陳 凱,陳求穩(wěn),于海燕,王備新,金小偉,王業(yè)耀,許人驥,蔡 琨 (1.南京農(nóng)業(yè)大學(xué)昆蟲(chóng)系,江蘇南京 1009;.南京水利科學(xué)研究院生態(tài)環(huán)境研究中心,江蘇 南京 100;.浙江省環(huán)境監(jiān)測(cè)中心生態(tài)所,浙江 杭州 1001;.中國(guó)環(huán)境監(jiān)測(cè)總站,北京 10001;.江蘇省環(huán)境監(jiān)測(cè)中心,江蘇 南京 1006)
河流生態(tài)健康評(píng)價(jià)是進(jìn)行水環(huán)境管理和保證水環(huán)境健康的重要內(nèi)容;開(kāi)展我國(guó)流域水生態(tài)完整性監(jiān)測(cè)和評(píng)價(jià)的方法及其體系構(gòu)建,是我國(guó)流域水質(zhì)目標(biāo)管理技術(shù)體系的重要組成,是實(shí)施水生態(tài)功能分區(qū)管理的重要基礎(chǔ),也是保障水生態(tài)可持續(xù)發(fā)展的重要前提[1].生物完整性是水生態(tài)系統(tǒng)健康評(píng)價(jià)的重要指標(biāo),是水生態(tài)完整性的關(guān)鍵組份.生物完整性是指與區(qū)域環(huán)境相適應(yīng)的,經(jīng)長(zhǎng)期進(jìn)化形成的生物群落組成、結(jié)構(gòu)和功能方面的屬性[3],其實(shí)質(zhì)是通過(guò)完整性指數(shù)測(cè)量河流生物學(xué)和生態(tài)學(xué)資源的現(xiàn)狀[2];基于生物群落特征的多參數(shù)指數(shù) MMI (通常也被稱為生物完整性指數(shù)IBI),及反映樣點(diǎn)觀測(cè)物種組成(觀測(cè)值O)與期望物種組成(期望值 E)差異性的 O/E指數(shù),均是應(yīng)用最廣泛的評(píng)價(jià)河流生物完整性的兩種指數(shù).
我國(guó)自2003年開(kāi)始底棲動(dòng)物完整性指數(shù)評(píng)價(jià)溪流健康的研究,當(dāng)前評(píng)價(jià)的指示生物類群已涉及魚(yú)類、浮游生物、微生物和水生植物,評(píng)價(jià)水體也拓展到河流、湖泊、水庫(kù)、濕地和河口等.但我國(guó)的MMI和O/E指數(shù)的研究和應(yīng)用工作大多直接借鑒國(guó)外已有方法,對(duì)生物完整性指數(shù)構(gòu)建方法與技術(shù)體系是否符合我國(guó)水生態(tài)系統(tǒng)特征的研究不夠深入,一定程度上限制了我國(guó)應(yīng)用生物完整性評(píng)價(jià)水生態(tài)健康的業(yè)務(wù)化運(yùn)行.本文就MMI和O/E指數(shù)概念、國(guó)內(nèi)外發(fā)展史和應(yīng)用現(xiàn)狀進(jìn)行詳細(xì)的綜述,提出構(gòu)建我國(guó)MMI和O/E指數(shù)評(píng)價(jià)體系亟待解決的問(wèn)題和研究建議,旨在進(jìn)一步推進(jìn)我國(guó)水生態(tài)系統(tǒng)健康評(píng)價(jià)的研究和業(yè)務(wù)化應(yīng)用.
多參數(shù)指數(shù)(MMI)方法上等同于生物完整性指數(shù)(IBI),試圖通過(guò)綜合生物群落組成、結(jié)構(gòu)、物種性狀和功能參數(shù)定量描述生物完整性[4].雖然IBI指數(shù)特指生物完整性指數(shù),但其已有構(gòu)建過(guò)程較難真實(shí)且全面的反映生物完整性(例如已有IBI對(duì)具有描述生態(tài)系統(tǒng)功能信息的生物性狀和功能參數(shù)的應(yīng)用極少),因此MMI的概念更為合適且正在被廣泛應(yīng)用.O/E指數(shù)基于河流無(wú)脊椎動(dòng)物預(yù)測(cè)與分類系統(tǒng)(RIVPACS)模型,計(jì)算物種觀測(cè)豐富度(O)和期望豐富度(E)的比值(即:觀測(cè)值/期望值,O/E)定量物種組成完整性[5-6],其比值反映期望物種組成在調(diào)查樣點(diǎn)的出現(xiàn)率,一定程度上表征樣點(diǎn)生物組成完整性的喪失程度[7].
1.1 MMI構(gòu)建方法
MMI構(gòu)建依據(jù)評(píng)價(jià)目的、水生態(tài)系統(tǒng)類型、指標(biāo)生物類群和可用數(shù)據(jù)類型等有所區(qū)別[8],但生物參數(shù)選擇是構(gòu)建具有高敏感性和響應(yīng)性MMI的核心要求[9],如基于概念模型選擇生物學(xué)意義較重要的參數(shù)(如Karr最初構(gòu)建IBI指數(shù)的方法[10]),和選擇判別能力(區(qū)分參照和受損健康狀況)最優(yōu)的參數(shù)(例如 Barbour等[11])是兩種基本方法.經(jīng)典MMI的構(gòu)建步驟(圖1)包括:
(1)確定參照、非參照和受損點(diǎn)位.通常通過(guò)設(shè)定土地利用和水化學(xué)的客觀閾值、結(jié)合專家判斷法確定參照點(diǎn)位和非參照點(diǎn)位[12].參照條件的定義包括:極小干擾條件、歷史狀況、較小干擾條件和現(xiàn)存最佳條件[12].參照點(diǎn)位的設(shè)定應(yīng)避免使用生物參數(shù)標(biāo)準(zhǔn),生物參數(shù)標(biāo)準(zhǔn)易造成 MMI構(gòu)建的“循環(huán)效應(yīng)”[12].
(2)標(biāo)準(zhǔn)化野外采樣、生物分類和數(shù)據(jù)整理工作.建立標(biāo)準(zhǔn)化的水生生物采樣工具[13-14]和采樣強(qiáng)度[15],實(shí)驗(yàn)室內(nèi)生物分類水平[15]與數(shù)據(jù)處理方法[15-17]等,以提高過(guò)程的質(zhì)量保證和結(jié)果的可靠性.
(3)候選生物參數(shù).候選生物參數(shù)類型可以分為:群落組成參數(shù)、豐富度/多樣性參數(shù)、敏感值/耐污值參數(shù)、營(yíng)養(yǎng)結(jié)構(gòu)/功能性參數(shù)、生物習(xí)性參數(shù)等[8-9].廖靜秋等[18]總結(jié)了魚(yú)類、底棲動(dòng)物和著生藻類MMI構(gòu)建的常用生物參數(shù).
(4)參數(shù)分布范圍檢驗(yàn).較窄的分布范圍說(shuō)明參數(shù)反映的自然梯度和人類脅迫梯度的范圍較窄[9],包括:所有點(diǎn)位或參照點(diǎn)位生物參數(shù)值的分布范圍極小,或大部分點(diǎn)位的參數(shù)值都為相同數(shù)值.分布范圍的檢驗(yàn)方法多樣,應(yīng)依據(jù)生物參數(shù)信息確定檢驗(yàn)分布范圍的方法和閾值[9].
(5)控制自然梯度影響.底棲動(dòng)物群落組成、結(jié)構(gòu)和功能等在時(shí)空尺度同時(shí)受到人類干擾和自然梯度的影響[19-20],混淆自然變異和人類干擾的影響,容易增加評(píng)價(jià)結(jié)果出現(xiàn)I型(Type I,將健康水體誤判為受損水體)和 II型(Type II,將受損水體誤判為健康水體)錯(cuò)誤的概率[21].為準(zhǔn)確評(píng)價(jià)人類干擾的影響,應(yīng)盡可能地降低自然梯度的影響.依據(jù)自然特征分區(qū)和預(yù)測(cè)模型是兩種主要方法.
