楊金鳳,馮愛萍,王雪蕾*,李新榮,王昌佐,田 壯
海河流域農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)識(shí)別方法
楊金鳳1,馮愛萍2,王雪蕾2*,李新榮1,王昌佐2,田 壯1
(1.北京市農(nóng)林科學(xué)院植物營(yíng)養(yǎng)與資源研究所,北京 100097;2.生態(tài)環(huán)境部衛(wèi)星環(huán)境應(yīng)用中心,北京 100094)
在綜合分析農(nóng)業(yè)面源污染風(fēng)險(xiǎn)源匯因子的基礎(chǔ)上,篩選出影響海河流域農(nóng)業(yè)面源污染的8個(gè)主要因子(年降水量、溶解態(tài)面源污染物入河系數(shù)、吸附態(tài)面源污染物入河系數(shù)、年植被覆蓋度、坡度、土壤可侵蝕性因子、農(nóng)田氮表觀平衡量和農(nóng)田磷表觀平衡量),建立了農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)識(shí)別指標(biāo)體系,采用多因子綜合分析法對(duì)海河流域農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)等級(jí)進(jìn)行評(píng)價(jià),并與DPeRS模型風(fēng)險(xiǎn)識(shí)別結(jié)果進(jìn)行偏差分析.結(jié)果表明,海河流域有61.91%的區(qū)域存在農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn),集中分布在流域的中部和南部地區(qū),高風(fēng)險(xiǎn)區(qū)主要分布在北京市東南部、天津市中部、流域山東段東北部和河南段南部等區(qū)域;與DPeRS模型識(shí)別結(jié)果對(duì)比驗(yàn)證,顯示同一風(fēng)險(xiǎn)等級(jí)面積相差不超過12%,且高風(fēng)險(xiǎn)級(jí)別面積相差僅為0.12%,97.17%以上的區(qū)域均為偏差小或無偏差,表明該識(shí)別方法具有與DPeRS模型法同等水平的農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)識(shí)別精準(zhǔn)度,可實(shí)現(xiàn)區(qū)域農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)的快速、高效識(shí)別.
農(nóng)業(yè)面源污染;指標(biāo)體系;潛在風(fēng)險(xiǎn)識(shí)別;海河流域
在來自工業(yè)和城市生活污水的點(diǎn)源污染得到控制后,面源污染逐步成為水體污染的主要污染源,其中農(nóng)業(yè)生產(chǎn)生活引起的面源污染是目前水體污染的主要原因之一.農(nóng)業(yè)面源污染源分散且隱蔽,污染發(fā)生的時(shí)間和空間具有隨機(jī)性和不確定性,監(jiān)測(cè)、控制難度大.識(shí)別農(nóng)業(yè)面源污染的高風(fēng)險(xiǎn)區(qū),將有限的資源投入到對(duì)水體危害可能性最大而范圍相對(duì)較小的區(qū)域進(jìn)行重點(diǎn)治理,可大大降低治理難度和提高治理成效.因此,建立一種農(nóng)業(yè)面源污染風(fēng)險(xiǎn)識(shí)別方法是農(nóng)業(yè)面源污染管理和控制的當(dāng)務(wù)之急.
目前識(shí)別農(nóng)業(yè)面源污染風(fēng)險(xiǎn)的方法[1-2]有輸出系數(shù)法、面源污染定量模型法、指標(biāo)體系法.輸出系數(shù)法[3]結(jié)構(gòu)簡(jiǎn)單,所需資料較少,可直接評(píng)估和預(yù)測(cè)農(nóng)業(yè)面源總氮和總磷的污染負(fù)荷量,但其在區(qū)域尺度上的應(yīng)用需要大量的實(shí)地監(jiān)測(cè)資料.面源污染定量模型包括SPARROW模型、AnnAGNPS模型、SWAT模型和HSPF模型等[4-7],需要參數(shù)較多,而目前農(nóng)業(yè)管理中的數(shù)據(jù)積累還不夠豐富,下墊面情況更復(fù)雜,區(qū)域性差異大,更是增加了地面基礎(chǔ)信息的獲取難度,國(guó)外模型直接移植的難度較大.指標(biāo)體系法可綜合分析影響農(nóng)業(yè)面源污染物流失的主要因子,能夠?yàn)檗r(nóng)業(yè)面源污染風(fēng)險(xiǎn)提供一個(gè)更為合理的評(píng)價(jià)框架,靈活性較強(qiáng).常用的方法有非點(diǎn)源污染潛力指數(shù)法(APPI)[8-9]和磷指數(shù)法(PI)[10-12].但常規(guī)農(nóng)業(yè)面源污染風(fēng)險(xiǎn)指標(biāo)體系法[13-14]中存在考慮的污染來源分類少、指標(biāo)選擇不全面、研究單元太粗等問題,因此,本研究從簡(jiǎn)單快速、低成本、精確性和適應(yīng)性強(qiáng)等角度,建立了識(shí)別農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)程度的指標(biāo)體系,確定指標(biāo)權(quán)重并劃分因子等級(jí),采用多因子綜合分析法計(jì)算海河流域農(nóng)業(yè)面源污染風(fēng)險(xiǎn)指數(shù),對(duì)其農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)進(jìn)行評(píng)價(jià),確定流域內(nèi)農(nóng)業(yè)面源污染發(fā)生風(fēng)險(xiǎn)高的區(qū)域?yàn)橹攸c(diǎn)控制區(qū),為農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)評(píng)價(jià)和快速篩查奠定了基礎(chǔ).
