Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
SouthͲEasternEuropeanJournal ofEarthObservationandGeomatics Issue Vo3,No2S,2014 LongͲtermhumanimpactonAlpineTundraͲ25yearsofchanges assessedbyaerialphotography JanaMüllerová*,MichaelaVítková InstituteofBotany,AcademyofSciencesoftheCzechRepublic,Prƽhonice,Czech Republic *Correspondingauthor:[email protected] Abstract:InourstudyweassessedthelongͲtermeffectsofalkalinegravelusedfortrail stabilizationuponadjacentalpinetundravegetationonnutrientpoorsubͲalpineplateau ofKrkonošeMtsNationalPark,CzechRepublic,usingaseriesofaerialMSS(1986,1989, 1997)andCIRimagery(2012),GPSmapping,andrepeatedsoilandvegetationsurveys. Aerial imagery was analyzed using both pixel and objectͲbased approach. During the studyperiodof25years,theareaofroadsidevegetationmorethendoubled,showing high rate of spread. In terrain depressions leading down the slope from alkaline stabilized roads it formed extensive lobes reflecting the nature of the terrain and reaching far into undisturbed vegetation (up to 156m). The spread of roadside vegetationwasspatiallyexaminedtodetermineitsdrivingforces,andwasfoundtobe significantly related to the type of the stabilizing material and the position relative to theroad(slopeposition,distancefromtheroad),indicatingtheeffectofrunoffwater. The latest imagery of 2012 documented reconstruction of trails (started in 2005). The conservation measure stopped the ecosystem alteration although the damage during reconstruction was extensive and ability of arcticͲalpine tundra to recover in the extremeclimaticconditionsveryslow.Longertimelagisthereforeneededtorecordthe recoveryprocess. Keywords: CIR Aerial Imagery, Change Detection, EnvironmentalMonitoring, Historical Analysis,RemoteSensing,RoadDisturbances,VegetationChanges. 1. Introduction Roads represent an important landscape element affecting both biotic and abiotic components. Disruption of the chemical environment (alteration of chemical soil properties) along roads affects plant growth and species diversity and composition (Cape et al., 2004). The impact is especially profound if alkaline gravel is used in predominantlynutrientͲpoorenvironmentdominatedbystresstolerantplants(Hilland Pickering,2006).Suchprocessesfacilitatecolonizationbymorecompetitive,oftennonͲ nativespecieswithgreaternutrientuptakeefficiencyandhigherbiomass(Dulièreetal., 1999).Thealpinetundrarepresentsanextremelyfragileecosystemsensitivetohuman disturbances. In such nutrient poor environment, adding nutrients (alteration of soil propertiesalongroads)isassumedtohaveverystrongimpactonvegetationstructure. 483 ®AristotleUniversityofThessaloniki,Greece PublishedonlineMay2014 SouthͲEasternEuropeanJournal ofEarthObservationandGeomatics 1986 Issue Vo3,No2S,2014 1989 1997 2012 Figure1.TheeffectofalkalineroadͲstabilizingmaterialontundravegetationtracedfromMSSandCIR aerialimageryinKrkonošeMts.,CzechRepublic. 2. Methodology OurstudyissituatedintheKrkonošeMtsNationalParkandBiosphereReserve(Figure 1),thehighestmountainrangeintheCzechRepublic.Uppermostareassituatedabove the timberline are covered by a particular ecosystem called "alpine" or "arcticͲalpine tundra",displayingaffinitiestobothsubͲarcticandhighͲmountainregions(Gordonetal. 2002), and hosting many endemic, glacial relic and rare species, and unique geomorphologiccomponents(Štursa1998).DespitetheN.P.status,in1970'sto1980's many paths and roads were stabilized to prevent erosion using unsuitable alkaline material.Inthisacidicenvironment,suchmeasuremarkedlyalteredbothsoilproperties andvegetationstructurealongsuchtrails.Whereasnormaltundravegetationisformed by short thin grasses and sessile herbs, along trails paved with alkaline nutrients, the native flora was replaced by mesoͲ to nitrophilous species and species of manͲmade habitats, mostly herbs with broader leaves and comparable higher biomass. This enabledtorecognizetheroadͲalteredvegetationonremotelysensedimagery. RoadͲrelated changes were reconstructed using a series of multispectral (1986, 1989, 1997) and CIR aerial imagery (2012), GPS mapping, and repeated soil and vegetation surveys.Theroadsidevegetationcouldbetracedonthemultispectraldatabecauseitis 484 ®AristotleUniversityofThessaloniki,Greece PublishedonlineMay2014 SouthͲEasternEuropeanJournal ofEarthObservationandGeomatics Issue Vo3,No2S,2014 muchtaller,hasbroaderleavesandproducesahigherbiomasscomparedtothenatural vegetation (Figure 2). It was classified using both pixel based (parallelepiped classification Ͳ nonͲparametric supervised method defining the lowest and the highest values of classes in every spectral band; for details see Müllerová 2004), and objectͲ based approach (a combination of automated hierarchical, iterative, and ruledͲbased classification). Seven permanent transects following the slope and perpendicular to a road/trail were selected randomly to analyze the effects of different road/trail constructionmaterialonvegetationcomposition(iny.2000and2004)andsoilphysicalͲ chemicalproperties(iny.1998,2000and2001;fordetailsseeMüllerováetal.2011). a) c) b) d) Figure2.RoadͲalteredvegetationon:a)MSSimagery(histogramequalized,roadͲalteredvegetationin violet),andb)CIRimagery(roadͲalteredvegetationinred).Bottompictures(bandd)showOBIA classificationresults. 