分區(qū)(classification或者regionalization)即利用生態(tài)區(qū)(Omernik[22-23])最小化參照和評(píng)價(jià)樣點(diǎn)間的自然特征差異性,但用于分區(qū)的自然特征并不一定是影響水生生物群落自然差異性的主要原因[24-25],而且該方法不能定量化自然梯度的影響[24,26].重復(fù)性檢驗(yàn) (例如:信號(hào)/噪聲比)是結(jié)合生態(tài)區(qū)方法常用的,檢驗(yàn)生物參數(shù)時(shí)空尺度穩(wěn)定性的常用指標(biāo).
預(yù)測(cè)模型方法通過(guò)構(gòu)建生物參數(shù)對(duì)自然梯度的響應(yīng)模型,計(jì)算模型殘差,達(dá)到定量化降低甚至消除自然梯度影響的目的[27],該方法理論上不受生物類群和研究水體的限制[25,27-29].模型構(gòu)建方法包括:逐步多元線性回歸[27,30-32]、分類與回歸樹(shù)[28]、隨機(jī)森林[25,33-34]、促進(jìn)回歸樹(shù)[35]等.其中,基于分類與回歸樹(shù)模型的優(yōu)點(diǎn)有:可以處理自變量和應(yīng)變量之間的復(fù)雜關(guān)系,計(jì)算數(shù)據(jù)不需要符合正態(tài)分布、模型不會(huì)出現(xiàn)過(guò)擬合、自變量共線性不會(huì)對(duì)計(jì)算結(jié)果產(chǎn)生影響等[36-38],已經(jīng)逐漸成為目前最廣泛應(yīng)用的模型方法.
(6)判別能力/響應(yīng)性分析.計(jì)算生物參數(shù)區(qū)分不同脅迫程度影響的能力,檢驗(yàn)方法包括:相關(guān)性分析[9]、散點(diǎn)圖[39]、箱式圖[40]、非參數(shù) t檢驗(yàn)[28]、有效判別系數(shù)[41]等.判斷生物參數(shù)對(duì)人類干擾梯度響應(yīng)方向是判別能力分析的必要前提步驟[11].
(7)冗余分析.指生物參數(shù)的生物信息相似,或具有高度相關(guān)性[9];生物參數(shù)高度冗余易顯著降低 MMI的可靠性.剔除相關(guān)性較強(qiáng)(例如相關(guān)系數(shù) r>|0.70|)的生物參數(shù)和減小MMI構(gòu)成參數(shù)相關(guān)系數(shù)平均值[34,42]是避免冗余的兩種主要方法[9].聚類分析[28]和主成分分析[33]是近年流行的降低生物參數(shù)相關(guān)系數(shù)平均值的方法.
(8)生物參數(shù)賦值.該步驟標(biāo)準(zhǔn)化生物參數(shù)的分布范圍和一致化對(duì)脅迫的響應(yīng)方向,并計(jì)算MMI最終值.生物參數(shù)賦值的方法包括不連續(xù)賦值法和連續(xù)賦值法.不連續(xù)賦值法(如 3分法[10]和4分法[43])易增加MMI的不穩(wěn)定性,而且限制了其在不同類型水體的應(yīng)用[44].連續(xù)賦值法不僅避免了這些缺點(diǎn),而且能夠避免主觀確定生物參數(shù)的健康條件;使用連續(xù)賦值法時(shí)通常使用參照點(diǎn)位和受損點(diǎn)位的95%和5%分位數(shù)值作為期望最佳值或最差值,避免異常值的影響[28,44].
(9)表現(xiàn)力評(píng)價(jià)和評(píng)價(jià)標(biāo)準(zhǔn)建立.通過(guò)精確度、準(zhǔn)確度/偏差、響應(yīng)性和敏感性等比較評(píng)價(jià)表現(xiàn)力[25].通常利用參照點(diǎn)位 MMI值的標(biāo)準(zhǔn)差(SD)衡量精確度,SD 越小則精確度越高.準(zhǔn)確度/偏差通過(guò)構(gòu)建自然環(huán)境梯度對(duì)參照點(diǎn)位MMI的解釋模型,殘留自然變異的解釋量超過(guò) 10%[25]則認(rèn)為偏差較大,準(zhǔn)確度較低.響應(yīng)性測(cè)量參照點(diǎn)位和嚴(yán)重受損點(diǎn)位 MMI值的差異性,差異性越大,說(shuō)明響應(yīng)性越高.
圖1 構(gòu)建多參數(shù)完整性指數(shù)(MMI)經(jīng)典步驟的流程Fig.1 Schematic overview of the required steps to develop a Multimetric Index (MMI)
敏感性是處于非參照等級(jí)(non-reference condition, NRC)的監(jiān)測(cè)點(diǎn)位百分比.敏感性和評(píng)價(jià)標(biāo)準(zhǔn)緊密相關(guān),用于計(jì)算評(píng)價(jià)標(biāo)準(zhǔn)的方法包括生態(tài)學(xué)方法、統(tǒng)計(jì)學(xué)方法、專家判斷法[45]等.生態(tài)學(xué)方法通過(guò)計(jì)算評(píng)價(jià)指數(shù)和人類脅迫的關(guān)系確定;統(tǒng)計(jì)學(xué)方法通過(guò)等間距方法確定健康等級(jí);專家判斷法主觀性最強(qiáng),通過(guò)專家經(jīng)驗(yàn)賦值健康等級(jí).健康標(biāo)準(zhǔn)閾值計(jì)算方法包括:通過(guò)計(jì)算生物參數(shù)和人類脅迫梯度關(guān)系的TITAN[46-47]模型法,以參照點(diǎn)位值的5%、10%、或25%分位數(shù)作為評(píng)價(jià)標(biāo)準(zhǔn)[25,28,43],通過(guò)參照點(diǎn)位 SD 確定標(biāo)準(zhǔn)[48-49],通過(guò)單尾非中心對(duì)稱的不等和等價(jià)測(cè)驗(yàn)計(jì)算標(biāo)準(zhǔn)[33-34,50];對(duì)小于該閾值的MMI分布范圍進(jìn)行3或4等分,確定健康等級(jí)(例如:亞健康、一般、差、較差).
近年來(lái) MMI構(gòu)建步驟不斷簡(jiǎn)化,較大判別能力和較低冗余性是選擇MMI核心構(gòu)成參數(shù)的重點(diǎn)步驟,因此將兩者相結(jié)合是主要步驟,例如:Vander Laan和Hawkins[33]將t檢驗(yàn)和PCA相結(jié)合篩選生物參數(shù),Chen等[17]利用t檢驗(yàn)和聚類分析相結(jié)合選擇生物參數(shù).但本文認(rèn)為,分布范圍檢驗(yàn)是所有步驟的基礎(chǔ),較窄的分布范圍說(shuō)明參照點(diǎn)位覆蓋的自然梯度范圍可能較窄,而且較多0值或相同值易影響預(yù)測(cè)模型構(gòu)建,從而影響最終評(píng)價(jià)結(jié)果.
1.2 O/E構(gòu)建方法
O/E指數(shù)構(gòu)建的參照點(diǎn)位確定、標(biāo)準(zhǔn)化野外采樣、生物分類和數(shù)據(jù)整理等方法同MMI指數(shù).主要步驟包括[49](圖2):
(1)參照點(diǎn)位聚類.剔除稀有物種[51]后,根據(jù)物種組成對(duì)參照點(diǎn)位進(jìn)行聚類分析.聚類方法包括:TWINSPAN 法[5,52-53]、Ward融合法[54]、K-means法[55]、和現(xiàn)階段常用的UPGMA flexible β方法[51,56-58].
