Page 220 - 《水资源与水工程学报》2025年第2期
P. 220
1
2 6 & ' ( ) & * + , - 2025 $
»¼óV; n, ¡üoQÌ>¡ 4 ûí 2021,385:114863.
%»¼È-, È-í%¡5üoQT [9]FATHOLOLOUMIS,VAEZIAR,ALAVIPANAHSK,et
yXYÈZ。Î$H>d, A_ al.Effectofmultitemporalsatelliteimagesonsoilmoisture
predictionusingadigitalsoilmappingapproach [J].Geo
í%T, ÒÓ。æ¯ÈZo~QT
derma ,2021,385:114901.
Þg, ÃV¡o~Q»¼°OTQ©
[10]WEIXin,ZHANGLulu,YANGHaoqing,etal.Machine
½, ÀóVÉ¡NTBXí%, K¾u!NªT
learningforporewaterpressuretimeseriesprediction:
()RÝ。
applicationofrecurrentneuralnetworks[J].Geoscience
6 f i Frontiers,2021,12(1):453467.
[11]HEDDAM S.3newformulationforpredictingsoilmois
(1) 4WpPo~QTP>?+e7æL turecontentusingonlysoiltemperatureaspredictor:mul
[.T, ´+$g@leÕ, tivariateadaptiveregressionsplinesversusrandomforest,
æ
ÆÑg、 6Ðg、 jaJÔI.。 multilayerperceptronneuralnetwork,M5Tree,andmulti
(2) Y}gT 4 ûo(20、40、60、80cm) plelinearregression [J].WaterEngineeringModelingand
MathematicTools ,2021:4562.DOI:10.1016/B9780
BÌ>Üæ 4 û÷Bg(1、12、24、48h),
1282064417.00027X.
(RÁæÈZí%ÞgJ_ÒÓÈç
[12]AHMEDAAM,DEORC,GHAHRAMANIA,etal.
NLVTyo。
LSTMintegratedwithBoruta-randomforestoptimiserfor
( 3) Fa5| éÕÌ>óVT LSTM-1
soilmoistureestimationunderRCP4.5andRCP8.5global
í%, BX LSTMí%º¡oU 12、24、48h T warmingscenarios [J].StochasticEnvironmentalResearch
2
!,R xÈ 2.42%、5.71%、12.26%; (í% andRiskAssessment,2021,35(9):18511881.
Ï:¯õÅÉ·Ûv4WpP=âüoQ [13]ZHANGFangfang,WUShiwen,LIUJie,etal.Predic
。 tingsoilmoisturecontentoverpartiallyvegetationcovered
surfacesfrom hyperspectraldatawithdeeplearning [J].
ÕÖKL: SoilScienceSocietyofAmericaJournal ,2021,85(4):
[1]îö3, _!d, 1ÿ, Q. 4WznÝûÂPIMN 9891001.
OPKLTIM,> .[J]. F[#$Z,2018,34 [14]PONKINAE,ILLIGERP,KROTOVAO,etal.DoARMA
(11):96104. modelsprovidebettergapfillingintimeseriesofsoiltempera
[2]$ `, é ¦, aêf, Q. }vÊbSo~I tureandsoilmoisture ?thecaseofarablelandintheKulunda
Q©[J]. o~Z,2021,58(6):14231435. Steppe ,Russia[J].Land,2021,10(6):579.
[3]í |, j !, ÕH, Q. lnh!qFGo~I(. [15]CARRANZAC,NOLETC,PEZIJM,etal.Rootzone
«¦PQ©[J]. o~,2019,51(2):390398. soilmoistureestimationwithrandomforest[J].Journalof
[4]î×2, 1^., ç#Ì, Q. o~IÑHQ©.( Hydrology ,2021,593:125840.
)Ûe[J]. o~Z,2019,56(1):2335. [16]HANH,CHOIC,KIMJ,etal.Multipledepthsoilmois
[5]ZHANGQiuru,SHILiangsheng,HOLZMANM,etal.A tureestimatesusingartificialneuralnetworkandlongshort
dynamicdatadrivenmethodfordealingwithmodelstruc termmemorymodels [J].Water,2021,13(18):2584.
turalerrorinsoilmoisturedataassimilation [J].Advances [17]ADEYEMIO,GROVEI,PEETSS,etal.Dynamicneural
inWaterResources,2019,132:103407. networkmodellingofsoilmoisturecontentforpredictiveirri
[6]ABIOYEEA,ABIDINMSZ,MAHMUDMSA,etal.IoT gationscheduling [J].Sensors,2018,18(10):3408.
basedmonitoringanddatadrivenmodellingofdripirrigation [18]FILIPOVI N,BRDARS,MIMI G,etal.Regionalsoil
systemformustardleafcultivationexperiment[J].Information moisturepredictionsystembasedonlongshorttermMemory
ProcessinginAgriculture ,2021,8(2):270283. network[J].BiosystemsEngineering,2022,213:3038.
[7]ADABH,MORBIDELLIR,SALTALIPPIC,etal.Ma [19]ELSAADANIM,HABIBE,ABDELHAMEEDAM,et
chinelearningtoestimatesurfacesoilmoisturefromremote al.Assessmentofaspatiotemporaldeeplearningapproach
sensingdata [J].Water,2020,12(11):3223. forsoilmoisturepredictionandfillingthegapsinbetween
[8]TANEJAP,VASAVAHK,DAGGUPATIP,etal.Multi soilmoistureobservations [J].FrontiersinArtificialIn
algorithmcomparisontopredictsoilorganicmatterandsoil telligence ,2021,4:636234.
moisturecontentfrom cellphoneimages [J].Geoderma, [20]Ò!G, /Gï, a ò, Q. cñëRI¬¬èºB