Page 22 - 《水资源与水工程学报》2025年第1期
P. 22
!36 "!1 # & ' ( ) & * + , - Vol.36No.1
2025 $ 2 % JournalofWaterResources&WaterEngineering Feb.,2025
DOI:10.11705/j.issn.1672-643X.2025.01.03
8_ ARIMA56P GMS fg00
1,2
1,2
1,2
506 , 789 , : ; , <=> 1,2
(1. 4!4a0( pU)4BZghchk
¼QÔ:?@pqr, pU 100083;
2. 4!4a0( pU)Z]^chk0G, pU 100083)
> ?: 24BZíîiº¦ðñôÎ4BZZ¦Ï, 4²³cÖ*+{ÊZc8íê。
;Ì}=>,
% ARIMAiºðñÊZc8ÏÎ@<íê, À®/ GMS 4BZJiº, ijXEJK4BZZ¦ú¼LD,
#}.Æ{4BZZ¦ðñ。eLµ¶XEJK 2000—2020 T{!ø\íê, s ARIMAiºðñ
2021 T{ÊZc8, °ðñ®«Àç GMS iº, Ù94BZZ¦ijpq。®«S¤:GMS iºXEJK
4BZZ¦{ij{«Ý, aÜí NSEîµ2¦ 0.71~0.96 |Î,RMSEî3¦ 0.05~0.45m|Î, g/ß~
é;ARIMAiºø\íê{ðñß~é, 8íê{ðñ{«»ÊZ; ®/ ARIMAiºc GMS iº{A
B¦ß~>ѬS0[Ý, 5K4BZ]^¯#0éê。AB#{§¯2iº
ôÎíêéÞÑà{ª³Ñ, v55Kðñ4BZZ¦q1。
1@A: 4BZZ¦; ÊZc8íê; ÏÎ@<µ¶ ARIMA;GMS; XEJK
BC<DE:TV213.4;P333 FGHIJ:A FKLE:1672643X(2025)01001811
SimulationsofgroundwaterlevelsintheYangRiverBasin
CombiningtheARIMAmethodwithGMS
1,2
1,2
1,2
1,2
SUNFubao ,TONGJuxiu ,LIANGChang ,TONGJinwei
(1.KeyLaboratoryofGroundwaterCirculationandEnvironmentalEvolution,MinistrayofEducation,ChinaUniversityofGeosciences(Beijing),
Beijing100083,China;2.SchoolofWaterResourcesandEnvironment,ChinaUniversityofGeosciences(Beijing),Beijing100083,China)
Abstract:Conventionalgroundwaternumericalmodelsoftenfacelimitationsinpredictingfuturegroundw
aterlevelsduetothedifficultyofobtainingprecipitationandevaporationdata.Toaddressthisissue,this
studyputsforwardanimprovedgroundwaterlevelpredictionmethod.Weemploytheautoregressiveinte
gratedmovingaverage (ARIMA)modeltopredicttimeseriesdataforprecipitationandevaporation,then
integratethepredictionresultswiththegroundwaterflowmodelofgroundwatermodelingsystem(GMS)
tosimulatechangesingroundwaterlevelsintheYangRiverBasin.Thisnewmethodisthenappliedto
analyzethehistoricalmeteorologicaldataoftheYangRiverBasinfrom2000to2020.TheARIMAmodel
isusedtopredictprecipitationandevaporationin2021,andthepredicteddataareinputintotheGMSto
conductgroundwaterlevelsimulationexperiments.TheresultsindicatethattheGMSperformswellin
simulatinggroundwaterlevelsintheYangRiverBasin,withmostNSEvaluesrangingfrom0.71to0.96,
andRMSEvaluesbetween0.05and0.45m,demonstratinghighoverallaccuracy.TheARIMAmodel
exhibitsstrongpredictiveperformanceformeteorologicaldata ,withpredictionaccuracyforevaporation
outperformingthatforprecipitation.ThecombinedapproachoftheARIMAandGMSmodelsdemonstrates
highaccuracyandapplicability ,whichcanprovidetechnicalsupportforregionalgroundwaterresource
management.Theproposedmethodcanovercomethelimitationsofdependenceonfuturedataavailability
facedbyconventionalmodels,offeringaviablereferenceforgroundwaterlevelpredictioninsimilarre
!"#$:20240711; %$:20240918
'()*: !"wx#0%»r¬(3(42072271); !õà(3(HBC327)
+,-.: xr(1999—), Ç£N, ÈPABS, VWXY4BZíîijAB。
qr+,: m(1981—), ±pÍN, OP, QR, OPSTU, VWXY4BZJ%\y{íîij、 p0
\íêÒ¼AB。