(2)點(diǎn)位權(quán)重計(jì)算.使用人類活動(dòng)難以改變的自然因子(例如:經(jīng)度、維度、海拔、河流等級(jí)等),建立參照點(diǎn)位聚類類群與自然因子的判別模型,計(jì)算各點(diǎn)位屬于參照點(diǎn)位聚類類群的權(quán)重.多元判別模型(MDA)是經(jīng)典的模型方法[25,52-53,59],邏輯斯蒂回歸[60]、人工神經(jīng)網(wǎng)絡(luò)(ANN)[61]、貝葉斯網(wǎng)絡(luò)[62]等也是較常用的模型,RF模型是近年來(lái)最流行的方法[25,33,63].
(3)物種權(quán)重計(jì)算.計(jì)算區(qū)域物種庫(kù)中每個(gè)物種屬于每個(gè)參照點(diǎn)位類群的權(quán)重.
(4)物種出現(xiàn)概率(Pc).將第 2和 3步的點(diǎn)位權(quán)重與物種權(quán)重進(jìn)行加權(quán),計(jì)算物種出現(xiàn)概率Pc.Pc值越大,說(shuō)明物種出現(xiàn)的概率越高,為常見(jiàn)物種;Pc值越小,說(shuō)明物種可能出現(xiàn)的概率越小,為稀有物種.通常使用 Pc≥0.5和 Pc>0兩種閾值用于 O/E 計(jì)算,Pc≥0.5(常見(jiàn)物種)較 Pc>0的 O/E 表現(xiàn)力更高[64-65].
(5)觀測(cè)值/期望值(O/E)比值.將大于設(shè)定 Pc閾值的所有出現(xiàn)物種的Pc值相加得到該點(diǎn)位的期望豐富度值(E),同時(shí)計(jì)算大于設(shè)定 Pc閾值的物種觀測(cè)豐富度值(O).最終O/E值在理論上介于0~1之間,但也有可能略大于1.
圖2 構(gòu)建和應(yīng)用基于RIVPACS模型的O/E指數(shù)經(jīng)典步驟的流程Fig.2 Schematic representation of the general steps followed in the RIVPACS approach for O/E construction and application
(6)評(píng)價(jià)表現(xiàn)力和模型誤差.O/E表現(xiàn)力的評(píng)價(jià)參數(shù)同 MMI評(píng)價(jià)表現(xiàn)力.但對(duì) O/E,還可以計(jì)算重復(fù)樣本標(biāo)準(zhǔn)差(RSSD)[57]占零模型標(biāo)準(zhǔn)差[57]的范圍百分比(PctRange)[57-58];零模型標(biāo)準(zhǔn)差代表理論上O/E的最低精確度,RSSD代表O/E指數(shù)的變異僅來(lái)源于重復(fù)采樣過(guò)程中的系統(tǒng)誤差,是理論上O/E最高精確度,PctRange值越大說(shuō)明O/E值的精確性越高.
(7)評(píng)價(jià)標(biāo)準(zhǔn).O/E≈1代表評(píng)價(jià)點(diǎn)位的物種組成完整性較高,O/E<<1表示物種組成完整性受到了破壞.評(píng)價(jià)標(biāo)準(zhǔn)構(gòu)建方法同 MMI評(píng)價(jià)標(biāo)準(zhǔn)建立方法.
MMI和O/E指數(shù)都測(cè)量生物完整性,理論基礎(chǔ)都是參照健康方法理論(RCA)[66-67],在標(biāo)準(zhǔn)化野外采樣、實(shí)驗(yàn)室處理和數(shù)據(jù)分析等工作的基礎(chǔ)上,聯(lián)系生物特征與環(huán)境特征,最后通過(guò)數(shù)值方法直觀表述健康得分,用于評(píng)估水體健康現(xiàn)狀和退化程度.
MMI和O/E的主要區(qū)別在對(duì)生物完整性的定義、對(duì)生物群落受損狀況表征和對(duì)參照點(diǎn)位分類3個(gè)方面[68].MMI定義的生物完整性包括個(gè)體、種群、群落和生態(tài)系統(tǒng)特征等,不同生物特征對(duì)不同脅迫類型的響應(yīng)具有差異性,因此MMI理論上對(duì)更多的脅迫類型存在響應(yīng);O/E指數(shù)根據(jù)物種組成相似性,通過(guò)表征物種缺失代表生物完整性的退化程度.與O/E相比,MMI對(duì)生物完整性的定義更全面;但在理論和實(shí)際操作中,O/E比MMI更容易理解.O/E對(duì)物種組成變化更加敏感,評(píng)價(jià)結(jié)果不僅能夠體現(xiàn)物種豐富度變化,而且能夠表明物種的替代;而MMI不能夠反映物種組成的變化,具有相同生態(tài)功能的物種替代可能不會(huì)影響 MMI.經(jīng)典MMI對(duì)所有點(diǎn)位進(jìn)行預(yù)先分區(qū),然后對(duì)處于同一生態(tài)區(qū)的點(diǎn)位進(jìn)行MMI的構(gòu)建和應(yīng)用;O/E則是通過(guò)后分區(qū)的方法,按照物種組成信息將具有相似物種組成的參照點(diǎn)位聚類在一起.
MMI構(gòu)建需要參照點(diǎn)位和嚴(yán)重受損點(diǎn)位的數(shù)據(jù),但O/E的構(gòu)建僅依據(jù)參照點(diǎn)位數(shù)據(jù)[25].O/E在構(gòu)建過(guò)程中定量檢驗(yàn)評(píng)價(jià)點(diǎn)位與參照點(diǎn)位空間關(guān)系,但MMI并不會(huì)定量檢驗(yàn)它們的關(guān)系[7,48].已有研究表明,MMI對(duì)中低程度的干擾較敏感,對(duì)中等到較嚴(yán)重干擾的響應(yīng)敏感性明顯降低,但O/E對(duì)不同受損程度都擁有較好的敏感性和響應(yīng)關(guān)系[25,33,69].
與MMI相比,O/E指數(shù)[68]:(1)不需要將所有點(diǎn)位進(jìn)行預(yù)先分組,(2)對(duì)參照點(diǎn)位是否來(lái)自于同一個(gè)研究區(qū)域不敏感,(3)利用相對(duì)獨(dú)立的數(shù)據(jù)進(jìn)行 O/E指數(shù)的計(jì)算,減少了未知誤差的影響,(4)加權(quán)點(diǎn)位可能性和物種可能性,保證結(jié)果的可靠性.但O/E指數(shù)構(gòu)建需要大量數(shù)據(jù)收集,且模型方法較復(fù)雜,不利于O/E的推廣和應(yīng)用.
3.1 MMI發(fā)展簡(jiǎn)史和國(guó)內(nèi)外應(yīng)用現(xiàn)狀
Karr[10]最早基于魚(yú)類群落建立半定量半定性的生物完整性指數(shù),評(píng)價(jià)人類干擾對(duì)黑溪及其對(duì)應(yīng)流域的影響(圖3).該評(píng)價(jià)方法得到許多研究者的認(rèn)可,在不斷完善其構(gòu)建和評(píng)價(jià)方法的基礎(chǔ)上,應(yīng)用在北美其他流域[70-71];并逐漸傳播到歐洲大陸,在相對(duì)更大的空間尺度上進(jìn)行應(yīng)用[30,72],同時(shí)候選生物參數(shù)增加了反映生態(tài)系統(tǒng)功能的參數(shù)[30].在此基礎(chǔ)上成功地應(yīng)用定量程度更高的MMI評(píng)價(jià)溪流[73-74]、湖泊[75-76]、濕地[77]和河口[78-79]等水體生態(tài)健康,評(píng)價(jià)指標(biāo)生物也由魚(yú)類拓展到底棲動(dòng)物[80]、藻類[28,81]、浮游動(dòng)物[82]、水生植物[83]及綜合生物類群[84-85].最初的 MMI用于較小空間尺度下的河流生態(tài)健康評(píng)價(jià)[10],在標(biāo)準(zhǔn)化和統(tǒng)一構(gòu)建方法的基礎(chǔ)上[9],MMI被逐漸應(yīng)用在區(qū)域尺度[39,86]、國(guó)家尺度[87]、以及不同大洲[88-89]的水體.但由于研究區(qū)域自然梯度和干擾類型的差異性,導(dǎo)致 MMI在較大空間尺度上無(wú)法進(jìn)行有效比較[7],因此預(yù)測(cè)模型 MMI[27]于 21世紀(jì)初開(kāi)始出現(xiàn),通過(guò)定量自然變量對(duì)生物群落的影響,提高 MMI在時(shí)空尺度的可比性和穩(wěn)定性[34](圖3).
