetated (<30% cover) areas on newly exposed sediments in recently drained lake basins. The surface form generally is nonpatterned due to the lack of ice wedge Lacustrine Barrens (not development. Soils are saturated to well-drained, sandy to loamy, lack surface organics, and are mapped) alkaline. Typical colonizers are Arctophila fulva, Carex aquatilis, Dupontia fisheri, Scorpidium scorpioides, and Calliergon spp. on wet sites and Poa alpigena, Senecio congestus, Salix ovalifolia, and Salix arctica on drier sites. Shallow (<1 m), permanent waterbodies with emergent aquatic sedges and grasses. Water and bottom sediments freeze completely during winter, but the ice melts in early June. The sediments range from sands to organics (10–50 cm deep) overlying silt loam. In deeper water (30– Lacustrine Marsh (not 100 cm), Arctophila fulva can form sparse to dense stands and is the predominant vegetation. In mapped) shallower (<30 cm) water, Carex aquatilis and Eriophorum angustifolium are dominant, and Utricularia vulgaris is common. This ecosystem type is important to waterbirds but could not be mapped separately and is included in both Lowland Lakes and Lowland Wet Sedge Tundra. RIVERINE ECOTYPES Flat areas on inactive floodplains subject to infrequent flooding with vegetation dominated by needleleaf trees. The late-successional forest has an open to closed tree canopy dominated by Picea glauca. Soils are well-drained, loamy to gravelly, have moderately thick surface organics, Riverine Spruce and are acidic. The understory is dominated by Alnus crispa, Vaccinium uliginosum, V. vitis- Forest idaea, Arctostaphylos rubra, Cornus canadensis, Viburnum edule, Rosa acicularis, Mertensia paniculata, and feathermosses (Hylocomium splendens, Rhytidiadelphus triquetrus, and Pleurozium schreberi). Flat areas on inactive floodplains subject to infrequent flooding with mixed forests co-dominated by needleleaf and broadleaf trees. The mid-successional forests have an open to closed tree canopy dominated by Picea glauca and Populus balsamifera. Soils are well-drained, loamy to Riverine Spruce- gravelly, have moderately thick surface organics, and are circumneutral to acidic. The understo- Balsam Poplar Forest ry is dominated by Alnus crispa, Vaccinium uliginosum, V. vitis-idaea, Arctostaphylos rubra, Cornus canadensis, Viburnum edule, Rosa acicularis, Equisetum arvense, Epilobium angustifolium, Calamagrostis canadensis, and feathermosses (Hylocomium splendens, Rhytidiadel- phus triquetrus, and Pleurozium schreberi). Flat areas on inactive floodplains subject to infrequent flooding and that have vegetation dominated by broadleaf forests. Soils are well-drained, loamy to gravelly, have thin surface organics, Riverine Balsam and are circumneutral. The mid-successional forest has an open to closed canopy dominated Poplar Forest by Populus balsamifera or occasionally Betula papyrifera. The understory has Alnus crispa, Rosa acicularis, Equisetum arvense, Epilobium angustifolium, Hedysarum alpinum, Cala- magrostis canadensis, Galium boreale, and Rhytidiadelphus triquetrus. Flat areas on active floodplains subject to frequent flooding that have vegetation dominated by tall shrubs in the boreal region. Soils are well-drained, loamy to gravelly, have very thin surface Riverine Tall Alder- organics, and are circumneutral. The early succession community has an open to closed tall Willow Shrub shrub canopy dominated by Salix alaxensis, S. arbusculoides, S. monticola, and Alnus crispa. The understory is dominated by Vaccinium uliginosum, Artemisia tilesii, Calamagrostis canadensis, Petasites frigidus, and Equisetum arvense. Mosses and lichens are not abundant. Flat to gently sloping areas on active and inactive floodplains in arctic regions subject to variable flooding frequency and that have vegetation dominated by tall and low shrubs. On the narrow zone close to the river, soils are frequently flooded, well-drained, lack organic accumulations, and have vegetation dominated by open tall (>1.5 m) Salix alaxensis, S. arbusculoides, Riverine Low Willow and S. glauca. Alnus crispa is uncommon. In the understory, Equisetum arvense, Astragalus Shrub Tundra alpinus, Aster sibericus, and Festuca rubra are common. On inactive floodplains, where soils have interbedded organic layers and are seasonally saturated, Salix lanata richardsonii and S. pulchra are dominant. Common understory species include Salix reticulata, Arctostaphylos rubra, Dryas integrifolia, Arctagrostis latifolia, Equisetum spp., legumes, Tomentypnum nitens, and other mosses. Flat areas on inactive floodplains subject to infrequent flooding and that have vegetation dominated by dwarf shrubs. Soils are well-drained, sandy to rocky, have thin surface organics, Riverine Dryas Dwarf are alkaline, and are underlain by ice-poor permafrost. The dwarf shrub Dryas integrifolia is Shrub Tundra dominant, and Salix reticulata, S. lanata richardsonii, Carex bigelowii, Arctagrostis latifolia, Astragalus spp., Oxytropis deflexa, and Equisetum scirpoides are common. Tomentypnum nitens and Distichium capillaceum are common mosses. Flat areas on inactive floodplains subject to infrequent flooding and that have vegetation co- dominated by sedges and low and/or dwarf shrubs. Soils are moderately well-drained, loamy, 124 Riverine Moist Sedge- have moderately thick surface organics, are circumneutral and underlain by ice-rich permafrost. Shrub Tundra Vegetation is dominated by Carex aquatilis and Eriophorum angustifolium with Dryas integrifolia, Salix lanata richardsonii, S. reticulata, and Carex bigelowii, Equisetum spp., Tomen- thypnum nitens, and Campylium stellatum as common associates. Class Description Flat areas on active and inactive floodplains subject to frequent or infrequent