崔遠(yuǎn)來,吳 迪,王士武,溫進(jìn)化,王賀龍
?
基于改進(jìn)SWAT模型的南方多水源灌區(qū)灌溉用水量模擬分析
崔遠(yuǎn)來1,吳 迪1,王士武2,溫進(jìn)化2,王賀龍2
(1. 武漢大學(xué)水資源與水電工程科學(xué)國(guó)家重點(diǎn)實(shí)驗(yàn)室,武漢 430072;2. 浙江省水利河口研究院,杭州 310020)
為提出一種合理有效的南方多水源灌區(qū)灌溉用水量模擬統(tǒng)計(jì)方法,該文針對(duì)南方多水源灌區(qū)水循環(huán)及灌溉取水特點(diǎn)對(duì)SWAT模型進(jìn)行改進(jìn),尤其添加了多水源自動(dòng)灌溉模塊用于模擬作物不同水源類型的灌水量,并統(tǒng)計(jì)推求灌區(qū)灌溉用水量。以浙江省浦江縣通濟(jì)橋水庫灌區(qū)為例,應(yīng)用改進(jìn)SWAT構(gòu)建灌區(qū)水循環(huán)模型,利用灌區(qū)出口實(shí)測(cè)月徑流數(shù)據(jù)及4條干渠渠首監(jiān)測(cè)的灌水量數(shù)據(jù)校正及驗(yàn)證模型,其中月徑流在驗(yàn)證期的Nash-Suttclife效率系數(shù)為0.89,干渠灌溉用水量模擬值與觀測(cè)值相對(duì)誤差的絕對(duì)值最大不超過20%,表明改進(jìn)SWAT模型具有良好的模擬效果。利用所建模型模擬分析通濟(jì)橋水庫灌區(qū)長(zhǎng)系列灌溉用水量,結(jié)果顯示灌區(qū)灌溉用水量呈現(xiàn)豐水年小、干旱年大的變化規(guī)律;除監(jiān)測(cè)的骨干水源通濟(jì)橋水庫及浦陽江取水以外,灌溉用水量的41.40%來源于灌區(qū)內(nèi)部的河道、塘堰及小型水庫,說明只監(jiān)測(cè)干渠渠首灌水量無法統(tǒng)計(jì)整個(gè)灌區(qū)灌溉用水量;隨著灌區(qū)節(jié)水改造投入,灌區(qū)灌溉水利用系數(shù)提高,其灌溉用水量減少?;诟倪M(jìn)SWAT模型進(jìn)行多水源灌區(qū)灌溉用水量模擬為灌區(qū)灌溉用水量統(tǒng)計(jì)分析提供了一種有效的方法。
灌溉;模型;水庫;多水源灌區(qū);改進(jìn)SWAT;不同水源類型
為提升中國(guó)水資源公報(bào)質(zhì)量和支撐最嚴(yán)格水資源管理制度考核工作,落實(shí)最嚴(yán)格水資源管理制度,開展水資源監(jiān)控及用水總量統(tǒng)計(jì)十分必要。在中國(guó),農(nóng)業(yè)用水量占總用水量的60%以上,其中,灌溉用水占農(nóng)業(yè)用水量的90%以上,是用水總量統(tǒng)計(jì)的重點(diǎn)。農(nóng)業(yè)用水指農(nóng)田灌溉用水、林果地灌溉用水、草地灌溉用水、漁塘補(bǔ)水和畜禽用水,其中灌溉用水量是指從各類水源取來用于灌區(qū)作物灌溉的水量之和。中國(guó)南方多水源灌區(qū)中存在多種水源,不同水源之間存在不同程度的互聯(lián)互通,屬于庫、塘、渠結(jié)合的“長(zhǎng)藤結(jié)瓜”灌溉系統(tǒng)[1-2]。由于水源種類多、分布復(fù)雜、取水的隨機(jī)性強(qiáng)、且存在重復(fù)利用,有時(shí)難以有效區(qū)分不同水源的灌水量,水量計(jì)量工作不僅量大,且存在相當(dāng)難度,并且只計(jì)量灌區(qū)渠首的取水量也不能代表整個(gè)灌區(qū)的灌溉用水量[3]。此外,灌區(qū)各個(gè)分區(qū)之間的土地利用、水源類型等存在差異性,采用典型調(diào)查或現(xiàn)行灌溉定額與實(shí)際灌溉面積數(shù)據(jù)進(jìn)行框算得到的灌溉用水量精度需要進(jìn)一步提高[4]。因此,尋求一種合理有效的灌溉用水量模擬統(tǒng)計(jì)方法十分必要。推求灌溉用水量的基礎(chǔ)是水量平衡原理[5],鑒于多水源灌區(qū)的空間異質(zhì)性,分布式水文模型是獲得水量平衡要素的一個(gè)有效工具。其中,SWAT(soil and water assessment tool)模型是一個(gè)具有物理基礎(chǔ)的分布式水文模型,具有自動(dòng)灌溉模塊可用于推求作物灌溉用水量[6-8]。其不僅能夠模擬日、月、年尺度流域水循環(huán)過程,還可用于人類活動(dòng)對(duì)水循環(huán)影響的分析研究,在灌區(qū)水循環(huán)等方面已有諸多應(yīng)用[9-14]。