我國(guó)的生物監(jiān)測(cè)工作始于20世紀(jì)70年代環(huán)境污染調(diào)查[1],但對(duì) MMI研究的起步較晚,王備新等[43,90]以天目山—大別山闊葉林生態(tài)區(qū)的溪流為研究對(duì)象,于2003年首次嘗試構(gòu)建底棲動(dòng)物MMI指標(biāo)體系.自此以來(lái),以魚(yú)類[91-92]、底棲動(dòng)物[93-94]和硅藻[95]為指標(biāo)生物的MMI指數(shù)被廣泛應(yīng)用于我國(guó)溪流[96]、河流[92,97]、湖泊[98]等水體的生態(tài)健康評(píng)價(jià).我國(guó)主要利用 MMI確定流域水生生物健康現(xiàn)狀(例如Huang等[99])、構(gòu)建區(qū)域尺度下特定流域的生物完整性評(píng)價(jià)體系與評(píng)價(jià)標(biāo)準(zhǔn)(例如張遠(yuǎn)等[96])、分析與MMI指數(shù)顯著相關(guān)的環(huán)境梯度(例如Li等[100])等.我國(guó)僅在近年開(kāi)始預(yù)測(cè)模型MMI的研究,Chen等[34]利用不同季節(jié)和年份的漓江流域河流底棲動(dòng)物數(shù)據(jù),陳凱等[101]利用浙江中北部東苕溪、西苕溪、錢塘江流域和臨安市河流共 4個(gè)子流域的底棲動(dòng)物和環(huán)境數(shù)據(jù),比較了預(yù)測(cè)模型和常規(guī) MMI的評(píng)價(jià)表現(xiàn)力;與國(guó)外已有研究結(jié)果類似,研究結(jié)果都發(fā)現(xiàn)預(yù)測(cè)模型和常規(guī)MMI的核心組成參數(shù)存在差異,預(yù)測(cè)模型 MMI的精確度和準(zhǔn)確度高于,但敏感性和響應(yīng)性低于常規(guī)MMI.
3.2 O/E發(fā)展簡(jiǎn)史和國(guó)內(nèi)外應(yīng)用現(xiàn)狀
英國(guó)首先提出基于底棲動(dòng)物的RIVPACS模型構(gòu)建O/E進(jìn)行河流生態(tài)健康評(píng)價(jià)的概念[5-6](圖3).90年代初RIVPACS II和RIVPACS III模型在標(biāo)準(zhǔn)化采樣方法、增加參照點(diǎn)位數(shù)量、評(píng)估不同分類等級(jí)結(jié)果、綜合多時(shí)間尺度數(shù)據(jù)、結(jié)合定性和定量數(shù)據(jù)構(gòu)建模型、研究點(diǎn)位聚類和預(yù)測(cè)新方法、評(píng)估模型輸出不確定性等方面得到了顯著的提高[102].隨著時(shí)間推移,RIVPACS模型方法和技術(shù)在美國(guó)得到了前所未有的推進(jìn)[103],例如:RIVPACS零模型構(gòu)建[57]和優(yōu)化判別預(yù)測(cè)模型選擇[104],通過(guò)監(jiān)測(cè)物種組成完整性評(píng)價(jià)水生態(tài)健康也逐漸成為 RIVPACS的核心目標(biāo)之一[105].在此基礎(chǔ)上, Reynoldson等[106]發(fā)展適合加拿大的BEAST模型,Smith等[107]構(gòu)建適合澳大利亞的AUSRIVAS模型.O/E指數(shù)評(píng)價(jià)的目標(biāo)生態(tài)系統(tǒng)也擴(kuò)展到湖泊和濕地[106,108];指標(biāo)生物也從底棲動(dòng)物擴(kuò)展到魚(yú)類[109]、底棲硅藻[110]、大型植物[111],綜合生物類群[84,112],甚至是棲息地組成[113]等.
在我國(guó),應(yīng)用RIVPACS模型的O/E指數(shù)評(píng)價(jià)河流生態(tài)健康的研究最初僅有綜述性文章介紹其基本概念,并分析優(yōu)缺點(diǎn)等[114-119].隨后,張杰等[120]嘗試構(gòu)建底棲動(dòng)物O/E指數(shù)評(píng)價(jià)漓江生態(tài)系統(tǒng)健康,用于預(yù)測(cè)模型試驗(yàn)研究;陳凱[69]應(yīng)用多季節(jié)和多年分的底棲動(dòng)物數(shù)據(jù)構(gòu)建浙江省多個(gè)河流流域的 O/E指數(shù),并比較其評(píng)價(jià)表現(xiàn)力;Chen等[17]研究了底棲動(dòng)物采樣強(qiáng)度對(duì)O/E指數(shù)評(píng)價(jià)表現(xiàn)力的影響;陳凱等[121]利用已有的季風(fēng)氣候區(qū)RIVPACS模型評(píng)價(jià)了淮河流域典型水體的底棲動(dòng)物組成完整性.
圖3 MMI和O/E指數(shù)發(fā)展簡(jiǎn)史(改自Hawkins等2010[66])Fig.3 Brief history of the development, application, and interpretation of MMI and O/E indices[66]
MMI和O/E指數(shù)的理論基礎(chǔ)和技術(shù)方法日漸成熟,體現(xiàn)出廣泛的應(yīng)用可行性.然而,目前我國(guó)MMI和O/E指數(shù)的應(yīng)用大多直接借鑒國(guó)外已有方法,缺乏適合我國(guó)水生態(tài)系統(tǒng)特征的生物完整性指數(shù)基礎(chǔ)方法和技術(shù)體系的深入研究.因此,結(jié)合我國(guó)不同區(qū)域的水環(huán)境污染和干擾類型的新特點(diǎn)[122]、水生生物區(qū)系組成和分布等研究,提出適用于我國(guó)多種水生態(tài)系統(tǒng)特征的生物評(píng)價(jià)方法和技術(shù)體系,對(duì)我國(guó)水體健康現(xiàn)狀評(píng)價(jià)、水生態(tài)系統(tǒng)退化原因診斷、水生態(tài)健康變化趨勢(shì)預(yù)測(cè)預(yù)警、全國(guó)尺度生物評(píng)價(jià)技術(shù)及規(guī)范應(yīng)用、推廣和比較等有重要的積極作用.本文認(rèn)為當(dāng)前我國(guó)MMI和O/E指數(shù)研究工作應(yīng)著重在:(1)生物和環(huán)境基礎(chǔ)數(shù)據(jù)收集和數(shù)據(jù)庫(kù)構(gòu)建;(2)構(gòu)建規(guī)范化方法和技術(shù)體系的適用性研究;(3)評(píng)價(jià)基準(zhǔn)和標(biāo)準(zhǔn).
