7 AD-A 3 440 LANOSAT-ASSISTED ENVIRONMENTAL

MAPPIND IN THE ARCTIC

NATIONAL WILDLIFE REFUGE ALASKA(U) COLORADO UNIV ATBOULDER D A WALKER ET AL. NOV 82 CRREL-82-37

UNCEASSIFIED GS-RN- 1 T

060-

0450 F/G 8/6 NL

.EEEIII •

Bil' 11. *2.O

IIIIL25 .4~r Jill1.--6

MICROCOPY RESOLUTION TEST CHARTNATIONAL BURIAU Of STANDARDS 1963 A

MA123440

US Army CorpsREPORT 82-37 of Engineers

Cold Regions Research &Engineering Laboratory

Landsat-assisted environmental mapping inthe Arctic National Wildlife Refuge, Alaska

1U

r- %* i i .Iterq*4 %

tons I111* so bum MA

%J3 (020

For conversion of SI metric units to U.S./British customary units of measurementconsult ASTM Standard E380, Metric Prac-tice Guide, published by the American Socie-ty for Testing and Materials, 1916 Race St.,Philadelphia, Pa. 19103.

Cover: Sadlerochit River just north of Sadlero-chit Spring. (Photograph by D. Atwood.)

CRREL Report 82-2?November 1982

Landsat-assisted environmental mapping inthe Arctic National Wildlife Refuge, Alaska

D.A. Walker, W. Acevedo, K.R. Everett,L. Gaydos, J. Brown and P.J, Webber

4

Prepared forU.S. GEOLOGICAL SURVEY t . ..

U.S. FISH AND WILDLIFE SERVICE ions,..ro-. .. .Approved for public release; dislribution unlimited.

UnclassifiedSECURITY CLASSIFICATION OF THIS PAGE (Whei Date Entered)

REPORT DOCUMENTATION PAGE READ INSTRUCTIONSBEFORE COMPLETING FORM

I. REPORT NUMBER 2. GOVT ACCESSION NO. 3. RECIPIENT'S CATALOG NUMBER

CRREL Report 82-37 IAD D,/,,3 t/O4. TITLE (md Subtitle) S. TYPE OF REPORT & PERIOD COVERED

LANDSAT.ASSISTED ENVIRONMENTAL MAPPING IN THEARCTIC NATIONAL WILDLIFE REFUGE, ALASKA 6. PERPORMING ORG. REPORT NUMBER

7. AUTHOR(e) 8. CONTRACT OR GRANT NUMBER(s)

D.A. Walker, W. Acevedo, K.R. Everett, Depai tment of die Interior GrantL. Gaydos, J. Brown and P.J. Webber No. GS RN#l -7060-0450

9. PERFORMING ORGANIZATION NAME AND ADDRESS 10. PROGRAM ELEMENT. PROJECT, TASK

University of Colorado U.S. Army Cold Regions Research & AREA & WORK UNIT NUMBERS

U.S. Geological Survey Engineering LaboratoryOhio State University (see next page)

1. CONTROLLING OFFICE NAME AND ADDRESS 12. REPORT DATENovember 1982

U.S. Geological Survey Nv. 1982

U.S. Fish and Wildlife Service 13. NUMBER OF PAGES

6814. MONITORING AGENCY NAME & ADDRESS(i1 different from Controlling Office) IS, SECURITY CLASS. (of this report)

U.S. Army Cold Regions Research and Engineering Laboratory UnclassifiedHanover, New Hampshire 03755 15. DECLASSIFICATION/OWNGRADING

SCHEDULE

16. DISTRIBUTION STATEMENT (of this Report)

Approved for public release: distribution unlimited.

17. DISTRIBUTION STATEMENT (of the absttrct entered In Block 20, If different from Report)

1B. SUPPLEMENTARY NOTES

19. KEY WORDS (Continue on reveree side if necessary and Identify by block number)

Arctic National Wildlife Refuge Vegetation classificationGeobotanical mappingLandsatMappingSoils

M9 AUSYNACT (Centilwas r eviere oif N neey and Idealt by block number)

This report presents a Landsat-derived land cover map of the northwest portion of the Arctic National WildlifeRefuge, Alaska. The report is divided into two parts. The first is devoted to the land cover map with detailed descrip-dons of the mapping methods and legend. The second part is a description of the study area. The classitication systemused for the maps is an improvement over existing methods of describing tundra vegetation. It is a comprehensivemethod of nomenclature that consistently applies the same criteria for all vegetation units. It is applicable for large-and small-scale mapping and is suitable for describing vegetation complexes, which are common in the patterned-groundterrain of the Alaskan Arctic. The system is applicable to Landsat-derived land cover classifications. The descriptionof the study area focuses on five primary terrain types: flat thaw-lake plains, hilly coastal plains, foothills, mountainous

JA 75 " Unclassified

SECURITY CLASSIFICATION OF THIS PAGE (When Date Entored)

UnclassifiedSECURITY CLASSIFICATION OF THIS PAOiErtlam Data Enterue

9. Performing Organization Name and Address (cont'd).

Institute of Arctic and Alpine Research U.S. Geological SurveyUniversity of Colorado NASA Ames Research CenterBoulder, Colorado 80309 Moffett Field, California 94035

Institute of Polar Studies U.S. Army Cold Regions ResearchOhio State University and Engineering LaboratoryColumbus, Ohio 43210 Hanover, New Hampshire 03755

20. Abstract (cont'd).

terrain, and river flood plains. Topography, landforms, soils and vegetation are described for each terrain type. Thereport also contains area summaries for the Landsat-derived map categories. The area summaries are generated forthe five terrain types and for the 89 townships within the study areas. Two land cover maps at 1:250,000 are included.

sits U

piIijt.smw&U1ty Codes

I~'1i~t1l 'or

ii UnclassifiedSECURITY CLASSIFICATION OF THIS PAGOE('When Data Entered)

PREFACE

This report was prepared by Dr. Donald A. Walker. Research Associate, Instituteof Arctic and Alpine Research (INSTAAR), University of Colorado; William Ace-vedo. Senior Data Analyst. Technicolor Government Services, NASA/Ames Re-search Center; Dr. K.R. Everett. Professor. Institute of Polar Studies. Ohio StateUniversity; Leonard Gaydos, Chief, Geographic Investigations Office, U.S. Geo-logical Survey (USGS), NASA/Ames Research Center. Dr. Jerry Brown, Chief,Earth Sciences Branch, Research Division. U.S. Army Cold Regions Research andEngineering Laboratory (CRREL); and Dr. Patrick J. Webber, Professor andDirector of INSTAAR, University of Colorado.

This report was prepared as input to the Environmental Impact Statement re-quired under Section 1002 of the Alaska National Interest Land Conservation Act(ANILCA) of 1980. The experience gained with Landsat and geobotanical mappingin other parts of northern Alaska over the past eight years was applied to this largeand relatively unknown region of northeastern Alaska. Much of the experience re-quired to accomplish the project was gained through studies supported by the De-partment of Energy, USGS and CRREL. This specific study was supported byUSGS and the U.S. Fish and Wildlife Service (USFWS).

James R. Wray, USGS, Reston, Virginia. was responsible for the design of theLandsat land cover map. Carol Hurr, USGS, Denver. and Dr. John Haugh. Depart-ment of the Interior, Reston. provided editorial and technical assistance in compil-ing early versions of the report. USFWS. under Dr. William Kirk's direction, organ-ized the logistics for the initial field ground check in August 1981. Bob Bartles of theUSFWS Barter Island field station and personnel of the Distant Ef rlv Warning(DEW) line site provided considerable assistance for the field work. •

The report was reviewed by Dr. Vera KomdrkovA, Research Assdiciate, INSTAAR;Dr. Paula Krebs, Bureau of Land Management, Anchorage; Paul Brooks,'Chief.USGS Alaska Program Office; Mark Shasby, Technicolor GovernMent Services,Alaska Operation. Anchorage; Carolyn Merry, CRREL; and Carol Simniorrs, Uni-versity of Colorado. John Hall of USFWS and Karen Howe of CRRELreviewed theNWI equivalents in Appendix C. The comments of the reviewers are grelnappre-ciated; they provided the basis for numerous changes in the original manuscript,parts of which are included in the Environmental Impact Statement and BaselineStudies for the Arctic National Wildlife Refuge.

The contents of the report are not to be used for advertising or promotional pur-poses. Citation of brand names does not constitute an official endorsenett or ap-proval of the use of such commercial products. I

iii

I

CONTENTSP'age

A bstract .. . . . . . . . . . . . . . . . . . . . . . . . . . . .P reface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iF~orew ord ... . . . . . . . . . . . . . . . . . . . . . . . . .. vIntroduction .. . . . . . . . . . . . . . . . . . . . . . . . . . . IA land cover map of the ANWR study area............................. I

Legend development............................................. 2Miapping method................................................ 4Results........................................................ 8Discussion..................................................... 8

Description of' the ANWR study area ................................. 9General description.............................................. 9Description of.specific terrain types ................................ 11

Conclusions ....................................... 33Literature cited .................................................. 34Appendix A: Descriptions of ILandsat land cover categories for ANWR ... 39Appendix B: Area summaries....................................... 47Appendix C: Aproximiate equivalent unit-. in sevcral systemis of land cover,

wetland and vegetation class ificat ions% used in northern Alaska .......... 51Appendix D: Soil taxonomyN ........................................ 55Appendix E3: Summary of principal 1.andsal land cos cr categories within the

ter-rain types of the ANWR 'stud%. area ............................... 57

ILLUSTRATIONS

FigureI . Location of the ANWR study area in northern Alaska ................ 22. Topography of the study area .................................. facing p.53. Cluster diagram for Scene No. 22008-20420, Bands 5 and 6 ............ 64. Flat thaw-lake terrain in the delta of the Canning River ................ 1I5. Thaw-lake plain terrain with strangmnoor .......................... 126. Histic Pergelic Cryaquept soil ................................... 127. Unit Ill, Wet Sedge Tundra..................................... 138. Unit IVa, Moist/Wet Sedge Tundra Complex ....................... 149. Ground view of Moist/Wet Sedge Tundra Complex................. 14

10. Unit Ila, Pond/Sedge Tundra Complex ........................... I1511. Unit 111b, Aquatic Tundra......................................I1512. Unit Illd. Wet Sedge Tundra.................................... 1613. Vertical aerial photograph of gently rolling coastal-plain terrain in the

vicinity of the Niguanak River, southeast of Barter Island ......... 1714. Unit Vb, Moist Sedge/Barren Tundra Complex ..................... 18I5. Foothills terrain south of Camden Bay ........................... 19

iv

Figure Page16. DisconltinuLous gravel outcrop in fihe foothills ........... 1917. Foothills terrain with Land Cover Unit IVa. Moisu"Wct Sedge Tundra. 2018. Foothills soils ............................................... 2019. Unit VIk., Moist Sedge Tussock, Dwarf Shrub/Wet Dwart'Shrub Tunl-

dra Comiplex ............................................ 2120. Soils of thle water track areas showing a complex pedon typical of such

terrain .................................................. 2221 . Frost-scat terrain in thle toot hills ................................. 2222. Common soil pedon consisting of' Pergelic Cryaquept and Histic Per-

gelic C ryaquoll .......................... ................ 2323. Unit Via, Moist Sedge Tussock. Dwarf Shrub Tundra onl foothills

south of Camrden BaN ...................................... 2424. Cottongrass tussocks in Unit VIa ................................ 2425. Unit Vila, Mloist Dwarf Shrub, Sedge Tussock Tundra ................ 2526. Unit Vill. Shrub Tundra....................................... 2527. Mountainous terrain in thle Sadlerochit Miountains .................. 2628. Sadlerochit Spring............................................ 2729. Idealized toposequence across Mountainous Terrain................. 2730. Pergelic Cry'umbrept soil....................................... 2831. The Canning River............................................ 2932. Idealized toposequence across River Flood Plains and Foothills ........ 2933. Pergelic Cryorthent soil onl a Di-i'as river terrace ..................... 3034. One of the numerous river icings along the Canning River. 27 July 1981 3135. Partially vegetated river bar oin thle Katakturuk River................ 3236. Meandering tundra streamn ..................................... 3237. Unit lVb. D)ry Prostrate Shrub. Forb Tundra....................... 33

Plate I. Vegetation and land cover. Arctic National Wildlife Refuge, CoastalPlain, Alaska........................................... inside back cover

Plate 2. Dominant landcover categories in the northern coastal portion of theArctic National Wildlife Refuge, Alaska ...................... inside back cover

TABLES

Table1. Four-level hierarchy for mapping tundra regions ..................... 42. Temperature, wind and precipitation data for Barter Island. Alaska ... 10

V

SUMMARY

This report describes the landforms, soils and vegetation in a 1.4-million-acre(5700-km) portion of the Arctic National Wildlife Refuge (ANWR) in Alaska. Thearea is being considered for seismic oil exploration activities scheduled to begin inDecember 1982. A colored land cover map of the study area at 1:250,000 scale wasderived from Landsat data as part of this project and was used extensively in the ter-rain analysis. Descriptions of the environment are based on a seven-day reconnais-sance trip in August 1981 that concentrated in four townships of the study area. Thefield data are supplemented with information from 1:60,000-scale, color-infraredand 1:18,000-scale, color aerial photographs. The report is divided into two parts.The first describes the Landsat map, the methods for making it, and the legend. Thesecond is a description of the ANWR study area based on the major terrain types.

The Landsat-derived land cover map consists of a digital mosaic of portions ofthree Landsat scenes. The land cover classification consists of the following mapcategories: 1) Water, 2) Pond/Sedge Tundra Complex or Aquatic Tundra or shallowwater, 3) Wet Sedge Tundra, 4) Moist/Wet Sedge Tundra Complex or Dry ProstrateShrub, Forb Tundra, 5) Moist Sedge, Prostrate Shrub Tundra or Moist Sedge/Bar-ren Tundra Complex (frost-scar tundra), 6) Moist Sedge Tussock, Dwarf ShrubTundra, 7) Moist Dwarf Shrub, Sedge Tussock Tundra or Moist Sedge Tussock,Dwarf Shrub/Wet Dwarf Shrub Complex (water track complex), 8) Shrub Tundra,9) Partially vegetated areas, 10) Barren gravel or rock, II) Wet gravel or mud, and12) Ice. The land cover classification system is briefly explained and equivalent unitsare given for six other vegetation, wetland and land cover classification systems.

The legend system is a solution to the current need for a tundra classification. Thesystem is still being perfected, but it already meets the following criteria:1) The system is applicable to both large- and small-scale mapping.2) It consistently applies the same criteria to naming all community types.3) It has a consistent method ot naming vegetation complexes.4) It is well suited for specific application to Landsat multispectral scanner data.The study area is divided into five terrain types and the ocean. These types deline-

ate relatively large areas with similar primary landforms. They are, with their per-centage of the study area in parentheses, foothills (44.7%), river flood plains(24.60), hilly coastal plains (22.4%), ocean (5.207o), flat thaw-lake plains (3.1%)and mountainous terrain (0.03%).

Foothills are the most common terrain type in the study area. Tundra with sedgetussocks and dwarf shrubs covers most foothill uplands. Dominant plants includesheathed cottongrass (Eriophorum vaginatuin), numerous ericaceous shrubs, dwarfbirch (Belula nana) and diamond-leafed willow (Salix planifolia ssp. pukchra).Shrub cover varies in upland tundra types and is responsible for much of the varia-tion in spectral reflectances from moist tundra vegetation. Stream valleys, south-facing slopes and water tracks (drainage channels on slopes) are likely to have well-developed shrub communities. Frost scars occur on most upland surfaces and maycover up to 900o of the surface. Pergelic Cryaquolls or Pergelic Cryaquepts underliemuch of the moist tundra, although Histic Pergelic Cryaquepts are common soils inwater tracks.

River flood plains, which cover large portions of the study area, are highly com-plex landscapes that include present flood plains, former braided drainages, riverdeltas, bluffs, terraces, wet tundra, gravel bars, dunes, mud flats and river icings.Willow communities are common along the rivers.

vi

OP

Hilly coastal plains occur north of the foothills, particularly east of the HulahulaRiver. The vegetation and soils are a combination of those found in the foothills andthose found in the flat thaw-lake plains. Wet sedge tundra covers about 23%0 of theunit, and complexes of moist and wet sedge tundra cover about 34%. Moist tundratypes cover about 40% of the unit. Wet tundra areas are mostly confined to depres-sions between low ridges that have an east-west orientation. Nearly flat hill crests of-ten have extensive thermokarst pits.

Flat thaw-lake plains are restricted to small areas near the coast and are dom-inated by wet sedge tundra and complexes of moist and wet sedge tundra. A largeportion of these areas is covered by lacustrine complexes. The vegetation and soilsare strongly controlled by microrelief associated with patterned ground, mainly ice-wedge polygons and strangmoor. The dominant plants are aquatic sedge (Carexaquatius), other sedges and mosses. The dominant soils are Pergelic Cryaquepts andHistic Pergelic Cryaquepts, with thick, fibrous, organic surface horizons. In wetterareas these soils often form complexes with true organic soils, mainly Pergelic Cryo-hemists. Moderately well drained areas often have mineral soils, Pergelic Crya-quolls, in association with frost boils. The vegetation near the coast is affected by asteep summer gradient, with low temperatures near the coast and higher tempera-tures inland. Beaches, lagoons, estuaries and low-lying areas, which are all inun-dated by sea water during storm surges, support saline-tolerant plant communitiesand haline soils.

Mountainous terrain occurs only in a small portion of the study area, near Sadle-rochit Spring. Surface forms consist of block fields, talus slopes, sorted stone nets,steep solifluction slopes, and rock outcrops. Soils are restricted to small areas wherefiner materials can collect. In less rocky areas, Pergelic Cryorthents form complexeswith Pergelic Cryochrepts, Pergelic Cryaquepts, and occasionally, Cryohemists insolifluction areas. The land cover in most mountainous areas is classed as either bar-ren or partially vegetated due to the dominance of rock, but fairly lush tundra vege-tation occurs locally on partially stable terrain.

vii

-1J

LANDSAT-ASSISTED ENVIRONMENTALMAPPING IN THE ARCTIC NATIONALWILDLIFE REFUGE, ALASKA

D.A. Walker, W. Acevedo, K.R. Everett, L. Gaydos,J. Brown and P.J. Webber

INTRODUCTION A LAND COVER MAP OFTHE ANW'R STLIDY AREA

The Alaska National Interest Lands Conser,,a-tion Act (ANILCA) of 1980 created the present There are two main objectives fot mappingboundaries of the Arctic National Wildlife Refuge project:(ANWR). Under this legislation, 1.4 million acres I) To produce a land cover map ,NWR(5700 km-) of the northern portion of the refuge study area that has wide applicatio. s\ildlifewere opened to oil and gas exploration at the dis- and land use studies.cretion of the Secretary of the Interior (Fig. I). 2) To develop a legend that is appropriate forThe area, henceforth referred to as the ANWR Landsat and that has application to other areas instudy area, is being considered for seismic oil ex- northern Alaska.ploration that would begin in December 1982. A Landsat-derived land cover classification was

An Environmental Impact Statement (EIS) was deemed the most suitable approach for this map-prepared to satisfy Section 1002d of ANILCA. A ping project for several reasons. First, the mapbaseline study was prepared under Section 1002c had to be prepared in less than one year; L.andsal-of ANILCA to serve as the basis for the EIS and assisted mapping is the quickest method for map-for regulating future exploration. The extent, lo- ping large areas. Second, the area is generally in-cation and carrying capacity of wildlife habitats accessible for detailed mapping on the ground.are major emphases of the baseline studies, and Third, funds were insufficient for conventionalvegetation maps are required for examining wild- mapping by aerial photointerpretation, and not alllife habitats. Because existing vegetation maps of the area was covered on aerial photographs.covering the study area were too general for habi- Finally, Landsat provides digital, geographicallytat studies, a new 1:250,000-scale land cover map referenced information that lends itself well towas suggested, and Landsat-assisted mapping was geographic-based information systems. Otherdeemed the most practical approach in the limited data bases can be registered to it, and analysestime available. A Landsat-based land cover map such as area measurements can be performed east-(inserted in the back of this report) was produced ly. Also, maps can be converted to different scalesbetween August 1981 and April 1982, and was or the boundaries changed.partially verified by the three senior authors. Our recent work in the Prudhoe Bay region

This report is divided into two parts. The first (Walker and Acevedo, in preparation) has shownpart is devoted to the land cover map, with discus- that excellent Landsat-derived land cover mapssions of the mapping methods and legend. The can be prepared for the tundra of the Arcticlegend is a step toward a Landsat classification Coastal Plain. The Coastal Plain has two attri-system for tundra. The second part consists of de- butes that aid in interpreting Landsat data. First,scriptions of the major terrain types within the the terrain is nearly flat or gently rolling, so deepANWR study area. shadows do not create problems in interpretation,

~It

A RC rI C 0 CE A N "st.o

CHUKCHI BfAtPoRr ........ A '

SfA4

1i'urre 1. L.ocation of ihe .-I,%''R siudv areu in torthern.4 laska.

as Ihes do in mountainous areas. Second, the %eg- status of the site, also affect the reflectance. Aelation is all o%\ growing; there are no trees to classification syslenflfor landsat-dersed mappinginask the terrain or other vegetation. ,hould recogni/e the effect of' moi,ture and shrub

cover on spectral reflectance. The secondar. fac-Legend development tors are more difTicult to classify consistentl., but

they should be considered during photointerprea-("i l .r,1W Landsal-deri'el iulu'a legeints tion.