SWAT模型是適用于自然流域的分布式水文模型[7]。然而,灌區(qū)水文循環(huán)過程相對(duì)于自然流域更加復(fù)雜多變,不僅受自然流域水平衡要素的影響,而且受人類活動(dòng)的影響[15-16]。因此,為更好地將SWAT模型應(yīng)用于灌區(qū)水循環(huán)過程的模擬,國(guó)內(nèi)外不少學(xué)者對(duì)其進(jìn)行了改進(jìn)。代俊峰等[17]針對(duì)中國(guó)南方丘陵水稻灌區(qū)的水文特點(diǎn),改進(jìn)SWAT模型的灌溉水分運(yùn)動(dòng)模塊、稻田水分循環(huán)模塊、稻田水量平衡各要素以及產(chǎn)量模擬的計(jì)算方法,增加了渠系滲漏模擬模塊及其對(duì)地下水的補(bǔ)給作用、塘堰的灌溉模塊等,并將其應(yīng)用于灌區(qū)水管理研究[18];陳強(qiáng)等[19]將PSO算法代替SWAT模型原有的SCE自動(dòng)率定算法,構(gòu)建新的SWAT模型參數(shù)自動(dòng)率定模塊,并將水資源配置模型的農(nóng)業(yè)灌溉用水輸入到改進(jìn)后的SWAT模型中,實(shí)現(xiàn)2個(gè)模型的松散耦合;Xie等[20]在稻田蒸發(fā)蒸騰、控制灌溉排水、塘堰實(shí)時(shí)灌溉等方面對(duì)SWAT模型進(jìn)行了改進(jìn);此外,還有其他學(xué)者對(duì)SWAT模型也做了相應(yīng)的改進(jìn)[21-26]。然而,上述改進(jìn)中均未涉及作物的多種水源聯(lián)合灌溉,同時(shí),由于水源選擇單一,SWAT模型未能模擬作物的多水源灌溉。
綜上,本文以SWAT模型為基礎(chǔ)工具,針對(duì)南方多水源灌區(qū)水文循環(huán)及作物灌溉特點(diǎn),整合代俊峰等[17]及Xie等[20]對(duì)SWAT模型稻田水循環(huán)模塊的改進(jìn),同時(shí)提出一種多水源自動(dòng)灌溉模塊并將其添加到SWAT模型,從而得到適用于南方多水源灌區(qū)的改進(jìn)SWAT模型。以浙江省浦江縣通濟(jì)橋水庫灌區(qū)為例對(duì)改進(jìn)SWAT模型進(jìn)行率定及驗(yàn)證,利用改進(jìn)SWAT模型模擬分析灌區(qū)不同水源的灌溉用水量及其變化規(guī)律,以期為灌區(qū)灌溉用水量模擬統(tǒng)計(jì)提供有效方法。
通濟(jì)橋水庫灌區(qū)位于浙江省浦江縣南部浦陽江盆地。灌區(qū)的灌溉水源以通濟(jì)橋水庫、浦陽江為骨干水源,灌區(qū)內(nèi)小型水庫(金獅嶺水庫、里塢水庫、岳塘水庫)、塘堰及河流為輔,骨干水源以4條干渠貫穿整個(gè)灌區(qū),構(gòu)成典型的南方多水源“長(zhǎng)藤結(jié)瓜”灌溉系統(tǒng),如圖1a所示。
由圖1a所示,從北至南依次為北干渠、中干渠、南干渠72線、南干渠80線,其中南干渠72線的水源為浦陽江,其余3條水源為通濟(jì)橋水庫。由于通濟(jì)橋水庫灌區(qū)并非閉合流域,因此選擇的建模區(qū)域較灌區(qū)范圍稍大,所選區(qū)域的土地利用類型主要為林地、城鎮(zhèn)、水稻田、葡萄地、水域及旱地,如圖1b所示。灌區(qū)屬亞熱帶季風(fēng)氣候,多年平均氣溫16.6 ℃,多年平均降雨量1 466 mm,年內(nèi)降水分布不均,多年平均水面蒸發(fā)量907 mm。灌區(qū)主要種植水稻、小麥等糧食作物及葡萄、草莓等經(jīng)濟(jì)作物,其中水稻與葡萄為主要灌溉作物,水稻種植面積為1 693.33 hm2,葡萄種植面積為1 893.33 hm2,且每年的6-9月為灌溉季節(jié)。
SWAT模型將研究區(qū)劃分為多個(gè)子流域,進(jìn)而將子流域劃分為多個(gè)水文響應(yīng)單元(HRU),以HRU為最小水文響應(yīng)單元進(jìn)行模擬。針對(duì)南方多水源灌區(qū)作物灌溉特點(diǎn)提出的多水源自動(dòng)灌溉模塊的結(jié)構(gòu)如圖2所示。
注:HRU為最小水文響應(yīng)單元。
結(jié)合圖2對(duì)多水源自動(dòng)灌溉模塊的說明如下:
1)原SWAT中灌溉用水量一般預(yù)先確定作為輸入條件或利用單一水源的自動(dòng)灌溉模擬[7],改進(jìn)SWAT可以根據(jù)作物適宜田間土壤含水率或稻田水層深度控制標(biāo)準(zhǔn),自動(dòng)觸發(fā)灌溉并確定每次田間凈灌水量,同時(shí)從不同水源取水灌溉,從而在水循環(huán)模型中同步實(shí)現(xiàn)不同水源類型灌溉用水量的自動(dòng)模擬。