flooding and that have vegetation dominated by sedges. Soils are poorly drained, loamy with moderately thick to thick surface organics, are circumneutral to alkaline, and are underlain by ice-rich perma- Riverine Wet Sedge frost. Surface forms vary from nonpatterned to low-relief, low-centered polygons; the latter are Tundra indicative of progressive ice wedge development. Vegetation is dominated by Carex aquatilis and Eriophorum angustifolium, although occasionally the willow Salix lanata richardsonii is a co-dominant. Other species include Dupontia fisheri, Equisetum variegatum, Pedicularis sudetica, Campylium stellatum, Scorpidium scorpioides, and Limprichtia revolvens. Shallow waterbodies (0.1–1.0 m) on active and inactive floodplains subject to occasional flooding with vegetation dominated by emergent aquatic grasses and sedges. Due to shallow water Riverine Marsh depths, the water freezes to the bottom in the winter, and the ice melts by early June. Arctoph- (not mapped) ila fulva usually is found in deeper water while Carex aquatilis is usually found in very shallow water. Hippuris vulgaris occasionally is present. Barren or partially vegetated (<30% cover) areas on active river channel deposits associated with meandering or braided rivers. Frequent sedimentation and scouring restricts establishment and growth of vegetation. Soils are poorly to excessively drained, sandy to gravelly, lack Riverine Barrens surface organics, are alkaline, and usually have ice-poor permafrost in arctic regions and lack permafrost is boreal regions. Typical pioneer plants include Salix alaxensis, Deschampsia caespitosa, Chrysanthemum bipinnatum, Epilobium latifolium, Artemisia arctica, Festuca rubra, Arctagrostis latifolia, and Trisetum spicatum. Permanently flooded channels of freshwater rivers and streams, and lakes on inactive floodplains that are subject to occasional flooding. Some stream water flows throughout the year. Peak flooding generally occurs during spring breakup, and the lowest water levels occur during Riverine Waters mid-summer. Riverbed materials can be either sand or gravel. Shallow (<1.5 m) or deep lakes usually are associated with old river channels, point bars, and meander scrolls, although some result from thawing of ice-rich permafrost on large floodplains. Some may have connecting channels that flood during high water. Shorelines usually are smooth (lack polygonization). COASTAL ECOTYPES Low-lying, salt-affected areas along the coast with vegetation dominated by either grasses or dwarf shrubs. Soils are well-drained, slightly saline, and alkaline. This class includes three vegetation types. On active dunes and beaches, vegetation includes Elymus arenarius, Chry- santhemum bipinnatum, Puccinellia spp., Artemisia tilesii, and Salix ovalifolia. Well-drained inactive tidal flats dominated by dwarf shrub vegetation have S. ovalifolia, Stellaria humifusa, Coastal Grass and Dwarf E. arenarius, Deschampsia caespitosa, Dupontia fisheri, Carex subspathacea, and A. tilesii. Shrub Tundra Inactive dunes along the Chukchi Sea with slightly saline sandy soils have dwarf shrub vegetation dominated by Empetrum nigrum, S. ovalifolia, E. arenarius, Lathyrus maritimus, C. bipinnatum, and lichens. Substantial areas of this mapped class would have been more accurately mapped as Lowland Moist Sedge-Shrub Tundra but could not be adequately differentiated spectrally or by modeling. Low-lying, salt-affected areas on tidal flats, deltas, and muddy beaches along the coast that are frequently flooded and have vegetation dominated by sedges. The surface is nonpatterned. Soils are poorly drained, clayey to loamy, usually lack surface organics, and are brackish and alkaline. Coastal Wet Sedge The soils are underlain by ice-poor permafrost. Vegetation is dominated by Carex subspathacea, Tundra Carex ursina, and Puccinellia phryganodes, with Dupontia fisheri, Puccinellia andersonii, Cochlearia officinalis, and Stellaria humifusa also common. Non-vascular plants usually are absent. Substantial areas of Lowland Wet Sedge Tundra are included in these mapped areas but could not be adequately differentiated. Barren or partially vegetated, low-lying, salt-affected areas on tidal flats, deltas, and muddy beaches along the coast that are frequently flooded. Soils are poorly drained, clayey to loamy, usually lack surface organics, and are brackish and acidic to alkaline. The soils are underlain by ice-poor permafrost. Common colonizing plants include Deschampsia caespitosa, Elymus arenarius, Salix ovalifolia, and Stellaria humifusa in well-drained areas, and Puccinellia phry- Coastal Barrens ganodes, Dupontia fisheri, and Carex subspathacea in wetter areas. This class also includes tundra that has been killed by saltwater intrusions from storm surges and is being colonized by salt-tolerant plants. Newly deposited sediments typically are found on top of a thick organic horizon. These areas have low pH, high salinity, and shallow thaw depths. Common colonizing plants include Puccinellia phryganodes, Stellaria humifusa, Cochlearia officinalis, and Salix ovalifolia. Shallow (~<2 m) estuaries, lagoons, embayments, and tidal ponds along the coast of the Beau- fort and Chukchi seas. Winds, tides, river discharge, and icing create dynamic changes in physical and chemical characteristics. Salinity ranges widely from nearly fresh near rivers to saline in Coastal Water unprotected areas. Tidal ranges normally are small (<0.2 m) along the Beaufort and moderate (0.5–1 m) along the Chukchi seas, but storm surges produced by winds may raise sea level as 125 much as 2–3 m. Bottom sediments are mostly unconsolidated mud and sand. The ice-free period extends from July until October. Winter freezing generally begins in late September.

