Estimation of N status in Spartina alterniflora (smooth cordgrass) using handheld spectrometry

摘要

Saltmarshesareamongoneofthemostproductiveecosystemsintheworldandareimportantcomponentsofestuarinesystems,sinceitprovidefoodandnutrientstobothestuarineandcoastaloceanconsumers,servesasahabitatforyoungandadultestuarineorganisms,andarefugeforlarvalandjuvenileorganismsandregulatingimportantcompoundsoftheestuarinechemicalcycle.ThelocationofsaltmarshesasabufferzonebetweenlandandseaandthecontinuouslyincreasingN-­‐loadfromlandmakeitaraisingconcernregardingmonitoringandestimationofitsvulnerabilitytoeutrophicationandinterestinitsabilitytoremoveNbeforeitsenterstheestuariesandcoastaloceanwaters,alongwithmonitoringofthecurrentNstatus.RemotesensingisaparticularhelpfultoolwhentryingtoextractinformationfromlargeareasandtoestimateNstatusofvegetation.ThespectralreflectancesignatureofSpartinaalterniflora(adominantsaltmarshspecies)wasinvestigatedin13siteswithvaryingNinput,withintwoNewEnglandsaltmarshes(PlumIslandSoundandGreatSippewissetMarsh,USA),tosurveyifremotesensingcanbeusetosensevegetationresponsetodifferentnutrientinput.Twodifferentremotesensingtoolswasused;aDuelChannelUnispec,whichmeasurecanopyreflectanceandaSPAD-­‐502chlorophyllmeter,whichmeasureleafreflectance.Threedifferentvegetationindexes(NDVI,GreenNDVIandEVI)wereusedtomodelvegetationbiophysicalvariables.ItwasnotpossibletoestimateifoneindexortheotherwouldbebetterforanoverallusetoestimateNstatusbuttheresultsindicatethefeasibilityinpredictingNstatus.SPADvaluesgiveanindirectindicationofchlorophyllandnitrogencontentinleafbiomassbutonlyalowcorrelationwasobservedthancorrelatedwithredandgreenreflectance.MoreemphasishastobegivingoncalibrationofSPADmeasurementstoobtainmorereliableresults.SpectralreflectancedataobtainedfromUnispecmeasurements,clearlyillustratedthatthevegetationstateinthetwositeswithhighestNinput(20and300-­‐foldlargerthanreferencesites)representedthehealthiestgreenvegetationwithahighplantbiomass,whichcorrelatedwiththeNinputreceived.Intheremainingsites,therewasnotobservedacleardistinguishbetweenthespectraldataandobservedNinput.Remotesensingcanprovideinformationaboutvariationsinvegetationandgiveaninsightintoimportantvegetationbiophysicalfeatures.ThereforeusingremotesensingtodeterminedNstatusofvegetationisalowcostusefulmethod,butemphasisonfuturestudiesinthisareashouldbeapriority.