Although Landsat-derived maps hae been Second, a classification system should be Ilexi-developed for several areas of northern Alaska ble enough to describe the great variety of tundia

(Belon et al. 1975, Nodler and LaPerriere 1977, communities. There has not yet been sufficient Cx-Nodler et al. 1978, Lyon and George 1979, \lot- perience with Landsat imagery of tundra environ-rissey and Ennis 1981), there has not yet been an inents to establish a final set of units \sith which toattempt to develop a comprehensive Landsat land categorize all tundra landscapes. Most approachescover legend applicable to all areas of the Alaskan have attempted to force Landsat interpretationsarctic tundra. There are, however, numerous clas- into rather rigid legend systems that ha\e ne\ersification systems that were examined for polen- been applied to detailed tundra mapping or thattial application to this mapping objective. These simply are not suitable for a Landsat approach.fit in three categories: remote-sensed land cover Third, a classification system should use cortsis-classifications (Anderson et al. 1976, Nodler and tent criteria in naming vegetation units. This isLaPerriere 1977), wetland classifications (Berg- emphasiz,-d in all major national and internationalman et al. 1977, Cowardin et al. 1979), and vege- mapping systems. Most of these systems recognizetation classifications (Fosberg 1970, UNESCO the importance of dominant growth forms. We1973, Viereck and Dyrness 1980, Driscoll el al. have found that a comb* ation of dominant1981). A full discussion of the merits and prob- growth forms and site moisture can be used con-ferns of each system is beyond the scope of this re- sistently for preparing maps of tundra vegetation.port; none satisfied the objective of this project. The system should also have a consistent method

A satisfactory legend for Landsat-assisted map- of describing vegetation complexes (i.e. areasping of tundra vegetation must meet several crite- where there are mixtures of vegetation communi-ria. First, the legend must be based on the charac- ties), which are particularly common in the pat-teristics of tundra vegetation that can be recog- terned-ground landscapes of the Arctic.nized on Landsat imagery. Our experience with Finally, the small-scale, Landsat-based legendLandsat imagery and tundra has shown that there units should be able to serve as a basis for moreare two primary aspects of arctic Alaskan coastal detailed community-level mapping without chang-tundra vegetation that affect its spectral rellec- i"g the system. It is also important that the systen:tance-the amount of water on the surface and the should be cross-r-ferenced to the other ciassifica-percentage of shrubs in the vegetation canopy. tion systems currently in use in Alaska.Numerous secondary factors, such as the openness The classification system developed here satis-of the graminoid layer (the layer of grasslike fies these criteria. The classification system forplants), the color of the substrate, the amount of small-scale maps, such as the 1:250,000-scale landerect, dead graminoid vegetation, and the nutrient cover map at the end of this report, is based on a

2

method of describing plant communities at large physiognomic descriptor hae all upper-case let-

scales (e.g. 1:6000 scale) thar was developed for ters, and the plant names are italici/ed. (The

Prudhoe Bay, Alaska (Walker et al. 1980, Walker upper-case lettering applies only it, large-scale

1981). This large-scale nomenclature system is coniniunit. names and is not used for small-scale

described first so that the basis of the smaller- ,egetation unit names.)

scale, Landsat-level map units can be better An example of a communit. name using this

understood. The map itself is a major step toward systeam is W-T ('arex aquatilis. Drepanocludu.s

producing accurate vegetation maps for the entire hreviloliu.s SEDGE TUNDRA. A more complex

Arctic Slope of Alaska. example with two dominant grosth forms isMOIST Lriophoruin friste. Drras inie.'/loli, Sa-

\omencluitre fr hlrge-scuh /i.x arctica, Toit Wfllh lV,)nll litlelrs, Th innoli

vegetation uawpping suhuljormi.s SEDGE. PROSTRATE SHRUB

Community nomenclatu11C for large-scale map- TUNDRA.ping (1:12.000 scale or larger) incorporates four Vegetation complexes are also described %%ith a

major parts: I) a site moisture term, 2) the doli- uniform nomenclature. A unit is considered a

nant plant species in each part of the vegetation complex if it contains tso or more distinct corn-

canopy. 3) the dominant plant gro th forms, and tunities. and each community covers at least 300 "o

4) all o.erall phvsiognornic descriptor. These parts of the area. In the Arctic. community mosaics oc-

are always arranged in this sequence. cur mainly as a response to initor elesation differ-

The site Moisture term can be DRY, MOIST. ences associated \%ith ice-ssedge polygons, frost

WET, or if the segenaned area is permanently co- boils, water tracks, strangtnoor, solifluction lobes

ered %%ith \,ater, AQUATIC. The site moisturc and other kind,, of patterned ground. Esen at the

term is followed by tile names of the dominant 1:6(XX) scale most of tile indisidual communities

plant taxa, usually one or more Iron each of tile are too small to map \%ithoul reference to segeta-

shrub, herb and cr~plogarni cotlpotlenlts of the ion complexes. In this nomenclature the term

vegetation. The number of taxa is kept to the min- COMPLEX is attached to the end of the land cov-

imum required to adequately distinguish the corn- er name, and the major elements of the complex

munity frot other,, on the map: the total tie\cr cx- are separated by a slash (0). For example, a map

ceeds ,ix. unit composed of wet sedge tundra and moist

The terms used for plant growth forms are: I) sedge tundra is called a WET/ MOIST SEDGE

TALL SHRUB (>2 m tall), 2) MEDIUM SHRUB TUNDRA COMPLEX. The most abundant unit

(0.5-2 m), 3) DWARF SHRUB (0.1-0.5 m), 4) of the complex is named first. A map unit with a

PROSTRATE SHRUB (<0.1 m), 5) TUSSOCK wet, dwarf-shrub community occurring in the

GRAMINOID, 6) NONTUSSOCK GRAMI- water tracks of sedge-tussock, dwarf-shrub tundra

NOID, 7) FORB, 8) MOSS and 9) LICHEN. All would be named MOIST SEDGE TUSSOCK,

low-growing woody plants (<0.1 m high), such as DWARF SHRUB/WET DWARF SHRUB TUN-

Dryas integrifolia and Arciostaphylos rubra, are DRA COMPLEX.

classed as prostrate shrubs. The graminoid vegeta- At large scales, complexes of vegetation can be

tion is further broken down into either sedge- or described by the nature of the surface form on

grass-dominated units. Cushion plants, such as sshich the complex occurs. For example, vegeta-

Saxifraga oppositifolia and O.xyropis nigrescens, tion on a foothill area silh water tracks w\ould

are included in the forb category. The lichen vege- have the followring explanation in the legend:

tation is further broken down into crustose- or MOIST SEDGE TUSSOCK. DWARF SHRUB,

fruticose/ foliose-lichen-dominated units. If a veg- WET DWARF SHRUB TUNDRA COMPLEX

etation unit is dominated by more than one (water track complex):

growth form, each covering at least 30% of the a) Interfluves and upland tundra areas: MOIST

ground, it will have more than one growth form in Eriophorum vaginatum, Betula nana, Salix plani-

its name (e.g. MOIST SEDGE TUSSOCK. folia, Ledum decumbens, Sphagnum sp., Cladina

DWARF SHRUB TUNDRA). sp. TUSSOCK SEDGE, DWARF SHRUB TUN-

The physiognomic descriptot TUNDRA is used DRA,

for all arctic and alpine ionfotested areas with b) Water tracks: WET Salix planifolia, Betula

generally continuous ground cover. The physiog- nana, Carex aquatilis, Sphagnum sp. DWARF

nomic term BARREN is applied to units with less SHRUB TUNDRA.

than 30016 vegetation cover. The site moisture The term "water track complex" could be used as

term, the dominant plant growth forms, and the a shorter synonym in general discussion.

1 3

V,1~

Table 1. Four-level hierarch) for mapping tundrat regions.

I2 3 4

Application: 'Co~ small scale m1aps Sm~all-scale. %ei , gecial. Small -scale. mote detailed. I atge-scale maps %tt I detailedI itdsat tIC l 111.1p. I andsat -lcel minaps gt ound tel erecmc data

%K.Ole: I :2,5(E,( of 'titallet I .5119),Mx ) 1:12.tNm)- I:5(EtN(Xt 1 4WM- 1: 1 2.MMIP

%omenclalure: 1 . Plh siognont k descr i ptors. Ph singonimic dewi i pti o . I'lti rog nomnic descript or, . I -Ph stop nom ic dew r ipt or.2. site ntot01tute 2. Site rir1rlstute 2. Site moisture

3. toinntatt platti gto%4;It 3. Doinat plant giot ilitot ittI tot is

4. Specie. comtposition

lKAMPlc'%: 1. 1 undra 1. \kct Ituidra i. V% et Sedge I titdra 1.- WI: Are iuiio

2. Blattetn 2. \IoIf' 1tida 2. NIlts Sedge. Protrate SLI~i: 1 1NL)RA

ShiubIfuIdra . fIOIS Lr itim'ernlo

3. Mloist \%et Sedge Itundra S~alt% un-itca, Tomnthtr-.oft ples ,tuo mioents. Thtuinntoli

sI/uhuh~s,remtI SE:D6E r4. tDr Irostrate Siub t'RosrRArTL SItLIJjI tidia Barren (oupte\ TUND)R A

Mlap categoi C. cait include Mlap categories canr include%egnetrion cornpleses of sepet ation conipleses thtatcomrbintationti t land- require cortptete: desct iption'coset t~pes. tot eash partt oI the coniplev

tihest -W Is) m lit.\lean annual rum' tall -1 cIt.

\ oount'Ia ire lor ',ifl-sca/(' amount of information asailabke on thle mrap).veL~eitio inupopitfrv Hos% eser-. in some cases, the %egetation ss ithin a

For small-scale. Landsat-lesel mlappin,. thle de- niap categorN may %ar% so s\ idel\ that it is not pos-flls ot' community composition canl t arc\hbe inl- ~ible it) use the nomertclat ure ssstemn at all. Par-chided, but the site moistutre teirm. thle dominant iall s egetated areas, are often of this naltrut .plant growth forms, and the phssiognomnic de- At small scales the s551cm is not rigid. It simipkscriptor can normall\ be ectained in filie land cos er offers a meanis of describing tundra land cos er as,titles. There arc, however.i numl~erouls escept otis. conlsistenit[\ as possible. The s\.sten for mappingitWith Landsat it is sometimes difficult to separate at small scales is strotigl\ rooted lin the tiomencla-rather distinct land cover types solely on the basis Ilure s%stctrts tot large-scale mnapping fin that thereof' spectral reflectance. atid it mua be tlecessar\ to is a nat ural nrest in, oft nomnetclat ue at toul Ics els,describe a map category (i.e. one color onl thie (0 able I.map) with several land cover t~pes,. lFor esample."Pond/Sedge Tundra Comples: or Aquatic "urt- Mapping methoddra; or shallow water" describes a mrap) categoi\ The mapping methods consisted of 1) acquiringconsisting of several very wet land coser typecs that ground referenice data from aerial phlotogaphscould not be distinctly separated in the Iandsat and field obsersations, 2) preparing the land coverdata. classificauioni from the digital [atidsat data, and 3)

Sometimes it may also be necessary to drop a composing and printing thle finial colored map.part of the land cover title. For example. it maN Computer-generated data pros ided detailed areanot be possible to distingutish Wet Medium Shrub summaries for each map catego y. lIn addition, aTundra fromt Wet Dwarf' Shrub Tundra or- Dry simplified, hand-drawit land cover mrap) wi I oly-Medium Shrub Tundra, so the termi Shrutb Tundra gonal map category boundaries was prepared for1could be used. Every attempt should be made to smaller-scale mapping programs.retain the complete title, since this increases thfe

4

" ~pt

771.b

44 F H

F

TE" t

LES

f'igurf 2. Topography oftstrTrhe la, ecircled numhfrs refer r

.J.

14 4

--------------

c N 8 A r

I DZN, DAY

~~ FP- ~ FP' ~* 4

I P.

FH * -

-. F H

N -~PIS.

ICAL SURVEY v .~*'

~', ~ I 7 -. 5C

143'

47.

H _

__________FP

______________

WILDERNES

-~ - '. 17

X7%0

Z 1

-, '4

-V - - - -

Ground refirence data vegetation from year to year. The manipulated ic-Ground reference data for the project were suits were later merged to create a land covet map

gathered during a USFWS-supported helicopter for the entire study area.survey, 12-17 August 1981. The survey was con- A clustering algorithm was used to deline dis-centrated in four townships (Fig. 2) thai included crete groups of pixels on the basis of their tel'lcc-the major landform features within the ANWR tance in the four Landsat spectral band,,. lhestudy area. This information was supplemented analyst selected the number of clusters based onwith extensive ground reference data gathered experience with similar l.andsat data and on an es-from similar Alaskan arctic areas outside ANWR timate of the desired number of clusters in tile fi-during the past several years. nal classilication. The numbers of clusters desig-

The field method consisted of locating large ho- nated for this study were 28 for scene 2570-20462,mogeneous areas of terrain on 1:60,000-scale, ae- 35 for scene 21633-20531, and 31 for scene 22(X)8-rial, color-infrared photographs and then describ- 20420. A systematic 10% sample of the pixelsing (from the air) the vegetation (e.g. wet sedge from each Landsat scene %%as used in the clusteiingtundra; dwarf shrub, sedge tussock tundra; par- algorithm. This algorithm sorts the Landsal datatially vegetated) and the landform (e.g. low-cen- into classes such that the pixels within each classtered polygon; strangmoor; water tracks). In par- are as similar to each other as possible and pixelsticularly homogeneous areas that appeared to be from different classes are as different from eachrepresentative of common units, we took detailed other as possible. These clusters describe in statis-notes of the vegetation, site factors and soils. tical terms (means and covarianies) tile spectralVegetation complexes were described if they were properties of Landsat data (Swain 1972).extensive and were associated with distinctive It was expected that these clusters would de-patterned-ground features. Each area was photo- scribe only general land cover units and not thegraphed from the air and the ground. specific units desired. Therefore, the ground refer-

We spent most of the field time describing map ence data were used to full advantage. Landsatunits from aircraft. This could be done readily, pixels from each of the four ground referencesince most of the vegetation and soils are similar areas were clustered separately. The resultingto those we have encountered in other areas of clusters were used to define vegetation classes,northern Alaska. This method allowed us to cover while the 10010 sample was used to select clustersas much terrain as possible with the limited represenmative of ice, water and barren land.amount of helicopter support. We took detailed At this point some vegetation classes were stillnotes regarding species composition in 35 stands not well defined by clusters. These were classesof vegetation. We hope that these data will be sup- that occurred rarely in the areas studied. It wasplemented with additional quantitative informa- likely that there were not enough pixels containingtion in future years and will be presented in a sepa- these vegetation types to allow them to be segre-rate report. gated into distinct clusters; hence, they were ini-

tially included in spectrally similar clusters. ToComputer classification handle these special cases, known occurrencesof the Landsat data were defined and mapped from the ground refer-

The methods employed for producing the land ence information. Pixels from these locations werecover classification from the Landsat data were then clustered independently, without considera-those available at the USGS Geographic Investiga- tion of other classes in the area, to produce sharp-tions Office at Ames Research Center, Moffett er spectral classes.Field, California (Morrissey and Ennis 1981). Thus, a final set of clusters, as defined by spec-Three Landsat scenes were used in this analysis. tral statistics, was produced for each scene. EachThe eastern two-thirds of the study area was coy- pixel in the scene was assigned to a cluster usingered by scene 22008-20420 (22 July 1980) acquired discriminant function analysis to determine thefrom the Canada Center for Remote Sensing. The cluster to which the pixel had the greatest proba-remainder of the region was covered by scene bility of belonging (Gaydos and Newland 1978).21633-20531 (13 July 1979), except for a small By viewing the results on a color image display,wedge in the southwest corner, which was coveked the analyst was able to enlarge small portions ofby scene 2570-20462(14 August 1976). Each scene the Landsat scene. He could also view only thehad to be treated separately in the analysis to ac- pixels in a selected spectral class by assigning a col-count for the variation in spectral response among or to that class. He then identified the vegetationscenes and because of phenological changes in in the spectral class based on field notes and high

5

100.0-

ICE

83.3 2

BARRENS,

z 66. 7 BRIGHT COLORED ..z 66.7-4[ GRAVELS(D LACK 40 1

IhJW BARENS, 0 12PARTIALLY

DARK COLORED MUD, VEGETATED AREAS, 50.0- WET GRAVEL VIOLETz

SREr 9 TUSSOCK SEDGE,DWARF

W. 6 L /cGHr pROWN

UJ33 .328 2 30 1I- 5 MOIST/ WET3 S~EDGE TUNDRA DAFSRB

2CONDLCOMPEXN TUSSOCK SEDGE

COMPLE GREENTUNDRA AN WATER

16.7- WATER D O4EWET SEDGE MOIST SEDGE TRACK TUNURA

L IcHT *LE TUNDRA T UND RA AND O,4/K ORO WN

0A4rK GREEN FROST BOIL

TUNDRA SHRUB TUNDRARED

4,N

O.CSEDGETUNDAN

COM0LE

REFLECTANCE-INFRARED SAND

Figure 3. Cluster diagram for Scene No. 22008-20420. Bands 5 (red) and 6 (near IR). Each ellipse encloses 80% o /thepixels assigned to the respective cluster. The land cover designafion. and map color' indicate howv the clu.iers were grouped ti thefinal classification.

resolution, color-infrared photography. This most units with different spectral properties (e.g. water.important interpretation step was aided immense- ice, barren and partially vegetated areas). In otherly by field experience, cases there were insufficient ground reference data

Each cluster was interpreted and assigned to a to define more categories.map category, i.e. one or more land cover units The major units described are the ones thatrepresented by a single color on the map (Fig. 3). could be distinguished during the time availableOften several clusters were included in a single for interpretation in late 1981. We hope that themap category. In some cases this was because a dissimilar vegetation types now within some mapmap category included a number of land cover categories can be separated by breaking single

6

spectral classes into several clusters. Digital terrain onal were assigned weights of I. Each map cale-data are being used to help separate classes on the gory showing in the window was rated by addingbasis of slope. Class consistency was a primary the weights, and the pixel in question (the centergoal. Care was taken to identify and resolve con- pixel of the group of 9) was assigned to the mapflicts between the three Landsat scenes as the final category with the highest rating. The window was12 map categories were developed, then moved to the next pixel. This 4-2-1 smooth-

ing process eliminated many isolated pixels whileColored niap preparation preserving the shape of the remaining units. A se-

Geoimetric correction. Ground control points, cond round of filtering was performed on the re-i.e. surface features identifiable both on 1:63,360- sults from the first with a 2-I-I smoothing, elimi-scale topographic maps and in the Landsat data, nating more single pixels and creating more homo-were selected for computing the parameters need- geneous map areas.ed for geometric correction of the Landsai data Base map. A base map for the final, colored,-And registration to a Universal Transverse Merca- Landsat-derived land cover map was compiledtor (UTNI) projection. About 30 well-distributed from a mosaic of the four USGS 1:250,000-scalecontrol points were selected for each scene. These quadrangles (Flaxman Island, Barter Island. Nit.points %%ere located on USGS 1:63,360-scale topo- Michelson and Demarcation Point). Projection in-graphic maps and on gray-scale line-printer maps formation, township boundaries, and geographicproduced from the Landsat data at a scale of ap- names within the ANWR study area are on thisproximately i:24,000. The points were used to de- map. This base map also contains the land coverfine second-order, least-squares polynomial trans- unit descriptions and information on how the mapformation equations relating latitude and longi- was prepared.tude on the maps to row and column positions on Printing. A tape of the Landsat data was usedthe Landsat scenes. The coefficients in the polyno- to drive a large-format laser plotter to generatemial equation weae used for correcting each scene four color-separation plates (yellow, cyan, magen-to a UTM projection (Zone 6). The geometric cor- ta and black) at 1:250,000 scale. Each pixel was re-rection resulted in pixels that were 50 meters on a produced as a matrix of 16 dots to reproduce theside. desired pattern for each color. Each 50-m pixel