Continued next page Table A2.1. Continued

Class Description OTHER ECOTYPES Deep (~>2 m) marine waters of the Beaufort and Chukchi seas outside of lagoons and barrier Marine Water islands. Ice coverage is highly variable from permanent pack ice to seasonally ice free areas. (not mapped) Small areas of Marine Water are included in Coastal Water for mapping purposes. Barren or partially vegetated areas resulting from human disturbance. As mapped, the human- Human Modified modified areas are predominantly roads, pads, and mine pits and overburden. Areas with clouds, snow, and ice. The Clouds and Ice Class was combined with the Shadow classes for the final map. Most of the original shadow classes in the input maps in the Brooks Cloud, Snow, and Ice Range were recoded to alpine classes based on modeling. Remaining shadow areas are primarily due to clouds in the Brooks Foothills. Aufeis on rivers was classified as Riverine Barrens to avoid creation of a separate Riverine Ice class.

126 Wildlife Response to Environmental Arctic Change (WildREACH): Predicting Future Habitats of Arctic Alaska Appendix 3: 17–18 November 2008 WildREACH Westmark Hotel, Fairbanks, Alaska

Day 1: Monday, 17 November 2008 Workshop

Registration: 7:30 a.m.–9:30 a.m. Agenda 7:30 a.m. Arrival and Coffee Service Plenary Session Held in Gold Room

8:00 a.m. Welcome Geoffrey Haskett, Regional Director, U.S. Fish and Wildlife Service 8:10 a.m. Introductions Philip Martin, U.S. Fish and Wildlife Service 8:30 a.m. Workshop Goals and Structure Philip Martin, U.S. Fish and Wildlife Service 8:50 a.m. Observed Climate Change in Northern Alaska Martha Shulski, University of Alaska Fairbanks 9:05 a.m. Projected Climate of Northern Alaska Peter Larsen, The Nature Conservancy 9:20 a.m. Coastal Processes David Atkinson, University of Alaska Fairbanks 9:35 a.m. Permafrost and Active Layer Dynamics Vladimir Romanovsky, University of Alaska Fairbanks 9:55 a.m. Permafrost-Influenced Geomorphic Processes Torre Jorgenson, ABR Environmental Services, Inc. 10:15 a.m. Plenary Discussion/Q&A 10:30 a.m. Break 10:50 a.m. Vegetation Change Eugenie Euskirchen, University of Alaska Fairbanks 11:05 a.m. Hydrologic Processes Amy Tidwell, University of Alaska Fairbanks 11:20 a.m. Hydrology Panel—Predicting Change Snow: Matthew Sturm, Cold Regions Research and Engineering Laboratory Water Balance: Doug Kane, University of Alaska Fairbanks Feedbacks: Larry Hinzman, University of Alaska Fairbanks Anna Liljedahl, University of Alaska Fairbanks 11:50 a.m. Plenary Discussion/Q&A 12:30 p.m. Lunch (on your own) 1:30 p.m. Integration – Potential Ecosystem Pathways Torre Jorgenson, ABR Environmental Services, Inc. 1:50 p.m. Charge to Working Groups Philip Martin, U.S. Fish and Wildlife Service 2:00 p.m. Working Groups Breakout Session I Working Groups will be presented with scenarios of climate and landscape change that would affect landscape-scale habitat availability, e.g., broad scale conversion from one habitat type to another. What species (or species attributes) would be sensitive indicators of the hypothesized changes? 3:30 p.m. Break 4:00 p.m. Return to Plenary Session: Working Group Reports Bird Working Group Fish Working Group Mammal Working Group Managers Working Group 5:00 p.m. Summary of Day’s Discussion; Plans for Day 2 5:15 p.m. Adjourn 5:30 p.m. Reception and Poster Session (light food and cash bar) All participants are encouraged to bring a poster that describes synthesis research, data, and modeling needs within their area of interest, or relevant research findings from integrated projects. 127 Day 2: Tuesday, 18 November 2008

8:00 a.m. Arrival and Coffee Service

8:15 a.m. Review of Day 1 Discussion, Day 2 Goal, Workshop Products 8:25 a.m. Trophic Systems: Herbivores Brad Griffith, University of Alaska Fairbanks 8:40 a.m. Trophic Systems: Aquatic Mark Wipfli, University of Alaska Fairbanks 8:55 a.m. Charge to Working Groups Philip Martin, U.S. Fish and Wildlife Service 9:10 a.m. Working Groups Breakout Session II Climate-associated processes may lead to changes in habitat suitability which cannot easily be equated with change in availability. Examples include: changes in water temperature that affect physiological processes of fish, enhanced food availability due to increased primary and secondary productivity, changing seasonality that results in asynchrony between optimal food availability and critical life history phases. Based on the scenarios of climate change outlined in the previous Breakout, what are the most important mechanisms by which climate would affect habitat suitability? Express the relationships and mechanisms in the format of “box-and- arrow” conceptual models. 11:15 a.m. Return to Plenary Session: Report of Working Groups Bird Working Group Fish Working Group Mammal Working Group Managers Working Group 12:00 p.m. Lunch (on your own) 1:00 p.m. Return to Plenary Session 1:15 p.m. Bayesian Network Modeling Erik Beever, U.S. Geological Survey 1:30 p.m. Empirical Temperature Downscaling: Improving Thermal Information Detail David Atkinson, University of Alaska Fairbanks 1:45 p.m. Charge to Working Groups Philip Martin, U.S. Fish and Wildlife Service 2:00 p.m. Working Groups Breakout Session III Focusing on the conceptual models identified in the previous breakout session, what are the key areas where reducing uncertainty in physical process models would enhance our ability to predict habitat change? What would we most like modelers and researchers in other disciplines to work on? 4:00 p.m. Return to Plenary Session: Report of Working Groups Bird Working Group Fish Working Group Mammal Working Group Managers Working Group 4:30 p.m. Identify Areas of Commonality Among Working Groups Panel Discussion: Working Group Leaders, Workshop Participants 5:00 p.m. Summary, Discussion, and Next Steps Philip Martin, U.S. Fish and Wildlife Service 5:15 p.m. Workshop Adjourns

Day 3: Wednesday, 19 November 2008

U.S. Fish and Wildlife Service Conference Room, Federal Building

9:00 a.m. Invited Writing Group U.S. Fish and Wildlife Service Staff & invited Workshop Participants Synthesis of Key Results and Themes from Workshop The third day is intended for a small group of participants who are willing to make a substantial commitment to drafting and reviewing the final document from this workshop. If you are interested in participating, please 128 contact Philip Martin at 907-456-0325 or [email protected]. 12:00 p.m. Adjourn