3)圖2中水源1、水源2等為HRU的灌溉水源及順序,即在模型運(yùn)行之前需要為HRU指定灌溉水源及順序。以子流域?yàn)閷?duì)象進(jìn)行指定,同一子流域內(nèi)HRU的灌溉水源及順序一致。對(duì)于南方多水源灌區(qū),每一個(gè)子流域可能存在的水源有子流域內(nèi)部河道、塘堰、中小型水庫、子流域外部河道及大型水庫。通常設(shè)定第一水源為子流域內(nèi)部河道,第二水源為子流域內(nèi)部塘堰,并在SWAT模型軟件界面上指定,而后面的中小型水庫、子流域外部河道及大型水庫則利用新增的輸入文件進(jìn)行指定。
4)水源灌溉可用水量是指可從水源取用的最大毛水量考慮灌溉水利用系數(shù)后折算成田間的凈水量。對(duì)于不同類型的水源,由于水源之間存在差異其計(jì)算有所不同,具體如下:
對(duì)于河道
對(duì)于塘堰
對(duì)于水庫
5)模擬多水源灌溉時(shí),從第一個(gè)水源開始,取第一水源實(shí)際凈灌水量為該水源灌溉可用水量與本次灌溉需水量的較小值,當(dāng)?shù)谝凰磳?shí)際凈灌水量小于灌溉需水量則計(jì)算缺水量,如圖2所示,繼而下一個(gè)水源取水灌溉,直至灌溉滿足要求或到最后一個(gè)水源。每個(gè)水源的實(shí)際凈灌水量除以灌溉水利用系數(shù)得到該水源毛灌溉用水量(簡(jiǎn)稱灌溉用水量)。
除多水源自動(dòng)灌溉模塊的添加外,針對(duì)“長(zhǎng)藤結(jié)瓜”灌溉系統(tǒng),對(duì)改進(jìn)SWAT模型增加了骨干水源對(duì)灌區(qū)內(nèi)部塘堰的補(bǔ)水功能,即灌溉季節(jié)若出現(xiàn)塘堰干涸情況,骨干水源將通過渠道對(duì)其進(jìn)行補(bǔ)水,以此來反映農(nóng)民干預(yù)調(diào)水的情況。此外,改進(jìn)還包括稻田水平衡要素計(jì)算的改進(jìn)及渠系水滲漏計(jì)算的添加等,具體內(nèi)容見參考文獻(xiàn)[17]和[20]。對(duì)于改進(jìn)SWAT模型的使用,模型框架的搭建在SWAT軟件界面上進(jìn)行,與原模型基本一致,此外僅需創(chuàng)建文本文件輸入水稻控制水深、灌溉水源及順序等參數(shù)即可。
1.3.1 研究區(qū)模型構(gòu)建
1.3.2 模型校正及驗(yàn)證
模型構(gòu)建之后需進(jìn)一步對(duì)其進(jìn)行校正及驗(yàn)證。參考前人研究中選擇的徑流敏感參數(shù)[27-29],結(jié)合研究區(qū)的特點(diǎn)選取參數(shù),采用SWATCUP(SWAT Calibration and Uncertainty Programs)軟件中的SUFI_2算法進(jìn)行參數(shù)敏感性分析[30],從而選擇10個(gè)參數(shù)作為徑流的敏感參數(shù)。利用研究區(qū)出口(見圖1a)1995-2007年的實(shí)測(cè)月徑流數(shù)據(jù)對(duì)敏感參數(shù)進(jìn)行率定從而校正徑流模擬過程,經(jīng)率定確定敏感參數(shù)的取值后,利用研究區(qū)出口2008-2015年的實(shí)測(cè)月徑流數(shù)據(jù)對(duì)徑流模擬過程進(jìn)行驗(yàn)證,并選取相對(duì)誤差(RE)、決定系數(shù)(2)和Nash-Suttclife效率系數(shù)(NS)來評(píng)估模型效率[31]。
經(jīng)過反復(fù)調(diào)整后確定徑流敏感參數(shù)取值,計(jì)算得到率定期徑流的評(píng)價(jià)指標(biāo)RE為11.83%、2為0.88以及NS為0.85,由此可知,通過參數(shù)率定,研究區(qū)出口徑流量模擬結(jié)果與實(shí)測(cè)值吻合性較好。驗(yàn)證期徑流的評(píng)價(jià)指標(biāo)RE為9.05%、2為0.90以及NS為0.89,表明率定后的模型性能較為穩(wěn)定。研究區(qū)出口徑流模擬值與實(shí)測(cè)值的月際動(dòng)態(tài)如圖3所示,圖3顯示模型模擬的月徑流過程與實(shí)測(cè)結(jié)果總體變化一致,說明模型適用于通濟(jì)橋水庫灌區(qū)的水量平衡模擬及分析。
表1 監(jiān)測(cè)點(diǎn)灌溉用水量模擬值與觀測(cè)值對(duì)比
