G.2094Page1GlobalResearchjournalofNaturalScience&Technology(GRJNST)Volume:04-Issue3(2026),2094ISSNP:2790-7643ISSNE:2790-7651www.grjnst.nethttps://doi.org/10.53762/grjnst.04.03.15RemoteSensing–BasedAssessmentofForestCarbonSequestrationandLandUseChangeDynamicsunderClimateVariabilityReceived:30March2026.Accepted:01May2026.Published:30May2026MuhammadHassanAli(CorrespondingAuthor)DepartmentofForestryandRangeManagement,ShaheedBenazirBhuttoUniversityofVeterinaryandAnimalSciences,Sarkrand,Sindh,Pakistansoilscience1070@gmail.comFatehAliChohanDepartmentofForestryandRangeManagement,ShaheedBenazirBhuttoUniversityofVeterinaryandAnimalSciences,Sarkrand,Sindh,Pakistanfateh.ali0335@gmail.comAsadullahDepartmentofForestryandRangeManagement,ShaheedBenazirBhuttoUniversityofVeterinaryandAnimalSciences,Sarkrand,Sindh,Pakistansaad.janii.94@gmail.comKalsoomDepartmentofEnglishliteratureUniversityofSindhJamshoro,Sindh,Pakistanmahachohan554@gmail.comMuhammadIsmailChohanDepartmentofForestryandRangeManagement,ShaheedBenazirBhuttoUniversityofVeterinaryandAnimalSciences,Sarkrand,Sindh,Pakistanimail40@gmail.comGRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15Copyright©2026GRJNST.ThisarticleispublishedunderanOpenAccessmodel.ItismadeavailabletothepublicunderthetermsoftheCreativeCommonsAttribution4.0International(CCBY4.0)license,whichpermitsunrestricteduseanddistributionG.2094Page2WalliullahLagahriDepartmentofForestryandRangeManagement,ShaheedBenazirBhuttoUniversityofVeterinaryandAnimalSciences,Sarkrand,Sindh,Pakistanraiswaliullah@gmail.comMuhammadAfzalDepartmentofForestryandRangeManagement,ShaheedBenazirBhuttoUniversityofVeterinaryandAnimalSciences,Sarkrand,Sindh,Pakistanafzalshar66@gmail.comAbstract:Forestecosystemsareamongthemostsignificantterrestrialcarbonreservoirsandplayacrucialroleinregulatingtheglobalcarboncyclethroughcarbonsequestration.However,rapidlanduseandlandcoverchange(LULC),combinedwithincreasingclimatevariability,andhassignificantlyalteredthestabilityandefficiencyoftheseecosystems.Thisreviewpaperexaminestheapplicationofremotesensingtechnologiesinassessingforestcarbonsequestrationandmonitoringlandusedynamicsunderchangingclimaticconditions.Thestudyhighlightshowurbanization,agriculturalexpansion,deforestation,andwildfireactivitiescontributetocarbonemissionsandecosystemdegradationacrosstropical,coastal,semi-arid,andurbanforestlandscapes.Specialemphasisisplacedontheintegrationofoptical,LiDAR,andSyntheticApertureRadar(SAR)systemsforestimatingabove-groundbiomass(AGB),soilorganiccarbon(SOC),andvegetationstructuralcharacteristics.Thereviewfurtherexplorestheroleofvegetationindices,multi-sensorfusion,andmachinelearningalgorithmssuchasRandomForest(RF),SupportVectorMachines(SVM),XGBoost,andConvolutionalNeuralGRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page3Networks(CNNs)inimprovingbiomassestimationaccuracyandcarbonaccounting.Additionally,thepaperdiscussestheinfluenceofclimaticvariables,includingtemperaturerise,alteredprecipitationregimes,droughts,andwildfires,onforestcarbonsequestrationpotentialandecosystemresilience.EmergingtechnologiessuchasESABiomass,NISARmissions,andDigitalTwinEarthsystemsarealsoevaluatedfortheirtransformativepotentialinglobalforestmonitoringandcarbonmanagement.Thefindingssuggestthatintegratingadvancedremotesensingwithartificialintelligenceandclimatemodelingprovidesaneffectiveframeworkforsustainableforestmanagement,climatemitigationstrategies,andaccuratecarbonmonitoringatregionalandglobalscales.Keywords:RemoteSensing,ForestCarbonSequestration,LandUseChange,ClimateVariability,Above-GroundBiomass,LiDAR,SAR,CarbonAccounting,MachineLearning,ForestMonitoringIntroductionTheterrestrialbiosphereconstitutesoneofthemostsignificantanddynamiccomponentsoftheglobalcarboncycle,withforestecosystemsalonestoringapproximately45%ofallterrestrialcarbon.Theseecosystemsfunctionascriticalnaturalsinks,sequesteringanestimated3.5PgCyr⁻¹fromtheatmosphere,therebyplayingapivotalroleinmitigatingtheprogressionofanthropogenicclimatechange(Xu,2025).However,thestabilityandefficacyofthissequestrationcapacityarecurrentlyunderunprecedentedpressurefromthetwindriversoflanduseandlandcoverchange(LULC)andintensifyingclimatevariability.ThetransformationofnaturallandscapesGRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page4drivenbyurbanization,agriculturalexpansion,andindustrialresourceexploitationaccountsfornearly25%ofhuman-causedgreenhousegas(GHG)emissions(Merrikhpouretal.,2025).Concurrently,theincreasingfrequencyofextremeclimaticevents,includingdroughtsandwildfires,hasbeguntocompromisethehistoricalresilienceofthesecarbonreservoirs,potentiallyshiftingthemfromnetsinkstonetsourcesofatmosphericcarbon(Xuetal.,2021).Inthiscontext,remotesensing(RS)technologieshaveemergedasindispensabletoolsforthemonitoringandassessmentofforestcarbondynamicsatmultiplespatialandtemporalscales.SincethelaunchofLandsat-1in1972,Earthobservationhasevolvedfromsimplelandcoverclassificationtosophisticated,multi-sensorframeworkscapableofquantifyingthree-dimensionalforeststructure,above-groundbiomass(AGB),andsoilorganiccarbon(SOC)(Wulderetal.,2019).Modernassessmentsnowintegratehigh-resolutionopticalimagery,activemicrowaveradar,andlaser-scanningsystemstoprovideacomprehensive,spatiallyexplicitunderstandingofhowLULCandclimatevariabilityinteracttoshapethefutureofterrestrialcarbonsequestration(H.Nguyenetal.,2019).SpatiotemporalDynamicsofLandUseChangeandCarbonFluxLanduseandlandcoverchangerepresentsthesecond-largestsourceofglobalCO₂emissions,trailingonlythecombustionoffossilfuels.Between2001and2023,theglobalforestsystemwasanetsinkof−5.5±8.1GtCO₂eyr⁻¹,afigurethatobscuresthevolatilebalancebetween−14.5±7.7GtCO₂eyr⁻¹ofcarbonremovalsand9.0±2.7GtCO₂eyr⁻¹ofgrossemissions.Themagnitudeofthesefluxesisheavilyinfluencedbyregionaldevelopmenttrajectoriesandthespecificnatureoflandtransitions(Luetal.,2014).GRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page5UrbanizationandPeri-UrbanCarbonDepletionRapidurbangrowth,particularlyindevelopingmetropolitanbasins,catalyzessignificantenvironmentaltransformationsthatoftenleadtoanetincreaseincarbonemissions(Ashaoluetal.,2019).IntheKağıthanebasinofIstanbul,forexample,theexpansionofresidentialandindustrialareashasbeendrivenbya16.59%populationincreaseoverthepastdecade.AnalysisusingSentinel-1andSentinel-2dataprocessedontheGoogleEarthEngine(GEE)platformrevealsthatwhilesomevegetationgrowthisobservedinmanagedurbangreenspaces,thereisasystemicdeclineinnaturalforestcoverandbarrenlands.Thisshiftisprojectedtoincreaseregionalcarbonemissionsbyupto13%between2035and2095(Kocaman&Ağaçcıoğlu,2025).Beyonddirectbiomassloss,urbanizationmodifieslocalhydrologicaldynamics,includingpeakdischargepatternsandsurfacerunoff,whichindirectlyaffectsthehealthandcarbonuptakeofsurroundingvegetation(Ashaoluetal.,2019).Theconversionofnaturallandcovertobuilt-upenvironmentsaltersthesurfaceenergybalance,oftenleadingtourbanheatisland(UHI)effectsthatcanbeaccuratelymappedusingthermalinfraredremotesensing.Thesethermalanomaliesexacerbatephysiologicalstressonremainingurbantrees,potentiallyreducingtheirlong-termcarbonsequestrationpotential(Yang,2025).AgriculturalExpansionandCoastalEcosystemVulnerabilityTropicalcoastalecosystemsrepresentsomeofthemostcarbon-denseenvironmentsonEarth,yettheyareincreasinglythreatenedbytheexpansionofindustrialagriculture.InPhangNgaBay,southernThailand,extensivelandusetransitionsbetween2000andGRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page62020wereprimarilydrivenbythereplacementofevergreenandpararubberforestswithoilpalmplantations(Aratrakornetal.,2006).Thisconversionledtoanetcarbonstoragelossofapproximately,withthemostintensivedegradationoccurringinuplandforestedareasbetweenandmetersinelevation(Alongi,2014).Theroleof"bluecarbon"ecosystems,specificallymangroves,isparticularlycriticalinthesecoastallandscapes.Despitecoveringonlyaboutone-fifthofthetotalareainPhangNgaBay,mangrovesconsistentlycontributedoveroftheregionalcarbonstorage.Thishighlightsacriticalspatialdisparity:whileagriculturalencroachmentcauseswidespreadlow-intensitycarbonlossacrossuplandareas,thelocalizeddestructionofmangroveorswampforestsresultsinmassive,immediatecarbonreleases(Murphy,Hall,&Jintana,2020).IntegratingtheInVEST(IntegratedValuationofEcosystemServicesandTradeoffs)modelwithhigh-resolutionsatellitedatahasallowedresearcherstodemonstratethatlocalizedgainsinmangrovesequestrationcanoccasionallyoffsetsomeconversionlosses,providedthattargetedconservationstrategiesareimplemented(Sharpetal.,2018).Table1.RemoteSensing–BasedEvaluationofLandUseTransitionsandAssociatedCarbonDynamicsLandUseTransitionNetCarbonPrimaryRSKeyDriversCategoryImpactMonitoring(Relative)StrategyForesttoUrbanHighLossOptical(Sentinel-Pop.Growth,(Direct+2)+ThermalInfrastructureGRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page7Indirect)ForesttoOilPalmHighBiomassSAR(Sentinel-1)CommercialLoss+GEEAgricultureCoastalForesttoCriticalCarbonLiDAR+Multi-FoodSecurity,AquacultureReleasespectralEconomicsReforestation/AfforestationGradualTime-seriesCarbonCredits,Accumulation(NDVI/EVI)PolicyForesttoGrassland/PastureModerate-HighLandsatGFCLivestock,Small-Loss(Hansenetal.)scalefarmingLong-termVariationsinTropicalandSemi-AridForestsInregionslikeSoutheastVietnam,theperiodbetween1990and2020sawsignificantfluctuationsincarbonservicesduetoecologicaldegradationandurbanexpansion.Whileecologicalsuccessionandforestrestorationeffortspartiallycompensatedforsomelosses,thecombinedanthropogenicimpactoutweighedthenaturalrecoverycapacity,leadingtoanetdeclineintotalcarbonstorage.Thistrendisechoedglobally,withtropicaldeforestationparticularlyinBrazilandIndonesiaaccountingfornearlyhalfoftheglobalforestlossduetolandconversion.Incontrast,India'ssemi-aridanddryforests,locatedinregionslikeRajasthanandGujarat,presentadifferentchallengeforcarbonsequestrationassessment.Theseecosystemspossesslowerbiomassaccumulationandcarbonstoragecapacitycomparedtohumidtropicalforestsduetowaterscarcityandhightemperatures.However,theyGRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page8oftencontainresilientbelow-groundcarbonpoolsthatcontributesignificantlytosoilorganiccarbon(SOC)underextremeclimaticconditions.Remotesensingofthesedrylandsrequiresashiftinfocusfromcanopygreennesstostructuralparametersandlitterdepositionrates,asfast-growingplantationspecieslikeTeakorEucalyptusareincreasinglyusedforreforestationonwastelands.RemoteSensingTechnologiesforCarbonAccountingTheevolutionofremotesensinghasprovidedamulti-layeredframeworkforassessingforestcarbon,movingfrombroadlandcovermappingtotheprecisequantificationofindividualtreecomponents(Pendletonetal.,2012)OpticalRemoteSensingandSpectralLimitationsOpticalsensors,includingLandsat8/9andSentinel-2,remainthefoundationofglobalmonitoringduetotheirconsistentspectrallibrariesandhighrevisitfrequencies.Theseplatformsallowforthecalculationofvegetationindicessuchasthenormalizeddifferencevegetationindex(NDVI),theenhancedvegetationindex(EVI),andthephotochemicalreflectanceindex(PRI).Theseindicesserveasproxiesforcanopygreenness,leafareaindex(LAI),andphotosyntheticvigor,whichareempiricallylinkedtoabove-groundbiomass(AGB)(Lietal.,2021).However,afundamentalchallengewithopticalremotesensingisthe"saturation"effect.Inhigh-biomasstropicalforests,thespectralresponseofvegetationindicesoftenreachesaplateauoncethecanopyisfullyclosed,typicallyatbiomasslevelsbetween150and200Mgha⁻¹.Thismakesitdifficulttodistinguishbetweenmoderatelydenseandveryold-growthforestsusingopticaldataalone(Sinhaetal.,2019).Tomitigatethis,newermissionssuchasSentinel-2utilizered-edgespectralbands,whichhaveshownhigherGRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page9sensitivitytochlorophyllcontentandnitrogenstatus,therebyimprovingAGBandsoilorganiccarbon(SOC)estimationsevenindensecanopies(Jagadishetal.,2024).ActiveSystems:LiDARandSARActiveremotesensingtechnologieslightdetectionandranging(LiDAR)andsyntheticapertureradar(SAR)haverevolutionizedbiomassestimationbyprovidinginformationontheverticalandthree-dimensionalstructureofforests(Choi,2024).LiDARsystemsuselaserpulsestomeasurethedistancebetweenthesensorandtheEarth'ssurface,generatinghigh-resolution"pointclouds"thatrepresenttheforest'sverticalprofile.Airbornelaserscanning(ALS)iscurrentlyconsideredthemostaccuratetechnologyforextractingmajorforestattributessuchastreeheight,canopydensity,andvolumeforabove-groundbiomass(AGB)estimation.Onafinerscale,terrestriallaserscanning(TLS)andclose-rangesensingfromunmannedaerialvehicles(UAVs)allowforthemeasurementofdiameteratbreastheight(DBH)andtrunkshapewithcentimeter-levelprecision(Xuetal.,2023).Ameta-analysisofclose-rangesensingaccuracyindicatesthatground-basedLiDARremainsthe"goldstandard"forsingle-treeandplot-levelassessments,thoughUAV-basedsystemsaremoreefficientforstand-scaleanalysis(Fayadetal.,2016).SARsystems,ontheotherhand,usemicrowaveenergytopenetratecloudcoverand,dependingonthewavelength,theforestcanopyitself.ThesensitivityofSARtobiomassisprimarilyafunctionofitswavelength:GRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page10X-bandandC-band(e.g.,Sentinel-1):Theseshorterwavelengthsinteractprimarilywithleavesandsmallbranchesintheuppercanopy.Theyareeffectiveforforestcoverchangedetectionbutsaturateatrelativelylowbiomasslevels.L-band(e.g.,ALOSPALSAR-2,NISAR):Thislongerwavelengthpenetratesdeeperintothecanopy,interactingwithlargerbranchesandtrunks.ItprovidesamorerobustproxyforAGBandislesspronetosaturationthanopticalorC-bandSAR(LeToanetal.,2024).P-band(e.g.,ESABiomassmission):Withawavelengthofapproximately70cm,P-bandSARcanpenetratetheentirecanopytointeractwiththemaintrunksandthegroundsurface,makingitthemosteffectivetoolformeasuringhigh-biomassforestsupto500Mgha⁻¹(HoTongMinhetal.,2016).Multi-SensorFusionandAdvancedModelingThemostaccurateforestcarbonassessmentsareincreasinglybasedonthe"fusion"ofmulti-sourcedata.Byintegratingopticalspectraldatawith3DstructuralinformationfromLiDARorSAR,researcherscanovercomethelimitationsofindividualsensors.Forexample,combiningSentinel-2spectralbandswithDigitalElevationModels(DEMs)andLiDAR-derivedcanopyheightshasbeenshowntosignificantlyimprovetheaccuracyofmachinelearningmodelsforAGBprediction(Ali,2025).ThisfusionapproachisparticularlyrelevantformeetingtheMeasurement,Reporting,andVerification(MRV)standardsrequiredbyinternationalclimateframeworkssuchasGRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page11theUNFCCCandREDD+.Integratedmodelscanleveragethebroadspatialcoverageofsatelliteimageryandthehigh-precision"truth"providedbyfieldplotsandairborneLiDARtogeneratewall-to-wallmapsofcarbonstocksandfluxes(Santos,2025).Table2.ComparisonofRemoteSensingSensorTypesforForestStructuralandBiomassAssessmentSensorTypeSpecificStructuralVariableAccuracyPrimaryPlatform/BandMeasuredOpticalSentinel-2(Red-Canopyedge)Greenness/LAILiDARALS(Airborne)TreeSARL-bandHeight/CanopyProfileLargeRange(R2)0.550.820.850.95Limitation–SpectralSaturation–HighAcquisitionCost0.60–SignalDe-(ALOS/NISAR)Branches/Trunks0.80correlationSARP-band(Biomass)MainStem/Trunk0.75–IonosphericVolume0.90InterferencePassiveSMOS/SMAPVegetationOptical0.50–CoarseSpatialMicrowave(VOD)Depth0.70Res(>9km)ClimateVariabilityanditsImpactonSequestrationPotentialGRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page12Climatevariabilityexertsaprofoundinfluenceonthephysiologicalprocessesandspatialdistributionofforestvegetation,therebyalteringitscarbonsequestrationpotential(CSP).CSPisdefinedasthedifferencebetweenanecosystem'smaximumcarboncarryingcapacity(CCC)anditsactualcarbonstock,representingthepotentialforfuturecarbonuptake(Xu,2025).TemperatureandPrecipitationThresholdsOngoingclimatechanges,characterizedbyshiftingtemperatureregimesandalteredprecipitationpatterns,canseverelyimpactforeststructureandfunction.InthemountainecosystemsofYunnanProvince,China,simulationmodelshaveshownthatthesuitabilityofforesthabitatsisprimarilylimitedbytheminimumtemperatureofthecoldestmonth(TMW)andtotalseasonalprecipitation(PRS).Forexample,a1°Cincreaseintemperaturecombinedwitha20\%decreaseinprecipitationcouldreducethepotentialdistributionareaofmajorforesttypesby12.41\%(Iheaturu,2026).Interestingly,thecombinedeffectofincreasedtemperatureanddecreasedprecipitationcan,insomespecificcases,increasetheCSPofcertainforesttypesbyacceleratingbiomassturnover,thoughthisisoftenatransientresponse.Generally,however,frequentandseveredroughteventssuchasthe"flashdroughts"observedinsoutheasternAustraliaortherecord-lowwaterlevelsintheAmazonRiverin2023poseasignificantthreattoforesthealth.Theseeventscauseunusuallyrapiddryingofsoilandvegetation,leadingtowidespreadtreemortalityandreducedcarbonuptake(Tianetal.,2023).BiogeophysicalandBiogeochemicalFeedbackLoopsTheinteractionbetweenforestsandtheclimatesystemoccursthroughtwoprimaryfeedbackpathways:biogeochemical(BGC)andbiogeophysical(BGP).TheGRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page13biogeochemicalpathwayinvolvestheexchangeofgreenhousegases,primarilyCO₂,betweentheforestandtheatmosphere.Deforestationandforestdegradationreleasestoredcarbon,increasingatmosphericCO₂concentrationsandcontributingtoglobalwarming(Boysenetal.,2020).Earthsystemmodels(ESMs)estimatethathistoricallanduseandlandcoverchange(LULC)-inducedcarbonemissionshaveresultedinaglobalwarmingofapproximately0.21±0.14°C.Thebiogeophysicalpathwayinvolveschangesinthephysicalcharacteristicsofthelandsurface,suchasalbedo,surfaceroughness,andevapotranspiration(ET)(Amalietal.,2024).Albedo:Convertingdark,absorbentforeststoreflectivepasturesorsnow-coveredgrasslandsincreasessurfacealbedo,reflectingmoresolarradiationbackintospace.Thishasacoolingeffect,particularlyinhigh-latitudeborealregions(Jiaoetal.,2023).Evapotranspiration:ForestsarehighlyefficientattransferringwaterfromthesoiltotheatmospherethroughET,aprocessthatcoolsthelocalenvironment.Intropicalregions,thelossofETduetodeforestationgenerallyleadstosignificantlocalwarmingthatcanovercompensateforanyalbedo-relatedcooling(Boysenetal.,2020).SurfaceRoughness:Forestshavehighsurfaceroughness,whichpromotesatmosphericturbulenceandefficientheattransfer(Presteleetal.,2016).Reducingthisroughnessthroughlandclearingcanleadtohighertemperaturegradientsatthesurface.Theneteffectofforestlandusechangeisoftenadelicatebalancebetweenthesetwopathways.Onaglobalscale,theBGCtemperatureeffectshistoricallydominatetheBGPeffects,meaningthattheoverallimpactofhistoricallandusechangehasbeentowarmtheclimate.However,atregionalscales,particularlyinthetropicsandhighlatitudes,theGRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page14BGPeffectscanbeequalinmagnitudetotheBGCeffects,highlightingtheneedforspatiallyexplicitclimatemitigationstrategies(Amalietal.,2024).Table3.BiogeochemicalandBiogeophysicalFeedbackPathwaysAssociatedwithLandUseChangeFeedbackPathwayMechanismClimateImpactClimateImpact(Global)(Regional)BiogeochemicalCO_{2}ConsistentVarieswithbiomass(BGC)Release/StorageWarmingfromdensityLossBiogeophysicalAlbedoShiftNegligibletoSlightCoolinginBoreal(BGP)Cooling(HighAlbedo)BiogeophysicalEvapotranspirationSlightWarmingIntenseWarmingin(BGP)fromLossTropicsBiogeophysicalSurfaceRoughnessSlightWarmingLocalizedHeatFlux(BGP)fromLosschangeTheWildfire-ClimateFeedbackLoopClimatechangeisincreasinglydrivingadangerous"two-waystreet"relationshipwithlandusethroughtheintensificationofwildfires.Warmingtemperaturesandlongerperiodsofdroughtcreatehotter,drierconditionsthatescalatefirerisk.Emissionsfromforestfireshaveincreasedby60\%globallysince2001,withfirenowaccountingforGRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page15one-thirdofalllandcoverchangeinsomeregions(Jiaoetal.,2023).Thiscreatesaself-reinforcingfeedbackloop:climatechangefuelsfires,whichreleasevastquantitiesofCO_2,whichinturnacceleratesfurtherwarming.Borealandhumidtropicalregions,whichcontaintheworld'slastgreattractsofnaturalforest,haveseenthemostdramaticincreasesinfire-drivenforestloss,withastrongcorrelationr^2=0.85intropicalregions)betweenglobaltemperatureanomaliesandfire-inducedforestloss(Papucci,2026).MachineLearningFrameworksforBiomassEstimationThetransitionfromtraditionalforestinventoriestoremotesensing-basedassessmenthasbeenacceleratedbytheapplicationofadvancedmachinelearning(ML)anddeeplearning(DL)algorithms.Thesemethodsareparticularlyeffectiveatcapturingthenon-linearandhigh-dimensionalrelationshipsbetweenremotesensingfeaturesandforestcarbonstocks(Fayadetal.,2016).AlgorithmSelectionandPerformanceMetricsAwiderangeofmachinelearning(ML)algorithmsiscurrentlyutilizedintheliterature,withrandomforest(RF),supportvectormachines(SVM),andextremegradientboosting(XGBoost)beingthemostprominent(Donatoetal.,2011).figure:2Characteristicextraction"leadstothefirstconcentricarcsegment,"ProblemCharacteristicSpace".GRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page16RandomForest(RF):RFisthemostfrequentlyusedalgorithm,appearinginapproximately88%ofrecentstudies.Itisanensemblemethodthatconstructsmultipledecisiontreesandaveragestheirpredictions,makingitrobustagainstoutliersandnoisypredictors.InXinjiang,China,theRFmodelcombinedwithGRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page17topographicandmeteorologicaldatasignificantlyimprovedabove-groundbiomass(AGB)estimationaccuracyacrossdiverseforesttypes,achievingvaluesgreaterthan0.65(Cohen&Goward,2004).XGBoost:WhileRFiscommon,XGBoosthasrecentlyshownsuperiorperformanceinapproximately75%ofthestudieswhereitwasdirectlycomparedwithothermethods.InastudyofLarixprincipis-rupprechtiiplantationsinnorthernChina,XGBoostachievedanof0.82andarootmeansquareerror(RMSE)of0.73Mgha⁻¹usingSentinel-2data,outperformingbothSVM()andRF().XGBoost'ssuccessisattributedtoitsefficienthandlingofnon-linearitiesanditsregularizationparametersthatpreventoverfitting(Lüetal.,2023).DeepLearning(DL):Convolutionalneuralnetworks(CNNs)areincreasinglyappliedtoextracttexturalandstructuralfeaturesfromhigh-resolutionimagery(e.g.,UAV-RGBorPlanetScope).DLmodelscanreducebiomassestimationerrorsby5%to20%comparedtotraditionalregressionmethods,particularlyincomplex,heterogeneousstands(Xu,2025).FeatureSelectionandVariableImportanceTheperformanceofMLmodelsishighlydependentontheselectionofcharacteristicvariables.Featureselectionalgorithms,suchastheBorutaalgorithmortheLeastAbsoluteShrinkageandSelectionOperator(LASSO),areusedtoidentifythemostrelevantpredictorsfromapoolofspectralbands,vegetationindices,andtexturefeatures(Yazaretal.,2023).GRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page18Inlarge-scaleprovincialmodels,climate(e.g.,meanannualtemperature,precipitation),topography(e.g.,slope,elevation),andtexturefactorsoftenemergeasmoresignificantthanindividualspectralbands.Forinstance,addingDigitalElevationModel(DEM)datatoopticalimageryfrequentlymakestheDEMthemostimportantpredictorvariable,asitaccountsfortheenvironmentalgradientsthatgoverntreegrowthandbiomassaccumulation(Amalietal.,2024).Table4.PerformanceComparisonofMachineLearningandProcess-BasedModelsinCarbonandBiomassAssessmentModelBestPerformancePrimaryDataSourceKeyFeatureforAccuracyType(R2)XGBoost0.82Sentinel-2Red-edge+VegetationIndicesRF0.75UAV-RGB/LiDARTexture+HeightmetricsSVM0.79Landsat-9NIRandSWIRbandsCNN0.85–0.98High-resDroneSpatialImagerypattern/SegmentationInVESTN/A(Process-LULC+BiomassCarbonpoolcoefficientsbased)mapsThe2025TechnologicalFrontier:DigitalTwinsandNext-GenMissionsGRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page19Thefieldofforestcarbonmonitoringiscurrentlyenteringatransformativephasecharacterizedbythelaunchofdedicatedbiomasssatellitesandtheoperationalizationof"DigitalTwin"Earthsystems.TheESABiomassandNISARMissionsTheyear2025marksthelaunchoftwocriticalradarmissionsdesignedtoquantifyglobalforeststructurewithunprecedentedprecision(Orlovetal.,2024).ESABIOMASSMission(scheduledlaunch:April29,2025):ThismissionwillutilizeaP-bandsyntheticapertureradar(SAR)todeliverglobalmapsofforestbiomassandtreeheighteverysevenmonths.Itsuniquewavelengthwillallowitto"see"throughdensetropicalcanopiestomeasuretheprimarywoodybiomass,achievingaprojectedcanopyheightrootmeansquareerror(RMSE)of1–2mandanabove-groundbiomass(AGB)RMSEof15–25Mgha⁻¹(Presteleetal.,2016).NASA-ISRONISARMission(scheduledlaunch:July30,2025):NISARisadual-frequency(L-bandandS-band)SARmission.TheL-bandishighlysensitivetolargebranchesandtrunks,whiletheS-bandisresponsivetouppercanopyfoliage.Theintegrationofthesetwofrequencieswillextendthedynamicrangeofbiomassretrievalandimprovecoherencestabilityforlong-termmonitoring(Presteleetal.,2016).DestinationEarth(DestinE)andForestDigitalTwinsTheEuropeanCommission’sDestinationEarth(DestinE)initiativeisbuildingahighlyaccuratedigitalreplicaoftheEarthsystem,poweredbyhigh-performancecomputing(HPC)andAI.AflagshipcomponentofthissystemistheForestDigitalTwin(ForestDTC),ledbyorganizationslikeVTTandECMWF.GRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page20TheForestDTCoperatesata10-meterresolution,primarilyutilizingSentinel-2datatoprovideacomprehensivedescriptionoftheforestsubsystem.Itintegratesseveralspecializedmodels:PRELES:AlightuseefficiencymodelthatoutputsGrossPrimaryProduction(GPP),ET,andNetEcosystemExchange(NEE)usingdailyweatherdata.CROBAS:Atreegrowthmodelthatusesstandvariables(species,density,DBH)tosimulatebiomassandlitterfall.YASSO15:Asoilcarbonmodelthatestimatessoilrespirationandcarbonaccumulationbasedonlitterfallandclimate(Jiaoetal.,2021).This"livingdigitaltwin"allowsuserstorun"what-if"scenarios,testingtheimpactofdifferentclimatepathwaysandforestmanagementstrategies(e.g.,selectiveloggingvs.afforestation)onfuturecarbonsequestration.Suchsystemsarerevolutionaryforcarbonmarkets,astheyprovideatransparent,investible,andauditableledgerofforesthealththatsurpassestraditionalmulti-yearverificationcycles(Boysenetal.,2020)MethodologicalChallengesandStrategicImplicationsDespitethesetechnologicalleaps,severalpersistentchallengescontinuetolimittheabsoluteaccuracyofremotesensing-basedcarbonassessments(Amalietal.,2024).GRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page21Figure1.Comparisonofground-basedtraditionalinventoriesandremotesensingestimationmethodswithintheforestcarbonaccountingframework.GRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page22ScaleInversionandGround-TruthingAsignificanttrade-offexistsbetweenspatialscaleandestimationaccuracy.Whileground-basedLiDARofferssub-centimeterprecisionatthesingle-treelevel,itsaccuracytendstodiminishwhenscaleduptoplot-levelorstand-levelanalysesduetocumulativeerrorsinsingle-treesegmentationandtheinterconversionofvariableslikeDBHandheight.Furthermore,thereisan"R&Dgap"inground-truthingfordiverseecologicalzones;mosthigh-precisionallometricmodelsaredevelopedfortemperateorplantationforests,leavingsignificantuncertaintieswhenappliedtonaturaltropicalorsemi-aridecosystems(Amalietal.,2024).Non-PermanenceRisksandPolicyAlignmentForforestcarbonprojectstoremainviableinglobalmarkets,theriskof"non-permanence"thepotentialforsequesteredcarbontobere-releasedduetowildfire,drought,orillegalloggingmustbeaddressed.Remotesensingallowsforthedetailedanalysisofmulti-yeardatasetstospotpatternsinprecipitationandtemperature,helpingdevelopersidentifywhetheralocationisexperiencingincreasingdroughtstressorshiftingtowardconditionsunsuitableforlong-termforestgrowth(Orlovetal.,2024).Additionally,thereisaneedtoalignsatellite-basedfluxestimateswithNationalGreenhouseGasInventories(NGHGIs)usedundertheParisAgreement.Currentbottom-upmodelsandatmospherictop-downobservationsoftenshowagapinestimatedanthropogeniclandusefluxes,primarilyduetodifferingdefinitionsof"natural"versus"anthropogenic"forestland.ReconcilingthesedifferencesusingEarthGRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