Scene inosaic. Once the data were corrected to a was plotted at 200-pm spot size to achieve thecommon base, a single data set was formed from 1:250,000 scale.the mosaic of the three scenes. Common tie-points The color-separation plates were registered towere identified where the scenes overlapped, and the base map and used to prepare the Landsat lith-they were used to set mosaic parameters. The re- ographs by a color additive process. The finalsuiting mosaic of the classified scenes showed no map, "Vegetation and Land Cover. Arctic Na-

apparent scene boundaries. This is a good qualita- tional Wildlife Refuge, Coastal Plain, Alaska."tive indicator of the Landsat classification process by W. Ncevedo, D.A. Walker, L. Gaydos, and J.and the consistency of the spectral class descrip- Wrav is available from the U.S. Geological Survetions. as Map 1-1443 and is included in this report as

Data generalization. The original pixel data had Plate I.a "salt and pepper" appearance, with many iso-lated pixels in otherwise homogeneous areas of Area sutomariescolor. This detracted from the appearance of the The areas covered by each of the 12 map cale-map without adding much information at the gories were calculated using a counting algorithm.1:250,000 scale. Thus, it was desirable to genera- Data were generated for the areas iA ithin three %Ctlize the Landsat pixel data for cartographic pre- of boundaries: I) the entire ANWR stud\ area. 2)sentation. the 89 townships within the study area, and 3) (te

Accordingly the data were filtered in an attempt boundaries of the regional terrain types (,,ee theto remove tiny occurrences of map categories one second part of this report, vhich describes tie ia-or two pixels in size. The filtering was done by sys- jor terrain types). This was accomplished b. digi-lematically moving a 3- , 3-pixel window across ti/ing the respective boundaries, registering thethe classified data. In the first round the center boundaries to the Landsat land cover map, andpixel (the one being considered for reclassifica- applying the counting algorithm. The area sum-(ion) was assigned a weight of 4, while its adjacent inaries were calculated prior to filtering I'o tapneighbors, both horizontally and vertically, were generaliation.assigned weights of 2, and the pixels on the diag-

7

Simplified land cover map in the five terrain types listed in Table B2 isA more generalized map was prepared at the re- 1,556,830 acres, and the sum for all map categor-

quest of USFWS. This map was derived from an ies in Table BI is 1,640,626 acres. The differenceearlier unfiltered version of the Landsat-derived of 83,796 acres is the amount of ocean within theland cover map, which contained more isolated ANWR study area. Table B3 contains the data forpixels than appear on Plate 1. Polygons were each of the 89 townships within the study area.drawn around large areas with similar dominantland cover. This map (Plate 2) is useful as input to Discussionsmall-scale regional or state-wide mapping pro- We see an urgent need for tundra map legendsgrams. One place where this map is being used ef- that are compatible at large and small scales; thefectively is in the 1:250,000-scale geographic infor- ultimate goal is a comprehensive mapping systemmation system of the North Slope Borough. Since that can be used both for the broad picture and forthe map information is in polygon format instead detailed site-specific studies.of pixels, the map lends itself better to incorpora- In the past ten years we have mapped manytion into the Integrated Terrain Unit Mapping ap- areas of the Alaskan Arctic Coastal Plain andproach that the Borough is using (Environmental have experimented with various map legends, pri-Systems Research Institute 1982). marily at the scale of 1:6000 (Walker 1977, Walker

and Webber 1980, Walker et al. 1980, Sohio Pet-Results roleum Co. 1981). The mapping system described

here has been designed to provide vegetation in-Land cover maps formation for composite mapping approaches

The Landsat-derived land cover map (Plate 1) such as geobotanical master maps (Everett et al.has 12 map categories, which are briefly described 1978, Walker et al. 1980) or the Integrated Terrainin the map legend and are explained more fully in Unit mapping approach (Environmental SystemsAppendix A. Several of the colors represent more Research Institute 1982), where additional charac-than one land cover category. For example, dark teristics of the landscape, such as soils, landformsblue represents three land cover types: Pond/ and slopes, are also mapped. Our system has theSedge Tundra Complex, Aquatic Tundra or shal- advantage that it contains only vegetation infor-low water. In most cases the land cover types with- mation and is independent of soil, landform,in a single map category are not radically differ- altitude, etc. Purists may want to separate the site-ent. The exception is the light-green category (Unit moisture descriptor, since this is a characteristic ofIV), which may represent Dry Prostrate Shrub, the site; this is easy to do because it always appearsForb Tundra (Dryas river terraces) or Moist/Wet in the same part of the vegetation titles.Sedge Tundra Complex. These units have also Plentiful detailed ground reference data are theproved difficult to separate in Landsat studies in key to the success of any Landsat mapping pro-the National Petroleum Reserve in Alaska.* Pho- gram. The mapping system and legend describedtographs of most of the major land cover units are here is a "from-the-ground-up" approach, withincluded in the second part of this report. its foundation in very detailed mapping programs

A thorough field check was conducted in the that rely on large amounts of accurate ground re-summer of 1982. The initial impression is that the ference data (Walker et al. 1980, Walker andmap is very accurate and that there are no major Webber 1980).problems with the classification. A full report of Vegetation classification in Alaska is in a statethe field check and statistical analysis will be made of healthy turmoil. Several schemes are currentlyseparately. being applied to Alaskan tundra. Probably none

of these will gain universal acceptance, but it isArea summaries safe to say that they will all contribute in some

The acreages and percentages for each map cat- way to a final statewide system. The current lackegory are given in Appendix B. Table BI contains of a classification applicable to all Alaskan arcticacreage summaries for the entire ANWR study tundra has prompted the development of the sys-area. Table B2 contains the data for each terrain tem described here. However, our system adds totype. The sum of the areas for all map categories the profusion of mapping approaches. Appendix

C cross-references this system to six other majorsystems, showing the equivalent units in two other

*Personal communication with Paula Krebs, Bureau of Land remote-sensor land cover classifications (Ander-Management, Anchorage, Alaska. son et al. 1976, Nodler -and LaPerriere 1977), two

8

wetland classifications (Bergman et al. 1977, Cow- rises more steeply from the sea, there are bluffs up,ildin et al. 1979) and two vegetation classifica- to 8 m high.iions (Vicreck et al. 1981, UNESCO 1973). Along the coast (here are small areas of flat

The ANWR mapping project was perfectly suit- coastal plain with oriented thaw lakes similar toed for the application of modern technology using the terrain of the Teshekpuk Lake section in theLandsat multispectral scanner data and advanced western part of the Arctic Coastal Plain (Wah-cartographic techniques. A high-qualiy, publisha- haftig 1965). Most of the coastal plain withinble map of the ANWR study area was produced ANWR is gently iolling. with many small lakes%%ithin eight months of the beginning of tihe pio- and wet ter rain mixed with small areas of uplands.ject. 'The map is a valuable reference for evaluat- Stream drainage patterns in the hilly coastal plainsing the potential impact of seismic operations ate belle developed than in the flat tha%%-lakewithin the wildlife refuge. The results can also be plains, and the lake basins are contained beleenused as a basis for more detailed habitat studies. intervening small iidges and terrain irregularities.Since the data are in digital format, the map can Hilly coastal-plain terrain is common between ihebe easily used in geographic information systems. Ftulahula River and Pokok Bay, and stretches foiThe area measurements are most useful for the en- about 35 km inland, where the true foothills be-vironmental descriptions of the major terrain gin.types, and will also be useful for future studies The foothills occur west of the Hulahula Riverwithin township-sized areas. to (he Canning and extend across the entire south-

The current legetld system is still being perfect- emn portion of the stvdy area, with maximum cle-

ed, but it answers an immediate need lot a tundia vations of about 360 in. A small amount of moun-classification by meeting the following criteria: tainous terrain, reaching an elevation of 975 mn,

I1) The system is applicable to both large- and ,ccurs in the vicinity of Sadlerochit Spring.small-scale mapping. There are many river systems. The Canning,

2) It consistently applies the same criteria to all famayariak, Katakturuk, Sadlerochit, Hulahula,community names. lago and Aichilik rivers have braided drainages

3) It has a consistent method of naming vegeta- and deltas that collectively cover a large portion oftion complexes, the ANWR study area.

4) It is well suited for specific application toLandsal multispectral scanner data. Climate

The only regularly collected climatic data withinthe study area are from Barter Island (Table 2).

DESCRIPTION OF THE This station lies within a bell of strong coastal in-ANWR STUDY AREA fluence. In summer it has fog about 25% of the

time, and the mean July temperature is only about

General description 5 'C. These conditions only occur in a small por-tion of the study area, however. The inland por-

Physiography tions experience higher summer temperatures,The study area covers 5700 km of the Arctic however, our observations during 1981 indicate

Slope north of 69034 '. It lies between the Canning that the inland areas are somewhat colder and wet-and Aichilik rivers (142-146°W), and is contained ter than areas at a similar distance from the sea inwithin the White Hills section of the Arctic Coast- the Prudhoe Bay region (Haugen and Brownal Plain physiographic province defined by Wahr- 1980), probably because of the higher elevations inhaftig (1965). The terrain is, for the most part, hilly ANWR.and dissected by numerous streams that originatein the Sadlerochit, Romanzof and Franklin moun- Soilstains of the Brooks Range. Nearly the entire study area falls within the

The coastal area includes a chain of barrier is- Coastal Plain Land Resource Region defined it,lands that form the seaward limit of the ANWR the Exploratory Soil Survey of Alaska (Rieger etstudy area. The coastline itself is irregular and al. 1979). Within this region the survey recogni/edcontains many small bays. lagoons and spits. Ex- two soil associations (i.e. "segments of the land-tensive mud flats are associated with the Canning, scape with distinctive topographic and soil pat-Hulahula, Okpilak and Jago deltas. Most of the terns"): I) Pergelic Cryaquolls-Histic Pergeliccoastline is low lying, with only small bluffs less Cryaquepts, with loamy-lextured mineral compo-than 3 m high. At Camden Bay, where the land nents occurring on nearly level to rolling topogia-

.:: ,:'N'::-'* " t.l~r9

Table 2. Temperature, wind and precipitation data for BarterIsland, Alaska (70'08'N, 143'35'W, 12 m elevation). (FromSearb.s 119681 and Brower et al. 119771.)

I'l'ew/t'la alLtw e (f * If fill/ Aill JII Pir(' lpilnlr lth)III/lI Afil. . lll. A fecitt IPm l'lU/ dl11VOI(+llt 11111PI1

Jan 23.2 30.9 27.1 2.1.9%k 1(.2

Stel 25.2 32.2 281.7 22.3%k S.9Mar 22.1 30.1 26.1 21.7% 51.\pr 13.1) 21.6 17.3 19.111= 4.3

%la 3.4 9.2 6.3 19.51: 6.4.lile 1.9 1.5 1.2 18.)1: 13.)Jul% 8.6 1.7 5.2 16.9L- 22.4.ug 7.1 1.5 4.4 K.4L 26.7

'sepl 2.2 2.1 0. 1 21.01: 23.;(l-t 5.2 11.3 8.2 22.1 21.1No% 14.2 21.2 17.7 24.3U 10.2Dc' 19. 27.3 23.6 22.6" 7.4

-linnual 9.7 15.3 12.0 24.11- 159.5

l1| ghi l - 25.6 U, lo el - 50 .

flighci - 149 km hi.

M ean anti al lt s , all-o" 15 cli.

ph and 2) Iet gclic 0 .aqucpt,, "ith %:\ giasell> Vegetationtltnci al conIponCnk1 OCCI ring on ncarl. IcN el lop- The general character of Alaskan arctic vegeta-oiaphl .I* Ihe tl lnil o.c1tts oil he broad foot- lion has been thoroughly described by Brittonhill, and Ihe coasial plain. and the second OCClIs (1957), Spetzman (1959) and Wiggins and Thomason he haidcd Met allc%s and IhCil a,,ociaicd (1962). The Arctic Foothills and Arctic CoastalIcitace and dellas. Plain of northern Alaska are in the Tundra region

The only regional soil surveys for areas north of of the Arctic as defined by Aleksandrova (1980).the Brooks Range are those describing the Prud- In this region, mesic habitats are mostly complete-hoe Bay production area and the Ogotoruk Creek ly covered with low-growing plants, such aswatershed (Holowaychuk et al. 1966, Everett sedges, grasses, mosses, lichens, small herbs and1980b, Everett and Brown 1982). Brown (1962, dwarf shrubs. Taller shrubs are restricted to areas1966, 1969) developed a soils and landform map with protected southern exposures, where thefor approximately 19 km- in the Okpilak River amount of solar radiation is maximized. On thearea and described soils between the Jago and Hu- 1:2,500,000-scale major-ecosystem map of Alaskalahula rivers. In general the soils of the flat thaw- (Joint Federal-State Land Use Planning Commis-lake plains are similar to those described for the sion for Alaska 1973) most of the study area isPrudhoe Bay region (Everett and Parkinson 1977. mapped as Moist Tundra, with several small areasParkinson 1978. Everett 1980b) and Barrow dominated by Wet Tundra and with High Shrub(Drew 1957, Drew and Tedrow 1957, 1961, Ger- occurring along the streams.sper et al. 1980). The soils of the Arctic Foothills Detailed vegetation studies in the Arctic Coastalhave been described in detail by Tedrow (1977) Plain of Alaska include those at Barrow (Tieszenand Everett (1981). In the route selection studies 1978, Brown et al. 1980), Fish Creek (Lawson etfor the Alaskan Arctic Gas Study Company in al. 1978), Atkasook (KomdrkovA and Webber1974, Janz (Hettinger and Janz 1974) described 1980) and Prudhoe Bay (Walker et al. 1980, Walk-several soils from the foothills near the Kongakut, er 1981). Major studies in the foothills includeAichilik and Okerokovik rivers and from the those at Franklin Bluffs (Koranda 1960), Umiatflood plain of the Canning River. (Churchill 1955), Cape Thompson (Johnson et al.

1966) and the Seward Peninsula (Sigafoos 1952,Racine 1975, Racine and Anderson 1979). Vegeta-

'The soil terminology follows the U.S. Soil Taxonomy (Soil tion on arctic flood plains has been described bySurvey Staff 19751. Appendix 1) gives a brief explanation of Bliss and Cantlon (1957) and Spetzman (1959).thi, 'system. There are also a few vegetation studies from

10

b -

... . . . " - " .... .. - ""- . --

within the ANWR study area (Nodler and LaPer- (Table B2) and observations from other similarriere 1977, Oldemeyer et al. 1978, Murray 1980, areas in northern Alaska.Machilda and Oldemeyer 1980, Weiler 1980, Appendix E summarizes the dominant land cov-Meyers 1981, Robus 1981). Much of the recent er units, landforms, soils and vegetation terrainwork in ANWR is related to USFWS habitat stud- types. The table, which can be used to compareies for caribou, muskox and waterfowl. The most regions, condenses much of the information pre-extensive work from within the study area is that senied here. It also contains more specific infor-of Hettinger (Hettinger and Janz 1974), who de- mation about the dominant plant communitiesscribed several vegetation types in the foothills within the various land cover units.ipear the Aichilik and Kongakut rivers and alongthe Canning River as part of the biological studies Flat thaw-lake plains

of the proposed Arctic Gas Pipeline route. Landforms and soils. The proximity of the Ro-manzof and Sadlerochit mountains to the coast re-

Descriptions of specific suits in a much narrower coastal plain than is

terrain types found farther west, such as along the Trans-Within the ANWR study area, there are five Alaska Pipeline and in the National Petroleum

major terrain types that can be defined on the ba- Reserve in Alaska. In ANWR, typical coastal-

sis of dominant landforms: flat thaw-lake plains, plain topography, with large, oriented thaw lakes,

hilly coastal plains, foothills, mountainous terrain drained-lake basins and expanses of low-centeredand river flood plains. These are treated as distinct ice-wedge polygons, is found only in a few small

ecological entities with representative suites of areas, primarily near the flat, braided deltas of

landforms, soils and vegetation. The boundaries rivers (Fig. 4). This topography is best developedof the terrain types (Fig. 2) were drawn from a in the delta of the Canning River and for some

base consisting of a combination of the Landsat- 12-15 km eastward in a narrow coastal belt, asderived land cover map (Plate 1) and the USGS well as in a small area southwest of Barter Island.1:250,000-scale topographic maps. The primary Flat thaw-lake plains compose only about 3076 oflandforms are easy to interpret on 1:60,000-scale the study area.

aerial photographs. The following geobotanical The thaw-lake plains appear to be remnants of asummaries are based on the field work during plain that was once more extensive. They are topo-

August 1981, Zhe results of the planimetry analysis graphically similar to the plains east of the Can-

Figure 4. Flat Ihaw-lake terrain in file delta of the Canning River. The lakeshere show only weak orientation.

II

4412

I ha t 4ae ph / wr fil/ 1,

;..

" p

Jigture 7. Unit i11, Wet Sedge Tundra. The areas between lakes and ponds are,itoslv wt' %t'di!e tundra. The enire area is contained within a large, partiallv drained

take hamin with dislunct ritis and .trang.l'attres. The interpond areas contain aboutl200% ,otsi tundra on he stranes and hiher ,nlcrosite.s.

ning River. being compo:,ed of more than 307o (4607 cover) (Fig. 7, Land Cover Unit Ill), Waterwater, mostly in small (generally <260 ha), shal- (19.50o cover), Moist/Wet Sedge Tundra Com-low, elliptical, oriented lakes. The areas between plex (12.60% cover) (Figs. 8 and 9, Land Coverthe lakes are poorly drained as a result of a very Unit IV), Aquatic Tundra and Pond/Sedge Tun-low surface hydrologic gradient and a thin active dra Complex (8.40760 cover) (Figs. 10 and 1I, Land-

layer. Some microrelief is nearly always present, Cover Unit 11). Elevation differences of less thanusually in the form of low-centered nonorthogon- 0.5 m are major influences on the distribution ofal polygons, strangmoor or low-centered polygons communities. Microtopography associated withwith pond complexes (Fig. 5). the rims, basins and troughs of ice-wedge poly-

The area is underlain by ice-rich permafrost at gons creates distinct patterns of plant communi-depths of about 40 cm. Except for polygon rims ties (Wiggins 1951, Britton 1957, Cantlon 1961,and slightly elevated microsites, the perched water Walker 1981).table is very close to or slightly above the soil sur- The patterns of plant succession in the flatface for most or all of the thaw period. Fibrous thaw-lake plains are, intimately linked to the ori-Histic Pergelic Cryaquepts (Fig. 6) or occasionally ented thaw-lake cycle (Hopkins 1949, BrittonHistic Pergelic Cryaquolls and Pergelic Crya- 1957, Carson and Hussey 1962, Billings and Peter-quepts are common in the wet and very wet areas. son 1980, Everett 1980a). Although much hasPergelic Cryaquolls occur on the mesic polygon been written about thaw-lake mechanisms, the cy-rims. The soils are mostly neutral to slightly alka- cle and successional patterns are still incompletelyline, even the saline fibrous Histic Pergelic Crya- understood (Mackay 1963). Probably the biggestquepts and Cryohemists along the coast. questions concern the time scale, i.e. how long the

Vegetation. The vegetation of the thaw-lake cycle takes to operate and how long the present,plains is similar to that described at Barrow (Brit- wet, coastal-plain environment has existed.ton 1957, Walker 1977, Tieszen 1978, Brown et al. A steep summer temperature gradient is of pri-1980), Atkasook (Komirkovi and Webber 1980), mary importance with respect to vegetation nearFish Creek (Lawson et al. 1978) and Prudhoe Bay the coast. Data from Prudhoe Bay and the Trans-(Webber and Walker 1975, Walker et al. 1980, Alaska Pipeline show that the mean July tempera-Walker 1981). The dominant Landsat map cate- tures at the coast are within a few degrees of freez-

gories (Plate I, App. A) are Wet Sedge Tundra ing due to the ice-covered Beaufort Sea. More

13

,I !

q420

/ ts'ure 8. OL'i I 1a. Moist Wet' Sediw Tundia Compl(ex. In ,l,,% tAallp(' t/h'

comple.% c onis~ts i abldi tqll/ /1)JIII hi'! andi mior, tiindru.

figure 9. Ground views of Moistl We ~t Sed(.'e Tundra Compldex. 7te relative/v-wsell drained area in, the./reiground~ has a dliv'e'flora wvith/i ~s'de.. primtrate wi'llow.her/is and lichens, while the si'eter areas (the dlarker itclus; it/ih cottmngras j have

inainli' sedges and mosses.