4.1 生物與環(huán)境基礎(chǔ)數(shù)據(jù)收集和數(shù)據(jù)庫(kù)構(gòu)建
我國(guó)地域遼闊,存在地形、地貌、氣候、溫度、水文條件等自然環(huán)境梯度從南至北和從東至西差異大、水生生物多樣性豐富的特點(diǎn),但不同水生生物的區(qū)域尺度和全國(guó)尺度多樣性及其分布格局依然不清.因此,基礎(chǔ)數(shù)據(jù)收集和數(shù)據(jù)庫(kù)構(gòu)建對(duì)我國(guó)MMI和O/E指數(shù)研究和推廣應(yīng)用顯得尤為重要.如參照樣點(diǎn)的數(shù)量及其空間分布是否覆蓋所有水生態(tài)功能區(qū)類型,是否包含了生物群落隨時(shí)間(年份、季節(jié)、月份)的變化特點(diǎn),均會(huì)對(duì)MMI和 O/E指數(shù)的準(zhǔn)確性和精確性產(chǎn)生影響.基礎(chǔ)數(shù)據(jù)收集和數(shù)據(jù)庫(kù)構(gòu)建工作包括:(1)歷史資料的收集與整理.主要是對(duì)流域內(nèi)或區(qū)域內(nèi)曾開(kāi)展生物監(jiān)測(cè)工作樣點(diǎn)及其已有生物監(jiān)測(cè)數(shù)據(jù)的收集和整理;(2)區(qū)域物種數(shù)據(jù)厘訂.根據(jù)最新發(fā)表的物種鑒定與分布的數(shù)據(jù)資料,結(jié)合形態(tài)學(xué)和分子生物學(xué)手段,如分子條形碼技術(shù),完善和明確區(qū)域物種名錄;(3)補(bǔ)充調(diào)查.結(jié)合自然環(huán)境特征,重點(diǎn)開(kāi)展符合參照條件樣點(diǎn)的環(huán)境和生物數(shù)據(jù)的補(bǔ)充調(diào)查工作;(4)建立具有檢索和常用生物指數(shù)運(yùn)算功能的樣點(diǎn)與物種數(shù)據(jù)庫(kù).
4.2 構(gòu)建規(guī)范化方法和技術(shù)體系的適用性研究
規(guī)范化MMI和O/E指數(shù)構(gòu)建方法和技術(shù)體系,并研究和提高其對(duì)不同水生態(tài)類型的適用性,有助于不同地區(qū)類似水生態(tài)系統(tǒng)健康評(píng)價(jià)體系的推廣應(yīng)用,提高評(píng)價(jià)結(jié)果可比性,統(tǒng)一評(píng)價(jià)基準(zhǔn)的構(gòu)建和評(píng)價(jià)標(biāo)準(zhǔn)等級(jí)的實(shí)施.我國(guó)不同地區(qū)生物完整性指數(shù)監(jiān)測(cè)技術(shù)方法發(fā)展不平衡,目前沒(méi)有適合我國(guó)不同水體類型特點(diǎn)的且通用的生物完整性指數(shù)監(jiān)測(cè)技術(shù)規(guī)范,缺乏樣品野外采集和實(shí)驗(yàn)室處理、指示生物類群選擇、指數(shù)構(gòu)建方法和流程選擇、空間尺度選擇等構(gòu)建方法和技術(shù)等方面的規(guī)范化的適用性標(biāo)準(zhǔn).
(1)野外采集和實(shí)驗(yàn)室樣品處理方法與規(guī)范.保證生物數(shù)據(jù)的有效性與準(zhǔn)確性是提高生物完整性指數(shù)及其評(píng)價(jià)基準(zhǔn)的基礎(chǔ)條件.結(jié)合我國(guó)水生態(tài)分區(qū)的河流及對(duì)應(yīng)流域的自然特征,建立適合不同水環(huán)境特征的水生生物樣品野外采集(如采樣工具選擇、目標(biāo)生境選擇、固定采樣面積、樣本重復(fù)數(shù)量),和實(shí)驗(yàn)室樣品處理(如抽樣個(gè)體數(shù)、水生生物分類等級(jí)、混合分類單元(OTU)使用和一致化分類等級(jí))的方法和技術(shù)規(guī)范.
(2)指示生物類群選擇.確定不同水生態(tài)分區(qū)和典型流域的水環(huán)境污染現(xiàn)狀特征、典型水環(huán)境問(wèn)題及其作用空間尺度,明確指示污染類型的單一或綜合生物類群,提高水生態(tài)現(xiàn)狀評(píng)價(jià)結(jié)果的全面性和可信度.結(jié)合水環(huán)境數(shù)據(jù)收集結(jié)果,綜合水生生物群落組成、結(jié)構(gòu)和生物性狀及其功能多樣性參數(shù)的響應(yīng)特征,探索生物類群權(quán)重賦值,明確評(píng)價(jià)指標(biāo)構(gòu)建的方法體系.
(3)候選生物參數(shù)數(shù)據(jù)庫(kù).依據(jù)收集的基礎(chǔ)數(shù)據(jù),建立我國(guó)水生態(tài)系統(tǒng)完整性評(píng)價(jià)的指導(dǎo)性常用候選生物參數(shù)數(shù)據(jù)庫(kù);增加能夠反映生態(tài)系統(tǒng)功能的水生生物性狀組成和功能多樣性參數(shù)列表;為全面構(gòu)建生物完整性提供基礎(chǔ)參數(shù)平臺(tái).
(4)指數(shù)構(gòu)建方法、流程選擇和空間尺度選擇方法和技術(shù)體系.對(duì)擁有復(fù)雜地質(zhì)學(xué)、地理地形學(xué)、河流形態(tài)學(xué)、景觀學(xué)等自然特征的不同河流類型,構(gòu)建利用我國(guó)數(shù)據(jù)驗(yàn)證的、適合不同流域和水功能分區(qū)的生物完整性指數(shù)構(gòu)建方法和技術(shù)體系,包括:合理的評(píng)價(jià)指數(shù)構(gòu)建過(guò)程、自然特征差異性消除方法、評(píng)價(jià)模型和預(yù)測(cè)模型的構(gòu)建和適用性檢驗(yàn)、合適的生物完整性評(píng)價(jià)空間尺度選擇、不同流域評(píng)價(jià)結(jié)果闡述內(nèi)容的一致性和水生態(tài)系統(tǒng)健康的變化趨勢(shì)等.