page23observation-basedframeworkslikeGlobalForestWatchisessentialforbuildingglobalconfidenceinforestcarbonaccounting(Luetal.,2014).ConclusionsRemotesensing–basedassessmentofforestcarbonsequestrationandlandusechangedynamicshasbecomeanessentialapproachforunderstandingtheinteractionbetweenterrestrialecosystemsandclimatechange.Thereviewdemonstratesthatforestsfunctionasmajorcarbonsinks,yettheirsequestrationcapacityisincreasinglythreatenedbyurbanization,agriculturalexpansion,deforestation,droughts,andwildfiredisturbances.Landuseandlandcoverchangessignificantlyalterecosystemstructureandcontributetogreenhousegasemissions,therebyintensifyingglobalwarmingandecologicalinstability.Advancedremotesensingtechnologies,includingopticalsensors,LiDAR,andSARsystems,havegreatlyenhancedtheabilitytomonitorforeststructure,biomassdistribution,andcarbondynamicsacrossmultiplespatialandtemporalscales.Theintegrationofmulti-sensordatawithmachinelearninganddeeplearningalgorithmshasfurtherimprovedtheaccuracyofabove-groundbiomassestimationandcarbonstockmapping.Moreover,climatevariabilitystronglyinfluencesforestproductivity,evapotranspiration,andcarbonsequestrationpotential,creatingcomplexbiogeophysicalandbiogeochemicalfeedbackmechanismswithintheEarthsystem.EmergingtechnologiessuchasESABiomass,NISAR,andForestDigitalTwinsystemsrepresentamajoradvancementinglobalcarbonmonitoringandsustainableforestmanagement.Despitethesedevelopments,challengesrelatedtoscale,sensorlimitations,groundvalidation,andnon-permanencerisksremainsignificant.Thestudyconcludesthatcombiningremotesensing,artificialintelligence,ecologicalmodeling,andpolicy-drivenGRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page24conservationstrategiesisessentialforeffectivecarbonaccounting,climatemitigation,andlong-termecosystemsustainabilityunderchangingenvironmentalconditions.ReferencesCohen,W.B.,&Goward,S.N.(2004).Landsat'sroleinecologicalapplicationsofremotesensing.BioScience,54(6),535.https://doi.org/10.1641/0006-3568(2004)054[0535:lrieao]2.0.co;2H.Nguyen,T.,Jones,S.,Soto-Berelov,M.,&Haywood,A.(2019).Landsattime-seriesforestimatingforestabovegroundbiomassanditsdynamicsacrossspaceandtime:Areview.RemoteSensing,12(1),98.https://doi.org/10.3390/rs12010098Lu,D.,Chen,Q.,Wang,G.,Liu,L.,Li,G.,&Moran,E.(2014).Asurveyofremotesensing-basedabovegroundbiomassestimationmethodsinforestecosystems.InternationalJournalofDigitalEarth,9(1),63-105.https://doi.org/10.1080/17538947.2014.990526Lü,F.,Song,Y.,&Yan,X.(2023).EvaluatingcarbonsinkpotentialofforestecosystemsunderdifferentclimatechangescenariosinYunnan,SouthwestChina.RemoteSensing,15(5),1442.https://doi.org/10.3390/rs15051442Merrikhpour,H.,Badamfirooz,J.,Zarandian,A.,&Mousazadeh,R.(2025).Assessingcarbonstorageandeconomicvaluationofstockedcarbonfunctionbasedonlanduse/landcoverusingtheInVESTmodel.EnvironmentalHealthEngineeringandManagementJournal,12(4).GRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page25Wulder,M.A.,Loveland,T.R.,Roy,D.P.,Crawford,C.J.,Masek,J.G.,Woodcock,C.E.,...&Lymburner,L.(2019).CurrentstatusofLandsatprogram,science,andapplications.RemoteSensingofEnvironment,225,127-147.https://doi.org/10.1016/j.rse.2019.02.015Xu,L.,Saatchi,S.S.,Yang,Y.,Yu,Y.,Pongratz,J.,Bloom,A.A.,...&Schimel,D.(2021).Changesinglobalterrestriallivebiomassoverthe21stcentury.ScienceAdvances,7(27),eabe9829.https://doi.org/10.1126/sciadv.abe9829Xu,X.(2025).Biocharapplicationforsoilcarbonsequestrationandgreenhousegasmitigationinforestecosystems:AbibliometricanalysisusingCiteSpace.MDPI.Ashaolu,E.D.,Olorunfemi,J.F.,&Ifabiyi,I.P.(2019).Assessingthespatio-temporalpatternoflanduseandlandcoverchangesinOsundrainagebasin,Nigeria.JournalofEnvironmentalGeography,12(1-2),41–50.https://doi.org/10.2478/jengeo-2019-0005Kocaman,B.,&Ağaçcıoğlu,H.(2025).Assessmentoftheimpactoflanduse/landcoverchangesoncarbonemissionsusingremotesensinganddeeplearning:AcasestudyoftheKağıthanebasin.Sustainability,17(23),10690.https://doi.org/10.3390/su172310690Yang,H.R.(2025).Exploringthecomplexrelationshipsbetweeninfluentialfactorsofurbanlanddevelopmentpatternsandurbanthermalenvironment:AstudyondowntownShanghai.Sustainability,17(19),8547.https://doi.org/10.3390/su17198547GRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page26Yazar,M.,DalogluCetinkaya,I.,Iban,M.C.,&Bilgilioglu,S.S.(2023).Thegreendivideandheatexposure:UrbantransformationprojectsinIstanbul.FrontiersinEnvironmentalScience,11,Article1265332.https://doi.org/10.3389/fenvs.2023.1265332Alongi,D.M.(2014).Carboncyclingandstorageinmangroveforests.AnnualReviewofMarineScience,6,195-219.Aratrakorn,S.,Thunhikorn,S.,&Donald,P.F.