圖1 研究區(qū)位置示意
海河流域位于我國(guó)華北地區(qū),總面積31.82萬km2,包括海河、灤河和徒駭馬頰河3大水系、7大河系、10條骨干河流,地跨北京、天津、河北、山西、山東、河南、內(nèi)蒙古和遼寧等8個(gè)省份,其中,北京、天津全部屬于海河流域,河北省91%、山西省38%、山東省20%、河南省9.2%的面積屬于海河流域,內(nèi)蒙古自治區(qū)1.36萬km2和遼寧省0.17萬km2屬于海河流域.該流域總體地勢(shì)是西北高東南低,西部為黃土高原和太行山區(qū),北部為蒙古高原和燕山山區(qū);流域?qū)儆跍貛|亞季風(fēng)氣候區(qū),年平均氣溫在1.5~14℃,年平均相對(duì)濕度50%~70%,多年平均降水量539mm(2011~2016年),屬半濕潤(rùn)半干旱地帶,年平均陸面蒸發(fā)量470mm,山區(qū)小于500mm,平原大于500mm,多年平均水面蒸發(fā)量850~1300mm,研究區(qū)位置見圖1.
1.2.1 農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)評(píng)價(jià)指標(biāo)體系 農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)指標(biāo)的選擇是否恰當(dāng)對(duì)整個(gè)研究過程及研究結(jié)果都存在較大的影響.農(nóng)業(yè)面源污染的發(fā)生受多方面因子的影響與控制,其中包括人類無法調(diào)控的自然因子,如降雨、地形、地貌等,同時(shí)也包括了人類活動(dòng)可以調(diào)控的許多因子,如植被覆蓋、農(nóng)藥化肥的使用、農(nóng)田灌溉等.所有氣候因素都對(duì)水土流失有相應(yīng)影響,其中降水最為重要,一般是年降水量越大,水土流失就越嚴(yán)重,地形地貌和土壤植被主要通過降雨和地表徑流影響面源污染.經(jīng)濟(jì)水平?jīng)Q定人的生產(chǎn)生活方式,主要通過社會(huì)經(jīng)濟(jì)活動(dòng),影響土地利用方式、農(nóng)業(yè)生產(chǎn)方式及管理水平、農(nóng)村庭院養(yǎng)殖集中程度和規(guī)模、居民環(huán)境保護(hù)意識(shí)等,農(nóng)村人口現(xiàn)狀及增長(zhǎng)速度直接影響耕地利用方式及利用程度、農(nóng)業(yè)面源污染物的產(chǎn)生總量.
本研究在充分考慮影響農(nóng)業(yè)面源污染的自然因素(氣象、地形地貌、土壤、植被、水文等)和人為因素(化肥農(nóng)藥施用、耕作、灌溉、畜禽養(yǎng)殖、農(nóng)村生活垃圾及污水排放等),兼顧污染物的產(chǎn)生、遷移和消減整個(gè)過程,結(jié)合現(xiàn)有資料,最終選擇能反映農(nóng)業(yè)面源潛在污染普遍特征的三大類指標(biāo):水文氣象指標(biāo)、土壤地形植被指標(biāo)和經(jīng)濟(jì)指標(biāo),水文氣象指標(biāo)具體包括年降水量、溶解態(tài)面源污染物入河系數(shù)和吸附態(tài)面源污染物入河系數(shù);土壤地形植被指標(biāo)具體包括年植被覆蓋度、坡度和土壤可侵蝕性因子;經(jīng)濟(jì)指標(biāo)具體包括農(nóng)田氮表觀平衡量和農(nóng)田磷表觀平衡量.三大類8個(gè)指標(biāo)的含義及算法具體如下:
(1)年降水量:降水是影響地表土壤侵蝕和面源擴(kuò)散的重要因素之一,因降水有時(shí)空變化,面源污染也有時(shí)空不同,受降水強(qiáng)度、持續(xù)性、數(shù)量和降雨頻率等因素影響.在這些因素中,對(duì)面源污染有重要影響的是降水量和降雨強(qiáng)度,其大小直接影響著徑流量的大小,進(jìn)而影響面源污染的程度[15].基于流域范圍內(nèi)氣象站的降水量數(shù)據(jù),以DEM作為協(xié)變量,利用薄板樣條滑動(dòng)平均法進(jìn)行降水量的空間插值[16],得到流域年降水量空間數(shù)據(jù).