Jeff Adams Mark Bertram Assessment and Monitoring U.S. Fish and Wildlife Service Appendix 4: U.S. Fish and Wildlife Service 101 12th Avenue 101 12th Avenue Fairbanks, AK 99701 WildREACH Fairbanks, AK 99701 Phone: 907-456-0446 Phone: 907-456-0218 [email protected] Fax: 907-456-0454 Workshop [email protected] Erica Betts University of Alaska Fairbanks Participants Stephen M. Arthur 869 Landon Lane Division of Wildlife Conservation Fairbanks, AK 99709 Alaska Department of Fish and Phone: 404-317-5312 Game [email protected] 1300 College Road Fairbanks, AK 99701-1599 Dixie Birch Phone: 907-459-7336 U.S. Fish and Wildlife Service Fax: 907-452-6410 1011 East Tudor Road [email protected] Anchorage, AK Phone: 907-786-3523 David E. Atkinson [email protected] International Arctic Research Center Karen Bollinger University of Alaska Fairbanks U.S. Fish and Wildlife Service 930 Koyukuk Drive, Room 313 101 12th Avenue PO Box 757340 Fairbanks, AK 99701 Fairbanks, AK 99775-7340 Phone: 907-456-0427 Phone: 907-474-1126 Fax: 907-456-0208 Fax: 907-474-2643 [email protected] [email protected] Bonnie Borba Scott D. Ayers Commercial Fisheries School of Fisheries and Ocean Sci- Alaska Department of Fish and ences Game University of Alaska Fairbanks 1300 College Road PO Box 757220 Fairbanks, AK 99701-1599 Fairbanks, AK 99775-7220 Phone: 907-459-7260 Phone: 907-474-2486 Fax: 907-459-7271 [email protected] [email protected]

Michael Baffrey Alan W. Brackney Office of the Secretary Arctic National Wildlife Refuge U.S. Department of the Interior U.S. Fish and Wildlife Service 1689 C Street, Suite 100 101 12th Avenue, Room 236 Anchorage, AK 99501 Fairbanks, AK 99701 Phone: 907-271-4399 Phone: 907-456-0450 Fax: 907-271-4102 Fax: 907-456-0428 [email protected] [email protected]

Earl Becker Scott Brainerd Division of Wildlife Conservation Division of Wildlife Conservation Alaska Department of Fish and Alaska Department of Fish and Game Game 333 Raspberry Road 1300 College Road Anchorage, AK 99518-1599 Fairbanks, AK 99701-1599 Phone: 907-267-2407 Phone: 907-459-7261 Fax: 907-267-2433 Fax: 907-459-7332 [email protected] [email protected]