14

J'ilre 10. L nu 1ID, J-11)111 .'wd~e lundil (wnplev. IN CUIan/l/ 01"I p' e

ani all ua/ilic c o1)PM1i1?lhilIC If I I/it w I ,, I / /7 1 w- cI I I ',, I Ii -f I rn I, I It '1 SC/ .

l'!( itrilii Shrub Tund/ra o)li/u the i~ riii 1%.

igure ii ii 1llh. A quiiic Tundra. 7/li circular (Piiii * I ve'L00'io)1 all' ,,,a,,ilvaqualic .wdge (Carc' aq ua ilis). At,' (1(1A er area'i% betwee l~''?c/l/% arei Iiewilf grass

(Arcrophila fuIk'a).

LJ

limurt 12. w'nil Iltl, I Set Sd'e Tundra (.l/i'/ ). I/t11 rea /I the delta ol/utt' amA ir Rit -'' v' ti l/i'dh I hqlint' wdvt' W('atr\ 'sUbspat hacca) andcp 'on! alah a1ira% 1:6tc11llia p1hrxganode), flit' siA are sahu' (I/ac iv' i Perel t( "liaqtlt ll J.

moderate temperatures are found inland (Conover vegetation in northern Alaska has been described1960, Cantlon 1961, Haugen and Brown 1980, by Jefferies (1977), Taylor (1981) and WalkerWalker 1981). Low summer temperatures and low (1981). Within ANWR it has been studied by Mey-levels of summer solar radiation associated with ers (1981) in the Beaufort Lagoon area.coastal fog are primarily responsible for a distinc- The immediate coastal areas are characterizedtive band of coastal vegetation that has few shrub by sand and gravel beaches, spits and bars. Thesespecies, limited tussock formation, reduced moss features compose only a small percentage of theand lichen growth, and fewer species in the total study area but are locally important for the nu-flora (Cantlon 1961, Clebsch and Shanks 1968, merous birds that use them. Vegetation is sparseWalker 1981). This band of coastal tundra, which due to the very unstable substrate. A few species,Cantion (1961) termed "littoral tundra," lies such as sea purslane (Honcken'a peploides),north of the 7 °C mean July isotherm. Worldwide, oyster leaf (Merlensia inaritima) and commonthis zone is equivalent to the arctic subregion of scurvy grass (Cochlearia officinalis), grow onAleksandrova's (1980) Tundra region. Near Bar- slightly protected beach gravels. Barter Island isrow the coastal strip is about 100 km wide; at mostly covered with tundra, which is a remnamPrudhoe Bay it is about 25 km wide. Within from when the island was part of the mainland.ANWR the littoral band and the coastal plain in The other islands, however, are barrier islandsgeneral are narrower, and the coastal temperature that were never part of the mainland.gradient is more compressed.

Along the northern limit of the littoral tundra Hill coastal plainsband, there is a narrow strip of vegetation that is Landfo rms and soils. Hilly coastal plains coverassociated with the saline soils found immediately 220 of the study area. Stretching inland betweenadjacent to the coast. This area is affected by the Hulahula and Jago rivers is a complex regionwind-blown salt spray and occasional storm surges of gently undulating tundra with small thaw lakesthat flood large areas of inland tundra. Taylor and pond complexes. This area is quite different(1981) divided this shoreline vegetation into six from the flat coastal plains. Stream drainages arehabitat types: tidal salt marsh (Fig. 12), upper- better defined, and large expanses of well-drainedstorm-zone salt marsh, gravelly beach, raised terrain border many of the streams.bench, coastal dunes and coastal bluffs. Coastal East of the Hulahula River, particularly be-

16

'Iw

opo

4;

Fiur 1. ericl eral ho gap (cutk/,00)(y col. olio v % (I-/I//If ~tit/Il dtvi/E 1. 1

Me~gm na Rie,-uha P1Bre tln.lpad al-41oll IIc% is 11 lm h-tl(l(m cle 1-/1deeSaIlc 1 vofalt n.*.,te trj l m oo livm(.eN 4i o r ieeii i ll k

171

tween the Okpilak and Jago rivers and between while soils in the depressions resemble those of thethe Niguanak and Sikrelurak rivers (Plate I1, there lat thaw-lake plains. The gently sloping (5Q0 orare many slightly elevated ridges and depressions less) interflues mostly hasc moist tussock tundrathat parallel the coast (Fig. 13); most have less with flat-centered polygons. Frost scars occupy upthan 10 m of relief contrast. Fhe origin of these to 300'o of most surfaces and have Pergelic Cr~a-ridges is uncertain. On aerial photographs they ap- quept soils with loam and fine sandy loam tes-pear to be either dunes or beach terraces from past tures. The soils between tie frost scars are Pergelicmarine transgressions. In 1982 a brief check of' Cryaquolls, commonly w ith 1-12 cm of sapric, or-these features showed that they contain gravelly ganic-rich material as a surface horion overlyingsoils, so they could not be dunes. They may be ex- mineral soil with a loam, or occasionally a siltpressions of the underlying geologic stratigraphy, loam, texture. In August the active layer is 35-45as are those found in the foothills. The ridges oc- cm deep. A water table does not develop in thecur up to elevations of 200 tn on the northern polygon centers but may develop in the narrowflank of the foothills. Flat areas between the troughs between polygons, where the soils are (fib-ridges contain complexes of wet and moist tundra ic) Histic Pergelic Cryaquepts.associated with poorly developed ice-wedgepolygons. Vegetation. The vegetation, like the landformsSome areas, such as those along the western side and soils, is a mix of that of the flat thaw-lake

of the Niguanak River and in the coastal area just plains and that of the foothills. The vegetation onsouth of Pokok Bay, have hillier terrain with well- the low ridges is mainly moist sedge tundra thatdeveloped drainages. The vegetation and land- may or may not contain cottongrass tussocks. Theforms closely resemble those of the foothills (de- dominant land cover categories (App. A) arescribed in the next section), except that the hills Moist Sedge, Prostrate Shrub Tundra and Moistare less steep and thermokarst pits are more exten- Sedge/Barren Tundra Complex (Land Cover Unitsive on the broad hill crests. Elevations in the re- V). The latter category is associated mainly withgion rise gradually from near 30 rn to about 100 frost-scar terrain (Fig. 14, Land Cover Unit Vb).m. This unit covers 37.6016 of the hilly coastal plain.

The soils on the ridges and gently sloping areas Moist/Wet Sedge Tundra Complex is also im-are similar to those in the foothills terrain type, portant, covering 33.50o of the hilly coastal

..

M

ligutre 14. Unii V'h. Moist Sedge/Barren Tundra Comph'v (tosit-scar tindra).This is a t ypical.ro.-sctar area on a gentle slope. Note the lack of cotionkrass ItUAock.%atltf dwarf .shrubhs. The i' esgetatji bet ween fro.M scars (oPt.I.5t minnl of 8is'huel A '.5 .'d!A'

(Carex bigelowfi). art ic avens (Dryas integrifolia). w'ide-leafi'd arctagromtA (Arcta-grostis latifolia). the ipois Tonienthypnum nilens and numerou.s .futicoe fIIchl.s.

18

I I Im 11 , [)" Ii 1ep 11' , i l III%:e 1111t oai a pLI me' sit Ikeiil IIJ l Ild LIIIdI I\ III l 0 ll I ciiIii I'li\kI Id~ \\CI I I Iilei I' It l Ii Iili 'Iils 221 o d l e and Shles lu ( ilaee i Of" If 01Ci 0 IL'

11atte1r L% lg lil ~ui~ l 111nit Ipilk IJU L ls/,,.* like celuieiils ilfile Ite Ilihe, 0I 1i*. I , ). \% llIL,

.'il 4' Of I lLC lil% .I L . I il ls l\ IllieLlill\ ieril t cIs.IIAik.' I 011ii Cdi 110111l 1 C1011l 111C Iti~lii IM~ l le betu eei HeiCeeu B:%c li e1. ( .liiiiiii aiind '~Ici '. IiiilL ik .ii' a lu~i~c I lie elc\,.iiiliis of ieJ iulliu011s u'M\ 1001111 l l t: 110111 ,ui ( 1111,ICII 1,i\ [ 1 IOf.L

.1 I d I I I11k.. i, l e ( lilii'e aie suhlilki c I mo li, a ~ Io i III 11-li, I lit: IVC5s ll 01 \ Ccliitl ,I IC tillk .

StflIIIp I Il SiCd 1i1 (lie uorih1-suuUtli patieil is onle et~l hai~e haii cii (Iite Wiic 161 ii c l

1 wr n Iu /5 (lipt ' l /A w rai wi ii,? iui illitell Bl v. Ic lipi l 'if/ 1 1, t- (i C,

,hl Pii il'i l hi /A t M/0 /h/ I it/ )t 1 1 It t u t .',hpill I tuu iw \ /ii , it i R i I i

lI iure 16. 1)i %co il lio~ll.i lU grit did hH i/ inl d ii' /1 u/A.

19

I. /t/ItI I old' tti/ i t h/ i t/I I andt U t ' ef I/I I I I a. loitIl I! Ic '/ c W"

hot l t Iopt k Now Ibt it/io~l' plt //IC l,// I/( h1t/ ! oftil%.

It l /S I I 0011)111% StiII't /it Itoi 4110% li/ l/till ) I P4.1I

it) lllo ill i %% tth /il t) I iii and I lanklitl BfC ' I Itiit it.2 :lii h pa al 0 k

I lie I a ai mia,k Rixei, Olet Kaakuiiuk R'i~ci. til. t, -i ilici 20 kill ttltit01111l1 L'\I~L'tl 10 11W It

2(0

/-igure 19. Unit Vlic, Moist Sedge Tussock, Dwarf Shruh Wei I)wurf ShrubTundra Coinple." (14 aler track coipmple.\). Ilit drainage .hunpiets are thinnuied hylwarl ./hrul. main/y diamoond-lealed willow (Salix planifolia s.,. pulchrap and divarlhitch (Betula nana s.p. exili).

posed with patternless areas, strangmoor and low- conduct snow meltwater and subsurface watercentered polygons. Thermokarst pits are common during the thaw season. Parallel and sub-paralleland indicate substantial amounts of wedge ice. water tracks are commonly present, giving theThe better-drained elements of the hill crests have topography a ribbed appearance (Fig. 19). Strang-Pergelic Cryaquolls or Histic Cryaquolls with he- moor is often found in the channels, suggestingmic- or sapric-textured organic horizons overlying slow mass movement of the saturated soil. Wil-dark-colored, organic-rich mineral material, lows and birches are concentrated in these fea-Cryaquepts occur in frost scars, which may oc- tures. The inter-water-track areas are tens to hun-cupy up to 40% of the surface. Mineral horizons dreds of meters wide, with relief from 0. 15 to I mof both soils are loams or fine sandy ioams with above the track. The water track portion of thevariable amounts of pebbles. The active layer slope presents a relatively smooth and gradedranges from about 30 cm thick beneath the cross section. Polygonal outlines are usually notAquolls to more than 60 cm thick beneath the apparent, although ice wedges may still be exten-Aquepts. A water table is absent or well below the sive beneath the slopes. The soils are Pergelic Cry,-surface. The wet areas commonly have Histic aquolls or Pergelic and Histic Pergelic Cryaquepts(>20 cm of fibrous organic material) Pergelic (Fig. 20). In most cases, 4-15 cm of Hemic er Sap-Cryaquolls or, if the colors and organic carbon ric organic matter overlie mineral material, .hichcontent below the histic epipedon do not conform is often mixed with organic materials. The activeto the criteria for a mollic epipedon, Histic Crya- layer thicknesses range from 30 to 50 cm. Thequepts. In either case a water table occurs at or presence of a water table is uncommon. Withinabove the surface, and the active layer is between the water tracks, Histic Pergelic Cryaquepts are40 and 45 cm deep. In the wettest areas there is the most common soils. The active layer depthsenough buoyancy in the fibrous, organic, root- range from 40 to 50 cm, and a water table is al-rich mat so that its true thickness is difficult to de- most always present, commonly within 10 cm oftermine. the surface.

The hill slopes are generally greater than 5% Frost scars are almost always a component ofand are covered with cottongrass tussocks. Areas tussock tundra and can compose up to 50% of areferred to as water tracks (or "horsetail drain- given surface, with anywhere from a few percentages" [Cantlon 1961]) are shallow channels that to 75% showing some activity (i.e. having bare

mineral soil exposed). Where slope breaks occur,21

PIP,

t .

i ptt 0 oo h ae ttcAaea hwn!acml. (o ~/01"

the H11ssiA (/cn110I if / ip l o1ic tI/tII WelV/ ( /auol All / PI vi.C/

Vigtwv 21. I)-os-.scur f'rraiIin ti f.1oi/.

both the density and tie activity of' frost scars W\here bedrock is \ci\ close to thie surf'ace, as,commonly inrae(Fig. 21) in teecases shown in igur f6 1rost scat,, and, or patterned65-8O0o of the surface masy be composed of' active barren grase Inma\' comnpose 70)-80%W of' the sur1-frost scars. Frost-scar soils (Fig. 22) are composed l'ace. Because these surf'aces are exposed. snox%of grayish brown. usually mottled, sandy-tCX(tured cover is thin or absent, and abrasion by blow ingmineral material. Under the present tawonmic snow% may be severe. However-, well-developledsystem, the soil pedon is considered a Ruptic- soils may occur beneath stable microsites,. TheAqueptic ('ryaquoll. soils, are lPergelic (lit hic) ('ryumbrept s if' uf'ficient

22

CrvuqJepl andl Iltstic Pert-geIic 0-'vcuiuoll. Ime .(a oil uo-

organic matter is present or Pergelic (lithic) Crvo- tess steep., urailyI south-f'acing slopes hia\C \%ell-chrepts if it is not. The active laser is greater than -etilope Shrub Iudr (i. 26.~ 1ldc(os cI mn deep, and ice volumes are generally low. Untit VlIII). Dense shrbill i cdeude lt

Solifluction forms, such as discontinuous stripes uiioW along mail\ stream mainits.of frost scars or lobes. are common downslopc rulssock tundr-a ca n have a i bess ilder i n an as I f0Nofrom some outcrops and where slope breaks exceed subty pes that are difficult to classits* . I hese are7-104ro. This is due principally to the water adC d related to lactors such as slope stabilils, .Soil nlors-

by melting snowbanks. Microrelief is relatively lure. cryoturbat ion and successional hlislors th legreat. T he high moisture environment and the ni- effect of, frost aictivity is of prIi mar \ inpor 1tlncecrorelief make such areas quite susceptible to ve- (Hopkins and Sigafoos 1951. Sigaloos 1952.hicular impacts. Churchill 1955. Racitie and Airder son 1979).

Vt'etation. The vegetation intile foothills is I-rost -scar I utdra is widespread ill flie toot hills, In

predotminantly Moist Sedge Tussock, Dwkarf Shrub arecas wvith netral oi slight ly alkaiiie soil,,. 11hi,Tundra (Iig. 23 and 24, l.and Cover Unit VI). has importatit implications t'mrtlie s euetlion. ipwhich cov'-rs 53.801 of the foot hills terrain type, lan~d tundra soiil- are norma lacidic due to thickdIn some areas shrubs arc dominant and thle land peat Ia\ crs and t[lie accuniulat on ol or garic acids.cover class is Moist Dwarf Shrub, Sedge Tussock Spilhagnrulu mloss. tluliierous er-icalceous" silt ills.Tundra (Fig. 25, Land Cover Unit VIla). and ot her bog species. such as cloudbeti\ (Ru/Ins

The vegetation in water tracks is often domi- c/umaemuorus)., are adaptied to an acidic ens it oit-

nated by dwarf shrubs, mainly dwarf birch (Rletu/a ment. Where thle soil is mlore alkallinle, as' inl placestiana) and diamond-leafed willo\4 (Salis planifidia %% here frost stirring has brought alkatline patretssp. pulelira). Land cover within water track com- material to thie surfalce, many\ of t hese species aicplexes is generally classed as Moist Sedge Tussock, replaced by at different suite of' upland tunldraDwarf Shrub/Wlei Dwarf Shrub Tundra Complex plaits. including aicl avenls Ira i~t'i 1 ).1

(water track complex) (Land Cover Unit Vile). A Iligeloss % edge (( lre'x higu'Iowltr,. ssooh> \\ illo)\\

23

if i.. ur 2.3. 1 nit I hti. .lii Scleel~c !u%o(.A. I l/,, hub I unduhu It pehI ties-%(),A 11/t1ti . tuIh c/th oll l1oth/lA sou/h of ( olui'll, /?(/. Ih' ael 'c

0,c~~aw hawc Wom So , / / alt Sel I&i

.I i.'iet 24. ( ,)//illh/% U1%%c n nit I 1(/. i/t fnni ulw oil i ielic //I(. milhul shrilbs arc d1,,,nc cud It a/c c 1 1-18i/011' ( SZ i\ jl i fiii l cI. ci)L IIIJ) (14. t /)1/(,r Ic 114 Ik i uli tiala ssy. t'~jlis). Icii, //c'cr i accnil iI I ( III IciJ ) cm/I 1lich itl l Ifl-

S(V". . sits-idaca I.

24

1i iiurt' 25. L nii I 11a. Monist Iw), 5/jiub, S'di.'c FilsnA i unda (Upj/aniddwarl -.~ Inu .v oc)A tundral). /lls will ocurill(1 /hit'(wa-ii i u Ibd iii'l 'hun su

liure 26. Lii 1"111, Shrub Tu,:dra. Thismieu is ona teimh/umQo(u/Uu rocky i' 4dLc netar the Jtq'( k, ver.

25s

and (he gra'' Allllf ' s~ts I ri t I Ca Il tl- t/aut turnl aniid IIih)/I mat1 / /n' Io tie

dra is inOs cointjotil onl siopeN antd i'" pa I itulatlx. soutl Ith s1 ccoitdai% lava notiniall ilae mte acid-coimlOn ill hti %we'lcrn patl[ of (lic sludx a: ca. iphilott, aldinc d h1CMI Ie laxa meITIon11led Il il e plc-

\\'ill] the presentl I and'ai itlap caict-itl ic'. if t' cCditig palagi apli.nll ot sbcltd~liei'hI fotl-scat tuiitdi (I MittCo\ ci U nil Vb ) I 101m N iolsi Scdge.' P1t I a(iC 'lstsiwaloull It/il nSil 11i) [un1dia (I anid (io~ci I nii \ .a). I Io%\Ccl . I *;01ol/s (nti sm/s. \Ist1ittiailltti ICE 1,01tt t

hIe latici i( pe OCCLnr1 alonet 'Ir1Cain1 di ainlai'N n1cal i' E[ in 10 11 ict fl at SIlliscliit '~ tiV and cofihe coast. Mihcre ius''ockl and dh%at I 'tir1Nb Zile ti~Aboi 0.03"1 (it til s'iudl anca (I Ic'.