(2)溶解態(tài)和吸附態(tài)面源污染物入河系數(shù):是指產(chǎn)生的面源污染物進(jìn)入河網(wǎng)的比例,是用來估算面源污染物入河排放量的重要參數(shù).按照溶解態(tài)和吸附態(tài)兩種污染物存在形式分為溶解態(tài)污染物入河系數(shù)和吸附態(tài)污染物入河系數(shù).其中溶解態(tài)入河系數(shù)由徑流系數(shù)決定,而吸附態(tài)入河系數(shù)由泥沙輸移系數(shù)決定[17].溶解態(tài)面源污染物入河系數(shù)為年徑流量與年降水量的比值,具體公式如下:
式中:CR為溶解態(tài)面源污染物入河系數(shù);Prec和Runoff分別為年降水量和年徑流量.
吸附態(tài)面源污染物入河系數(shù)為年泥沙含量與年土壤侵蝕量的比值,具體公式如下:
式中:SDR為吸附態(tài)面源污染物入河系數(shù);Sed為年泥沙含量;Sel為年土壤侵蝕量;為降雨侵蝕力因子;為土壤可侵蝕性因子;分別為坡長(zhǎng)因子和坡度因子,無量綱;為生物措施因子,無量綱;為工程措施措施因子,無量綱.
(3)年植被覆蓋度:植被指數(shù)是一種無量綱的輻射測(cè)度,用來反映綠色植被的相對(duì)豐度及其活動(dòng),其中以歸一化植被指數(shù)(NVDI)應(yīng)用最為廣泛,而且經(jīng)過驗(yàn)證,植被指數(shù)與植被覆蓋度有較好的相關(guān)性,用它來計(jì)算植被覆蓋度比較合適.該因子與耕作管理密切相關(guān),直接影響著土壤侵蝕速度.可利用遙感數(shù)據(jù),采用最大最小值定量反演算法進(jìn)行流域植被覆蓋度反演[18-19].
(4)坡度:坡度是形成土壤侵蝕的根本原因,對(duì)侵蝕強(qiáng)度的影響也非常大,一般來說,地形的坡度越大,侵蝕的可能性也越大[20].基于DEM高程數(shù)據(jù),利用ArcGIS軟件功能模塊計(jì)算流域坡度.
(5)土壤可侵蝕性因子():是土壤潛在侵蝕性的量度,它受土壤物理性質(zhì)的影響,如與土壤機(jī)械組成、有機(jī)質(zhì)含量、土壤結(jié)構(gòu)、土壤滲透性等有關(guān),值越大,土壤就容易遭受侵蝕.因子采用EPIC模型計(jì)算,并對(duì)其計(jì)算結(jié)果進(jìn)行糾正[21-22].具體公式如下:
(6)農(nóng)田氮表觀平衡量和農(nóng)田磷表觀平衡量:定義為氮磷輸入項(xiàng)與輸出項(xiàng)之差,當(dāng)平衡量為負(fù)值時(shí)表示土壤養(yǎng)分輸出大于輸入,處于虧損狀態(tài);當(dāng)平衡量為正值時(shí),表示土壤養(yǎng)分輸入大于輸出,處于盈余狀態(tài),盈余的氮磷會(huì)增加農(nóng)田面源污染的風(fēng)險(xiǎn)[17].根據(jù)縣級(jí)的化肥施用量、畜禽養(yǎng)殖量、農(nóng)作物產(chǎn)量、農(nóng)業(yè)人口等51個(gè)統(tǒng)計(jì)指標(biāo)數(shù)據(jù),采用輸入輸出法[23]計(jì)算,具體公式如下:
式中:bal為所述農(nóng)田氮磷平衡量或所述農(nóng)田磷平衡量;area為耕地面積和園地面積之和;1000為單位轉(zhuǎn)換系數(shù);Balance為養(yǎng)分平衡量;Input為養(yǎng)分輸入量;Output為養(yǎng)分輸出量;Ftlz為化肥養(yǎng)分輸入量;Mnr為有機(jī)肥養(yǎng)分輸入量;Irg為灌溉養(yǎng)分輸入量;Seed為種子養(yǎng)分輸入量;Bnf為生物固氮氮輸入量;Dpzt為干濕沉降養(yǎng)分輸入量;Hvst為作物帶走養(yǎng)分輸出量.
1.2.2 農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)評(píng)價(jià)指標(biāo)權(quán)重 不同指標(biāo)對(duì)農(nóng)業(yè)面源污染的潛在危害程度不同,因此需要確定各指標(biāo)的權(quán)重以獲得更準(zhǔn)確的污染風(fēng)險(xiǎn)等級(jí)[24].由于農(nóng)業(yè)面源污染受多因素共同作用,且具有隨機(jī)性、廣泛性、模糊性和滯后性等特點(diǎn),較適合采用層次分析法,本研究采取層次分析法中的冪法確定各指標(biāo)的權(quán)重,詳見表1.
