RemoteSensingofEnvironment112(2008)4034–4047
ContentslistsavailableatScienceDirect
RemoteSensingofEnvironment
journalhomepage:doc.guandang.net
Identi cationofinvasivevegetationusinghyperspectralremotesensingintheCaliforniaDeltaecosystem
ErinL.Hestira, ,ShrutiKhannaa,MargaretE.Andrewa,MariaJ.Santosa,JoshuaH.Viersb,JonathanA.Greenberga,SepalikaS.Rajapaksea,SusanL.Ustina
ab
DepartmentofLand,Air,andWaterResources,UniversityofCalifornia,Davis,One,ShieldsAve.,Davis,CA95616,UnitedStatesDepartmentofEnvironmentalScienceandPolicy,UniversityofCalifornia,Davis,OneShieldsAve.,Davis,CA95616,UnitedStates
ARTICLEINFOABSTRACT
Estuariesareamongthemostinvadedecosystemsontheplanet.Suchinvasionshaveledinpart,totheformationofamassive$1billionrestorationeffortinCalifornia'sSacramento–SanJoaquinRiverDelta.However,invasionsofweedsintoriparian, oodplain,andaquatichabitatsthreatenthesuccessofrestorationeffortswithinthewatershedandjeopardizeeconomicactivities.Thedoctrineofearlydetectionandrapidresponsetoinvasionshasbeenadoptedbylandandwaterresourcemanagers,andremotesensingisthelogicaltoolofchoiceforidenti cationanddetection.Howevermeteorological,physical,andbiologicalheterogeneityinthislargesystempresentuniquechallengestosuccessfullydetectinginvasiveweeds.Wepresentthreehyperspectralcasestudieswhichillustratethechallenges,andpotentialsolutions,tomappinginvasiveweedsinwetlandsystems:1)PerennialpepperweedwasmappedoveroneportionoftheDeltausingalogisticregressionmodeltopredictweedoccurrence.2)Waterhyacinthand3)submergedaquaticvegetation(SAV),primarilycomposedofBrazilianwaterweed,weremappedovertheentireDeltausingabinarydecisiontreethatincorporatedspectralmixtureanalysis(SMA),spectralanglemapping(SAM),bandindexes,andcontinuumremovalproducts.Perennialpepperweeddetectionwasmoderatelysuccessful;phenologicalstagein uenceddetectionrates.Waterhyacinthwasmappedwithmodestaccuracies,andSAVwasmappedwithhighaccuracies.Perennialpepperweedandwaterhyacinthbothexhibitedsigni cantspectralvariationrelatedtoplantphenology.Suchvariationmustbeaccountedforinordertooptimallymapthesespecies,andthiswasdoneforthewaterhyacinthcasestudy.Submergedaquaticvegetationwasnotmappedtothespecieslevelduetocomplexnon-linearmixingproblemsbetweenthewatercolumnanditsconstituents,whichwasbeyondthescopeofthecurrentstudy.Wediscussourstudyinthecontextofprovidingguidelinesforfutureremotesensingstudiesofaquaticsystems.
©2008ElsevierInc.Allrightsreserved.
Articlehistory:
Received6April2007
Receivedinrevisedform17January2008Accepted26January2008Keywords:
AquaticandwetlandweedsLogisticregressionmodelSpectralmixtureanalysisSpectralAngleMapperHyperspectralHyMapPhenology
1.Introduction
Invasionsofaquaticweedsintofreshwater,estuarine,and oodplainhabitatscandecreasebiodiversity,threatencriticalhabitat,alternut-rientcycles,anddegradewaterquality.AnestimatedUSD$100millionperyearisspentoncontrolanderadicationprogramstargetingaquaticweedsintheUnitedStates(Pimenteletal.,2000).Systematic,com-prehensivemonitoringprogramsareneededtodetectinvasionsinordertoeffectivelycontrolaquaticweeds.Traditionalmethodsofmonitoringweedinfestationsarecostly,timeconsuming,andoftenrequiredirectcontactwiththeweedswhichcanresultinfurtherweeddispersal(Bossardetal.,2000).Additionally,aquaticecosystemsareofteninac-cessibleorlogisticallydif cultfor eld-basedmonitoringmethods.Remotesensingprovidesasynopticsolutionformonitoringaquatic
Correspondingauthor.Tel.:+15307525092;fax:+15307545491.E-mailaddress:elhestir@ucdavis.edu(E.L.

Hestir).0034-4257/$–seefrontmatter©2008ElsevierInc.Allrightsreserved.doi:

10.1016/j.rse.2008.01.022
weedinfestationsoverlargespatialareas(AcklesonandKlemas,1987).Tobesuccessful,aremotesensingapproachmustbeaccurate,repeat-ableoverspaceandtime,andaccountfortheinherentspatialandenvironmentalheterogeneityofasystem.
Wepresentthreecasestudiesthatillustratehowtheseproblemscanbeaddressedtodevelopregional-scalemonitoringofinvasiveaquaticandwetlandweedsintheSacramento–SanJoaquinDelta:theterrestrialriparianweed,perennialpepperweed(Lepidiumlatifolium);the oatingaquaticweed,waterhyacinth(Eichhorniacrassipes);andthesubmergedaquaticweed,Brazilianwaterweed(Egeriadensa).Eachcasestudyhigh-lightsthecomplexitiesofremotesensingofaquaticandwetlandsys-tems.Theyalsodemonstratearangeoftechniquesthatcanbeused,andoftenintegrated,toproduceaccuratemapsforwetlandandestuarineresourcemanagers.Ourexamplesarefromastudytomaptheseinvasivespeciesathighspatialresolution(3m)overaregionalareaof2139km2.Thechallengeofanalyzingairbornehyperspectralremotesensingoveralargestudyregionnotwithstanding,wedemonstratethatthetechni-quespresentedcanbeappliedconsistentlytoall ightlinesinthestudy
Vegetation Ecology, Institute of Botany, the Chinese Academy of Sciences, ...ndings support that America is the primary geographical origin of invasive ...
invasive spread of exotic plants in native vegetation can pose serious ...imperative for conservation, management and the identification of research ...
are considered low maintenance and not invasive.... and vegetation to slow and detain stormwater, ... identification of the commissioning team, and ...
Using eMODIS Vegetation Indices for Operational Drought Monitoring 隐藏>> ... burn scar identification, ecosystem evolution, invasive species potential, ...
Invasive potential of the weed Parthenium hysterophorus 隐藏>> Invasive ...It forms pure stands at the expense of native vegetation, adversely affects...
Identification of Catechol and Hydroquinone_工学_高等教育_教育专区。158 Chem...(7), vegetation (8), and, due to their lipophilicity, in the tissues ...
(curly pondweed) is a cosmopolitan aquatic macrophyte considered invasive in ...ed as vegetation occurring in areas of open water were reclassi?ed as ...
( 2011) 03 - 0084 - 05 Morphological and molecular identification of ...( Jiangsu Key Laboratory for Prevention and Control of Invasive Species, ...
? Graffiti Trail braiding Obstacles in path Evidence of picnicking Litter Damage to vegetation Invasive vegetation Clarity of signposting ? ? ? ? ? ? ? ...
Identification of priority areas for controlling soil erosion_法律资料_人文...Quantitative analysis of the effect of vegetation cover on non-point ...
热门文档