Hot caclj~C iid 2s. 1I1 li-.atca t1 itlo~ill abo\C (m) itl and ix

Fihe bload top', of' gent1le hill', al-e Lusnalix ,(able uIilICIlaii h\ na (JM111c 'and'iotw' Ithat itp'and ha~e peat accutmulation'. [he ,Oils are conle- pilM101 iou'a tile sadle Iochl M\litaiii'. I ltc'cqIUell! l acidic, supporting thick Sphiagnumi peal Xi, \a' let t)Ot \1l'ti dinTI i nt1CI It)' 11Il idk~iK.x,%ilth cotlongrass tussocks (Lrio/haniun vaieina- \Ia't at theC coillllleiil lici c ac vend iat and atlemiln) aitd d,.\arft shrubs. SUCh a-, d\Narf' birch (Rtlu- based oln iiitaiillatioll itathlced tont 'iitmlal at ca'Ia iana), diamond-lealed \%iIlos\ (Stil-\ ii'i/a/iia %\C1i oft ilc -1iid alca. An ideali/cd '0cti01i otINp. ;tu/chra.). Labrador lea I Ledillu dtu17ll/'fl% lettli~oi cI ali tx i t MI inl I ILte il 29.bog blteberry I ~I cciniumu u/irgino.%tl,?) and moun- Ridue ci c'i' hasc ejl at poradic ac~cti nc' (it[lin cranberr\ I. vilbo-iu'). (oil' anld \cu-elaliol an.sii tcitittpi LI itd rat -

[aI I( atihe 1ltilahutla Riseit hIde i. at noticeale letiiI aiid to ack nid. I hie Noils o aiidute'Zate miatni

bonit11darx beisecit Units V and VI ciesponiding, Pet .li i I\oillle] ot (~Vube I' I!-,. Itot itapptosittac to1 tile break, ilt Nilpe bcels fC ile Itile [C\xiC Zilae 111 te ionith. llittie \\ftilt it1 its ta-

tlaicr plain' and tile toolltillN tIPlate It. \01r l Mtires. Lppei . Nleepet 110r6011S itt mosl alpiiiClhii bolldar\ tIle "Oil" are chat acteri't call\ ICNN Nhopes ate ittatledC b\ 'cree or block~ !alu. I licc

acidlic thtan 1o tile ontll. IJo tilte no0 t. ll'cic: tileN depos'its L 0itti1t01t1\ displa\ block ,tripe' Zitld Ot1\ picall\ lItas cot lolirass lu sitcks isitt basiplti- block-bordLIEed lerraces. Ice kc0itttoit11 till' tile Ill-Ians laxa Nuch l a acliIc a\Ctts tIrlia.s lli'elj/liq.ie'lce of he fitter cobble aitd eras el-sued t*1ILis ool'. \iiloss I.SaiN hialitau). tl(- ciited \i tloss (S. tiettis belO\s te laNe 'tit ac blocks. P'ocket, it1

NVt l hui alitd p11-11 11rp Ic itt tsi ill sasa hraue .Vixt - Per gctic Cr V um blqms, otfcrc ' o oacca-/lirttaf/t ntl/aia. s ilt Ntisc uch a' Faonew- 'tout. Pergelic 0(rvaq 01ntl occurI itl t-iHtte-lest iicI

f-iodrre 27. Mounino~ius terra-in ill d/id, .%ih/te rt/tIt I MtltliiuI -. t /tlt' ftiidtt

26

Iiiur(' 28. Sad/c'roct .Spring. M~any anusual pilat coimmuniles art, associated ti ilthe' .s/lrit .

D, 100P CrYumbreptP Cryorthent D, 100

A! P Ruptic -Entic CryumbreptB21

B22 I O. C Aft -T

M, B0,W, 20A,2'0 2

Cryhemst HIstic P Cryaquept Efl8. IIC2 10 1P Cryoquept mc -. ! C

M, 80, W, 20 ~-E 2o82Ohl V_

Histic P Cryorthent - h B~ 21 -j

M00uesaus(t0dy Mmlt -e

I~~ownslope from the talus (oigture)statuhere th dryd M-oled and owet tikndogai oio

vegetation coverri comonl copee aid topose c Theos activ ayeru Tkessan te sopsvre

hianks develop, the slopes display a variety of soli- considerably, ranging from 70 cm or more in thefluction forms, including turf-banked terraces, wectter areas to less than 30 cm onl some of' thelobes arid stripes. A complex of soils is found oit better-drained microtopographic elements withthese slopes, including Pergelic or Histic Perigelic organic-rich soils. Where coarse, blocky talus unl-(iryaquepts in %Net depressions, Pergelic (irya- derlies the solifluction slope, moving water ma%

27

distribute mineral material over the surface ol River flood plains coser 24 .6 0'o of the study area.otherise organic-rich soils. Solifluction slopes The present river flood plain consists of the actiseare naturally unstable, ,, ith a complex of perched channel and usually one or more terraces (-ig. 31 ).s, ater tables and subsurface drainage. lhey are An idealized section across a rier channel is,Cl susceptible to %,ehicle traffic. ,.,hich can pro- shon in Figure 32.duce mechanical and thernal erosion. The major rivers within the study area ha,.e

I e('w10ton. The segetation communities in al- braided channels ranging in s, idh Ii rom about 0. Ipine areas are complex and are interspersed wkith to 4 ki. The diamond-shaped islands betsseenunsegetated rock and talus slopes. The character channels are probably inundated at least sporadi-of the egetated slopes varies considerably. Alpine cally in most years during the period of meltolfloras are excepionally diserse due to tile wide (usually late May to early June). To tpes of is-range of' lithologies, soils, altitude, snow, depths, lands are recognized. The first consists of unege-exposure to \,ind, and site moisture. fhe more rated grasels and gra\elly sands (or silts in tile del-compleel> .egetated areas have extensive moss ta region). These areas are flooded annually andmats s.ith uunerous small shrubs,. such as mnoun- are subject to intensive svater and ice scouring.rain asens (I)rvas ocloeitla), prostrate \illos, s There is not any soil or vegetation on these fea-(e.g. Salix arctica, S. chamissonis, and S. phleho- tures, and they are described as ri\er wash. [he sc-pity/Ia) and small forbs (Land Cover Unit lXb; cond type of island is somewhat higher abose thcPartially vegetated areas. Alpine tundra). There main channel. The soils consist of sarious thick-are few, detailed vegetation studies of the Brooks nesses of silt, silt loam, loam and fine sand\ loamRange; the most relevant is that of' Batten (1977), oser grasel and gravelly sands. In the most stablewho described the vegetation of the L.ake Peters ,ituations a thin Al horizon has developed, andarea. On the L.andsat classification, mountainous some mottling occtrs in the fine sediments. Theareas are depicted as either partially vegetated acti\e layer normally exceeds I in. The soils are(Land Cover Unit IXb) or barren (Land Coser classified as Crsorthents. Most islands are corn-Unit X). plexes of Cryorthents and river s"ash.

In a less cases, sshere fine sandy surface sedi-Riverflood plains ruents have been r e\s orked into los. discontinuous

Land/ornts atnd soils. This terrain type includes ,and dunes. Cr opsaniment soils occur. s\ ith t has

the present channels and braided drainages as well depths in excess of I in. Sand dunes are, hosieser,as the adjacent abandoned channels and deltas. rare in tile stud\ area and are confined tIiostly to

Figure 30. Pergelic 'vuviohrept soil. This soil is'ommon io alprine crest areas.

28

- '-.a-. .-- 4

?I~

Figure 31. The C'anning River. Such areas contain varied habitats including barrenantd partiall " vegetated river gravels in the tcbanniel areas; willows and tjryaN terracetcoonouies are in the older chaninels to die rig/it.

Moisture Status (%) W- wet, M -moist, D-dryTerm is not currently recognized in U.S. Soil Taxonomy

W 30, M60 W 20. M80P Cryoduept W 0 0P CryacquoI a

W 30, D 70 W 10, M 50, 0 40 istic P Cryoquept Pi Rut~ -90 i rauo P Cryoquepts

P Cryorthent P Cryotluvent' 01 0

Ci - 01 H Ohl 0ciVC 01 Cg

R20 e - an Comle Iu gI

Figur 32. tdaie oeuneacosRvrFodPan aFo/tls

theOo detso h ann n aorsr.Teemnyrti h atrso hnesadilnsfetue ar smll motl lesta .5mhg. Fr h otprt h ol o h erce r r

an r opoe lfnesn ih infcat otets(i.3) n r el rie.0h vrsilt~~~~~~~~~~~1 co net lyn ie aeilsmyb 2 mo mr hc

Beyon the onfins ofthe baidedrive chan and re cmmonl beddd.CTecserlaes/rnel cmple area nuber o paied ad nopaird comosedof fne snds. Te re iluve n

teracemst F icre daoved te influence f~s sie lso ore Plains n clotmates.oeeesi

snoe-elt ofloodiCn igadorvr. Theonettra es o-aonoy oes oth pallenofo Fhanees nd perla-s

fetuesae mal mstyles hn .5mhih, Fr hemotpatte oisofth eracsa29Cy

and re ompoed f fie snd wth sigifiant orthms Fig.33)and re ell raied. he ver

Figure 33. Pergelic Crvorthent soil on a Dryas riverterrace (Unit I I'b). This soil reseinhles a I-lu vent o. o ortsouthern areas.

frost areas. The organic-rich surface (A) horizon traces of the river island pattern. The soils of themay be up to 15 cm thick. riverine areas. %,ith f'es e\ceptions, are ncail\ nete-

The channel areas are poorly drained, with fi- tral in pH or are moderately alkaline.brous organic (0) horizons 20 cm or more thick o\ - Steep bluff slopes ( 240) along the ri ors and aerlying mottled gray silts and silt-loams. Perma- narrow strip along their crests are included %,,ithinfrost is usually encountered within 0.5 m. These the riverine area (Fig. 32). Bluffs may be undei-soils are Pergelic Cryaquepts. A surface pattern going active erosion or the% may be fossil, in ihatconsisting of I- to 2-m-diameter polygon cells (re- they rise above long inactive river terraces. lI ci-ticulate pattern of Everett [1980aJ) is common on ther case they are subject to rapid natural failurethe better drained island elements. Some terraces by mudflow and/or solifluction. The result ismay lack any surface pattern and consist mostly of commonly a comple\ of soils. Solifluction lobeswet graminoid tundra, while others may have with over-thickened (cumlic) A horizons t1aweakly expressed polygons with disjunct rims stand 50 cm or more above the surrounding slope.and/or strangmoor. Wetter areas occur upslope from the solifluction

Away from the river (Fig. 32) there is poorer lobes. The soils are mostly Pergelic Cryaquolls.drainage, with Cryorthents and Fluvent-like soils but Pergelic Cryaquepts may be important locally.giving way to Pergelic and Histic Pergelic Crya- Steep drainage gradients associated with bluffsquepts in which the surface organic horizon is often produce a microtopographic reversal, i.e.composed of fibrous sedge peat and roots. The ac- the conversion of low-centered polygons to high-tive layer also decreases in response to the highly centered polygons due to erosion of the polygonorganic soils and poorer drainage. Disjunct low- troughs (Everett 1980a). During the course of thiscentered polygons and/or strangmoor mask the reversal the low, highly organic-rich centers (often

30

Pergelic Cryaquepts or Histic Pergelic Crya- supply a constant source of water during thequepts) undergo drainage and oxidation. Corn- winter (Childers et al. 1977).monly the soils are enriched with mineral materi- Vegetation. The vegetation types associatedals eroded from the exposed bluff or derived as with river systems are quite variable. The braidedeolian material from the river-island complex be- channels (Fig. 31) are subject to intense disturb-low. The resulting soils are well-drained Pergelic ance during spring breakup, and meanderingCryoborolls. The organic-rich surface horizons streams and braided rivers are constantly changingare underlain by variable thicknesses of oxidized, their channels. The first plants to colonize riversandy-textured materials that thaw to depths of I gravels include the river beauty (E')i1ohium lati-m or more. The processes of topographic reversal folium) and arctic wormwood (Artemisia ar'tica).are quickly attenuated inland from the bluff edge; Slightly more stable areas are often only partiallywithin a distance of 100 m, low-centered polygons vegetated but may contain a wide variety of taxa(or some other surface pattern) are generally well (Land Cover Unit IXa; Partially vegetated areas.developed. The addition of wind-blown fine sand River bars). These are among the most floristicallycan usually be recognized for distances well be- rich sites in the region (Fig. 35).yond 100 m. Bluff crests and their soils undergo Willows (SaliA spp.) are common on partiallynatural profile disruption by cryoturbation and vegetated gravel bars, and may form dense shrubwind abrasion. Because they are exposed, they are thickets; however, these thickets are not nearly ascommonly snow-free during winter, extensive as the willow communities along rivers

Another important feature of the larger braided farther west, such as on the Sagavanirktok, Ku-rivers is aufeis, or river icings (Fig. 34). These paruk and Colville rivers. This is apparent on thelarge ice bodies form during winter in sections of Landsat classification and color-infrared photos.the riers where there is constriction between the This situation contrasts markedly with impres-river bed and the overlying ice. The resulting hy- sions gained from earlier maps of the region (Vi-drostatic pressure cracks the ice and permits water ereck and Little 1972, University of Alaska, Arcticto flow over the surface, where it freezes in a thin Environmental Information and Data Centerlayer. In normal winters, numerous layers will 1975). The major ecosystem map of Alaska (Jointform thick aufeis deposits, which do not entirely Federal-State Land Use Planning Commission formelt the following summer. These features often Alaska 1973) shows extensive high-brush corn-occur downstream from perennial springs, which munities covering about 20076 of the study area.

Figure 34. One of te numerous river icim.. along the (inmin River. 27 ,hiy1981.

31

.. . . _._ .,

I 'ii'35~. I(Iu/I. vei'(1/t'l/le'( ivelr burt on te Auctkuitn tAIt' ". I/ic mum(1,~'mtii cI~iiI'/JitC' 'cIhIVI/IIsII0 K~' NIItlJa Itdi

I imuntrc 36. Afecaltderim! tu / it It/I(II/u/. \(Pci the' ivi/li-N. 1,1,(Irtat/ (1/I1/ I/h IC/c

Wt //If beA

tlo\%c er. oil [Ihe I atid~at Classification (Table 11l) tingu'tiill h.etmeel thle rker areas that Zile \cgetatedihriic kle,,, thtan 0.21i co~ ci of ,hitib tundra \%ill fail "illo", and those that Zile partiallk eeCL-\%ithin thle entlie tc d area. D~ense ,Iittuh, \%thinl lated \itl prostrate NIhruhsb and or fork, (lip. 35).mlajofr ri %el drIirtfagcs (la t e 113) co% er otlv %1 i f allI th Ie p1arI t Ily '.C vegtIC ted a Su t11 iO(N1"o 01 tile itid\ at-ca. \lost \%l(\ Cj)ljlll ll terraiit type \\Cie domlinatedL h 1,1110 ~ il~~~lie-, ahitt thle majorI rI'els are tairlk open., %\ilil %illii" '"i',iblatrd" wNould total rio more thianlarge CoollI n1T\C1' 01f grae arid riplari all fork". ()n I1.25"l'i of the ica U able 103).thle I alld~at Clav~liCatiloll it i' not po""ible to di"- Dll"Cn. \cillo\\ eonrnrnnllitie' al~i occur alcrir

.32

Fir~'ure 37. U'nit l1,7b, DI-1- Pi-rsttvte Shfrub. loh Tuntdra (Oryas II er, W/watvc).)"is 1111 UnU ipallY ()((-UIA oift iewlh-an r e r Ic'rauc'. Hlvlh vni,,oll i poll-gons20-4() (Inifl (iauuth'. rr. /oi/st~id area e i'omillol. Ihe /iriJmii1 laval hereare arcite avlns (Dlrsas iniegrifotia). blackish mterviiue (Oxytrlopis nligreCsccns). illlini('miA -vet'ch IAst ragalus a) pin us) and coamil Colf(nrIaxs (I riophltiin ii isic).

somi l ofithe smnalkei drajinages outside the major (1i ci r eaces rorri fihe Itroic abuindant 11b11 less 11(11-f'lood plains (Fig. 36), %%here the soik, arec some- iskicaalls an ir tcll'ich Mioist %%el Sedewhat miore ,table. Ili most cases these drainages Tnia otpies.li, ils may be p)OSSible Oil I aind-are niros and Uinlikely it) appear on tie tLandsat sat scene,. takeni lace illi[lhe sluninici Lii Iall. s% heriimage. Smaller st reanis and inter chanlnel areas o, there is a color con iasi blet weci t hese [%% 0 es

the larer risers hiave Ilush sedge and(. willowkstands. ire hieits of' the willowvs %ar' . accordiitgito tilie amtounrt of' wviitei snow cover and flie stitl- (O(N('I 'SION Sfiner tenripeil Lre ireginie. Willows near [ lie coastare mostlIy prostrate; nicar filie southternr boutnda *~ Ni Tis anal ys is has e ii phtasi/cd ses era Iii itla iiof the siltidv area, shrubs cait) exceed 2 in in freight . f'eatLiies of tilie A NW R siuLds area:

Di-viis-dotiiinated riter terrace,, are Coiion I ) Iliass-lake plains cos ci onls about 3%l' of i liel'eatures associated with the larger rivers Wi.3.study area. buit Liiiit%, 55icie %\t tutidia habitats ai cL-and Cover Unit lVb). These terraces are uliall\ likely ito occur (I and Cose c i is 1, 11, Ill aiid IV)just aboc fithe main channels anid are especially cos er about 40411 of [ lie l egioni. I liese ai cas alccomiimon along thieCanning River. On thle Iaiidsat pai t icuila Is' seiisiiisc to siface csplofition andclassificat ion these terraces have spectral sigia- areC xaluable as 5' ater f~ossI Itabmwatlures simuilar to t hat of' Moist / Wet Sedge Tundra 2) Thec area is itiosi I\ hi ll\ aitd co ciCL 11d biroisi

* Complex (Land Cover Unit lVa). Thle loss reflect- iudra of, var ivii character. Units IV, V anid \ Ialice of these terraces is due to [lie iiat of' arctic are tIlie dominitant land cos ci t v pcs. cos ci rig 5514avens (Di-avs in, rtififi) arid thle lack of' light - of tilie region . I ost -scar uridra cos cr5 large arealscolored, er - t, dead vegetal ion. These teni aces a e oI tilie Luplanids. Ilos e c. thter e is ciii ent l\ Spec-

a i in iport anit f'or imiany .species of' wildlife; a retic ir-a con Llu si Li bets' cci I' osi-sea i urndlri anid oilier* ~groutnd squirrels, l'oxes arid lemmings are at(- land co ei Litts. We hope t hat this pi obleiti can

tracted to the dry, sandy soils for burrowing sites, be eliminirated] b\ clrster ing thie data arid iI,and grizzly bears, are often seen hunt ing for thle Lise of1 digital elevationl Miodels MF)1 N .smaller aninials in these areas. Oti the aridsat 3) River Flood plainis arid braided river deltasclassificaion it would he helpf'ul to separate thle coiver nearly\ 25 Il i'iit lie st ud\ area. -I Iis has iii-

33

porrantl implications flt fite mlany% % idlife species lirifion. M.E. (1957) Vegetation ol thie Arctic [fill-t hat tili/e habitat associated %%itl it rer. dra. lIt Arc~~tic Ihllot' (I. P. H ansen, I-d.). Cot -

4) Ripaiiatt shrub comunties are niot neat ly sallis, Oregon: Oi~goti State Universit ['lress. pp.as cesii c as inldicated off pre% bous mlaps of tile 67-130.r egiont. ()nll\ 0.20*t of thre area has, dense shrubs. Brower, W.A.. I1.1. D~iaz, A.S. Prechiel. 111A%Open dulitu cotmmnities along thle rivers account Searhy and J.I.. Wise (1 977) (lhn/imac,/1das o.//thelot no mote titan 1.250%t of thle area. Outer Co'ntinental S'he'//H Baler nd ( oa.stal Re-

WC t commen110d ulSingi the results of this Srtudy as t. 'wns of.-I la.%ka. Vol. III: (ihu11czi- Beaui'/ir, .se(.a basis lot select ing areas wit hit eachi of thle major National Oceanic atid Ar mospheric Adiiita-tell ai t vpes lot derailed studies. itncluding milk- tion. Alaska Outer Cont inetral Shell' Lttsiron -

ing rnit Inaps atnd investigating the soils. %cege- miental Assessment P~rogram, 409 pp.tan on. ueonrot phic processes and wildif Il se. The Brown, .J. (1962) Soils of* the titlhert Brooks,lot atCd \%C %s esitedI inl 1981 Would be gIood st ud\ Range, Alaska. Phl.DI. thesis, Rutgers, Uinis rsits,sites. and wse t ecotumend additng ain alp-ine site New Jerse.te~at SadICI OClijt Spi itg and a flat t ha%-lake plaini Brown. J. (1966) Soils of file Ok pilak, Ris et

Nite soutlissestiof Batitet Island. These stuldies regiotn. Alaska. CRREI Research Report 188. 49\%illd htelp iii es aluatiig tile oserall reliability oit pp. AD 639691.lie I atidsat mlap aitd ss ould aid itt evalulat itg filie Brown. J. (1969) Soils of the Okpilak River re-

potcill al Ilpa tit'o seismic esplorat iotl \it hiti t[lie gion. Alaska. lin The Periglacial Entvirolnen,ss Idtlite tctlice. Past and Present (T.L. ti. Ed.). Ottawa: Arc-

tic Institute of' North America. pp. 93-128.Brown, J.. K.R. E~erefl. P.J. \%rebber, S.F.

