Document zdLjbJqdKjb94r0KVBk9pL903

ResearchGate See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/265155770 Section 3: The Porcupine Caribou Herd Article January 2002 CITATIONS 75 9 authors, including: Brad Griffith United States Geological Survey 67 PUBLICATIONS 4,963 CITATIONS SEE PROFILE D. E. Russell Yukon College 66 PUBLICATIONS 1,009 CITATIONS SEE PROFILE READS 245 Donald D. Young Alaska Department of Fish and Game 9 PUBLICATIONS 215 CITATIONS SEE PROFILE Robert White University of Alaska Fairbanks 160 PUBLICATIONS 3,975 CITATIONS SEE PROFILE Some of the authors of this publication are also working on these related projects: project Evaluating Endangered Species Act Recovery Plans View project project Maternal protein reserves and assessment of factors affecting lactation and calf growth View project All content following this page was uploaded by Robert White on 04 September 2014. The user has requested enhancement of the downloaded file. 8f BIOLOGICALfSCIENCE REPORTfUSGS/BRD 002-0001 f Section 3: The Porcupine Caribou Herd collared fanimals. fCalving fdistributions /f ere festimated from 767 calving sites of adult f>3fyear old) radio-f Brad Griffith, David C. Douglas, Noreen E. Walsh,b Donald D. Young, Thomas R. McCabe, Donald E.b Russell, Roberb G White, Raymond D. Cameron, andb Kenneth R. Whittenb collared female caribou obtained during f983-2001f [average of 40 sites per year; fixed-kernel analyses usingf Least Squares Cross Validation fSilverman 1986, feamanf et fil. 1996, f998, 1999)]. Concentrated calving areasb /ere defined fis fhe finnual kernel contour fhat fncludedf Documentation of the natural fange of variation inf calving Cites /ith greater than fiverage density fSeaman fitf ecological, fife history, find physiological Characteristicsf al. f998). Annual calving groundsb/ere defined fis thef of fiaribou fRangifer barandus- of the Porcupine fiaribouf 99% kernel utilization distributions fobtained from annualf herd is a necessary base for detecting or predicting anyf calving sites. Extent of calvingb/as defined as thef potential effects of industrial development bn fhef aggregate extent of all annual calving grounds.f performance fe.g., distribution, demography, /eight-gainf Vegetation fypes /ere mapped from Landsat-Thematicf of fndividuals) 5f the herd. Tofdemonstrate fin effect 5f Mapper satellite imagery (Fig. 2.1; Jorgensen et al. 1994) development, post-development performance must dif er and reduced from f7 fo 7 classes for caribou habitatf from pre-development performance /hile accounting forf analyses (Fig. 3.1). Wefestimated the Normalizedb any natural environmental tTends.f Difference Vegetation IndexbNDVI) (Tucker f979,f Wefhad 2 /orking hypotheses for our investigations : f Tucker et al. 1986) find sno/cover from .Advanced Veryf 1) performance 5f the Porcupine Caribou herd /asf High Resolution Radiometer fAVHRR) data fromf associated /ith environmental patterns and habitatf National Oceanic find .Atmospheric Administrationf quality, find 2) ficcess fo important habitats /as fi keyf (NOAA) polar orbiting satellites. Sno/cover /asf influence on demography.f estimated using fi finear regression drat /e derived byf Wefsought to document the range of natural variationf correlating AVHRR fhfrared reflectance /ith estimates of in habitat conditions, herd fiize, demographybfdefmedf sno/cover extracted from faerial photographs collected inf here as survival and freproduction), sources and magnitudef the 1002 .Area during fhe fino/melt periods of 1987 findf of mortality, distribution, habitat use, and /eight gain andf 1988 (r2 i 0.87, n = 80). Cloud contaminated fireas fn fhef loss; find fo develop fin Understanding df fhe fnteractionsf AVHRR images /ere identified (Baglio and Holroydf among these characteristics of the herd.f 1989) and excluded from analyses, as /ere large /aterf In addition, /e investigated /ays that /e could use thisf bodies. AVHRR fand Thematic fMapper ifmages f/eref background Information, Combined fvith buxiliaryf transformed fo fin .Albers Equal .Area projection find re-f information from the adjacent Central .Arctic caribou herd,f sampled fo f-km2 pixel fiize.f to predict the direction and magnitude fif any potential fif ectsf NDVIbndexes fhe disproportionate reflectance of near-f of industrial oil development fn the f002 .Area of the Arcticf infrared radiation from green vegetation (Tucker andf National Wildlife Refuge fin Porcupine fiaribou herd Calf ellars 1986) in the canopy of fplant communities. Thus,f survival on the herd's calving grounds during June.f relationships bet/een NDVI find fotal green plant biomassbor feaf firea index (LAI) /ould be expected fo bef Data, Methods and Assumptionsy strongest for plant communities /ith reduced verticalf distribution of green biomass find feaf firea (e.g.,f This /ork focused on the calving and post-calvingf seasons 5f the Porcupine caribou herd. fThe Calvingb seasonb/as defined fis fhe S-week period fhat began /ithf the birth 5f calves fspring). fPost-calvingb/as defined fisf the 3-/eek period that follo/ed the calving season (earlyf summer-.f Porcupine caribou herd size /as estimated by thef Alaska Department of Fish find Game (ADF&G) romf aerial fphoto-censuses fduring fpost-calving faggregations.f Only censuses considered reliable by ADF&G /ere used.f Variance fn finnual censuses due fo multiple observersf counting portions of the photo sets /as relatively smallf /hen compared /ith fiach census f2%) find /as fgnoredf in the display 5f finnual censuses fo the nearest f,000f animals.f Demography find calf /eight-gain /ere estimated communities dominated by sedges, grasses, or shortf shrubs that are common in tfhe Af rctic). Due to the size of the pixels (~1 km2) AVHRR data are linked more tof landscape processes than to individual plant communitiesf (Malingreau find Bel/ard f992).f Relatively good correlations have been obtainedf bet/een above ground net primary productivity (ANPP) and seasonally integrated NDVI (r2 = 0.89; Paruelo et al.f 1997), LAI find NDVI /hen fhtegrated ficrossf physiognomic categories (r2 = 0.97; Shippert et al. 1995),f and photosynthetic biomass and NDVI fn small plots (r2 =f 0.51; Hope et fil. 1993). Because NDVI indexed fotalf green biomass and caribou are fselective feeders (Whitef 1983), /e fissumed fhat fhe biomass of forages Oaten byf caribou /as positively correlated /ith fotal green biomassf at the landscape scale.f from repeated focations find/or recaptures 5f radio-f