表1 農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)識(shí)別指標(biāo)權(quán)重
1.2.3 農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)指數(shù)計(jì)算及風(fēng)險(xiǎn)分級(jí) 根據(jù)《土地利用現(xiàn)狀調(diào)查技術(shù)規(guī)程》[25]、區(qū)域降水分布規(guī)律及土壤侵蝕強(qiáng)度分級(jí)的參考指標(biāo)等,結(jié)合GIS的自然間斷點(diǎn)分級(jí)法(Jenks),充分考慮數(shù)據(jù)的均值、方差等統(tǒng)計(jì)結(jié)果,對(duì)8個(gè)潛在污染風(fēng)險(xiǎn)指標(biāo)進(jìn)行了分級(jí)并賦值1~4,各指標(biāo)分級(jí)標(biāo)準(zhǔn)及其賦值詳見表2.
在農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)指標(biāo)體系構(gòu)建基礎(chǔ)上,結(jié)合指標(biāo)權(quán)重和指標(biāo)賦值,最終建立了農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)指數(shù)算法,具體如下:
式中:NPSPRI為農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)指數(shù);為潛在風(fēng)險(xiǎn)指標(biāo)在指標(biāo)體系中的權(quán)重值,其值范圍為0~1;為指標(biāo)賦值,其值范圍為1~4;為指標(biāo)體系中的指標(biāo).
表2 農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)識(shí)別指標(biāo)分級(jí)標(biāo)準(zhǔn)
對(duì)農(nóng)業(yè)面源污染風(fēng)險(xiǎn)指數(shù)NPSPRI進(jìn)行自然間斷點(diǎn)分級(jí)法(Jenks)分級(jí)為4個(gè)風(fēng)險(xiǎn)等級(jí),詳細(xì)分級(jí)標(biāo)準(zhǔn)為:無風(fēng)險(xiǎn)(0,2.208]、低風(fēng)險(xiǎn)(2.208,2.704]、中風(fēng)險(xiǎn)(2.704,3]和高風(fēng)險(xiǎn)(3,4),并將各個(gè)級(jí)別數(shù)值范圍賦予1~4,即:無風(fēng)險(xiǎn)賦值1,低風(fēng)險(xiǎn)賦值2,中風(fēng)險(xiǎn)賦值3,高風(fēng)險(xiǎn)賦值4,得到農(nóng)業(yè)面源污染綜合指標(biāo)評(píng)價(jià)體系(CIES)潛在風(fēng)險(xiǎn)空間分布圖.
表3 主要數(shù)據(jù)
本文采用的數(shù)據(jù)主要包括海河流域2015年的遙感數(shù)據(jù)、氣象數(shù)據(jù)、水文數(shù)據(jù)、高程數(shù)據(jù)、土壤數(shù)據(jù)和農(nóng)業(yè)統(tǒng)計(jì)數(shù)據(jù)等,具體數(shù)據(jù)來源見表3.
采用前文表2提出的農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)指標(biāo)分級(jí)標(biāo)準(zhǔn)對(duì)海河流域農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)識(shí)別指標(biāo)體系中的8個(gè)指標(biāo)進(jìn)行分級(jí),結(jié)果表明,海河流域中部、東北部和南部的部分區(qū)域達(dá)到500mm以上的降雨量,且植被覆蓋度相對(duì)較高;除流域西北部和天津的部分區(qū)域外,其余地區(qū)溶解態(tài)和吸附態(tài)面源污染物入河系數(shù)相對(duì)較高;流域的山地丘陵地帶坡度相對(duì)較大,坡度小的區(qū)域土壤可侵蝕性因子值相對(duì)較高;流域的中部和南部地區(qū)農(nóng)田氮表觀平衡量和磷表觀平衡量較高,北部相對(duì)較低,8個(gè)指標(biāo)分級(jí)的空間分布詳見圖2.
圖2 海河流域農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)識(shí)別指標(biāo)空間分布
海河流域農(nóng)業(yè)面源污染風(fēng)險(xiǎn)等級(jí)圖及面積統(tǒng)計(jì)表詳見圖3和表4.
應(yīng)用農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)指數(shù)對(duì)海河流域農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)區(qū)域進(jìn)行識(shí)別(圖3(a)),結(jié)果表明:海河流域有61.91%的區(qū)域存在農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn),集中分布在流域的中部和南部地區(qū),高風(fēng)險(xiǎn)區(qū)面積占比為1.61%,主要分布在北京市東南部、天津市中部、山東段東北部和河南段南部等區(qū)域,此外,在河北段的部分區(qū)域也有零星分布.