IlFERATURE C(run) MacI~ean and I).F. Murrai (1980)) 1 lie coastalutndra at Baiross . lin AI, AIrelic Licos Isteulz: /it,

Aleksandrova. V.1). (1980) Fte Arctic and AInt- (iustal Ttondra tit Barrowt, A1 lsAa (. . Hio\\ i,(1-li.: jh('ir IDiimSon into (;e(lilotalj((,l AIrea.. P.C. Miller. [.1 . Ties/ett and 1. [. Biunntell,

Catmbridge: Cambridge 'nkisi rsit Press. 247 lp. Fds.). Srtroudsburg, iPa.: lDo\ deti. H-utchinisotiAnderson. J.R., I.E. Hard%, J.1. Roach and atid Ross, pp. 1-29.R.E. Witmer (1976) A land ulse and land coser Brown, J. and R. Berg (Eds.) (1980) Environment-classification sstem for Ilse w\ithl remlote senlsot al engineering and ecological baseline investiga-data. LISGS Professional Paper 969. 28 pp. tiotis along the Yukon River-Prudhoe Bay haulBatten, A.R. (1977) The vaseular floristics. major road. CRREL Report 80-19. AD 094497.vegetation units, and phytogeography of the Lake Cantlon, ,I.E. (1961) Plant cover in relation toPeters area, northeastern Alaska. Master's thesis, macro-, meso-, and micro-relief. Final Report.University of Alaska. Fairbanks, Alaska. 330 pp. Office of Naval Research, Grants No. ONR-208Belon, A.E., J.M. Miller and W.J. Stringer (1975) and 216, 128 pp.Environmental assessment and resource develop- Carson, C.E. and K.M. Husse (1962) The ori-ment in the Alaskan coastal zone based on Land- ented lakes of arctic Alaska. Journal of Geologv.sat imagery. Arctic Project Bulletin No. 7. Re- 70: 417-439.mote-Sensing Data for OCS Studies. NOAA/OCS Childers, J. M.. C'.E. Sloan, J. P. Nfeckel and J. W.Project Office. Geophysical Institute. University Nauman (1977) Hydrologic reconnaissance of theof Alaska. Fairbanks. eastern North Slope, Alaska, 1975. U.S. Geologi-Bergman, R.D)., R.I.. Howard. hiF. Abraha and cal Survey Open-File Report 77-492.M.w. Weller (1977)%Water birds and theit %selland Churchill. E.D. (1955) Phyrosociological and en-resources ti relation to oil deselopinti at Slot k- vironnienral characteristics of some plant corn-ersen Poinit. Alaska. U.S. D~ept. of' lmieior. I ish munities in th_ Uniiar region of Alaska. Ecoilogyr,and Wildlife Service. Reoutrce P'ublicatioti 129. 36: 606-627.38 pli. Clebsch, E.E.C. and R.E. Shanks (1968) SunmmerBillings, 1A.l. and K.M. Peterson (1980) V'egeta- climatic gradients and vegetation near Barrow,I otal change atid ice-wedge polygotns tlit otigl t lie Alaska. Arctic. 21: 161-171.

thas%-iake cycle in arctic Alaska.rtic and <I/- Conover, J.H. (1960) Macro- and micro-climiatei, Research. 12: 413-432. of the Arctic Slope of Alaska. Quartermaster Re-

Bliss. L,.C. and J.E. (anllon (1957) Successioni ill search and Engineering Center, Natick, Massa-Iikei allutium inl northertn Alaska. Amlerican Mid- chusetts, Technical Report EP-189, 65 pp.1(111( ,\~uturaMt, 58: 452-469.

34

Cowardin, L.M., V. Carter, F.C. Golet and E.T. 10-13 July 1978. Vol I. Ottawa: National Re-LaRoe (1979) Classification of wetlands and deep- search Council of Canada, pp. 359-365.water habitats of the United States. U.S. Fish and Everett, K.R. and J. Brown (1982) Some recentWildlife Service, OBS-79/31. trends in the physical and chemical characteriza-Drew. J.V. (1957) A pedologic study of Arctic tion and mapping of' tundra soils, arctic slope ofCoastal Plain soils near Point Barrow, Alaska. Alaska. Soil Science, 133(5): 264-280.Ph.D. thesis, Rutgers University, New Jersey. Fosberg, F.R. (1970) A classification of vegetationDrew. J.V. and J.C.F. Tedrow (1957) Pedology of for general purposes. In Guide to the check sheetan arctic brown profile near Point Barrow. Soil for IBP areas (G. Peterken, Ed.). IBP HandbookScience Society of Anerica Proceedings, 21: No. 4. Oxford: Blackwell Scientific Publishing.336-339. Gaydos, L. and W. Newland (1978) Inventory ofDrew, J.V. and J.C.F. Tedrow (1962) Arctic soil land use and land cover of the Puget Sound regionclassification and patterned ground. Arctic, 15: using Landsat digital data. U.S. Geological Sur-109-116. vey Journal of Research, 6: 807-814.Driscoll, R.S., D.L. Merkel, J.S. Hagihara and Gersper, P.L., V. Alexander, S.A. Barkley, R.J.D.L. Radloff (1981) A component land classifica- Barsdate and P.S. Flint (1980) The soils and theirtion for the United States: Status and plans. Un- nutrients. In An Arctic Ecosystem: The Coastalpublished draft, Rock), Mountain Forest and Tundra at Barrow, Alaska (J. Brown, P.C. Miller,Range Experiment Station, U.S. Forest Service, L.L. Tieszen and F.L. Bunnell, Eds.). Strouds-Fort Collins, Colorado, 32 pp. burg, Pa.: Dowden, Hutchinson and Ross, pp.Environmental Systems Research Institute (1980) 219-254.Susitna River basin, Alaska, automated geograph- Haugen, R.K. and J. Brown (1980) Coastal-inlandic information system, NASA/USDA/ADNR, fi- distribution of summer air temperature and pre-nal report: Willow subbasin, land capabil- cipitation in northern Alaska. Arctic and Alpineity/suitability analysis. ESRI, Redlands, Califor- Research, 12: 403-412.nia, 150 pp. Hettinger, L.R. and A.J. Janz (1974) VegetationEnvironmental Systems Research Institute (1982) and soils of northeastern Alaska. Alaskan ArcticAutomated geographic information systems, Gas Study Company, Biological Report Series No.North Slope Borough, Alaska, Colville Delta, 21, 207 pp.Milne Point. Unpublished draft report prepared Holowaychuk, N., J.H. Petro, H.R. Pinney, R.S.for the North Slope Borough Coastal Manage- Farnham and P.L. Gersper (1966) Soils of Ogo-ment Program, Barrow, Alaska. ESRI, Redlands, toruk Creek watershed. In Environment of theCalifornia, 174 pp. Cape Thompson Region, Alaska (N.J. Wilimov-Everett, K.R. (1980a) Landforms. In Geo- sky and J.N. Wolfe, Eds.). U.S. Atomic Energybotanical Atlas of the Prudhoe Bay Region, Commission, PNE-481, pp. 221-273.Alaska (D.A. Walker, K.R. Everett, P.J. Webber Hopkins, D.M. (1949) Thaw lakes and thaw sinksand J. Brown, Eds.). CRREL Report 80-14, pp. in the Imuruk Lake area, Seward Peninsula, Alas-14-19. ka. Journal of Geology, 57: 119-131.Everett, K.R. (1980b) Soils. In GeobotanicalAtlas Hopkins, D.M. and R.A. Sigafoos (1951) Frostof the Prudhoe Bay Region, Alaska (D.A. Walk- action and vegetation patterns on Seward Penin-er, K.R. Everett, P.J. Webber and J. Brown, sula, Alaska. U.S. Geological Survey Bulletin No.Eds.). CRREL Report 80-14, pp. 20-23. 974C, pp. 51-100.Everett, K.R. (1981) Soil-landscape relations at se- Jefferies, R.L. (1977)The vegetation of salt marsh-lected sites along environmental gradients in es at some coastal sites in arctic North America.northern Alaska. Final Report to U.S. Army Re- Journal of Ecology, 65: 661-672.search Office, Research Triangle Park, North Joint Federal-State Land Use Planning Commis-Carolina, Contract No. DAAG29-79-0160, 357 sion for Alaska (1973) Major Ecosystems ofpp. Alaska. Fold-out map.Everett, K.R. and R.J. Parkinson (1977) Soil and Johnson, A.W., L.A. Viereck, R.E. Johnson andlandform assoications, Prudhoe Bay area, Alaska. H. Melchior (1966) Vegetation and flora. In Envi-Arctic and Alpine Research, 9: 1-19. ronment of the Cape Thompson Region, AlaskaEverett, K.R., P.J. Webber, D.A. Walker, R.J. (N.J. Wilimovsky and J.N. Wolfe, Eds.). U.S.Parkinson and J. Brown (1978) A geoecological Atomic Energy Commission, PNE-481, pp. 277-mapping scheme for Alaskan coastal tundra. In 354.Proceedings of the Third International Confer- Komirkovi, V. and P.J. Webber (1980) Two lowen(e on Permafrost, Edmonton, Alberta, Canada, arctic vegetation maps near Atkasook, Alaska.

35

Arctic and Alpine Research, 12: 447-472. relation to topography, soils and frost action inKoranda, J.J. (1960) The plant ecology of the the Chukchi-Imuruk area, Seward Peninsula;Franklin Bluffs area, Alaska. Ph.D. Thesis, Uni- Alaska. Presented to the American Institute of Bi-

versity of Tennessee, 235 pp. ological Sciences Symposium, Corvallis. Oregon,

Lawson, D.E., J. Brown, K.R. Everett, A.W. 22 pp.Johnson, V. KonirkovA. B.M. Murray. D.F. Racine, C.H. and J.H. Anderson (1979) Flora andMurray and P.J. Webber (1978) Tundra disturb- vegetation of the Chukchi-lmuruk Lake area,ances and recovery following the 1949 exploratory Alaska. In Biological Survei' of Bering Land

drilling. Fish Creek, northern Alaska. CRREL Bridge National Monument (H. Melchior, Ed.).Report 78-28. 91 pp. ADA 065192. Report to the National Park Service, Contract No.Lyon, J.G. and T.L. George (1979) Vegetation CX-9000-3-031b, pp. 38-113.mapping in the Gates of the Arctic National Park. Rieger, S., D.B. Schoephorster and C.E. FurbushIn Proceedings, 45th Annual Meeting of the (1979) Exploratory soil survey of Alaska. U.S. De-American Society of Photogrommetry, Washing- partment of Agriculture, Soil Conservation Ser-ton, D.C., March 22, 1979, pp. 483-497. vice, 213 pp.Machilda, S. and J.L. Oldemeyer (1980) Caribou Robus, M.A. (1981) Muskox habitat and use pat-habitat use dynamics of the calving grounds of the terns in northeastern Alaska. M.S. thesis, Univer-William 0. Douglas National Wildlife Refuge. sity of Alaska, Fairbanks.Unpublished manuscript, U.S. Fish and Wildlife Searby, H.W. (1968) Climates of states: Alaska.Service, Fairbanks, Alaska, 24 pp. U.S. Department of Commerce, EnvironmentalMackay, J.R. (1963) The Mackenzie Delta area. Data Service. Environmental Sciences Adminis-N.W.T. Geographical Branch, Mines and Techni- tration.cal Surveys, Memoir 8, 202 pp. Sigafoos, R.S. (1952) Frost action as a primaryMeyers. C.R. (1981) Provisional checklist to the physical factor in tundra plant communities. Ecol-vascular flora of the Beaufort Lagoon coastal en- og., 33: 480-487.vironment, Arctic National Wildlife Refuge. Uni- Sohio Petroleum Co. (1981) Prudhoe Bay oilfieldversify of Alaska Museum, Fairbanks, 7 pp. geobotanical master map. Scale 1:6000, 23 mapMorrissey, L.A. and R.A. Ennis (1981) Vegeta- sheets (unpublished).tion mapping of the National Petroleum Reserve Soil Survey Staff (1975) Soil Ta.xonomv. U.S. De-in Alaska using Landsat digital data. U.S. Geolog- partment of Agriculture, Handbook 436, 754 pp.ical Survey Open File Report 81-315, Reston, Vir- Spetzman, L.A. (1959) Vegetation of the Arcticginia. Slope of Alaska: Exploration of Naval PetroleumMurray. D.F. (1980) Balsam poplar in arctic Alas- Reserve No. 4 and adjacent areas, Northern Alas-ka. Canadian Journal of Anthropology, 11(): ka, 1944-53. U.S. Geological Survey, Profession-29-32. al Paper 302-B, pp. 19-58.Nodler, F.A. and A.J. LaPerriere (1977) Landsat Swain, P.H. (1972) Pattern recognition-A basismapping-North Slope Arctic National Wildlife for remote sensing data analysis. Purdue Universi-Refuge. Mimeographed reprint, U.S. Fish and ty Laboratory for Application of Remote SensingWildlife Service , Fairbanks, Alaska, 16 pp. Information, Note 11572, Purdue University,Nodler, F.A.. A.J. LaPerriere and D.R. Klein West Lafayette, Indiana. 40 pp.(1978) Vegetation type mapping in northwestern Ta~lor, R.J. (1981) Shoreline vegetation of theAlaska in relation to caribou and reindeer range Arctic Alaska coast. Arctic, 34: 37-42.potentials. Alaska Cooperative Wildlife Research redrow. J.C.F. (1977) SoiA of /he Polar Land-Unit. University of Alaska, Special Report No. 2. scapoi. NeA Brunswick, N.J.: Rutgers Universit.33 pp. Press, 638 pp.

Oldemeer. J.L.. t).E. Biggens, (.J. Brecken- rieszen, I.L. (Ed.) (1978) Vegetalion and Produc-ridge and S. Machilda (1978) Caribou habitat lion li'colg.l. o Ian .llaskanA .Alrctic Tundra. Ne,studies on the Arctic National Wildlife Refuge. York: Springer-Verlag, 686 pp.Unpublished report. U.S. Fish ar d Wildlife Scr- Ilrich, A. and P. Gersper (1978) Plant nutrientvice, Fairbanks, Alaska. 28 pp. limitations of tundra plant growth. In IVegetationParkinson. R.J. (19-'4 Genesis and classification and Production Ecolog' of an .. lskat Arcticof Arctic Coastal Plain %oil%. Prudhoe Bay, Alas- lundra (Ties/en, L.LI_ Ed.). New York: Springer-ka. Institute of Polar Studies, Ohio State Univer- Verlag, pp. 457-481.sity, Columbus, Ohio, Report No. 68. 147 pp. t NESC() (1973) International classification andRacine, C.H. (1975) Tussock tundra vegetation in mapping of vegetation. Paris, 93 pp.

36

b.-,,.. . . . = . i , . . . . . .I I I I " - - _ .

t'niiersit) of Alaska, Arctic E:nvironmental Infor- 1.andsat-based land cover maps of Beechey Pointmation and D~ata C'enter (1975) Alaska, ReeLiontd quadrangle, Alaska (1:250,000 and 1:63,360I'role: .lrcuw Region (I A1.. Selkreg. Fd.). of- scales). Universitv of Colorado, Boulder.fice of Ithc (Io% ernor. 218 pp. Walker, D.A. and P.J. Webber (1980) VegetationI iereck. L.A. and EA..* Little. Jr. (1972) Alaska mapping. In Etitironitnenrtal Enginieerinig und &-o-11 Cs and shrubs. Forest Service, U .S. D~ept. of logical Baseline Invesligations Along tlhe Yukotn*\er culture, Agriculture Handbook No. 410, 265 River-Prudhoe Ba fluid Road (J. Brown andPP1. R.L. Berg, Eds.). CRREL Report 80-19, pp.Viereck, L..A. and C.T. I)%rness (1980) A prelimi- 45-48. ADA 094497.nary classificaiion systemi for vegetation of' Alas- Walker, D.A., K.R. Everett, P.J. Webber and J.ka. U.S. D~epartment oft Agriculture, Forest Set- Brown (1980) Geobotanical atlas of the PrudhoeNice, Pacific Northwecst Forest and Range Esperi- Bay region, Alaska. CRREL Report 80-14, 69 pp.nient Station. General 'Technical Report 1PNW\- Webber, P.J. and D).A. Walker (1975) Vegetation106, 38 lp. and landscape analysis at Prudhoe Bay, Alaska: AViereck, L.A., C.T. I)rness and A.R. Batten vegetation map of' the Tundra Biome study area.

1981) Revision of prelimninary classification S\'s- In Eco/t~gical In vesflgaflons ol the Tundra Bionietemn for \egetat ion of' Alaska. Unpublished mnanni- itn the Prudhoe Ba ' Region, Alaska (J . Brow~n,script. U.S. Forest Service. Instiue of' Northern Ed.). Biological Papers of the University ofForestr'.. Fairbanks. Alaska, 64 pp. Alaska, Special Report No. 2. pp. 81-91.'Wahrhaflig. C. (1965) Phvsiographic disisions of' Weller, (i.J. (1980) Caribou occurrence on Land-

Alaska. U.S. Geological Surs cx. Professional Pa- sat vegetation types on the coastal plain of theper 482, 52 pp. Arctic National Wildlife Refulge, May-June,Walker, D.A. (1977) The analysis of the effective- 1980. U.S. Fish and Wildlife Serv ice, Arctic Na-ness of a television scanning densitometer for indi- tional Wildlife Refuge, Fairbanks, Alaska, 36 pp.cating geobotanical features in an ice-wvedge poly- Wiggins. I.L. (1951) The distribution of vasculargon complex at Barrow, Alaska. Master's thesi's, plants onl polygonal ground near Point Barrows,University of Colorado, Boulder, 129 pp. Alaska. Contributions of/ the DudleY Herbarium,Walker, D.A. (1981) The vegetation and environ- Stanford Univer-sity' , 4: 41-56.mental gradients of the Prudhoe Bay region, Alas- Wiggins. I.L. and J.H. Thomas (1962). A Flora ofka. Ph.D. thesis, University of Colorado, Bould- the Alaskan Arctic Slope. Toronto: University of'er, 484 pp. Toronto Press, 425 pp.Walker, D.A. and W. Acevedo (in preparation)

37

#

-,, V---

- -

m -f- - -

a. r

:Z 7

0-A C- E -. * - - - -

CC7

-z Z

C) j

Z ~- * ~-

E 0.

-39

-77

EM

-'~~z z,- -0.

z -ii -c c :1' c-~; - - 7

CL-

C6

40

7, V- 7, 2:

- -, - - .41

42

E 72

z 71 -

t t -

4.1~

E* -Z

7s Z3

-3'1 --. 3 3

5. -3G-

El -Z -

Z33 tz- , -

CL E- -'c 3 E

- i3 .- z: 7

73 72 z

z. 3 z Z3 z

E 5

ig~~ z:z

, -~ ~ -44

m. '' V.

-C

M3-.~- -

C- u~.

-- 72 -

-- 3 ~ - 11, t;7r*3 --

55- - - 3 . '~- . -- -

C.

-3 73

-~r c-44., - 3

C- -- -73 -

3'- E - 3 -

3'C -3 . 3 :z -- 3 '- '- 4 '- - -

z' .4-- z3 tu-

-~ -

3'~~i c-- ~ 2 -

2 ~o Ij ol .r~ z-*-

-oI ~- -a zc.*30

45

-j

3 e3 7

33

L j

-3j .

E 3

go t

Z46

APPENDIX B: AREA SUMMARIES

Table BI. Acreage summaries of the vegetation of the CoastalPlain portion of the Arctic National Wildlife Refuge.The %ater category includes ocean within the study area boundar%.