Erik A. Beever Rena Bryan Alaska Science Center International Arctic Research U.S. Geological Survey Center 4210 University Drive University of Alaska Fairbanks Anchorage, AK 99508-4626 PO Box 757340 Phone: 907-786-7085 Fairbanks, AK 99775 129 Fax: 907-786-7150 Phone: 907-474-1556 [email protected] [email protected] Greta Burkart Judy Fahnestock Geoffrey Haskett Arctic Network Inventory and Arctic Research Consortium of the Alaska Regional Office Monitoring Program U.S. U.S. Fish and Wildlife Service National Park Service 3535 College Road, Suite 101 1011 East Tudor Road 4175 Geist Road Fairbanks, AK 99709 Anchorage, AK 99503 Fairbanks, AK 99709-3420 Phone: 907-474-1600 Phone: 907-786-3542 Phone: 907-455-0669 Fax: 907-474-1604 Fax: 907-786-3495 Fax: 907-455-0601 [email protected] [email protected] [email protected] Craig L. Fleener Larry D. Hinzman Kristina R. Creek Division of Subsistence International Arctic Research Arctic Research Consortium of the Alaska Department of Fish and Center U.S. Game University of Alaska Fairbanks 3535 College Road, Suite 101 1255 W 8th Street 930 Koyukuk Drive Fairbanks, AK 99709 PO BOX 115526 PO Box 757335 Phone: 907-474-1600 Juneau, AK 99802-5526 Fairbanks, AK 99775-7340 Fax: 907-474-1604 Phone: 907-465-4147 Phone: 907-474-7331 [email protected] Fax: 907-465-2066 Fax: 907-474-1578 [email protected] [email protected] Cathy Curby Arctic National Wildlife Refuge Brad Griffith Leslie Holland-Bartels U.S. Fish and Wildlife Service Alaska Cooperative Fish and Wild- Alaska Science Center 101 12th Avenue life Research Unit U.S. Geological Survey Fairbanks, AK 99709 University of Alaska Fairbanks 4230 University Drive, Suite 201 Phone: 907-456-0500 PO Box 757020 Anchorage, AK 99508-4664 [email protected] Fairbanks, AK 99775-7020 Phone: 907-786-7000 Phone: 907-474-5067 Fax: 907-786-7150 David Daum Fax: 907-474-7872 [email protected] Fishery Resource Office [email protected] U.S. Fish and Wildlife Service Birte M. Horn-Hanssen 101 12th Avenue Julie Griswold Arctic Research Consortium of the Fairbanks, AK 99701 Arctic Research Consortium of the U.S. Phone: 907-456-0290 U.S. 3535 College Road, Suite 101 Fax: 907-456-0454 3535 College Road, Suite 101 Fairbanks, AK 99709-3710 [email protected] Fairbanks, AK 99709-3710 Phone: 907-474-1600 Phone: 907-474-1600 Fax: 907-474-1604 Tony DeGange Fax: 907-474-1604 [email protected] Alaska Science Center [email protected] U.S. Geological Survey Judy Jacobs 4210 University Drive Guido Grosse U.S. Fish and Wildlife Service Anchorage, AK 99508-4626 Geophysical Institute 1011 E. Tudor Road Phone: 907-786-7046 University of Alaska Fairbanks Anchorage, AK 99503 Fax: 907-786-7021 903 Koyukuk Drive Phone: 907-786-3472 [email protected] Fairbanks, AK 99775 [email protected] Phone: 907-474-7548 Laurel Devaney [email protected] Jennifer Jenkins Fairbanks Fisheries Resource Office U.S. Fish and Wildlife Service U.S. Fish and Wildlife Service Raymond Hander 101 12th Avenue 101 12th Avenue, Room 222 U.S. Fish and Wildlife Service Fairbanks, AK Fairbanks, AK 99701 101 12th Avenue, Room 110 Phone: 907-456-0327 Phone: 907-456-0558 Fairbanks, AK 99701-6267 [email protected] Fax: 907-456-0454 Phone: 907-456-0402 [email protected] Fax: 907-456-0454 Tim Jennings [email protected] U.S. Fish and Wildlife Service Eugenie Euskirchen 1011 East Tudor Road Institute of Arctic Biology Melanie Hans Anchorage, AK 99503 University of Alaska Fairbanks U.S. Fish and Wildlife Service Phone: 907-786-3505 311 Irving Building I 101 Front Street [email protected] Fairbanks, AK 99775 Galena, AK 99741 Phone: 907-474-1958 Phone: 907-656-1231 Danielle Jerry [email protected] [email protected] U.S. Fish and Wildlife Service 1011 East Tudor Road Anchorage, AK 99503 130 Phone: 907-786-3335 Fax: 907-786-3901 [email protected] Benjamin Jones Richard B. Lanctot Joseph Margraf Alaska Science Center Migratory Bird Management Alaska Cooperative Fish and U.S. Geological Survey U.S. Fish and Wildlife Service Wildlife Research Unit 4230 University Drive, Suite 201 1011 East Tudor Road, MS 201 University of Alaska Fairbanks Anchorage, AK 99508-4650 Anchorage, AK 99503-6199 PO Box 757020 Phone: 907-786-7033 Phone: 907-786-3609 Fairbanks, AK 99775-7020 Fax: 907-786-7036 Fax: 907-786-3641 Phone: 907-474-7661 [email protected] [email protected] Fax: 907-474-6716 [email protected] Janet C. Jorgenson Peter H. Larsen Arctic National Wildlife Refuge The Nature Conservancy Philip D. Martin U.S. Fish and Wildlife Service 715 L Street, Suite 100 Ecological Services Fairbanks 101 12th Avenue, Room 236 Anchorage, AK 99501 U.S. Fish and Wildlife Service Fairbanks, AK 99701 Phone: 907-276-3133, ext 113 101 12th Avenue, Room 110 Phone: 907-456-0216 Fax: 907-276-2584 Fairbanks, AK 99701 Fax: 907-456-0428 [email protected] Phone: 907-456-0325 [email protected] Fax: 907-456-0208 Brian E. Lawhead [email protected] M. Torre Jorgenson ABR, Inc.—Environmental ABR, Inc.—Environmental Research and Services Angela Matz Research and Services PO Box 80410 Northern Alaska Ecological Services PO Box 80410 Fairbanks, AK 99708 U.S. Fish and Wildlife Service Fairbanks, AK 99709 Phone: 907-455-6777 101 12th Avenue, Room 110 Phone: 907-455-6777 Fax: 907-455-6781 Fairbanks, AK 99701 Fax: 907-455-6374 [email protected] Phone: 907-456-0442 [email protected] Fax: 907-456-0208 Jim P. Lawler [email protected] Douglas L. Kane National Park Service Water and Environmental Research 4175 Geist Road A. David McGuire Center Fairbanks, AK 99709-3420 Alaska Cooperative Fish and University of Alaska Fairbanks Phone: 907-455-0624 Wildlife Research Unit PO Box 755860 Fax: 907-455-0601 University of Alaska Fairbanks Fairbanks, AK 99775-5860 [email protected] 214 Irving Building I Phone: 907-474-7808 Fairbanks, AK 99775 Fax: 907-474-7041 Anna K. Liljedahl Phone: 907-474-6242 [email protected] International Arctic Research Fax: 907-474-7872 Center [email protected] Lon Kelly University of Alaska Fairbanks Northern Field Office PO Box 757340 Neil Mochnacz Bureau of Land Management Fairbanks, AK 99775 Arctic Science Division, Central and 1150 University Avenue Phone: 907-474-1951 Arctic Region Fairbanks, AK 99709 [email protected] Fisheries and Oceans Canada Phone: 907-474-2368 501 University Crescent Fax: 907-474-2280 Mark Lisac Winnipeg, MB R3T 2N6 [email protected] U.S. Fish and Wildlife Service Canada PO Box 270 Phone: 204-984-3147 Steve Kendall Dillingham, AK Fax: 204-984-2403 U.S. Fish and Wildlife Service Phone: 907-842-1063 [email protected] 101 12th Avenue, Room 236 [email protected] Fairbanks, AK 99701 William Morris Phone: 907-456-0303 Wendy M. Loya Habitat Division Fax: 907-456-0428 The Wilderness Society Alaska Department of Fish and [email protected] 705 Christensen Drive Game Anchorage, AK 99501 1300 College Road David R. Klein [email protected] Fairbanks, AK 99701-1599 Institute of Arctic Biology Phone: 907-459-7282 University of Alaska Fairbanks Edward J. Mallek Fax: 907-459-7303 PO Box 757020 Waterfowl Management [email protected] Fairbanks, AK 99775-7020 U.S. Fish and Wildlife Service Phone: 907-474-6674 101 12th Avenue Larry Moulton Fax: 907-474-6967 Fairbanks, AK 99701 MJM Research [email protected] Phone: 907-456-0341 1012 Shoreland Drive [email protected] Lopez Island, WA 98261 Phone: 360-468-4821 131 [email protected]