(2006).ChangesinbirdcommunitiesfollowingconversionoflowlandforesttooilpalmandrubberplantationsinsouthernThailand.BirdConservationInternational,16(1),71-82.Donato,D.C.,Kauffman,J.B.,Murdiyarso,D.,Kurnianto,S.,Stidham,M.,&Kanninen,M.(2011).Mangrovesamongthemostcarbon-richforestsinthetropics.NatureGeoscience,4(5),293-297.Murphy,D.J.,Hall,J.,&Jintana,V.(2020).LandusechangeandcarbonlossesinSoutheastAsiantropicalpeatlands.WetlandsEcologyandManagement,28(3),457-472.Pendleton,L.,Donato,D.C.,Murray,B.C.,Crooks,S.,Jenkins,W.A.,Sifleet,S.,...&Baldera,A.(2012).Estimatingglobal“bluecarbon”emissionsfromconversionanddegradationofvegetatedcoastalecosystems.PLoSONE,7(9),e43542.Sharp,R.,Tallis,H.T.,Ricketts,T.,Guerry,A.D.,Wood,S.A.,Chaplin-Kramer,R.,...&Vogl,A.L.(2018).InVEST3.7.0user’sguide.TheNaturalCapitalProject,StanfordUniversity.GRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page27Fayad,I.,Baghdadi,N.,Guitet,S.,Bailly,J.-S.,Hérault,B.,Gond,V.,ElHajj,M.,&TongMinh,D.H.(2016).AbovegroundbiomassmappinginFrenchGuianabycombiningremotesensing,forestinventoriesandenvironmentaldata.InternationalJournalofAppliedEarthObservationandGeoinformation,52,502–514.https://doi.org/10.1016/j.jag.2016.07.015Jagadish,B.,Behera,M.D.,Prakash,A.J.,Paramanik,S.,Ghosh,S.M.,Patnaik,C.,&Das,A.(2024).Indicatingsaturationlimitsofmulti-sensorsatellitedatainestimatingabovegroundbiomassofamangroveforest.JournaloftheIndianSocietyofRemoteSensing,52,2483–2500.https://doi.org/10.1007/s12524-024-01968-1Li,C.,Zhou,L.,&Xu,W.(2021).EstimatingabovegroundbiomassusingSentinel-2MSIdataandensemblealgorithmsforgrasslandintheShengjinLakeWetland,China.RemoteSensing,13(8),1595.https://doi.org/10.3390/rs13081595Sinha,S.,Mohan,Brun,S.,Das,A.K.,Sharma,L.K.,Jeganathan,C.,Santra,A.,SantraMitra,S.,&Nathawat,M.S.(2019).Multi-sensorapproachintegratingopticalandmulti-frequencysyntheticapertureradarforcarbonstockestimationoveratropicaldeciduousforestinIndia.CarbonManagement,11(1),39–55.https://doi.org/10.1080/17583004.2019.1686931Choi,C.(2024).CombiningTanDEM-XInterferometricSARandGEDILidarMeasurementsforImprovingForestHeight,StructureandBiomassEstimates.DeutschesZentrumfürLuft-undRaumfahrt(DLR).GRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page28HoTongMinh,D.,LeToan,T.,Rocca,F.,Tebaldini,S.,Villard,L.,Réjou-Méchain,M.,Phillips,O.L.,Feldpausch,T.R.,Dubois-Fernandez,P.,Scipal,K.,&Chave,J.(2016).SARtomographyfortheretrievalofforestbiomassandheight:Cross-validationattwotropicalforestsitesinFrenchGuiana.RemoteSensingofEnvironment,175,138-147.https://doi.org/10.1016/j.rse.2015.12.037LeToan,T.,Villard,L.,HoTongMinh,D.,Doblas,J.,Mermoz,S.,Ferro-Famil,L.,Koleck,T.,Bouvet,A.,Planells,M.,&Polidori,L.(2024).TacklinghighbiomassintropicalforeststhroughtheBIOMASSmission.TheInternationalArchivesofthePhotogrammetry,RemoteSensingandSpatialInformationSciences,XLVIII-3-2024,287-293.https://doi.org/10.5194/isprs-archives-xlviii-3-2024-287-2024Xu,Y.,Lu,Z.,Bürgmann,R.,Hensley,S.,Fielding,E.,&Kim,J.(2023).P-bandSARforgrounddeformationsurveying:Advantagesandchallenges.RemoteSensingofEnvironment,287,113474.https://doi.org/10.1016/j.rse.2023.113474Ali,J.(2025).Remotesensingandintegrationofmachinelearningalgorithmsforabove-groundbiomassestimationinLarixprincipis-rupprechtiiMayrplantations:acasestudyusingSentinel-2andLandsat-9datainnorthernChina.FrontiersinEnvironmentalScience.Iheaturu,C.J.(2026).Remotesensingoftropicalforestrecovery.RemoteSensingofEnvironment,335,115257.Jiao,Y.,Wang,B.,Yao,X.,Wang,S.,Chi,T.,&Meng,Y.(2023).ForestemissionsreductionassessmentusingopticalsatelliteimageryandspaceLiDARfusionforGRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15G.2094Page29carbonstockestimation.RemoteSensing,15(5),1410.https://doi.org/10.3390/rs15051410Papucci,E.(2026).Reviewofforestbiomassassessmentsbasedonremotesensingrevealsprogressinmethodologicalqualitybutmajorchallengesremain.Forestry.Santos,D.T.D.(2025).Multi-sensordatafusionforoperationalmonitoringofdeforestationandforestcarbonstocksintropicalecosystems.InternationalJournalofScience,Architecture,Technology,andEnvironment,2(12).Tian,L.,Wu,X.,Tao,Y.,Li,M.,Qian,C.,Liao,L.,&Fu,W.(2023).Reviewofremotesensing-basedmethodsforforestabovegroundbiomassestimation:Progress,challenges,andprospects.Forests,14(6),1086.https://doi.org/10.3390/f14061086Xu,X.(2025).EstimationofforestabovegroundbiomassusingSentinel-1/2synergizedwithextrapolatedparametersfromLiDARdataandanalysisofitsecologicaldrivingfactors.RemoteSensing.GRJNST,Volume:04-Issue3(2026)/ISSNP:2790-7643ArticleID:2094https://doi.org/10.53762/grjnst.04.03.15