4.3 評(píng)價(jià)基準(zhǔn)和標(biāo)準(zhǔn)
評(píng)價(jià)基準(zhǔn)與標(biāo)準(zhǔn)是有效實(shí)施水生生物完整性目標(biāo)的主要基礎(chǔ)和管理依據(jù).生物評(píng)價(jià)基準(zhǔn)是指未受人類干擾和污染的自然狀態(tài)下水生生物完整性狀況,是基于科學(xué)調(diào)查、實(shí)驗(yàn)和科學(xué)推論的客觀結(jié)果;生物評(píng)價(jià)標(biāo)準(zhǔn)是以評(píng)價(jià)基準(zhǔn)為理論依據(jù),綜合考慮自然條件和人類干擾等因素制定的,生物對(duì)不同類型和不同程度人類干擾響應(yīng)敏感性的直觀評(píng)價(jià).評(píng)價(jià)基準(zhǔn)是制定評(píng)價(jià)標(biāo)準(zhǔn)的理論基礎(chǔ),決定著評(píng)價(jià)標(biāo)準(zhǔn)的科學(xué)性、準(zhǔn)確性和可靠性,而評(píng)價(jià)標(biāo)準(zhǔn)是水環(huán)境管理的基礎(chǔ)和目標(biāo),也是判斷生物群落退化程度、評(píng)估生物群落受影響程度和確定技術(shù)方法進(jìn)行管理等的依據(jù)[123],是健康評(píng)價(jià)最重要的步驟之一[45].生物評(píng)價(jià)基準(zhǔn)確定和評(píng)價(jià)標(biāo)準(zhǔn)等級(jí)構(gòu)建的科學(xué)性和系統(tǒng)性內(nèi)容包括:(1)參照體系.參照點(diǎn)位質(zhì)量是評(píng)價(jià)基準(zhǔn)建立的根本,根據(jù)我國(guó)水生態(tài)分區(qū)的水體及其流域特征,確定合適的參照點(diǎn)位的定義,設(shè)定參照點(diǎn)位選擇的土地利用和物理化學(xué)指標(biāo)及其閾值標(biāo)準(zhǔn),明確基于特定參照點(diǎn)位類群的評(píng)價(jià)標(biāo)準(zhǔn)可應(yīng)用的空間尺度范圍;(2)評(píng)價(jià)標(biāo)準(zhǔn)計(jì)算方法.根據(jù)我國(guó)不同水生態(tài)分區(qū)的生物完整性指數(shù)分布特點(diǎn),科學(xué)選擇評(píng)價(jià)標(biāo)準(zhǔn)計(jì)算方法,制定適合評(píng)價(jià)不同空間尺度水體的健康評(píng)價(jià)標(biāo)準(zhǔn)等級(jí);(3)生物基準(zhǔn)和標(biāo)準(zhǔn)與污染物總量控制.生物完整性基準(zhǔn)和標(biāo)準(zhǔn)成為關(guān)系到容量總量控制能否全面實(shí)施的關(guān)鍵要素之一,明確污染的生態(tài)效應(yīng),基于污染物總量控制制定相應(yīng)的合理生物完整性指數(shù)標(biāo)準(zhǔn),為污染物總量控制提供數(shù)據(jù)積累和理論支持.
參考文獻(xiàn):
[1]金小偉,王業(yè)耀,王備新,等.我國(guó)流域水生態(tài)完整性評(píng)價(jià)方法構(gòu)建 [J]. 中國(guó)環(huán)境監(jiān)測(cè), 2017,33(1):75-81.
[2]Hawkins C P, Olson J R, Hill R A. The reference condition:predicting benchmarks for ecological and water-quality assessments [J]. Journal of the North American Benthological Society, 2010,1(29):312-343.
[3]Karr J R. Biological integrity: a long-neglected aspect of water resource management [J]. Ecological Applications, 1991,1(1):66-84.
[4]Karr J R, Chu E W. Introduction: Sustaining living rivers [J].Hydrobiologia, 2000,422/423:1-14.
[5]Wright J, Moss D, Armitage P, et al. A preliminary classification of running-water sites in Great Britain based on macroinvertebrate species and the prediction of community type using environmental data [J]. Freshwater Biology, 1984,3(14):221-256.
[6]Furse M T, Moss D, Wright J F, et al. The influence of seasonal and taxonomic factors on the ordination and classification of running-water sites in Great Britain and on the prediction of their macro-invertebrate communities [J]. Freshwater Biology, 1984,14(3):257-280.
[7]Hawkins C P. Quantifying biological integrity by taxonomic completeness: its utility in regional and global assessments [J].Ecological Applications, 2006,4(16):1277-1294.
[8]Hering D, Feld C, Moog O, et al. Cook book for the development of a multimetric Index for biological condition of aquatic ecosystems: experiences from the European AQEM and STAR projects and related initiatives [J]. Hydrobiologia, 2006,566:311-324.
[9]Stoddard J L, Herlihy A T, Peck D V, et al. A process for creating multimetric indices for large-scale aquatic surveys [J]. Journal of the North American Benthological Society, 2008,4(27):878-891.
[10]Karr J R. Assessment of biotic integrity using fish communities[J]. Fisheries, 1981,6(6):21-27.
[11]Barbour M T, Gerritsen J, Snyder B D, et al. Rapid bioassessment protocols for use in streams and wadeable rivers: periphyton,benthic macroinvertebrates and fish [M]. Washington:Environmental Protection Agency, 1999.
[12]Stoddard J L, Larsen D P, Hawkins C P, et al. Setting expectations for the ecological condition of streams: the concept of reference condition [J]. Ecological Applications, 2006,4(16):1267-1276.
[13]王備新,楊蓮芳.溪流底棲動(dòng)物定量與半定量采樣法比較研究[J]. 應(yīng)用與環(huán)境生物學(xué)報(bào), 2006,12(5):719-721.
[14]趙 靜,章守宇,林 軍,等.不同采樣設(shè)計(jì)評(píng)估魚(yú)類群落效果比較 [J]. 應(yīng)用生態(tài)學(xué)報(bào), 2014,25(4):1181-1187.
[15]Cao Y, Hawkins C P. The comparability of bioassessments: a review of conceptual and methodological issues 1 [J]. Journal of the North American Benthological Society, 2011,30(3):680-701.
[16]Ostermiller J D, Hawkins C P. Effects of sampling error on bioassessments of stream ecosystems: application to RIVPACS-type models [J]. Journal of the North American Benthological Society, 2004,2(23):363-382.
[17]Chen K, Hughes R M, Wang B. Effects of fixed-count size on macroinvertebrate richness, site separation, and bioassessment of Chinese monsoonal streams [J]. Ecological Indicators, 2015,53:162-170.
[18]廖靜秋,黃 藝.應(yīng)用生物完整性指數(shù)評(píng)價(jià)水生態(tài)系統(tǒng)健康的研究進(jìn)展 [J]. 應(yīng)用生態(tài)學(xué)報(bào), 2013,24(1):295-302.
[19]Vannote R L, Minshall G W, Cummins K W, et al. The river continuum concept [J]. Canadian Journal of Fisheries and Aquatic Sciences, 1980,1(37):130-137.
[20]Marchant R, Hirst A, Norris R, et al. Classification of macroinvertebrate communities across drainage basins in Victoria,Australia: consequences of sampling on a broad spatial scale for predictive modelling [J]. Freshwater Biology, 1999,41(2):253-268.
[21]Vander Laan J J. Environmental assessment of streams: linking land use, instream stressors, and biological indices to infer likely causes of ecological impairment [D]. Logan:Utah State University, 2012.
[22]Omernik J M. Ecoregions of the conterminous United States [J].Annals of the Association of American geographers, 1987,77(1):118-125.
[23]Omernik J M. Ecoregions: a framework for managing ecosystems[J]. Journal of the American Water Resources Association, 1995,33(5):925-949.
[24]Hawkins C P, Vinson M R. Weak correspondence between landscape classifications and stream invertebrate assemblages:implications for bioassessment [J]. Journal of the North American Benthological Society, 2000,19(3):501-517.
[25]Hawkins C P, Cao Y, Roper B. Method of predicting reference condition biota affects the performance and interpretation of ecological indices [J]. Freshwater Biology, 2010,5(55):1066-1085.
[26]Hawkins C P, Norris R H. Performance of different landscape classifications for aquatic bioassessments: introduction to the series [J]. Journal of the North American Benthological Society,2000,19(3):367-369.
[27]Oberdorff T, Pont D, Hugueny B, et al. Development and validation of a fish-based index for the assessment of ‘river health’ in France [J]. Freshwater Biology, 2002,47(9):1720-1734.
[28]Cao Y, Hawkins C P, Olson J, et al. Modeling natural environmental gradients improves the accuracy and precision of diatom-based indicators [J]. Journal of the North American Benthological Society, 2007,26(3):566-585.
[29]Jones R C. Modeling to improve vegetation-based wetland biological assessment [D]. Logan:Utah State University, 2014.
[30]Pont D, Hugueny B, Beier U, et al. Assessing river biotic condition at a continental scale: a European approach using functional metrics and fish assemblages [J]. Journal of Applied Ecology, 2006,43(1):70-80.
[31]Moya N, Tomanova S, Oberdorff T. Initial development of a multi-metric index based on aquatic macroinvertebrates to assess streams condition in the Upper Isiboro-Sécure Basin, Bolivian Amazon [J]. Hydrobiologia, 2007,589(1):107-116.