利用改進(jìn)SWAT模型模擬通濟(jì)橋水庫灌區(qū)1995—2017年的水循環(huán)過程,特別是其中的多水源自動(dòng)灌溉模塊模擬灌區(qū)水稻及葡萄的灌溉用水,基于灌區(qū)范圍內(nèi)子流域輸出的灌溉模擬結(jié)果統(tǒng)計(jì)得到通濟(jì)橋水庫灌區(qū)不同水源類型的灌溉用水量,匯總得到整個(gè)灌區(qū)的灌溉用水量,其中不同水源類型包括子流域內(nèi)部河道、塘堰、小型水庫、浦陽江以及通濟(jì)橋水庫。將1995-2017年灌溉季節(jié)(6-9月)的降雨量進(jìn)行排頻,豐水年(25%)、平水年(50%)、干旱年(75%)以及特旱年(95%)的灌溉用水量模擬結(jié)果如圖4所示。圖4表明,灌區(qū)灌溉用水量基本呈現(xiàn)豐水年小、干旱年大的變化規(guī)律,分析其原因是豐水年降水較多,使得稻田和旱地出現(xiàn)缺水的情況較少,從而整個(gè)生育期內(nèi)的灌溉需水量就較小,因此灌區(qū)豐水年的灌溉用水量較小,而干旱年則相反。但圖中顯示干旱年的灌溉用水量卻大于特旱年,其主要原因是特旱年降雨量過少,導(dǎo)致河道及塘堰中蓄積的水量較少,也即無水可取,使子流域內(nèi)部河道、塘堰以及浦陽江的灌水量偏少,因此其整體灌溉用水量會(huì)少于干旱年。從另一個(gè)角度分析,表明特旱年產(chǎn)生了水源缺水。
圖4 不同水平年灌溉用水量模擬值
不同水源類型的灌溉供水比例是指各類水源用于作物灌溉的水量占總的灌溉用水量的百分比。利用模型模擬得到1995-2017年灌區(qū)不同水源類型的灌溉用水量及供水比例的多年平均值如表2所示。
表2 不同水源類型灌溉用水量模擬值及供水比例的多年平均值
表2表明,通濟(jì)橋水庫灌區(qū)的作物灌溉主要由通濟(jì)橋水庫供水,其次是子流域內(nèi)部塘堰、河道以及浦陽江,小型水庫供水最少。除骨干水源通濟(jì)橋水庫及浦陽江以外,灌區(qū)灌溉用水有41.40%來源于灌區(qū)內(nèi)部的河道、塘堰及小型水庫,因此若采用渠首計(jì)量的措施統(tǒng)計(jì)灌區(qū)灌溉用水量,則會(huì)產(chǎn)生較大的誤差,而其他水源供水由于點(diǎn)多量大且具有隨機(jī)性,又不便于計(jì)量,給灌溉用水量統(tǒng)計(jì)帶來困難。利用本文的改進(jìn)SWAT模型則能模擬統(tǒng)計(jì)出灌區(qū)每一年不同水源類型的灌溉用水量,從而較為合理的推求出灌區(qū)總灌溉用水量。
對(duì)于灌區(qū)節(jié)水改造,其最終的表現(xiàn)形式體現(xiàn)為對(duì)灌溉水利用系數(shù)的影響,改進(jìn)SWAT模型將灌溉水利用系數(shù)作為輸入的變量,因此,通過變動(dòng)灌溉水利用系數(shù)則可模擬節(jié)水改造對(duì)灌溉用水量的影響。通濟(jì)橋水庫灌區(qū)現(xiàn)狀條件下的灌溉水利用系數(shù)為0.54,在此基礎(chǔ)上增加0.05來模擬節(jié)水改造后的灌溉用水量,見表3。
表3 多年平均情況下節(jié)水改造前后的灌溉用水量模擬值
表3表明,灌溉水利用系數(shù)增加0.05,通濟(jì)橋水庫灌區(qū)不同水源類型的灌溉用水量均減少,且主要體現(xiàn)在骨干水源通濟(jì)橋水庫的灌溉用水量減少,其主要原因是灌溉水利用系數(shù)提高后,骨干水源的輸配水損失減少更多,其他水源由于輸水線路短,減少幅度較少。
此外,對(duì)于水稻種植面積較多的灌區(qū),除采取變動(dòng)灌溉水利用系數(shù)的方法模擬不同節(jié)水情景的影響之外,改進(jìn)SWAT模型還可輸入水稻不同灌溉模式的3個(gè)控制水深[5],從而模擬采用水稻不同節(jié)水灌溉模式的影響,因此該模型可用于分析節(jié)水改造對(duì)灌溉用水量及灌區(qū)水循環(huán)的影響。
1)針對(duì)南方多水源灌區(qū)水循環(huán)及灌溉取水特點(diǎn)對(duì)SWAT模型進(jìn)行改進(jìn),尤其添加了多水源自動(dòng)灌溉模塊用于模擬作物不同水源類型的灌水量?;诟倪M(jìn)SWAT模型進(jìn)行多水源灌區(qū)灌溉用水量模擬為灌區(qū)灌溉用水量統(tǒng)計(jì)分析提供了一種有效的方法。
2)改進(jìn)SWAT模型構(gòu)建的通濟(jì)橋水庫灌區(qū)水循環(huán)模型模擬的月徑流與實(shí)測(cè)值比較,率定期及驗(yàn)證期的相對(duì)誤差均低于12%、決定系數(shù)均≥0.88、納什效率系數(shù)均≥0.85;4條干渠灌水量模擬值與觀測(cè)值相對(duì)誤差的絕對(duì)值最大不超過20%,校正時(shí)間段灌水量占全年灌水量比例模擬值與觀測(cè)值的相對(duì)誤差為7.64%,表明改進(jìn)SWAT模型具有良好的模擬效果且可用于模擬灌區(qū)不同水源類型的灌水量。