Karen Murphy David Payer Nora Rojek U.S. Fish and Wildlife Service Arctic National Wildlife Refuge Endangered Species Program 1011 E Tudor Road U.S. Fish and Wildlife Service U.S. Fish and Wildlife Service Anchorage, AK 99503 101 12th Avenue, Room 236 101 12th Avenue, Room 110 Phone: 907-786-3501 Fairbanks, AK 99701 Fairbanks, AK 99701 Fax: 907-786-3501 Phone: 907-455-1830 Phone: 907-456-0276 [email protected] Fax: 907-456-0428 Fax: 907-456-0208 [email protected] [email protected] Stephen M. Murphy ABR, Inc.—Environmental John F. Payne Vladimir E. Romanovsky Research and Services North Slope Science Initiative Geophysical Institute PO Box 80410 Bureau of Land Management University of Alaska Fairbanks Fairbanks, AK 99708 222 West 7th Avenue, #13 PO Box 757320 Phone: 907-455-6777 Anchorage, AK 99513-7599 Fairbanks, AK 99775-7320 Fax: 907-455-6781 Phone: 907-271-3431 Phone: 907-474-7459 [email protected] Fax: 907-271-4596 Fax: 907-474-7290 [email protected] [email protected] Matt Nolan Institute of Northern Engineering B. Zeb Polly Diane M. Sanzone University of Alaska Fairbanks Arctic Research Consortium of the Environmental Studies Group 455 Duckering Bldg U.S. BP Exploration (Alaska), Inc. PO Box 755910 3535 College Road, Suite 101 900 East Benson Boulevard Fairbanks, AK 99775-5910 Fairbanks, AK 99709-3710 PO Box 196612 Phone: 907-474-2467 Phone: 907-474-1600 Anchorage, AK 99519-6612 Fax: 907-474-7979 Fax: 907-474-1604 Phone: 907-564-4857 [email protected] [email protected] Fax: 907-564-5020 [email protected] Alvin Ott Alex Prichard Habitat Division ABR, Inc.—Environmental Brendan Scanlon Department of Fish and Game Research and Services Sport Fish Division 1300 College Road PO Box 80410 Alaska Department of Fish and Fairbanks, AK 99701-1599 Fairbanks, AK 99708 Game Phone: 907-459-7279 Phone: 907-455-6777 1300 College Road Fax: 907-459-7303 [email protected] Fairbanks, AK 99701-1599 E-Mail: [email protected] Phone: 907-459-7268 Mary Rabe [email protected] Ronnie Owens Alaska Department of Fish and Arctic Research Consortium of the Game Joel Schmutz U.S. PO Box 115526 Alaska Science Center 3535 College Road, Suite 101 Juneau, AK 99811 U.S. Geological Survey Fairbanks, AK 99709-3710 Phone: 907-465-6195 4210 University Drive Phone: 907-474-1600 [email protected] Anchorage, AK 99508-4626 Fax: 907-474-1604 Phone: 907-786-7186 [email protected] Kumi Rattenbury Fax: 907-786-7021 National Park Service [email protected] Tom Paragi 4175 Geist Road Division of Wildlife Conservation Fairbanks, AK 99709-3420 Robert W. Schneider Alaska Department of Fish and Phone: 907-455-0673 Northern Field Office Game Fax: 907-455-0601 Bureau of Land Management 1300 College Road [email protected] 1150 University Avenue Fairbanks, AK 99701 Fairbanks, AK 99709 Phone: 907-459-7327 Deborah Rocque Phone: 907-474-2302 Fax: 907-452-6410 Fairbanks Field Office Fax: 907-474-2282 [email protected] U.S. Fish and Wildlife Service [email protected] 101 12th Avenue, Room 110 Lincoln Parrett Fairbanks, AK 99701 Stan Senner Division of Wildlife Conservation Phone: 907-456-0272 Alaska State Office Alaska Department of Fish and Fax: 907-456-0208 National Audubon Society Game [email protected] 715 L Street, Suite 200 1300 College Road Anchorage, AK 99501 Fairbanks, AK 99701-1551 Phone: 907-276-7034 Phone: 907-459-7366 Fax: 907-276-5069 Fax: 907-459-7332 [email protected] 132 [email protected] Mark B. Shasby Charla Sterne Wendy K. Warnick Alaska Science Center U.S. Fish and Wildlife Service Arctic Research Consortium of the U.S. Geological Survey 1011 East Tudor Road, Stop 331 U.S. 4210 University Drive Anchorage, AK 99503 3535 College Road, Suite 101 Anchorage, AK 99508-4650 Phone: 907-786-3471 Fairbanks, AK 99709-3710 Phone: 907-786-7065 [email protected] Phone: 907-450-1600 Fax: 907-786-7150 Fax: 907-474-1604 [email protected] Matthew Sturm [email protected] Cold Regions Research and Richard Shideler Engineering Laboratory - Alaska Tara Wertz Alaska Department of Fish and PO Box 35170 Ecological Services Game Fort Wainwright, AK 99703-0170 U.S. Fish and Wildlife Service 1300 College Road Phone: 907-361-5183 101 12th Avenue, Room 236 Fairbanks, AK 99701 Fax: 907-353-5142 Fairbanks, AK 99701 Phone: 907-459-7283 [email protected] Phone: 907-456-0444 Fax: 907-456-3091 [email protected] [email protected] Eric J. Taylor National Wildlife Refuge System Matthew Whitman Reija S. Shnoro U.S. Fish and Wildlife Service Northern Field Office Arctic Research Consortium of the 1011 East Tudor Road, MS 221 Bureau of Land Management U.S. Anchorage, AK 99503-6199 1150 University Avenue 3535 College Road, Suite 101 Phone: 907-786-3846 Fairbanks, AK 99709 Fairbanks, AK 99709-3710 Fax: 907-786-3905 Phone: 907-474-2249 Phone: 907-474-1600 [email protected] Fax: 907-474-2280 Fax: 907-474-1604 [email protected] [email protected] Amy Tidwell Institute of Northern Engineering Helen V. Wiggins Martha Shulski University of Alaska Fairbanks Arctic Research Consortium of the Alaska Climate Research Center PO Box 755860 U.S. University of Alaska Fairbanks Fairbanks, AK 99775-5860 3535 College Road, Suite 101 903 Koyukuk Drive Phone: 907-474-2783 Fairbanks, AK 99709-3710 Fairbanks, AK 99775 Fax: 907-474-7979 Phone: 907-474-1600 Phone: 907-474-7885 [email protected] Fax: 907-474-1604 [email protected] [email protected] Declan Troy Trey Simmons Troy Ecological Research Associates Robert A. Winfree National Park Service 2322 E 16th Avenue Alaska Regional Office 4175 Geist Road Anchorage, AK 99508-2905 National Park Service Fairbanks, AK 99709-3420 Phone: 907-276-3436 240 W 5th Avenue, Office 521 Phone: 907-455-0666 [email protected] Anchorage, AK 99501 [email protected] Phone: 907-644-3516 Tim Viavant Fax: 907-644-3816 LaVerne Smith Alaska Department of Fish and [email protected] U.S. Fish and Wildlife Service Game 1011 East Tudor Road, MS361 1300 College Road Mark Wipfli Anchorage, AK 99503 Fairbanks, AK 99701-1599 Alaska Cooperative Fish and Wild- Phone: 907-786-3544 Phone: 907-459-7266 life Research Unit Fax: 907-786-3848 Fax: 907-456-2259 University of Alaska Fairbanks [email protected] [email protected] PO Box 757020 Fairbanks, AK 99775-7020 Melanie Smith Cashell Villa Phone: 907-474-6654 Audubon Alaska Arctic National Wildlife Refuge Fax: 907-474-6716 441 W 5th Avenue, Suite 300 U.S. Fish and Wildlife Service [email protected] Anchorage, AK 101 12th Avenue, Room 236 Phone: 907-276-7034 Fairbanks, AK 99701 Eve Witten [email protected] Phone: 907-456-0275 The Nature Conservancy [email protected] 715 L Street, Suite 100 Neesha Stellrecht Anchorage, AK 99501 U.S. Fish and Wildlife Service Douglas Vincent-Lang Phone: 907-276-3153 101 12th Avenue, Room 222 Sport Fish Division Fax: 907-276-2584 Fairbanks, AK 99701 Alaska Department of Fish and [email protected] Phone: 907-456-0297 Game Fax: 907-456-0208 333 Raspberry Road [email protected] Anchorage, AK 99518-1599 133 Phone: 907-267-2339 [email protected] David A. Yokel Other Contributors James D. Reist Arctic Field Office Arctic Fish Ecology and Assessment Bureau of Land Management Jonathan Benstead Research Section 1150 University Avenue Department of Biological Sciences Fisheries and Oceans Canada Fairbanks, AK 99709 University of Alabama 501 University Crescent Phone: 907-474-2314 PO Box 870206 Winnipeg, MB R3T 2N6 Fax: 907-474-2282 Tuscaloosa, AL 35487 Canada [email protected] Phone: 205-348-9034 Phone: 204-983-5032 Fax: 205-348-1403 Fax: 204-984-2403 Alison D. York [email protected] [email protected] Arctic Research Consortium of the U.S. Stephanie M. Clemens Ted Swem 3535 College Road, Suite 101 [email protected] U.S. Fish and Wildlife Service Fairbanks, AK 99709-3710 101 12th Avenue Phone: 907-474-1600 Fred DeCicco Fairbanks, AK 99701 Fax: 907-474-1604 Fairbanks, AK Phone: 907-456-0441 [email protected] [email protected] [email protected]