同時(shí),應(yīng)用DPeRS模型[27-30]對(duì)海河流域農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)區(qū)域進(jìn)行定量識(shí)別.DPeRS模型是一種基于二元結(jié)構(gòu)的半經(jīng)驗(yàn)半機(jī)理過程的模型,既考慮降水、植被覆蓋、地形地貌等自然因素,同時(shí)也考慮了施肥利用效率、人口、牲畜和家禽等社會(huì)經(jīng)濟(jì)因素,并耦合定量遙感技術(shù),可以對(duì)流域尺度面源污染負(fù)荷的時(shí)空動(dòng)態(tài)進(jìn)行精確的定量評(píng)估.為實(shí)現(xiàn)兩者的可比性,對(duì)DPeRS模型模擬的污染負(fù)荷結(jié)果也采用前文方法中提出的風(fēng)險(xiǎn)分級(jí)標(biāo)準(zhǔn)進(jìn)行劃分,即也分為4個(gè)風(fēng)險(xiǎn)等級(jí)(即無風(fēng)險(xiǎn)、低風(fēng)險(xiǎn)、中風(fēng)險(xiǎn)以及高風(fēng)險(xiǎn)),將4個(gè)等級(jí)數(shù)值范圍賦值1~4,得到模型風(fēng)險(xiǎn)等級(jí)圖(圖3(b)),結(jié)果表明:海河流域有50.86%的區(qū)域存在農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn),高風(fēng)險(xiǎn)區(qū)域面積占比為1.73%.農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)區(qū)域集中分布在流域的中部和南部地區(qū),中風(fēng)險(xiǎn)和高風(fēng)險(xiǎn)區(qū)主要分布在北京市東南部、天津市中北部、山東段東北部和河南段西部等區(qū)域,此外,在河北段的部分區(qū)域也有零星分布.
表4 海河流域農(nóng)業(yè)面源污染風(fēng)險(xiǎn)等級(jí)面積統(tǒng)計(jì)
圖3 海河流域農(nóng)業(yè)面源污染風(fēng)險(xiǎn)等級(jí)
表5 CIES等級(jí)結(jié)果與DPeRS模型結(jié)果偏差情況表
為驗(yàn)證所構(gòu)建的農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)指標(biāo)識(shí)別方法的精準(zhǔn)度,將綜合指標(biāo)評(píng)價(jià)體系風(fēng)險(xiǎn)等級(jí)圖與DPeRS模型風(fēng)險(xiǎn)等級(jí)圖的識(shí)別結(jié)果進(jìn)行偏差分析(圖4).
對(duì)比海河流域農(nóng)業(yè)面源污染綜合指標(biāo)評(píng)價(jià)體系風(fēng)險(xiǎn)等級(jí)結(jié)果與DPeRS模型識(shí)別結(jié)果,得出:同一風(fēng)險(xiǎn)等級(jí)面積相差不超過12%,尤其是高風(fēng)險(xiǎn)級(jí)別面積相差僅為0.12%(表4);97.17%以上的區(qū)域均為偏差小或無偏差(表5).表明本文提出的農(nóng)業(yè)面源潛在污染風(fēng)險(xiǎn)識(shí)別方法具有與DPeRS模型法同等水平的風(fēng)險(xiǎn)識(shí)別精準(zhǔn)度,且該識(shí)別方法不需要模型法所必須的復(fù)雜的指標(biāo)以及數(shù)據(jù)積累,也不需要考慮復(fù)雜的下墊面情況,可以實(shí)現(xiàn)簡(jiǎn)單快速、高精度的農(nóng)業(yè)面源污染風(fēng)險(xiǎn)識(shí)別.
設(shè)置了5種指標(biāo)組合情形,與DPeRS模型模擬的海河流域面源污染進(jìn)行偏差分析,篩選農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)指標(biāo)的最優(yōu)組合.結(jié)果表明,本文采用的8項(xiàng)指標(biāo)與DPeRS模型對(duì)海河流域面源污染風(fēng)險(xiǎn)的識(shí)別結(jié)果偏差是最小的,無偏差和偏差小的比例達(dá)到97.17%,說明本文構(gòu)建的風(fēng)險(xiǎn)評(píng)價(jià)指標(biāo)體系可以快速、高精度地識(shí)別農(nóng)業(yè)面源污染潛在高風(fēng)險(xiǎn)區(qū).農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)指標(biāo)篩選情形詳見表6.
圖4 海河流域CIES風(fēng)險(xiǎn)等級(jí)結(jié)果與DPeRS模型結(jié)果的偏差分析
表6 農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)指標(biāo)篩選
注:與DPeRS模型的偏差分析結(jié)果比較的是無偏差或偏差小的比例和(%).
3.1 在綜合分析海河流域農(nóng)業(yè)面源污染風(fēng)險(xiǎn)源匯因子的基礎(chǔ)上,篩選出8個(gè)主要影響因子,即年降水量、溶解態(tài)面源污染物入河系數(shù)、吸附態(tài)面源污染物入河系數(shù)、年植被覆蓋度、坡度、土壤可侵蝕性因子、農(nóng)田氮表觀平衡量和農(nóng)田磷表觀平衡量,建立了農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)識(shí)別指標(biāo)體系,應(yīng)用農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)指數(shù),可實(shí)現(xiàn)海河流域農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn)區(qū)的快速、高效識(shí)別.