3)模擬分析表明,通濟(jì)橋水庫灌區(qū)的灌溉用水量呈現(xiàn)豐水年小、干旱年大的變化規(guī)律;除監(jiān)測(cè)的骨干水源通濟(jì)橋水庫及浦陽江以外,灌溉用水量41.40%來源于灌區(qū)內(nèi)部的河道、塘堰及小型水庫,說明只監(jiān)測(cè)干渠渠首灌水量無法統(tǒng)計(jì)整個(gè)灌區(qū)灌溉用水量;此外,隨著灌區(qū)節(jié)水改造投入,灌區(qū)灌溉水利用系數(shù)提高,其灌溉用水量減少。
相對(duì)于采用典型監(jiān)測(cè)或現(xiàn)行灌溉定額與實(shí)際灌溉面積數(shù)據(jù)進(jìn)行框算的方法,針對(duì)存在多種水源的灌區(qū),利用分布式水文模型模擬統(tǒng)計(jì)灌溉用水量的方法更為準(zhǔn)確有效。灌區(qū)分布式水文模型的研究處于起步階段,如何將灌區(qū)自然特征與人類活動(dòng)有效結(jié)合并使得模型更加適應(yīng)于灌區(qū)特點(diǎn)需要進(jìn)一步研究。本文所述改進(jìn)SWAT模型中的多水源自動(dòng)灌溉模塊中僅考慮到多個(gè)地表水源的灌溉,故若用于地表水與地下水聯(lián)合應(yīng)用灌區(qū),則需要進(jìn)一步改進(jìn);并且在設(shè)定河道灌溉水可利用系數(shù)與塘堰死庫容占比時(shí),整個(gè)研究區(qū)采用相同的值,沒有考慮灌區(qū)空間異質(zhì)性對(duì)該值的影響。
[1] 郭元裕. 南方丘陵地區(qū)庫、塘、渠網(wǎng)系統(tǒng)渠道設(shè)計(jì)流量的計(jì)算方法[J]. 水利學(xué)報(bào),1965(4):62-66. Guo Yuanyu. Calculation method of channel design flow in reservoirs, ponds and canal network in hilly area in southern China[J]. Journal of Hydraulic Engineering, 1965(4): 62-66. (in Chinese with English abstract)
[2] 王士武,鄭世宗. 南方多水源型灌區(qū)灌溉水利用系數(shù)確定方法研究[J]. 中國(guó)農(nóng)村水利水電,2016(8):109-112. Wang Shiwu, Zheng Shizong. Study on determining method of irrigation water use efficiency in multiple water source irrigation area in South China[J]. China Rural Water and Hydropower, 2016(8): 109-112. (in Chinese with English abstract)
[3] 沈瑩瑩,吉曄,張紹強(qiáng). 我國(guó)農(nóng)業(yè)用水量統(tǒng)計(jì)工作面臨的問題及建議[J]. 中國(guó)水利,2016(7):50-52. Shen Yingying, Ji Ye, Zhang Shaoqiang. Challenges faced by statistics of agricultural water use in China and advices[J]. China Water Resource, 2016(7): 50-52. (in Chinese with English abstract)
[4] 沈瑩瑩,張紹強(qiáng),吉曄. 我國(guó)農(nóng)業(yè)灌溉用水量統(tǒng)計(jì)方法的確定及工作開展情況[J]. 中國(guó)農(nóng)村水利水電,2016(11):133-134. Shen Yingying, Zhang Shaoqiang, Ji Ye. The determination of the statistical methods for agricultural water use in China and its progressive situation[J]. China Rural Water and Hydropower, 2016(11): 133-134. (in Chinese with English abstract)
[5] 郭元裕. 農(nóng)田水利學(xué)[M]. 第三版. 北京:中國(guó)水利水電出版社,2007.