Land Cover Unit Acres Percentage of Study

Area

I. Water 101,355 6.2

it. Pond/Sedge Tundra Complex; Aquatic Tundra; 16,964 1.0

or shallow water

ill. Wet Sedge Tundra 260,057 15.8

IV. Moist/Wet Sedge Tundra Comrplex; or 210,565 16.5

Dry Prostrate Shrub, Forb Tundra

(Dryas river terraces)

V. Moist Sedge, Prostrate Shrub Tundra; 434,512 26.5

or Moist Sedge/Barren Tundra Complex

(frost-scar tundra)

Vt. Moist Sedge Tussock, Dwarf Shrub Tundra 414,550 25.3

ViI. Moist Dwarf Shrub, Sedge Tussock Tundra; 51,148 3.1or Moist Sedge Tussock, Dwarf Shrub/Wet

Dwarf Shrub Complex (water track complex)

Viii. Shrub Tundra 3,142 0.2

IX. Partially vegetated area 27,678 1.7

X. Barren gravel or rock 27,642 1.7

Xi. Wet gravel or mud 28,402 1.7

Xii,. ice 4612 0.31,640,626 100.0

47

C4 m N41 : ;

' ~ m Z .4fa ar -I Z -01CO -100 -;4 c

m 4 c;

n'm. r 1 . 0 0: .4m t--4 .4 .'44CN . 4C -

C ~~~~A 4N *4a

"4 a N%C 0 C -m.C 4C C 0 4 C ON C C C

m4 4(1 C; 1 0 0

a a;

"4 C 0

04 a a 00a0 C; a a;1 ;V

a a

o4 a) a 0

ua 48

k.0 Z- F-L

4-0-0> >N N o o- 0 0 N 4 0

*0~~~ N 00 ONO 0.00 ON 0N-4. N 1N N 440 '0-~~~~~4 c; 44 4; AN. . . . r 4. . .4 oe. H. . . I ,

N;N aQ N . . c-N 0' . .. . .N.

m NC' 0 0

o.ur- o 0 oO ~-004l It 0 0 000r 0 ON or ..-40 N cc 0

4-.N0 C;.. ON 04- 0 0 0 ON OO 0w m 0 00 0 00 0O N 00

NN0 NO N0 .0: N: 1-r N -

000040..4 N N 00 CONZ N rC 0' O N'0 0' 0'.N 00 00 r a~N Or-.

.. 4NN c Nc 0.N N, N0 .4 9 1 N:1

N0 In a

.4 . . . 0.c . . . . . . . 0. 04 >4. . . 4 r

.. 0- N ft w 'N 0 FA

0 0 N

Na

wC. 0 r 4 NO 00 "0 4 0 N 00Z

00 N4 0:r 6 00 QNO w 00 -4.NN N 0 L

co - N-wN N a' CM

.. 40N.. V 0'0 - a all m ON .4 M-0 0 . (0

N 0 u0 4

,

-4 00* 0 0 ;C 4A . ;I

.. 4ON! 4 0!..00 04 NOON ON 0 00 : '

- 0 Q 4 ..4 N; 14 N ; N; -; L. 4

NWN0 .40N aON N 0 -. Z ct ff 3a

NW .N .4N0 N Na...049

0

c 0 - -0% - -~I

* I-

I It

* --

-o

--a40 --- 44 0 -- 4 -0 00CS C

;z 0 0

~~~~ C v Cn0 0j 10.~ t 0 N 4- . C - N

- I - - 1

6 00

0.-.

0 .0 0

U~c .- .. O

ooS 90u04- CU z; D U C ~ ~9* e CL eC

110.0.0L;

*~e 4- 0 &O a 4

Is CC CcC. C

C~C o~ U

(L ( . aso' ~ O * ~ - . ~ 4

aM a.

0to c,

7o k0 T. *C ~-

CC ~~00~ 0.- - ;~ -. 0 0W C Z

A 4- o-o

4- 3 ,3 o~ a. . 4-2 o4 4o S Cc v

0 m

0 C

30 c34* m 4

-& le,4 o

CL C06 0

4-0000 (351

f f

0 A 4 U4 4

0 0~-Ca -It

3co 2 OCC C.

0a ~~~ Imsa C

S ~ ~ 1 11. 2 *.C mv~ U C

1.5 CU 1.1 MO

C1 C aea

CV 2

le c

u 01 a 52

* 0 0 0 0C C C C Co 0 0 0 07 7 7 7 7

0 C'

C. - 0

7 0 U 0~5-o - 0 0 0C ~J C C

~

7 - 5.7a N N5. 7 5)N ~ C~

* J77

777 7 7a 7 * 7 -

* V 7 V0. C C * C7 0 Oa Ca 07 - -~ -5- CN > 07 ON 0I = DI DI 7

0 0 0 0 0C C C C Co 0 0 0 07 Z 7~ 7 7

* 0

0 0 7

0 0 07 7 C* 0 0

- - 7 7 50 0 0 0 C- - U U 7- 0- U t.LV U 0 -00 *X ~ 7 0

>7 >7 - -

o V V UO C 0 00 0 0a a a 5O C0 0 2U U U -0 0 0 0

7 7 7 7 -0 C

V V C2 02 - - - 0

7 I I I C.N N N N 0 0

- C.>0 00 U-- >0 00770 -@7 0 0C U C V C

>0 U S

0 0 0 U

*~7 -7 5 7

tE t ~!~ !~ -

o 0 * -'C 'C * - -- - CC CC CC

53

APPENDIX D: SOIL TAXONOMY

The word "soil," as used in this report, refers the soils are termed Histic Pergelic Cryaquepts.to the thawed layer of ground observed in August. Wet Inceptisols lacking such an epipedon are sir-The nomenclature is that of the U.S. Department ply termed Pergelic Cryaquepts. Most Cryaqueptsof Agriculture. Soil Conservation Service Manual show some quantity of organic matter mixed in439, 1981 Draft and Handbook 436 (Soil Survey the subsurface horizons, presumably concentrat-Staff 1975) unless otherwise noted in the text. ing at or near the seasonal permafrost table. Man,

Of the ten soil orders recognized in Soil Taxon- Cryaquepts developed in silt, silt loam, or fineownv (Soil Survey Staff 1975), four are found in sandy loam materials display thixotropic charac-ANWR (App. A). Soils belonging to the order teristics and will flow spontaneously upon vibra-Mollisols are extensive. They occur in all terrain tion. Cryaquepts are common components of soiltypes, particularly in the Foothills and Hilly complexes and are often associated with Mollisols,Coastal Plain. Mollisols are dark-colored mineral particularly where they occur in frost scars. Insoils. The dark color is a reflection of at least such cases the complex is termed Ruptic-(inter-2.50'o organic carbon in the upper 18 cm of the rupted) Cryaqueptic Cryaquoll. The term Enticprofile. In addition to the dark color, Mollisols (Enlisol) may replace Cryaqueptic if frost-scahave a base saturation greater than 50076, i.e. Mol- soils show little or no horizon development.lisols are neutral or slightly alkaline in reaction Relatively well drained and stable sites, especial-(pH). Wet Mollisols are classified as Aquolls, aqu ly on ridge crests, have Pergelic Cryumbrepts,indicating an aquic (reducing) moisture regime which are Inceptisols with an umbric (dark-col-and oll indicating the soil order Mollisols. The ored, base-poor) surface horizon underlain by anprefix Cry appears at the Great Group level of acid (base-poor) subhorizon. Similar soils lackingclassification for all soils found in ANWR and in- the umbric surface horizon are usually designateddicates a mean annual temperature less than 8C. Cryochrepts.The term Pergelic (permanently frozen) appears at Entisols are soils with little or no horizon desel-the subgroup level of all ANWR soils. Thus, Per- opment. They are common on unstable sites suchgelic Cryaquoll defines a cold, wet, dark-colored, as sand dunes, flood plains and talus areas. Cry-base- and organic-rich mineral soil underlain by opsamments are sandy Entisols that are subject topermanently frozen material. Pergelic Cryaquolls blowing and drifting. Cryorthents are coarser andoccurring in very wet areas may have a surface are composed of gravels, rocks and materialshorizon with greater than 20 but less than 40 cm of found in recent mudflows, glacial deposits andorganic material. These soils are designated as stabilized river alluvium.Histic Pergelic Cryaquolls (the term Histic indi- Perhaps the least extensive and least predictablecates an organic surface horizon greater than 20 in occurrence are the organic soils (Histosols).cm thick). Mollisols on well-drained sites are These soils have a surface horizon composed oftermed Pergelic Cryoborolls. greater than 40 cm of organic materials and gen-

Inceptisols are mineral soils that have only erally greater than 600o organic matter overlyingweakly differentiated soil horizons. This is due gray, sometimes mottled, fine-textured mineralprimarily to the ineffectiveness of the leaching materials. Normally these soils are very wet, to theprocess in the cold, wet tundra. Organic carbon in extent that the organic materials are buoyant orthe Inceptisols is not evenly distributed in a dis- partially so. They occur on flat areas, either ontinct mollic epipedon (dark-colored, base-rich sur- ridge crests or in lowlands. Three taxa of coldface horizon) as in the Mollisols. Aquepts (ept in- Histosols are recognized: Pergelic Cryofibristsdicates the soil order) are wet Inceptisols that com- (fibrous, low-density organic matter), Pergelicmonly show some degree of mottling (iron oxida- Cryosaprists (nonfibrous, highly decomposed andtion) in the mineral soil below the organic surface dense organic matter) and Cryohemists (intermne-horizon. Mineral soils generally are gray, reflect- diate organic matter characteristics). Histosolsing a saturated, reducing environment. In some may have any pH, although in ANWR most arecases an organic surface horizon greater than 20 nearly neutral in reaction.cm thick forms the epipedon (surface horizon) and

55

hIECLD1NJ PAGK BLAK-NOT F] isZ

. .. . .. i| | -- -- - .. .... -- .. . . - - _-.... ..

I E

0 4

u00 0C -

2 . .2 .21 0.-

Cl *I- 4 100 1 o

1j0L T0C t

Cl-.0 14 10e 1-000 C oi I

1 ' L

10 A

C 0

4 4

4;~~~ 0 l - 4

-~~~o l O CC O 4C

94444 a 4aC 4C C a> > a7

014' uj' PAG CL a - O CCC 04 C. CL CC

Cl I5a

- SI

-. Ao~i -=

2 uI' 2 I.

'I~ ~~~~~~ - i4 I 1S0- ~ .

1

4; ~1 0 0 0 Z~

oo a e- o

'o m leo10 lo. C e a>o IaI ) cc ~~o~ C CU

C~~ o S S S C CI.55 U C S U C S C f

CC C U --S C C 2

a I CI

16 -(0 A 41 0 54- C C

oC

*~c C - C

* ~ ~~~ t C

*o u Cm C C o

c 0 C-U; .mU 0- a ~

C O. U C.4

Z *UC

U U LU U Uc U

00 00 00 U 000

000 U 0 U U C C 000f

C8 - CC CCCC 0

0100, 1 ,5-= 058 0

LUeCEL PAG -LN-O MLALU454DU

.5 5Ac .5I~A~ IIA~

;'g2:zz~~5 -l.a

zI~ o~aUWA-UL~1~: ~- sIII~:I3iA- 2j2* I ~~i~i~is* 1 ZEI *I~ A..~.

-IA~ 4~ 4 SS c-ne -

A- 0 0 2

S IA

5- 4. C- C C

* IA 5.I > a

5 0 0- A- 5-

I. I~.. IA U U.o CC IJ@ 0 IaS -C-

CI SC S SC 01001 01- I. S.C U.

0.00. a.

I. S* A-S IAo coo on

C .IA IA* n@- 011 0

C C C3 CL IA CC C

S.C IAIA -- S (I

3 LV U.00 0

Ifl C C C S

2 In 2

I~II a55 0

* - 511

* C

o VC C 0c s 4-V - C

AC 4-OS C 0014- 0 Cl*

~II1 & C

4-4- 0 4- 1.IA C In C I.-c S C ~~ 0. C

00

0

59

-I. -

A facsimile catalog card in Library of Congress MARCformat is reproduced below.

Walker, D.A.Landsat-assisted environmental mapping in the Arctic

National Wildlife Refuge, Alaska / by D.A. Walker,W. Acevedo, K.R. Everett, L. Caydos, J. Brown andP.J. Webber. Hanover, N.H.: U.S. Cold Regions Re-search and Engineering Laboratorv; Springfield, Va.:available from National Technical Information Ser-vice, 1982.

vii, 68 p., illus.; 28 cm. ( CRREL Report 82-37.Prepared for U.S. Geological Survey and U.S. Fish

and 'Wildlife Service / by U.S. Army Cold RegionsResearch and Engineering Laboratory.

Bibliography: p. 34.

1. Arctic National Wildlife Refuge. 2. Geobotan-ical mapping. 3. Landsat. 4. Mapping. 5. Soils.6. Vegetation classification,

(see Card 2)

Walker, D.A.

Landsat-assisted environmental mapping ....1982 (Card 2)

I. Acevedo, U. II. Everett, K.R. III. Gaydos, L.IV. Brown, J. V. Webber, P.J. VI. United States.Armv. Corps of Engineers. VII.- Cold Regions Re-search and Engineering Laboratorv, Hanover, N.H.VIII. Series: CRREL Report 82-37.

- I

UNITED STATES GEOLOGICAL SURVEY

53 km 1 146 00 550 01 E

7780

, Seaward limnit for area rneasurer'e'l .)j wuter

----------------------------------- ----------------

7770-

Roi'da,i ot studt, areu *'

7000

R7

N

7760 -T7 T s~ 7N

R27R28

T EON T6N;, 4S

T6~ ~ .r6N F~~g%. T6N T6NR2( ~ R2 \ .R26C

W27E R28E

* ./ ..-.- 4

T5f 23 TN5N.. ~T5 T%.R25E3 ET5 N11

R24E~ T 2E

7740 7,E P7,E

Land cover class, dominant vegetation, and map surface area in acres. hectares, and percent

The total surface area shown is 1,640.627 acres (2.563 sq. mi.. or 663,952 hectares (6.640 sq. km.).

Water. -Ocean surfaces. and lake and river surfaces larger IMoist Dwarf Shrub. Sedge Tussock Tundira; or Moist \(

7730 -than 1 acre. (Area measurement excludes water surface Sedge Tussock, Dwarf Shrub Wet Dwarf Shrub Com--

outside seaward limit of study area.)I plex (water track complex). -- Upland tundro kk:ih

S101.355 a (41.018 ha) (62 pct) shrubs to 50 cm high. or upland tuss~ock tondta kkitli

Pond'Sedge Tundra Complex; Aquatic Tundra; or shallow I husi ae rcs 5,4 2.10h)( i

water. - Very wet tundra areas with ponds and, or emer - Shrub-Tundra. -South-facing slope, in foiothills Tkt

gent communiies of Carex spp or Arctophila. and up 1. alpine, with willow. birch, alder to 2m. or dense OTh t

,50% moist or wet tundra 16.964 a (6,865 ha) Q10 pct). Water tracks; 3,.142a 1 2721 1i 1

Wet Sedge Tundra. -Wet tundra with little standing water Prilyvgttdaes-Dvrehbtt su

or with up to half of surface area water-coveredor emer- arsalpinveeatundras n mosse ats sihtin ludr, ti

with salt water L 260.057 a 1105.244 ha) (158 pct. taud. lihn-cvrd sre tntosid I,

7720 ~Moist, Wet Sedge Tundra Complex; or Dry Prostratemuflt276,ia11211it ,i

770Shrub, Forb Tundra (Dryas river Terraces). -Moist IV" Barren gravel or rock. - Bate Isit lbi Ises us, 1

sedge tundra with up to 40'r wet sedge tundra. Or dense - gravel and sand sp ~. alpinc harivi)- ialpy ,sll

prostrate mat of Dryas on river terraces mitl, and cultural barrens sroard ,T itisrra i .- ,

270,565 a 1109.496 hal (16b5pct). rich but sparse floras 27.642,1 l1~t 10' -

- ~~~~Moist Sedge. Prostrate Shrub Tundra: or Moist Sedge/ V Wtgae rmd Etniebre ntiirsstlaB~arren Tundra Complex (firost-scar tundra). - BetterWegrvlomu.-Etniebin uIint i

drained areas on rolling terrain sometimes with tussocks, and wet or dark-colored gravel on beaches or nivet beds.

0r sparsely vegftated frost-scar tundra ......... or dark

PREPARED IN COOPERATION WITH THE UNITED STATES FISH AND WILDLIFE

/.' UNITED STATES ARMY COLD REGIONS RESEARCH AND ENGINEERINC

145 00 600 km E UTM Grid Zorie 6 144 00 UTM Grid

B E A U FORT SE ASIN

Ary 321

remnt of water

as

A

CAM1DEN' B A Y

-- -- - - - - - - - - -

j Korigarievik Pta

T814 18N Ande,501a P1

. :R28ER3E4 'U

Cmir sn P1

e. j7N T-~

7.5 *VR28E -i .(I TNTRTN- T7N R31E5 3 -p

r R'9ER30E T~ T

9 16

R28 R29E R3ER3FR3' R33E

Pi.T6N T5PIN N T6N

R28E R9 ,RER3E'" '? 32E R&3

R30 31E %L/A&

Tuda or Mos '-a U N A

Dwar Shu Com R3) i, R31ER323R3

pln tundr with

,64 h. I 12-'--0ah 576

Vill C441 1, ',h -i'l--I

0iqn m) ai a qli ,ri Touhle W4 Tp4 . T4

owrf Sehru , Con -RI 2E *R3

(16272 hat 32 pct) ~h 06

it foi th ir iveh

h harr -kM and Index tor, mu b,,. t,'i

(11211 ,, Pt'Landsat 133-210531

orvd 1151SftravelI X scenes 13 July 1979

way). often neith ARCTIC NATIONAL-W WILDLIF

FISH AND WILDLIFE SERVICE ANDTH [HEE

H AND ENGINEERING LABORATORY

Grid Zonle 6 14 e UTM Grid Zone 7 400 ki E 143 00

9 0 '7P VEGETATION,A s1 Ze ~ 9N Barter 4

lll Island kk R4 ~ R6 O S~

R34E

T9N R36R33mdIP.> *rCOASTA

'44

N8

A, 'A.,

T7N T7 W4 E RRR33E *i34E 4. R36E3E R

R r.,, Rrile

N T5N N3 0 R3 T)N. 5

T3N TOTi

thet

450 kil I

7790NVEGETATION AND LAND COVER k

ARCTIC NATIONAL WILDLIFE REFUGE6E ~ COASTAL PLAIN, ALASKA

4 7770

776

7740

E,6 458

Introuctio

m ;Ldcdi otucl hd TVO iIkOd npc

m~ ~~~~~~~~~~~~~ ssmli wteUSFhdiiWlF~ 'wi UF~ono i xlriP' Ih otcNtioi Au ilf ,fii 73

p 0 ldsificitio o i~ei'ttiiii diiil~iil Lver ~ 7750i

at ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~n UNSuigoe idwpeneaiecnetoa ehn

calscren, ad ho mie poc ki clor.R

-- Shrub runra. -South-facdng slopes in foothills or sub-gent communities of Carex spp. or Arctophila: and up to alpine, with willow, birch, alder to 2m: or dense shrubs in50% moist or wet tundra.. 16.964 a (6,865 ha) (1 0 pct). water tracks ... 3.142 a (1.272 ha) (0 2 pct).

Wet Sedge Tundra. - Wet tundra with little standing water I1I Partially vegetated areas. - Diverse habitats including riveror with up to half of surface area water-covered or emer-gent vegetation, or coastal areas periodically covered bars, alpine tundra and moss mats w ath barren rock and

with salt water ..... 260,057 a (105,244 ha) (15.8 pt). talus. lichen-covered, sorted stone nets and beach or

Moist/Wet Sedge Tundra Complex, or Dry Prostrate mud flats .... 27.678 a ( 11201 hal f1 7 pct)

1720 Shrub. Forb Tundra (Dryoas river Terraces). -Moist IV Barren gravel or rock. -Bare light-colored river gravel. Xsedge tundra with up to 40% wet sedge tundra, or dense gravel and sand spits, alpine barrens (especially dolo-prostrate mat of Dryas on river terraces ........ . I mite), and cultural barrens (road or runway). often with

................. 270.565 a (109.496 ha) (16 5 pctl, rich but sparse floras 27,642 a (11.186 ha) (1 7 pct)Moist Sedge, Prostrate Shrub Tundra; or Moist Sedge/

Barren Tundra Complex (frost-scar tundra). - Better. V Wet gravel or mud.- Extensive barren mud in river deltasdrained areas on rolling terrain sometimes with tussocks: and wet or dark-colored gravel on beaches or river beds.or sparsely vegetated frost-scar tundra or dark-colored barren rock in mourtains.. .