3.2 農(nóng)業(yè)面源污染綜合指標(biāo)評(píng)價(jià)體系風(fēng)險(xiǎn)識(shí)別結(jié)果表明,海河流域有61.91%的區(qū)域存在農(nóng)業(yè)面源污染潛在風(fēng)險(xiǎn),集中分布在流域的中部和南部地區(qū),高風(fēng)險(xiǎn)區(qū)面積占比為1.61%,主要分布在北京市東南部、天津市中部、山東段東北部和河南段南部等區(qū)域.
3.3 對(duì)比海河流域農(nóng)業(yè)面源污染綜合指標(biāo)評(píng)價(jià)體系風(fēng)險(xiǎn)等級(jí)結(jié)果與DPeRS模型模擬結(jié)果,同一風(fēng)險(xiǎn)等級(jí)面積相差不超過12%,尤其是高風(fēng)險(xiǎn)級(jí)別面積相差僅為0.12%,97.17%以上的區(qū)域均為偏差小或無偏差,表明該識(shí)別方法具有與DPeRS模型法同等水平的風(fēng)險(xiǎn)識(shí)別精準(zhǔn)度.
[1] Ouyang W, Gao X, Wei P, et al. A review of diffuse pollution modeling and associated implications for watershed management in China [J]. Journal of Soils and Sediments, 2017,17(6):1527-1536.
[2] Wei P, Ouyang W, Hao F H, et al. Combined impacts of precipitation and temperature on diffuse phosphorus pollution loading and critical source area identification in a freeze-thaw area [J]. Science of the Total Environment, 2016,553(3):607-616.
[3] 朱 梅,吳敬學(xué),張希三.海河流域種植業(yè)非點(diǎn)源污染負(fù)荷量估算[J]. 農(nóng)業(yè)環(huán)境科學(xué)學(xué)報(bào), 2010,29(10):1907-1915.
Zhu M, Wu J X, Zhang X S. Estimation on non-point source pollution loads of crop farming in Hai Basin [J]. Journal of Agro-Environment Science, 2010,29(10):1907-1915.
[4] 徐 敏,陳 巖,趙琰鑫,等.流域空間屬性關(guān)聯(lián)模型(SPARROW模型)理論方法與應(yīng)用案例[M]. 北京:中國(guó)環(huán)境出版社, 2013.
Xu M, Chen Y, Zhao Y X, et al. Theoretical method and application case of watershed spatial attribute association model (SPARROW model) [M]. Beijing: China Environment press, 2013.
[5] 李開明,任秀文,黃國(guó)如,等.基于AnnAGNPS模型泗合水流域非點(diǎn)源污染模擬研究[J]. 中國(guó)環(huán)境科學(xué), 2013,33(S1):54-59.
Li K M, Ren X W, Huang G R, et al. Simulation of non-point source pollution in Sihe watershed with Ann AGNPS [J]. China Environmental Science, 2013,33(S1):54-59.
[6] 秦耀民,胥彥玲,李懷恩.基于SWAT模型的黑河流域不同土地利用情景的非點(diǎn)源污染研究[J]. 環(huán)境科學(xué)學(xué)報(bào), 2009,29(2):440-448.
Qin Y M, Xu Y L, Li H E.SWAT model of non-point source pollution under different land use scenarios in the Heihe river basin [J]. Acta Scientiae Circumstantiae, 2009,29(2):440-448.
[7] Chang C L, Li M Y.Predictions of diffuse pollution by the HSPF model and the back-propagation neural network model [J]. Water Environment Research, 2017,8:732-738.
[8] 周徐海,王 寧,郭紅巖,等.農(nóng)業(yè)非點(diǎn)源污染潛力指數(shù)系統(tǒng)(APPI)在太湖典型區(qū)域的應(yīng)用[J]. 農(nóng)業(yè)環(huán)境科學(xué)學(xué)報(bào), 2006,25(4):1029- 1034.
Zhou X H, Wang N, Guo H Y, et al. Preliminary application of agricultural non- point source pollution potential index in typical area of Taihu Lake [J]. Journal of Agro-Environment Science, 2006,25(4): 1029-1034.
[9] 曹昕鑫.基于APPI指數(shù)系統(tǒng)的旱作農(nóng)田面源污染發(fā)生潛力及優(yōu)先控制區(qū)識(shí)別—以沙潁河流域?yàn)槔齕D]. 合肥:安徽大學(xué), 2013.
Cao X X. Analysis on characteristics and evaluation of agricultural non-point source pollution potential and in typical dry farmland area of Shayin river catchment based on APPI [D]. Hefei:Anhui University, 2013.
[10] Zhang W W, Ma Y H, Lu Q, et al. Nutrient loss from farmland: research on and application of phosphorus index method [J]. Agricultural Science &technology, 2015,16(2):262-265.
[11] 李 琪,陳利頂,齊 鑫,等.媯水河流域農(nóng)耕區(qū)非點(diǎn)源磷污染危險(xiǎn)性評(píng)價(jià)與關(guān)鍵源區(qū)識(shí)別[J]. 環(huán)境科學(xué), 2008,29(1):32-37.