[6] Arnold J G, Srinivasan R, Muttiah R S, et al. Large area hydrological modeling and assessment part I: Model development[J]. Journal of the American Water Resources Association, 1998, 34(1): 91-101.
[7] Neitsch S, Arnold J, Kiniry J, et al. Soil and water assessment tool theoretical documentation, version 2009[R]. Texas: Texas Water Resources Institute, 2011.
[8] 王中根,劉昌明,黃友波. SWAT模型的原理、結(jié)構(gòu)及應(yīng)用研究[J]. 地理科學(xué)進(jìn)展,2003,22(1):79-86. Wang Zhonggen, Liu Changming, Huang Youbo. The theory of SWAT model and its application in Haihe Basin[J]. Progress in Geography, 2003, 22(1): 79-86. (in Chinese with English abstract)
[9] Ahmad K, Gassman P W, Kanwar R. Evaluation of the tile flow component of the SWAT model under different management systems[R]. Iowa: Center for Agricultural and Rural Development, 2002.
[10] Gosain A K, Rao S, Srinivasan R, et al. Return‐flow assessment for irrigation command in the Palleru river basin using SWAT model[J]. Hydrological Processes, 2005, 19(3): 673-682.
[11] 鄭捷,李光永,韓振中,等. SWAT模型在水資源管理中研究進(jìn)展與灌區(qū)應(yīng)用展望[J]. 土壤,2009,41(5):689-695. Zheng Jie, Li Guangyong, Han Zhenzhong, et al. Advance of SWAT in regional water resource management and application prospect in irrigation district[J]. Soils, 2009, 41(5): 689-695. (in Chinese with English abstract)
[12] 李穎,王康,周祖昊. 基于SWAT模型的東北水稻灌區(qū)水文及面源污染過程模擬[J]. 農(nóng)業(yè)工程學(xué)報(bào),2014,30(7):42-53. Li Ying, Wang Kang, Zhou Zuhao. Simulation of drainage and agricultural non-point source pollutions transport processes in paddy irrigation district in North-East China using SWAT[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2014, 30(7): 42-53. (in Chinese with English abstract)
[13] 樊慶鋅,孟婷婷,李金夢(mèng),等. 江川灌區(qū)旱田改水田加劇水體氮磷污染[J]. 農(nóng)業(yè)工程學(xué)報(bào),2014,30(12):79-86. Fan Qingxin, Meng Tingting, Li Jinmeng, et al. Changing from dry field to paddy field intensifying water pollution by nitrogen and phosphorus loads in Jiangchuan irrigation area[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2014, 30(12): 79-86. (in Chinese with English abstract)
[14] Ahmadzadeh H, Morid S, Delavar M, et al. Using the SWAT model to assess the impacts of changing irrigation from surface to pressurized systems on water productivity and water saving in the Zarrineh Rud catchment[J]. Agricultural Water Management, 2015, 175: 15-28.