[32]Moya N, Hughes R M, Domínguez E, et al. Macroinvertebratebased multimetric predictive models for evaluating the human impact on biotic condition of Bolivian streams [J]. Ecological Indicators, 2011,11(3):840-847.
[33]Vander Laan J J, Hawkins C P. Enhancing the performance and interpretation of freshwater biological indices: An application in arid zone streams [J]. Ecological Indicators, 2014,36:470-482.
[34]Chen K, Hughes R M, Xu S, et al. Evaluating performance of macroinvertebrate-based adjusted and unadjusted multi-metric indices (MMI) using multi-season and multi-year samples [J].Ecological Indicators, 2014,36:142-151.
[35]Esselman P C, Infante D M, Wang L, et al. Regional fish community indicators of landscape disturbance to catchments of the conterminous United States [J]. Ecological indicators, 2013,26:163-173.
[36]Cutler D R, Edwards Jr T C, Beard K H, et al. Random forests for classification in ecology [J]. Ecology, 2007,11(88):2783-2792.
[37]Breiman L. Random forests [J]. Machine Learning, 2001,1(45):5-32.
[38]Elith J, Leathwick J R, Hastie T. A working guide to boosted regression trees [J]. Journal of Animal Ecology, 2008,77(4):802-813.
[39]Mccormick F H, Hughes R M, Kaufmann P R, et al. Development of an index of biotic integrity for the Mid-Atlantic Highlands region [J]. Transactions of the American Fisheries Society, 2001,130(5):857-877.
[40]Barbour M T, Gerritsen J, Griffith G E, et al. A framework for biological criteria for Florida streams using benthic macroinvertebrates [J]. Journal of the North American Benthological Society, 1996,2(15):185-211.
[41]Stribling J B, Jessup B K, White J S, et al. Development of a benthic index of biotic integrity for Maryland streams [M].Annapolis: Maryland Department of Natural Resources, 1998.
[42]Van Sickle J. Correlated metrics yield multimetric indices with inferior performance [J]. Transactions of the American Fisheries Society, 2010,139(6):1802-1817.
[43]王備新.大型底棲無(wú)脊椎動(dòng)物水質(zhì)生物評(píng)價(jià)研究 [D]. 南京:南京農(nóng)業(yè)大學(xué), 2003.
[44]Blocksom K A. A performance comparison of metric scoring methods for a multimetric index for Mid-Atlantic Highlands streams [J]. Environmental Management, 2003,31(5):670-682.
[45]Birk S, Bonne W, Borja A, et al. Three hundred ways to assess Europe's surface waters: an almost complete overview of biological methods to implement the Water Framework Directive[J]. Ecological Indicators, 2012,18:31-41.
[46]King R S, Baker M E. Considerations for analyzing ecological community thresholds in response to anthropogenic environmental gradients [J]. Journal of the North American Benthological Society, 2010,29(3):998-1008.
[47]Mykr? H, Saarinen T, Tolkkinen M, et al. Spatial and temporal variability of diatom and macroinvertebrate communities: How representative are ecological classifications within a river system?[J]. Ecological Indicators, 2012,18:208-217.
[48]Clarke R T, Furse M T, Wright J F, et al. Derivation of a biological quality index for river sites: Comparison of the observed with the expected fauna [J]. Journal of Applied Statistics, 1996,2-3(23):311-332.
[49]Hawkins C P, Carlisle D M. Use of predictive models for assessing the biological integrity of wetlands and other aquatic habitats [M]. New York: John Wiley and Son, 2001:59-83.
[50]Bowman M F, Somers K M. Evaluating a novel Test Site Analysis (TSA) bioassessment approach [J]. Journal of the North American Benthological Society, 2006,25(3):712-727.
[51]Cao Y, Larsen D P, Thorne R S. Rare species in multivariate analysis for bioassessment: some considerations [J]. Journal of the North American Benthological Society, 2001,20(1):144-153.
[52]Moss D, Furse M T, Wright J F, et al. The prediction of the macro-invertebrate fauna of unpolluted running-water sites in Great Britain using environmental data [J]. Freshwater Biology,1987,1(17):41-52.
[53]Clarke R T, Wright J F, Furse M T. RIVPACS models for predicting the expected macroinvertebrate fauna and assessing the ecological quality of rivers [J]. Ecological Modelling, 2003,3(160):219-233.
[54]Ward Jr J H. Hierarchical grouping to optimize an objective function [J]. Journal of the American statistical association,1963,58(301):236-244.
[55]Moss D. An initial classification of 10-km squares in Great Britain from a land characteristic data bank [J]. Applied Geography, 1985,5(2):131-150.
[56]Mendes T, Calapez A R, Elias C L, et al. Comparing alternatives for combining invertebrate and diatom assessment in stream quality classification [J]. Marine and Freshwater Research, 2014,65(7):612-623.
[57]Van Sickle J, Hawkins C P, Larsen D P, et al. A null model for the expected macroinvertebrate assemblage in streams [J]. Journal of the North American Benthological Society, 2005,1(24):178-191.
[58]Hill R A, Hawkins C P. Using modelled stream temperatures to predict macro-spatial patterns of stream invertebrate biodiversity[J]. Freshwater Biology, 2014,59(12):2632-2644.
[59]Moss D. Evolution of statistical methods in RIVPACS [M].Ambleside: Freshwater Biological Association, 2000:25–38.
[60]Linke S, Bailey R C, Schwindt J. Temporal variability of stream bioassessments using benthic macroinvertebrates [J]. Freshwater Biology, 1999,42(3):575-584.
[61]Joy M K, Death R G. Predictive modelling and spatial mapping of freshwater fish and decapod assemblages using GIS and neural networks [J]. Freshwater Biology, 2004,49(8):1036-1052.
[62]Adriaenssens V, Goethals P, Charles J, et al. Application of Bayesian Belief Networks for the prediction of macroinvertebrate taxa in rivers [J]. Annales de Limnologie e International Journal of Limnology, 2004,40(3):181-191.
[63]Waite I R, Kennen J G, May J T, et al. Comparison of Stream Invertebrate Response Models for Bioassessment Metrics1 [J].Journal of the American Water Resources Association, 2012,48(3):570-583.
[64]Clarke R T, Murphy J F. Effects of locally rare taxa on the precision and sensitivity of RIVPACS bioassessment of freshwaters [J]. Freshwater Biology, 2006,51(10):1924-1940.
[65]Van Sickle J, Larsen D P, Hawkins C P. Exclusion of rare taxa affects performance of the O/E index in bioassessments [J].Journal of the North American Benthological Society, 2007,2(26):319-331.
[66]Hawkins C P, Olson J R, Hill R A. The reference condition:predicting benchmarks for ecological and water-quality assessments [J]. Journal of the North American Benthological Society, 2010,1(29):312-343.
[67]Reynoldson T B, Norris R H, Resh V H, et al. The reference condition: a comparison of multimetric and multivariate approaches to assess water-quality impairment using benthic macroinvertebrates [J]. Journal of the North American Benthological Society, 1997,16(4):833-852.
[68]Norris R H, Hawkins C P. Monitoring river health [J].Hydrobiologia, 2000,435(1-3):5-17.
[69]陳 凱.提高底棲動(dòng)物多參數(shù)(MMI)和多變量(O/E)指數(shù)評(píng)價(jià)溪流水質(zhì)能力的研究 [D]. 南京:南京農(nóng)業(yè)大學(xué), 2014.
[70]Miller D L, Hughes R M, Karr J R, et al. Regional applications of an index of biotic integrity for use in water resource management[J]. Fisheries, 1988,13(5):12-20.
[71]Steedman R J. Modification and assessment of an index of biotic integrity to quantify stream quality in southern Ontario [J].Canadian Journal of Fisheries and Aquatic Sciences, 1988,45(3):492-501.