............... 434,512 a (175,845 ha) (26.5 pct) .... ... 28.402 a (11.494 ha) 1 7 pct)Moist Sedge Tussock, Dwarf Shrub Tundra. - Well- VI X11

drained upland tussock tundra in foothills with high Ice. -River icin in the braided stream channels of most7710 percentage of cottongrass tussocks and dwarf or prostrate . larger rivers 4.612 a 1. 866 ha) (0 3 pct)

km N shrubs . .. 414,550 a (167.766 ha) (25.3 pt). l rs....... 46

69 30k

530 km E 1460 550 km E

/I

(es yiaI doab- _ _ _l

runway), often with ARCTIC NATIONAL WILDLIFE REFUGEa 118 a (. pt. X1 COASTAL PLAIN, ALASKA

put mud in river deltas/Sbeaches or river beds, 22008- 20420

Mtains ............. .. 2570-20462 22 July 1980A (11.494 ha) (1.7 pct). 14 Augugt 1976 (from Canadian Ground Station) r

xii Mic

mchannels of most A

a (1,866 ha) (0.3 pct). etan

145 00' 600 kmn E UTM Grid Zone 6 144 0t T Gi

and ALeona

-TJN T3N crelFR*37 E as"r;

ment AV.( the d%

- *Cal SC

Base is adapted from USGS 1 250.000-scale topo- meterand 10,000- meter intervals. Tickecrosses index DiStigraphic maps of Flaxman Island, Barter Island, Mt geographic grid Land lines and townships. num-Michelson, and Demarcation Point quadrangles. bered from Umiat Meridian and Base Line. represent 5 0Alaska. Universal Transverse Mercator IUTM) Pro- unsurveyed and unmarked locations predeterminedjection. 1927 North American datum, with UTM rec- by Bureau of Land Management. The 1955 mag- 5 0tangular grid shown for Zones 6 and 7 at 50.000- netic declination varies from 33 to 36- East ________

iZone 6 14400 UTM Grid Zone 1 400 km E 143 00 West of Greenwich

and Aloine Research). Design of map and statistical products is byLeonard Gaydlos and James R. Wray (USGS). Color separation andscreening for four-color process printing are done on a large-format-laser plotter using dot screens and angles developed on that equip-ment and plotted directly onto the separations as data are read from -7720the digital file, The resulting colors replicate those otherwise achievedat USGS using open -wi ndow plate nega fives :-onventional mechani-cal screens, and the same process ink colors.

Distance scales 1:250.000

5 0 5 10 Miles

5 0 5 10 15 Kilometers____________________________________________________7710

69 30'r West of Greenwch 450 kmi E

For sale bV Branch of Disribution U S Geciogica. c~r~ey

Box 25286i Feoe al Cerrie' Denver CO 81 -'

4!

AREA MEASUREMENTS, by Survey Township and Land Cover Class, in Percent and Acres-*rRangeE RangeE RangeE RangeE RangeE RangeE RangeE RangeE RangeE RangeE RangeE RangeE RangeE RangeE RangeE RangeE RangeE RangeE Township

and andCls 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Class

21.9 33.4 89.4 88.7 85.2 49.1 62.3 97.6 ,I1 4.2 8.4 0.7 0.2 0.5 4.7 5.1 -III 64.3 42.0 7.0 - 6.2 17.0 16.4 - I11IV 0.2 7.1 0.1 - 3.5 1.3 9.6 - IV

T v 0.4 0.3 0.4 1.3 - v TVI - . - VI

N Vil _ - _ - Vill NIx 0.3 0.4 1.0 1.1 4.9 2.5 - Ix

x 2.9 3.0 0.6 5.3 2.0 3.7 1.1 1.4 xXI 6.2 5.0 2.2 4.8 1.2 18.9 1.7 1.0 XIxIl - 0.3 - - - 0.0 0.0 - XII

Percent 100 100 100 100 100 100 100 100 PercentAcres 10,954 16,148 1,755 2,655 13,718 13,876 10,795 2,146 Acres

r 3.2 3.9 27.8 55.8 71.2 46.2 33.6 29.3 16.8 9.1 1.6 6.1 18.5 8.4 III 1.9 2.3 8.1 4.1 5.6 2.5 1.4 1.7 3.3 3.3 1.3 5.6 2.0 1.0 11I1 70.1 64.1 43.8 21.7 13.4 8.5 9. ; 28.7 36.7 24.3 45.0 31.7 38.0 23.7 IlIIV 2.4 2.6 6.4 5.4 2.6 27.5 33.5 13.9 25.6 39.0 33.4 23.8 24.8 22.3

T v 0.0 0.6 2.2 (.8 3.4 11.9 17.4 3.1 7.7 23.3 11.1 28.5 13.6 36.2 v TVI - 0.0 0.2 0.6 0.4 0.4 1.9 - 0.1 0.3 0.0 - - - VI8 VII - - 0.0 0.0 - - - - - VIl 8N ViII - - 0.0 0.0 0.0 - - - ViI N

Ix 5.0 5.4 1.2 0.1 0.0 3.0 1.5 7.2 3.7 0.5 2.8 2.1 0.9 3.8 IxX 2.4 2.3 2.0 0.2 0.3 - 0.1 1.9 1.4 0.1 3.0 1.6 1.1 3.1 XXI 14.7 18.8 8.1 2.7 3.0 - 1.1 14.2 4.7 0.1 1.8 0.6 1.1 1.0 X1XlI 0.3 0.0 0.2 3.6 0.1 - - - - 0.0 0.0 0.0 0.0 0.5 Xll

Percent 100 100 100 100 100 100 100 100 100 100 100 100 100 100 PercentAcres 15,724 22,910 22,773 15.491 3,838 235 10,593 20,280 22,152 22,469 22,314 22,602 18,013 5,286 Acres

I 3.1 1.6 0.6 0.1 1.5 4.4 18.1 5.3 0.5 1.3 1.2 2.4 4.2 1.1 2.4 8.0 44.8 III 0.4 0.8 0.6 0.3 0.8 0.1 0.5 0.1 0.5 0.3 1.3 1.6 3.2 0.8 2.1 0.8 1.4 IIIl1 42.7 25.8 13.6 26.8 8.1 29.7 10.0 0.6 2.9 33.7 20.1 23.6 30.9 7.6 26.1 12.8 21.5 I1IV 16.4 9.8 12.8 18.4 9.2 19.1 11.1 8.4 17.1 39.1 45.2 43.9 35.0 22.8 17.9 16.0 18.0 IV

T V 3.1 19.2 36.2 28.3 48.3 29.6 38.9 52.4 47.3 16.4 27.3 27.3 17.8 65.9 47.1 59.0 7.9 Vr TVI 0.6 11.6 23.0 15.3 27.5 9.9 14.2 30.8 28.0 1.4 0.3 1.1 0.0 0.3 0.2 - - VI7 VII 0.0 2.6 8.2 4.8 3.8 1.3 2.2 0.1 0.0 - - - VII 7N Vil - 0.6 1.3 0.S 0.8 0.1 0.3 - - - - - VlI N

IX 6.1 5.9 0.4 0.2 - 0.3 0.3 0.9 2.5 4.0 2.0 0.0 2.2 0.8 3.1 1.8 3.1 IxX 6.1 2.4 2.1 5.0 0.0 5.2 2.1 1.2 0.3 1.4 1.8 0.1 4.8 0.6 1.0 1.2 1.5 XXI 11.7 11.3 1.2 0.0 - 0.3 1.7 0.2 0.9 2.4 0.8 - 1.3 0.0 0.1 0.2 1.7 X)XII 9.8 8.4 0.0 0.3 - 0.0 0.6 0.0 - - - 0.0 0.6 0.1 0.0 0.2 0.1 XIl

Percent 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 PercentAcres 7,915 23,126 22,873 22,840 22,828 21,071 19,867 17,217 22,118 22,406 21,951 22,202 22,055 22,137 22,167 21,657 7,607 Acres

I 4.0 0.1 - - - 0.1 - 0.3 0.5 0.7 0.0 3.0 0.6 1.7 1.8 0.1 9.7 56.9 III - - 0.0 - - 0.0 - 0.0 0.8 0.1 0.1 1.7 0.8 1.3 1.1 0.1 1.0 4.8 I1Ifl 11.8 0.4 0.2 1.5 2.2 2.2 1.7 0.3 2.4 16.6 3.1 29.8 18.6 22.1 25.5 27.6 49.7 31.9 IfIIV 5.9 0.1 1.0 4.3 4.6 12.3 6.5 2.1 9.2 43.4 18.2 33.5 34.9 24.2 21.7 32.1 24.9 3.0 IV

T V 4S.4 25.4 30.7 36.8 32.6 36.1 32.6 21.5 35.9 26.6 65.8 29.6 37.4 48.8 46.4 35.6 12.4 0.1 V TVI 26.2 67.9 52.7 47.0 45.6 34.3 51.9 74.7 46.3 5.3 10.6 0.9 0.1 0.0 - - 0.0 - VI T6 VIl 0.5 3.5 12.9 7.S 11.8 10.9 S .2 0.1 0.0 . .. - VIl 6.

N Viii 0.2 2.6 1.7 0.4 0.4 0.2 0.2 - - - - - - - - - - - Vil NIX - - - - 0.0 0.1 C.0 0.2 3.5 3.5 1.0 0.4 2.2 0.7 1.5 2.4 1.8 1.1 IXX 0.3 0.0 0.8 2.5 2.8 3.8 1.9 0.8 0.8 1.7 1.1 1.0 4.6 1.1 1.9 1.8 0.3 0.4 XXI 5.7 - 0.0 - 0.0 0.0 - - 0.6 2.1 0.1 0.1 0.7 0.0 0.0 - 0.2 1.8 XIXII - - - 0.0 0.0 - 0.0 0.0 0.0 0.0 0.1 0.1 0.1 0.3 - - XIl

Percent 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 PercentAcres 15,856 22,395 23,166 23,164123,017 23,011 23,052 22,487 22,309 22.519 22,137 22,342 22,287 22,378 22,360 22,256 22,139 7,411 Acres

I 2.3 0.7 1.2 0.1 - - - - 0.1 0.3 0.4 1.7 1.0 0.9 0.1 - - 0.7 I

II - - - 0.1 - - - - 0.4 0.1 0.2 0.8 0.6 0.4 0.1 0.0 0.0 1.6 III1 4.7 0.6 3.3 1.7 4.9 0.3 0.1 0.4 4.8 8.1 6.3 14.5 14.2 8.1 10.9 36.2 58.9 63.1 IIIIV 2.2 0.1 0.2 7.0 10.4 4.9 2.5 3.4 20.5 38.6 19.6 30.2 30.2 21.6 36.9 44.4 35.8 28.6 IV

T v 92.3 32.4 42.9 29.5 23.5 18.8 11.5 6.1 30.0 25.0 38.8 37.0 43.4 61.0 50.1 18.0 3.7 0.2 v TVi 20.8 63.3 51.0 51.7 32.7 35.2 60.8 84.9 35.2 19.3 31.4 13.0 2.2 2.5 1.3 0.2 - - VI 5VII 1.8 2.0 0.8 6.9 22.4 36.6 21.4 4.1 0.1 0.0 0.0 - - - - - - - VII

N Vill 0.5 0.9 0.5 0.1 0.7 3.7 0.4 - - - - - - - - - - - ViII NIX - - - 0.0 0.0 - 1.0 0.8 6.3 2.5 2.1 1.2 3.7 1.9 0.5 0.8 1.5 4.2 IXX 0.4 - 0.1 2.9 5.4 0.5 2.3 0.3 1.9 3.7 0.8 1.0 4.0 2.9 0.1 0.4 - - XXI 15.0 - 0.0 - - - - - 0.5 1.2 0.4 0.6 0.7 0.1 - 0.0 0.1 1.6 XIXII - - 0.0 - 0.0 - 0.0 - 0.2 1.2 - 0.0 0.0 0.6 - 0.0 - - XII

Percent 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 PercentAcres 13,089 22,658 22,463 23,185 23,220 23,000 22,537 22,464 22,220 22,409 22,149 2, 02 22,151 22,379 22,343 21,322 12,197 1,894 Acres

2.4 0.0 - 0.1 0.0 0.0 - - - III 2.3 0.1 - 0.2 0.0 0.0 0.0 0.0 - II

Ill 3.5 0.6 0.6 5.7 8.4 2.8 0.5 13.7 28.2 IIIIV 9.4 2.9 3.3 15.8 22.3 11.5 3.0 25.7 53.3 IV

19.9 12.6 8.7 28.1 20.5 22.8 28.6 30.2 11.8 v TV!48.8 76.3 83.2 45.3 43.4 59.9 66.7 29.1 - VI

4 VII 2.0 4 3.9 0.4 0.5 0.4 1.0 0.2 - VII 4

N V VEGETATION AND LAND COVER 0 1 0.0 0.0 - - - ViII NIx 5.3 1.5 0.1 3.4 2.5 1.6 0.0 0.8 6.1 IXX ARCTIC NATIONAL WILDLIFE REFUGE 4.6 0.8 0.2 0.9 2.1 0.9 0.2 0.3 0.6 xXI 0.3 0.2 - 0.1 0.3 0.1 - 0.0 - XIxl' COASTAL PLAIN. ALASKA 1.4 - - - 0.0 - - - - XIl

Percent U S Geooqical Survey Map 1-1443 1982 100 100 100 100 100 100 100 100 100 PercentAcres 22,163 22,39 21,763 22,177 121,986 22,189 21,976 15,274 359 Acres3mm m. '. . - I - _1w

3.5 0.6 0.6 5.7 8.4 .8 0.5 3.7 28.211IF 9.4 2.9 3.3 15.8 22.3 11.5 3.0 25.7 53.3 IV

19.9 12.6 8.7 28.1 20.5 22.8 28.6 30.2 11.8Vl 48.8 76.3 83.2 45.3 43.4 59.9 66.7 29.1 - VI

4 vII 2.0 4.9 3.. 0.4 0.5 0.4 1.0 0.2 - vii 4N vIIl VEGETATION AND LAND COVER 0.1 0.1 0.0 - - 0.0 - - - Vi Nix 5.3 1.5 0.1 3.4 2.5 1.6 0.0 0.8 6.1 IxX ARCTIC NATIONAL WILDLIFE REFUGE 4.6 0.8 0.2 0.9 2.1 0.9 0.2 0.3 0.6 x

XI 0.3 0.2 - 0.1 0.3 0.1 - 0.0 - x1xIl COASTAL PLAIN. ALASKA 1.4 - - - 0.0 - - - - x1I

Percent U S Geological Survey Map 1-1443, 1982 100 100 100 100 100 100 100 100 100 PercentAcres 22,163 22,389 21.783 22,177 21,986 22,189 21,976 15,274 359 Acres

0.0 - - 0.3 -I Class number and order correspond to that in map 0.0 - 0.0 0.0 -[

III legend Afulltownshipof36sq ml contains23,040acres 0.4 0.0 0.8 1.1 12.5 I1IV (9.331 hectares) Map legend includes study area totals 1.8 1.0 4.8 2.1 31.2 IV

T v andarea conversion factors From the land cover data base 8.2 3.8 11.3 9.8 17.8 V TV in digital format many other map and statistical products 71.7 75.7 77.8 84.5 37-0 VI3

' VII are possible 14.5 18.6 3.9 2.2 1.5 VIIN VilI 1.0 0.4 0.0 - - vin N

IX 1.6 0.3 1.2 - - IXx 0.8 0.2 0.2 0.0 - XXl I 0.0 0.0 0.0 - - XIXII - - - - - XII

Perent R~llltaE RanpIE Rant*Et RanVlE RanlieE RanpE RangleE R-ngleE RanigeE RangeE RanigeE 100 100 100 100 io0 ge 39 40 Ace eAc. 23 2 25 26 27 28 29 30 31 32 33 22,352 22,072 22,345 18,514 1,7164 Acres

P 1 00e

2' '..1

- I,

70'00,

db4

Ladoercas n oinn egtto

Lando eeret comutis ofCed sp orinn vrct to

W a.-ca suu rce and p k tod rive sufcslremoist orwettudr Shrub , Tnr. Shfdig Tusoes indota; or Moist

Pond/Sdge Tndra Cmplex Aqalpinen with whr l ow brcmhg aldr toan tmsor k ense

Wet Sedge Tundra.-Wet tundra with little standing in water tracks.water, or with up to hailf of surface area water coveredor emeirgent vegetation, or coastal areas periodically Partially vegetated areas. - Diverse habitats in, & covered with salt watier. niver bars, alpine tundra and moss mats with harren ft ll

and talus, lichen-covered sot ied stone nets, and N.,ai,Moist/Wet Sedge Tundra Complex; or Civ Prostrate mud flats, often with rich but Soarse floras

Shrub, Fosb Tundra (Divas river terraces). -Moistsedge tundra with up to 40% wet sedge tundra, or Barren gravel or rock.-Bare light colored rivetcr qrUi4.densee prostrate mat of Dryas On river terraces,. gravel and sand spits, alpine barrens tespecial d( kmnr,-e

and cultural barrens linad or runwayt, often wth ri,Moist Sedge, Prostrate Shrub Tundra; or Moist Sedge/ but sparse floras.

Baren. Tundlra Complex (frost-scar tundvial.-Bet,ter drained areas on rolling terrain sometimes with Wet gravel or mud. -Extensive barren mud in niver delt ascottonrass tussock*; or sparsely vegetated frost-scat and wet or dark-colored gravel on beaches or river beds.tun*&a at_ __6hawg

145' 44

BEAUFORT SEA

Camden Bay

Tundra. or Moist

).-Upland tundraknd tusok tundra

In, or dense shrubs

hah,,s in udingIt A - rr rock

143' West of Greenwich

DOMINANTIr in CATE(

ithe northern cc

ARCTIC rWILDLIFI

Ala

Geo fimaph Bs

g Ecology Lan

by William Ac

d a va D Wsale nce

Greenwich

INANT LANDCOVERCATEGORIES

in the northern coastal portion of the

ARCTIC NATIONALWILDUFE REFUGE

Alaska

70-

SSO • ",oundary of -

study area 690

000

CreditsS This map is derived from a Landsat classification entitled "Vegeta-S tion and Landcover of the Coastal Portion of the Arctic National

* Wildlife Refuge, Alaska," which was produced at the USGS/ C ography Branch Moffett Field, Calif. for CRREL and the USFWS

/ U., William Acevedo (Technicolor Graphic Services) and Donald A./ Walker (Institute of Arctic and Alpine Research (INSTAAR), Uni

I versity of Colorado). This map was produced in the INSTAAR PlantDt Ecology Laboratory by Donald A. Walker and Patrick J. Webber

with funding provided by the U.S. Army Cold Regions Research andEngineering Laboratory, Hanover, N.H. Cartography is by MarthaBramhall. January, 1982.

Distance Scales 1:250,000

oar e mgm. vltbon, or coastal areas periodically Partially vegetated areas.-Diverse habitats including

covered with salt water. river bars, alpine tundra and moss mats with barren rockand talus, lichen-covered sorted stone-nets, and beach or

Moist/Wet Sedge Tundra Complex; or Dry Prostrate f mud flats, often with rich but sparse floras.Shrub, Forb Tundra (Drym river terraces).-Moistsedge tundra with up to 40% wet sedge tundra; or Barren gravel or rock.-Bare light-colored nver gravel,dense prostrate mat of Dry.s on river terraces. gravel and sand spits, alpine barrens (especialy dolomute),

and cultural barrens (road or runway), often with richMoist Sedge, Prostrate Shrub Tundra, or Moist Sedge/ but sparse floras.

Barren Tundra Complex (frost-scar tundra).-Bet-ter-drained areas on rolling terrain sometimes with Wet gravel or mud. -Extensive barren mud in river deltascottongrass tussocks; or sparsely vegetated frost-scar and wet or dark-colored gravel on beaches or river beds,tundra, or dark-colored barren rock in mountains.

Mois Sedge Tusock, Dwarf Shrub Tundra.-Well- Ice. -River icings in the braided stream channels of most ofdrained upland tussock tundra in foothills with high the larger rivers.percentage of cottongrass tussocks and dwarf or pros-

30N trate hrub146

I .I -

j

- - -- - .-.--. I -- -- - -. ----- -I,

Pu/

d

- - I I

Geography Branch Moffett Field, Calif, for CRREL and the USFWSby William Acevedo (Technicolor Graphic Services) and Donald A.Walker (institute of Arctic and Alpine Research (INSTAAR), Uni-versity of Colorado). This map was produced in the INSTAAR PlantEcology Laboratory by Donald A. Walker and Patrick J. Webberwith funding provided by the U.S. Army Cold Regions Research andEngineering Laboratory, Hanover, N.H. Cartography is by MarthaBramhall. January, 1982.

Distance Scales 1:250,000

5 0 5 10km ======I 1Miles

5 0 5 10 15== IKilometers

690West of Greenwich 30' N

*U.S. Government Printing Office.1982--578-253