[15] 王建鵬,崔遠(yuǎn)來. 水稻灌區(qū)水量轉(zhuǎn)化模型及其模擬效率分析[J]. 農(nóng)業(yè)工程學(xué)報(bào),2011,27(1):22-28. Wang Jianpeng, Cui Yuanlai. Modified SWAT for rice-based irrigation system and its assessment [J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2011, 27(1): 22-28. (in Chinese with English abstract)
[16] 于穎多,焦平金,許迪,等. 排水循環(huán)灌溉驅(qū)動(dòng)的稻區(qū)水循環(huán)模型與評(píng)價(jià)[J]. 農(nóng)業(yè)工程學(xué)報(bào),2016,32(11):138-143. Yu Yingduo, Jiao Pingjin, Xu Di, et al. Water cycle model and its assessment under cyclic irrigation of drainage water in paddy district[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2016, 32(11): 138-143. (in Chinese with English abstract)
[17] 代俊峰,崔遠(yuǎn)來. 基于SWAT的灌區(qū)分布式水文模型:Ⅰ.模型構(gòu)建的原理與方法[J]. 水利學(xué)報(bào),2009,40(2):145-152. Dai Junfeng, Cui Yuanlai. Distributed hydrological model for irrigation area based on SWAT Ⅰ. Principle and method[J]. Journal of Hydraulic Engineering, 2009, 40(2): 145-152. (in Chinese with English abstract)
[18] 代俊峰,崔遠(yuǎn)來. 基于SWAT的灌區(qū)分布式水文模型:Ⅱ.模型應(yīng)用[J]. 水利學(xué)報(bào),2009,40(3):311-318. Dai Junfeng, Cui Yuanlai. Distributed hydrological model for irrigation area based on SWAT II. Model application[J]. Journal of Hydraulic Engineering, 2009, 40(3): 311-318. (in Chinese with English abstract)
[19] 陳強(qiáng),秦大庸,茍思,等. SWAT模型與水資源配置模型的耦合研究[J]. 灌溉排水學(xué)報(bào),2010,29(1):19-22. Chen Qiang, Qin Dayong, Gou Si, et al. Research of the coupling of SWAT model and water allocation model[J]. Journal of Irrigation and Drainage, 2010, 29(1): 19-22. (in Chinese with English abstract)
[20] Xie Xianhong, Cui Yuanlai. Development and test of SWAT for modeling hydrological processes in irrigation districts with paddy rice[J]. Journal of Hydrology, 2011, 396(1): 61-71.
[21] Kang M S, Park S W, Lee J J, et al. Applying SWAT for TMDL programs to a small watershed containing rice paddy fields[J]. Agricultural Water Management, 2006, 79(1): 72-92.
[22] 仕玉治,張弛,周惠成,等. SWAT模型在水稻灌區(qū)的改進(jìn)及應(yīng)用研究[J]. 水電能源科學(xué),2010,28(7):24-28. Shi Yuzhi, Zhang Chi, Zhou Huicheng, et al. Development and application of SWAT model to paddy district in watershed scale[J]. Water Resources & Power, 2010, 28(7): 24-28. (in Chinese with English abstract)
[23] 鄭捷,李光永,韓振中,等. 改進(jìn)的SWAT模型在平原灌區(qū)的應(yīng)用[J]. 水利學(xué)報(bào),2011,42(1):88-97. Zheng Jie, Li Guangyong, Han Zhenzhong, et al. Application of modified SWAT model in plain irrigation district[J]. Journal of Hydraulic Engineering, 2011, 42(1): 88-97. (in Chinese with English abstract)
[24] Dechmi F, Burguete J, Skhiri A. SWAT application in intensive irrigation systems: Model modification, calibration and validation[J]. Journal of Hydrology, 2012, 470(14): 227-238.
[25] Sakaguchi A, Eguchi S, Kato T, et al. Development and evaluation of a paddy module for improving hydrological simulation in SWAT[J]. Agricultural Water Management, 2014, 137(5): 116-122.
[26] Chen Y, Marek G W, Marek T H, et al. Improving SWAT auto-irrigation functions for simulating agricultural irrigation management using long-term lysimeter field data[J]. Environmental Modelling & Software, 2018, 99: 25-38.
[27] 李碩,賴正清,王橋,等. 基于SWAT模型的平原河網(wǎng)區(qū)水文過程分布式模擬[J]. 農(nóng)業(yè)工程學(xué)報(bào),2013,29(6):106-112. Li Shuo, Lai Zhengqing, Wang Qiao, et al. Distributed simulation for hydrological process in Plain River network region using SWAT model[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2013, 29(6): 106-112. (in Chinese with English abstract)
[28] 孫麗,陳曦?zé)?,裴志遠(yuǎn). 基于SWAT模型的清江流域中上游旱災(zāi)監(jiān)測(cè)[J]. 農(nóng)業(yè)工程學(xué)報(bào),2014,30(21):129-137. Sun Li, Chen Xiwei, Pei Zhiyuan. Drought monitoring of upper and middle reaches of Qingjiang Basin based on SWAT model[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2014, 30(21): 129-137. (in Chinese with English abstract)
[29] 程艷,敖天其,黎小東,等. 基于參數(shù)移植法的SWAT模型模擬嘉陵江無資料地區(qū)徑流[J]. 農(nóng)業(yè)工程學(xué)報(bào),2016,32(13):81-86. Cheng Yan, Ao Tianqi, Li Xiaodong, et al. Runoff simulation by SWAT model based on parameters transfer method in ungauged catchments of middle reaches of Jialing River [J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2016, 32(13): 81-86. (in Chinese with English abstract)
[30] Abbaspour K C, Vejdani M, Haghighat S. SWAT-CUP calibration and uncertainty programs for SWAT[J]. Modsim International Congress on Modelling & Simulation Land Water & Environmental Management Integrated Systems for Sustainability, 2007, 364(3): 1603-1609.