Steve Zack Linda Deegan Ken Tape North America Program The Ecosystems Center Institute of Arctic Biology Wildlife Conservation Society Marine Biological Laboratory University of Alaska Fairbanks 718 SW Alder Street, Suite 210 7 MBL Street PO Box 80425 Portland, OR 97205 Starr 337 Fairbanks, AK 99708 Phone: 503-241-3743 Woods Hole, MA 02543 Phone: 907-374-9000 [email protected] Phone: 508-289-7487 Fax: 907-353-5142 Fax: 508-457-1548 [email protected] James Zelenak [email protected] U.S. Fish and Wildlife Service Øivind Tien 101 12th Avenue, Room 110 Vadim Fedorov Institute of Arctic Biology Fairbanks, AK 99701 Institute of Arctic Biology University of Alaska Fairbanks Phone: 907-456-0354 University of Alaska Fairbanks PO Box 757000 Fax: 907-456-0208 PO Box 757000 Fairbanks, AK 99775 [email protected] Fairbanks, AK 99775 Phone: 907-474-6843 Phone: 907-474-1533 [email protected] Chris Zimmerman E-Mail: [email protected] Alaska Science Center John E. Walsh U.S. Geological Survey Richard Flanders International Arctic Research 4210 University Drive Fairbanks, AK Center Anchorage, AK 99508-4626 [email protected] University of Alaska Fairbanks Phone: 907-786-7071 PO Box 757340 Fax: 907-786-7150 Marie-Luce Hubert Fairbanks, AK 99775-7340 [email protected] Jean-Louis Klein Phone: 907-474-2677 http://www.klein-hubert-photo.com Fax: 907-474-2643 [email protected] Russ Mitchell Inkworks Ken Whitten 2324 Waterford Road Fairbanks, AK Fairbanks, AK 99709 [email protected] Phone: 907-455-6378 Fax: 907-455-4740 [email protected]

Mitch Osborne U.S. Fish and Wildlife Service 101 12th Avenue Fairbanks, AK 99709 Phone: 907-456-0209 [email protected]

134 Working Group Breakout Session Charges Appendix 5: Sideboards and Assumptions Geographic Scope: The geographic scope includes terrestrial and freshwa- Working Group ter ecosystems from the crest of the Brooks Range northward. The bound- ary with the marine system is recognized as fuzzy; coastal processes are acknowledged to affect terrestrial and freshwater systems and vice versa. Charges and Processes that affect nearshore coastal environments may be considered for those species that cross ecosystem boundaries. Members

Priority Issues: It is recognized that climate change may influence organisms in multiple and complex ways. We intend to focus the discussion primarily on the following climate change effects: •Change in relative abundance and distribution of habitat types on the landscape. •Change in structural (including plant community structure) or physical characteristics of habitat. •Change in trophic systems, including primary and secondary productivity, phenology, and forage/prey availability.