3. Results Both pixel and objectͲbased image classifications enabled successfully analyze roadͲ related vegetation changes. Vegetation along roads paved with alkaline material grew considerablyfrom2.5%ofthestudysubsetin1986upto4.1%in2012(Figure1).Both soil and vegetation changes depended significantly on the type of the rodͲstabilizing materialandpositionrelativetotheslope(Müllerováetal.2011).Interraindepressions leading down the slope from alkaline stabilized roads the roadside vegetation formed lobesreflectingthenatureoftheterrainandreachingfarintoundisturbedvegetation, providingevidenceoftheroleofwatererosion.Alongalkalineroads,soilpHincreased from3.9upto7.6,basesaturationfrom9Ͳ30%upto100%,andlocalstressͲtolerant lowcompetitivetundraspeciesdisappeared(includingrareandprotectedspecies),and were replaced by mesoͲ to nitrophilous species and species preferring manͲmade habitats. 485 ®AristotleUniversityofThessaloniki,Greece PublishedonlineMay2014 SouthͲEasternEuropeanJournal ofEarthObservationandGeomatics Issue Vo3,No2S,2014 Based on increase between years 1886/97 measured by Müllerová et al. (2011), roadsidevegetationwasestimatedtocoverover7%in2012.Thispredictionwasnot metandtundraecosystemalterationsloweddownthankstotheconservationmeasures undertakenbyNPAuthoritiessince2004.Alkalineroadmaterialwasremovedandtrails stabilized using local building material. Today, most of the trails are reconstructed exceptfromtheasphaltroadbuiltwiththicklayerofdolomiteandmelaphyre.During the reconstruction, roadsides (ca 2 meters wide) were often disturbed and cleared of vegetation. From the latest imagery, we could document the process of vegetation recovery,althoughitwasnotthereforealwayspossibletodistinguishfromtheimagery whethervegetationalongroadsisdamagedbythereconstructionorreturnstonatural compositionoflowcoverandbiomass. 4. Conclusions Aerial imagery proved to be a good source of information on roadͲrelated changes in tundra vegetation, mainly thanks to its different size and biomass compared to native flora.Itwasevenpossibletosomeextendtodocumentthereconstructionofthetrails. Althoughtheappliedcontrolmeasureslimitexpansionofundesirablevegetation,they represent a serious disturbance in this fragile arcticͲalpine tundra ecosystem with limited ability to recover in the extreme climatic conditions, and future studies are thereforeneededtoassesstheprocessofecosystemrestorationinlongertimespan. References Cape J.N., Tang Y.S., van Dijk N., Love L., Sutton M.A., Palmer S.C.F., 2004, Concentrationsofammoniaandnitrogendioxideatroadsideverges,andtheir contribution to nitrogen deposition. Environmental Pollution, Volume 132, Issue3:469Ͳ478. DulièreJ.F.,CarnolM.,DalemS.,RemacleJ.,MalaisseF.,1999,Impactofdolomitelime on the ground vegetation and on potential net N transformations in Norway spruce(Piceaabies(L.)Karst.)andsessileoak(Quercuspeteraea(Matt.)Lieb.) standsintheBelgianArdenne,AnnalsofForestScience,Volume56:361Ͳ370. Gordon J.E., DvoƎák I.J., Jonasson C., Josefsson M., Kociánová M., Thompson D.B.A., 2002, GeoͲecology and management of sensitive montane landscapes. GeografiskaAnnaler,Volume84A,Issue3Ͳ4:193Ͳ203. HillW.,PickeringC.M.,2006,Vegetationassociatedwithdifferentwalkingtracktypesin the Kosciuszko alpine area, Australia, Journal of Environmental Management, Volume78:24–34. ŠtursaJ.,1998,ResearchandmanagementoftheGiantMountains'arcticͲalpinetundra (CzechRepublic).Ambio,Volume27:358Ͳ360. MüllerováJ.,2004,UseofdigitalaerialphotographyforsubͲalpinevegetationmapping: A case study from the Krkonoše Mts., Czech Republic, Plant Ecology, Volume 175,Issue2:259Ͳ272. 486 ®AristotleUniversityofThessaloniki,Greece PublishedonlineMay2014 SouthͲEasternEuropeanJournal ofEarthObservationandGeomatics Issue Vo3,No2S,2014 Müllerová J., Vítková, M., and Vítek, O., 2011, The impacts of road and walking trails upon adjacent vegetation: Effects of road building materials on species compositioninanutrientpoorenvironment,ScienceoftheTotalEnvironment, Volume19:3839Ͳ3849. 487 ®AristotleUniversityofThessaloniki,Greece PublishedonlineMay2014