Li Q, Chen LD, Qi X, et al. Identification of critical area of phosphorus loss in agricultural areas of Guishui river watershed by phosphorus loss risk assessment [J]. Environmental Science, 2008, 29(1):32-37.
[12] 宋月君,吳勝軍,劉永美,等.基于GIS技術(shù)的農(nóng)用地非點(diǎn)源磷污染危險(xiǎn)性評(píng)價(jià)—以長(zhǎng)江流域?yàn)槔齕J]. 測(cè)繪科學(xué), 2009,34(3):164-166.
Song Y J, Wu S J, Liu Y M, et al. Risk appraisal to the non-point source phosphorus pollution of agriculture land based on the GIS technology—A case study of Yangtze basin [J]. Science of Surveying and Mapping, 2009,34(3):164-166.
[13] 劉建昌,嚴(yán) 巖,劉 峰,等.基于多因子指數(shù)集成的流域面源污染風(fēng)險(xiǎn)研究[J]. 環(huán)境科學(xué), 2008,29(3):599-606.
Liu J C, Yan Y, Liu F, et al. Risk assessment and safety evaluation using system normative indexes integration method for non-point source pollution on watershed scale [J]. Environmental Science, 2008,29(3):599-606.
[14] 呂 川,張洪鈺,齊 琳.遼河源頭區(qū)流域農(nóng)業(yè)非點(diǎn)源污染風(fēng)險(xiǎn)評(píng)價(jià)與關(guān)鍵源區(qū)識(shí)別[J]. 江西農(nóng)業(yè)大學(xué)學(xué)報(bào), 2014,36(3):670-677.
Lv C, Zhang H Y, Qi L. Critical source area identification and risk assessment of agricultural non-point source pollution of the source areas of Liaohe river watershed [J]. Acta Agriculturae Universitatis Jiangxiensis, 2014,36(3):670-677.
[15] 王曉燕.非點(diǎn)源污染及其管理[M]. 北京:海洋出版社, 2003.
Wang X Y. Non-point source pollution and management [M]. Beijing: China Ocean Press, 2003.
[16] 劉志紅,Li L T, McVicar T R,等.專用氣候數(shù)據(jù)空間插值軟件ANUSPLIN及其應(yīng)用[J]. 氣象, 2008,34(2):92-100.
Liu Z H, Li L T, McVicar T R, et al. introduction of the professional interpolation software for meteorology data:ANUSPLIN [J]. Meteorological Monthly, 2008,34(2):92-100.
[17] 王雪蕾,王 橋,吳傳慶,等.國(guó)家尺度面源污染業(yè)務(wù)評(píng)估與應(yīng)用示范[M]. 北京:科學(xué)出版社, 2015:10-115.
Wang X L, Wang Q, Wu C Q, et al. Assessment and application of national diffuse pollution management [M]. Beijing: Science Press, 2015:10-115.
[18] Jiménez-Mu?oz J C, Sobrino J A, Plaza A, et al. Comparison between fractional vegetation cover retrievals from vegetation indices and spectral mixture analysis: Case study of PROBA/CHRIS data over an agricultural area [J]. Sensors, 2009,9(2):768-793.
[19] 王雪蕾,吳傳慶,馮愛萍,等.利用DPeRS模型估算巢湖流域氨氮和化學(xué)需氧量的面源污染負(fù)荷[J]. 環(huán)境科學(xué)學(xué)報(bào), 2015,35(9):2883- 2891.
Wang X L, Wu C Q, Feng A P, et al. Application of DPeRS model on estimation of non-point source pollution load of ammonia nitrogen and chemical oxygen demand in Chao Lake basin [J]. Acta Scientiae Circumstantiae, 2015,35(9):2883-2891.
[20] PoliD. General model for airborne and spacebome linear array sensors [J]. International Archives of Photogrammetry and Remote Sensing, 2002,34(B1):177-182.
[21] Zhang K L, Peng W Y and Yang H L. Soil erodibility and its estimation for agricultral soil in China [J]. Acta Pedologica Sinica, 2007,44(1):7-13.
[22] 王雪蕾,王新新,朱 利,等.巢湖流域氮磷面源污染與水華空間分布遙感解析[J]. 中國(guó)環(huán)境科學(xué), 2015,35(5):1511-1519.
Wang X L, Wang X X, Zhu L, et al. Spatial analysis on diffuse pollution and algal bloom characteristic with remote sensing in Chao Lake Basin [J]. China Environmental Science, 2015,35(5):1511-1519.
[23] Wang X L, Feng A P, Wang Q, et al. Spatial variability of the nutrient balance and related NPSP risk analysis for agro-ecosystems in China in 2010 [J]. Agriculture, Ecosystems and Environment, 2014,193:42- 52.
[24] Gburek W J, Sharpley A N, Heathwaite L, et al. Phosphorus management at the watershed scale: A modification of the phosphorus index [J]. Journal of Environmental Quality, 2000,29:135-140.
[25] 全國(guó)農(nóng)業(yè)區(qū)劃委員會(huì).土地利用現(xiàn)狀調(diào)查技術(shù)規(guī)程 [EB/OL]. https://max.book118.com/html/2018/1231/6111142204001242.shtm, 1984-09-09/2020-10-04.
National Agricultural Zoning Commission. Technical regulations for land use status investigation [EB/OL]. https://max.book118.com/ html/2018/1231/6111142204001242.shtm, 1984–09–09/2020–10–04.
[26] 符素華,劉寶元,周貴云,等.坡長(zhǎng)坡度因子計(jì)算工具[J]. 中國(guó)水土保持科學(xué), 2015,13(5):105-110.
Fu S H, Liu B Y, Zhou G Y, et al. Calculation tool of topographic factors [J]. Science of Soil and Water Conservation, 2015,13(5):105- 110.
[27] Wang X L, Wang Q, Wu C Q, et al.A method coupled with remote sensing data to evaluate non-point source pollution in the Xin'anjiang catchment of China [J]. Science of the Total Environment, 2012, 430:132-143.
[28] 馮愛萍,吳傳慶,王雪蕾,等.海河流域氮磷面源污染空間特征遙感解析[J]. 中國(guó)環(huán)境科學(xué), 2019,39(7):2999-3008.
Feng A P, Wu C Q, Wang X L, et al. Spatial character analysis on nitrogen and phosphorus diffuse pollution in Haihe River Basin by remote sensing [J]. China Environmental Science, 2019,39(7):2999- 3008.
[29] 王雪蕾,蔡明勇,鐘部卿,等.遼河流域非點(diǎn)源污染空間特征遙感解析[J]. 環(huán)境科學(xué), 2013,34(10):3788-3796.
Wang X L, CAI M Y, Zhong B Q, et al. Research on spatial characteristic of non-point source pollution in Liaohe River Basin [J]. Environmental Science, 2013,34(10):3788-3796.
[30] 馮愛萍,王雪蕾,徐 逸,等.基于DPeRS模型的海河流域面源污染潛在風(fēng)險(xiǎn)評(píng)估[J]. 環(huán)境科學(xué), 2020,41(10):4555-4563.
Feng A P, Wang X L, XU Y, et al. Assessment of potential risk of diffuse pollution in Haihe River Basin based using DPeRS model [J]. Environmental Science, 2020,41(10):4555-4563.
An identification method of potential risk for agricultural non-point source pollution in the Haihe River Basin.
YANG Jin-feng1, FENG Ai-ping2, WANG Xue-lei2*, Li Xin-rong1, WANG Chang-zuo2, TIAN Zhuang1
(1.Beijing Academy of Agriculture and Forestry Sciences, Institute of Plant Nutrition and Resources, Beijing 100097, China;2.Ministry of Ecology and Environment Center for Satellite Application on Ecology and Environment, Beijing 100094, China)., 2021,41(10):4782~4791
Based on a comprehensive analysis of the source and sink factors of agricultural non-point source pollution risk, eight main factors leading to the agricultural non-point source pollution, which were annual precipitation, dissolved non-point source pollutant inflow coefficient, granular non-point source pollutant inflow coefficient, annual vegetation coverage, slope, soil erodibility factor, nitrogen and phosphorus balance of farmland, were identified. Moreover, an identification index system of the potential risk for agricultural non-point source pollution was established. The multi-factor comprehensive analysis method was used to evaluate the risk of agricultural non-point source pollution in the Haihe River Basin. The results were compared with the risk identification results of Diffuse pollution estimation with remote sensing (DPeRS) model using deviation analysis. It showed that 61.91% of the Haihe River Basin had potential risks of agricultural non-point source pollution. It was concentrated in the central and southern regions of the basin. Among which were high-risk areas including the southeast of Beijing, the central part of Tianjin, the northeastern part of Shandong and the southern part of the Henan within the basin. Comparing the results with that of DPeRS model, the area of the same risk level regions differed by no more than 12%, and the area of high-risk regions differed by only 0.12%, and there were little or no deviation for more than 97.17% of the regions. It was shown that the identification method had the same accuracy as the DPeRS model method for identifying potential risks of agricultural non-point source pollution, making rapid and efficient identification of potential risks of regional agricultural non-point source pollution possible.
agricultural non-point source pollution;index system;potential risk identification;Haihe River Basin
X522
A
1000-6923(2021)10-4782-10
楊金鳳(1981-),女,山西朔州人,助理研究員,博士,主要從事農(nóng)業(yè)面源污染防治.發(fā)表論文20余篇.
2021-02-25
國(guó)家重點(diǎn)研發(fā)計(jì)劃項(xiàng)目(2016YFD0800903);國(guó)家自然科學(xué)基金資助項(xiàng)目(41871346)
* 責(zé)任作者, 研究員, wxlbnu@163.com