The importance of other potential effects of climate change—such as competitive interactions, invasive species, prevalence of disease, and contaminants—are recognized but are of secondary priority for this workshop.

Working Group Breakout Session I (Monday afternoon) The goals of this session are to: • Identify species or species groups that are expected to be sensitive indicators of the changes hypothesized in the climate scenarios and ecosystem pathway models. •For each species, develop hypotheses regarding positive or negative response to changes in the availability of habitat, based on knowledge of species’ life histories and habitat requirements. •For each species, identify specific parameters (e.g., distribution, abundance, demography, body condition, growth rate, etc.) that would be affected.

Groups should focus on landscape-level changes in habitat availability. Species should be selected for their value as indicators of climate change. Groups should NOT feel constrained to only those species or parameters that are easily measured. This breakout session is an opportunity for discussion on any and all species in the region that will be affected by climate change. Later in the workshop, we will consider feasibility issues when prioritizing the research and modeling needs that arise from the breakout group sessions.

Working Group Breakout Session II (Tuesday morning) The goal of this session is to refine draft conceptual models of climate effects on species, including processes that are not captured by habitat change models. Each working group will be presented with models drafted as a result of the scoping meetings. Groups are encouraged to consider “reasonable worst-case scenarios,” i.e., scenarios that are within the expected range of potential change AND would have the greatest magnitude effects, thus most detectable. For example: •Birds: Trophic system shifts for consumers of invertebrates, coastal habitat availability. •Mammals: Forage quantity and quality in summer, snow conditions, forage availability in winter. •Fish: Stream system flow regimes, availability of river delta overwintering habitat.

135 Working Group Breakout Session III (Tuesday afternoon) The goal of this final breakout session is to review the products from the first two breakout sessions and identify the most critical gaps in data and modeling needed to predict future habitats of arctic Alaska. Gaps could be of at least two types: • Models have not been constructed. •Models exist but are not supported by adequate data.

Identified gaps should focus on the underlying ecological and physical processes that may affect species in all three groups of interest (birds, fish, and mammals). The goal is to go beyond identification of data gaps for arctic species biology and refine our thinking of what is needed to gain predictive ability of the system-level physical processes and ecosystem functions.

Breakout session results will form the core content for a five-year strategic plan that identifies the priority research, modeling, and synthesis activities needed to predict climate-related impacts to fish and wildlife populations in the Arctic.

Fish Working Group Jeff Adams (leader) U.S. Fish and Wildlife Service Jennifer Jenkins (recorder) U.S. Fish and Wildlife Service David Atkinson University of Alaska Bonnie Borba Alaska Department of Fish and Game Guido Grosse University of Alaska Fairbanks Douglas Kane University of Alaska Fairbanks Anna Liljedahl (Monday) University of Alaska Fairbanks Joseph Margraf University of Alaska Fairbanks Neil Mochnacz Fisheries and Oceans Canada William Morris Alaska Department of Fish and Game Larry Moulton MJM Research Matt Nolan University of Alaska Fairbanks Amanda Rosenberger University of Alaska Fairbanks T. Scott Rupp University of Alaska Fairbanks Brendan Scanlon Alaska Department of Fish and Game Rod Simmons U.S. Fish and Wildlife Service Tim Viavant Alaska Department of Fish and Game Matthew Whitman Bureau of Land Management Mark Wipfli University of Alaska Fairbanks Chris Zimmerman U.S. Geological Survey Kristina Creek ARCUS Staff

136 Birds Working Group Philip Martin (leader) U.S. Fish and Wildlife Service James Zelenak (recorder) U.S. Fish and Wildlife Service Alan Brackney U.S. Fish and Wildlife Service Syndonia Bret-Harte University of Alaska Fairbanks Steve Kendall U.S. Fish and Wildlife Service Richard Lanctot U.S. Fish and Wildlife Service Peter Larsen The Nature Conservancy A. David McGuire University of Alaska Fairbanks Vladimir Romanovsky University of Alaska Fairbanks Joel Schmutz U.S. Geological Survey Stan Senner National Audubon Society Eric Taylor U.S. Fish and Wildlife Service Amy Tidwell University of Alaska Fairbanks Declan Troy Troy Ecological Research Steve Zack Wildlife Conservation Society Judy Fahnestock ARCUS Staff

Mammals Working Group David Payer (leader) U.S. Fish and Wildlife Service Tara Wertz (recorder) U.S. Fish and Wildlife Service Stephen Arthur Alaska Department of Fish and Game Erik Beever U.S. Geological Survey Scott Brainerd Alaska Department of Fish and Game Eugenie Euskirchen University of Alaska Fairbanks Brad Griffith University of Alaska Fairbanks Larry Hinzman University of Alaska Fairbanks David Klein University of Alaska Fairbanks Brian Lawhead ABR, Inc. Anna Liljedahl (Tuesday) University of Alaska Fairbanks Stephen Murphy ABR, Inc. Tom Paragi Alaska Department of Fish and Game Lincoln Parrett Alaska Department of Fish and Game Alex Prichard ABR, Inc. Richard Shideler Alaska Department of Fish and Game Martha Shulski University of Alaska Fairbanks Matthew Sturm Cold Regions Research and Engineering Lab David Yokel Bureau of Land Management Alison York ARCUS Staff

137 Managers Working Group Deborah Rocque (leader) U.S. Fish and Wildlife Service Michael Baffrey U.S. Department of the Interior Dixie Birch U.S. Fish and Wildlife Service Tony DeGange U.S. Geological Survey Geoffrey Haskett U.S. Fish and Wildlife Service Leslie Holland-Bartels U.S. Geological Survey Tim Jennings U.S. Fish and Wildlife Service Lon Kelly Bureau of Land Management Dennis Lassuy U.S. Fish and Wildlife Service John Payne North Slope Science Initiative Mark Shasby U.S. Geological Survey Robert Schneider Bureau of Land Management LaVerne Smith U.S. Fish and Wildlife Service Douglas Vincent-Lang Alaska Department of Fish and Game Robert Winfree National Park Service Helen Wiggins ARCUS Staff

138