[31] Moriasi D N, Arnold J G, Van Liew M W, et al. Model evaluation guidelines for systematic quantification of accuracy in watershed simulations[J]. Transactions of the ASABE, 2007, 50(3): 885-900.
Simulation and analysis of irrigation water consumption in multi-source water irrigation districts in Southern China based on modified SWAT model
Cui Yuanlai1, Wu Di1, Wang Shiwu2, Wen Jinhua2, Wang Helong2
(1.430072,; 2.310020,)
The statistics of irrigation water consumption in irrigation districts are of great significance to implement the most stringent water resources management system. On account of the impossibility of complete measurement, it is quite difficult to accurately calculate the irrigation water consumption in multi-source water irrigation districts in the south of China. Therefore, a precise and effective way is needed to estimate irrigation water consumption in multi-source water irrigation districts. In this study, the Soil and Water Assessment Tool (SWAT) was modified according to the characteristics of hydrologic cycle and irrigation operation in the multi-source water irrigation district in the south of China for accurately estimating irrigation water consumption. The water balance modules of paddy field were modified in SWAT model, in addition, a canal seepage loss calculation was added to SWAT model. Specifically, a multi-source water auto-irrigation module was added as one of the components of SWAT model to estimate the irrigation water consumptions from different types of water sources. Furthermore, the modified SWAT model with a digital elevation model (DEM), a soil map, a land cover map and multi-year meteorological data, was applied to build a distributed hydrological model of Tongjiqiao Reservoir Irrigation District (TID) in Zhejiang Province. Moreover, the observed monthly runoff was used to calibrate (1995-2007) and validate (2008-2015) the simulated runoff via SWAT Calibration and Uncertainty Programs (SWATCUP), and the observed irrigation water consumptions of 4 main irrigation canals in 2017 were used to calibrate the simulated irrigation water consumptions. The results showed that the simulated monthly runoff matched well with the observed values in calibration and validation periods, the absolute relative errors (RE) were less than 12%, the coefficients of determination (2) were greater than or equal to 0.88, and the Nash–Sutcliffe efficiencycoefficients (NS) were greater than or equal to 0.85 in both periods; in addition, the maximum of the absolute relative errors between simulated irrigation water consumptions and the observed values of 4 main irrigation canals was less than 20%, indicated that the modified SWAT model has a good performance in the multi-source water irrigation districts. Additionally, the irrigation water consumptions in different hydrological years in TID, multi-year averages of simulated irrigation water consumptions and water supply proportions of different types of water source were simulated and calculated based on the modified SWAT model, in addition, the effect of water saving reform on irrigation water consumption was also analyzed. And the results indicated that the irrigation water consumption is small in wet year and large in dry year. Moreover, in addition to the key water sources (namely the Tongjiqiao Reservoir and the Puyang River), 41.40% of the irrigation water consumption came from the rivers inside sub-basins, ponds and small-sized reservoirs, indicating that the amounts of water monitored at the head of canals fetching water from the key water sources did not represent the irrigation water consumption in irrigation districts. Beyond that, with the development of water saving reform in irrigation district, the irrigation water use efficiency increased so that the irrigation water consumption decreased. Consequently, the modified SWAT model can be used to simulate and analyze the irrigation water consumption in multi-source water irrigation districts accurately and reasonably, and the simulation of irrigation water consumption in multi-source water irrigation districts based on the modified SWAT model is an effective and rational method for calculation and analysis of irrigation water consumption in irrigation districts in the south of China, which satisfied the requirements of the total amount of water statistics and the most stringent water resources management system.
irrigation; models; reservoirs; multi-source water irrigation district; modified SWAT; different types of water sources
崔遠(yuǎn)來,吳 迪,王士武,溫進(jìn)化,王賀龍. 基于改進(jìn)SWAT模型的南方多水源灌區(qū)灌溉用水量模擬分析[J]. 農(nóng)業(yè)工程學(xué)報(bào),2018,34(14):94-100. doi:10.11975/j.issn.1002-6819.2018.14.012 http://www.tcsae.org
Cui Yuanlai, Wu Di, Wang Shiwu, Wen Jinhua, Wang Helong. Simulation and analysis of irrigation water consumption in multi-source water irrigation districts in Southern China based on modified SWAT model [J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2018, 34(14): 94-100. (in Chinese with English abstract) doi:10.11975/j.issn.1002-6819.2018.14.012 http://www.tcsae.org
2018-01-31
2018-05-10
國(guó)家自然科學(xué)基金(51579184);浙江省水利科技計(jì)劃項(xiàng)目(RC1712)
崔遠(yuǎn)來,江西武寧人,教授,主要從事節(jié)水灌溉理論與技術(shù)研究。Email:YLCui@whu.edu.cn
10.11975/j.issn.1002-6819.2018.14.012
S274.2
A
1002-6819(2018)-14-0094-07