[72]Schmutz S, Cowx I G, Haidvogl G, et al. Fish-based methods for assessing European running waters: a synthesis [J]. Fisheries Management and Ecology, 2007,14(6):369-380.
[73]Stoddard J L, Peck D V, Olsen A R, et al. Environmental Monitoring and Assessment Program (EMAP): western streams and rivers statistical summary [M]. Washington: Environmental Protection Agency, 2005.
[74]Rehn A C, Ode P R, Hawkins C P. Comparisons of targeted-riffle and reach-wide benthic macroinvertebrate samples: implications for data sharing in stream-condition assessments [J]. Journal of the North American Benthological Society, 2007,26(2):332-348.
[75]Beck M W, Hatch L K. A review of research on the development of lake indices of biotic integrity [J]. Environmental Reviews,2009,17:21-44.
[76]Drake M T, Valley R D. Validation and application of a fish-based index of biotic integrity for small central Minnesota lakes [J]. North American Journal of Fisheries Management,2005,25(3):1095-1111.
[77]Lunde K B, Resh V H. Development and validation of a macroinvertebrate index of biotic integrity (IBI) for assessing urban impacts to Northern California freshwater wetlands [J].Environmental Monitoring and Assessment, 2012,184(6):3653-3674.
[78]Weisberg S B, Ranasinghe J A, Dauer D M, et al. An estuarine benthic index of biotic integrity (B-IBI) for Chesapeake Bay [J].Estuaries, 1997,20(1):149-158.
[79]Williams M R, Filoso S, Longstaff B J, et al. Long-term trends of water quality and biotic metrics in Chesapeake Bay: 1986 to 2008[J]. Estuaries and Coasts, 2010,33(6):1279-1299.
[80]Klemm D J, Blocksom K A, Fulk F A, et al. Development and evaluation of a macroinvertebrate biotic integrity index (MBII)for regionally assessing Mid-Atlantic Highlands streams [J].Environmental Management, 2003,31(5):656-669.
[81]Wang Y, Stevenson R J, Metzmeier L. Development and evaluation of a diatom-based Index of Biotic Integrity for the Interior Plateau Ecoregion, USA [J]. Journal of the North American Benthological Society, 2005,24(4):990-1008.
[82]Kane D D, Gordon S I, Munawar M, et al. The Planktonic Index of Biotic Integrity (P-IBI): an approach for assessing lake ecosystem health [J]. Ecological Indicators, 2009,9(6):1234-1247.
[83]Feio M J, Aguiar F C, Almeida S, et al. AQUAFLORA: A predictive model based on diatoms and macrophytes for streams water quality assessment [J]. Ecological Indicators, 2012,18:586-598.
[84]Clapcott J E, Goodwin E O, Young R G, et al. A multimetric approach for predicting the ecological integrity of New Zealand streams [J]. Knowledge and Management of Aquatic Ecosystems,2014,415:3.
[85]Chen K, Hughes R M, Brito J G, et al. A multi-assemblage,multi-metric biological condition index for eastern Amazonia streams [J]. Ecological Indicators, 2017,78:48-61.
[86]Bramblett R G, Johnson T R, Zale A V, et al. Development and evaluation of a fish assemblage index of biotic integrity for northwestern Great Plains streams [J]. Transactions of the American Fisheries Society, 2005,134(3):624-640.
[87]Paulsen S, Stoddard J, Holdsworth S, et al. Wadeable streams assessment: a collaborative survey of the Nation's streams [M].Washington: Environmental Protection Agency, 2006.
[88]Harris J H, Silveira R. Large-scale assessments of river health using an Index of Biotic Integrity with low-diversity fish communities [J]. Freshwater Biology, 1999,41(2):235-252.
[89]Hering D, Moog O, Sandin L, et al. Overview and application of the AQEM assessment system [J]. Hydrobiologia, 2004,516(1-3):1-20.
[90]王備新,楊蓮芳,胡本進(jìn),等.應(yīng)用底棲動(dòng)物完整性指數(shù) B-IBI評(píng)價(jià)溪流健康 [J]. 生態(tài)學(xué)報(bào), 2005,25(6):1481-1490.
[91]鄭海濤.怒江中上游魚(yú)類生物完整性評(píng)價(jià) [D]. 武漢:華中農(nóng)業(yè)大學(xué), 2006.
[92]裴雪姣,牛翠娟,高 欣,等.應(yīng)用魚(yú)類完整性評(píng)價(jià)體系評(píng)價(jià)遼河流域健康 [J]. 生態(tài)學(xué)報(bào), 2010,30(21):5736-5746.
[93]曹艷霞,張 杰,蔡德所,等.應(yīng)用底棲無(wú)脊椎動(dòng)物完整性指數(shù)評(píng)價(jià)漓江水系健康狀況 [J]. 水資源保護(hù), 2010,26(2):13-17.
[94]張方方,張 萌,劉足根,等.基于底棲生物完整性指數(shù)的贛江流域河流健康評(píng)價(jià) [J]. 水生生物學(xué)報(bào), 2011,35(6):963-971.
[95]Wu N, Cai Q, Fohrer N. Development and evaluation of a diatom-based index of biotic integrity (D-IBI) for rivers impacted by run-of-river dams [J]. Ecological Indicators, 2012,18:108-117.
[96]張 遠(yuǎn),徐成斌,馬溪平,等.遼河流域河流底棲動(dòng)物完整性評(píng)價(jià)指標(biāo)與標(biāo)準(zhǔn) [J]. 環(huán)境科學(xué)學(xué)報(bào), 2007,27(6):919-927.
[97]渠曉東.香溪河大型底棲動(dòng)物時(shí)空動(dòng)態(tài)、生物完整性及小水電站的影響研究 [D]. 武漢:中國(guó)科學(xué)院水生生物研究所, 2006.
[98]蔡 琨,張 杰,徐兆安,等.應(yīng)用底棲動(dòng)物完整性指數(shù)評(píng)價(jià)太湖生態(tài)健康 [J]. 湖泊科學(xué), 2014,26(1):74-82.
[99]Huang Q, Gao J, Cai Y, et al. Development and application of benthic macroinvertebrate-based multimetric indices for the assessment of streams and rivers in the Taihu Basin, China [J].Ecological Indicators, 2015,48:649-659.
[100]Li F, Cai Q, Ye L. Developing a Benthic Index of Biological Integrity and some relationships to environmental factors in the subtropical Xiangxi River, China [J]. International Review of Hydrobiology, 2010,95(2):171-189.
[101]陳 凱,于海燕,張汲偉,等.基于底棲動(dòng)物預(yù)測(cè)模型構(gòu)建生物完整性指數(shù)(MMI)評(píng)價(jià)河流健康 [J]. 應(yīng)用生態(tài)學(xué)報(bào), 2017,6(28):1993-2002.
[102]Feio M J, Poquet J M. Predictive models for freshwater biological assessment: statistical approaches, biological elements and the Iberian Peninsula experience: a review [J]. International Review of Hydrobiology, 2011,96(4):321-346.
[103]Hawkins C P, Norris R H, Hogue J N, et al. Development and evaluation of predictive models for measuring the biological integrity of streams [J]. Ecological Applications, 2000,5(10):1456-1477.
[104]Sickle J V, Huff D D, Hawkins C P. Selecting discriminant function models for predicting the expected richness of aquatic macroinvertebrates [J]. Freshwater Biology, 2006,51(2):359-372.
[105]Karr, Ellen W. Chu J R. Sustaining living rivers [J].Hydrobiologia, 2000,422/423:1-14.
[106]Reynoldson T B, Bailey R C, Day K E, et al. Biological guidelines for freshwater sediment based on BEnthic Assessment of SedimenT (the BEAST) using a multivariate approach for predicting biological state [J]. Australian Journal of Ecology,1995,20(1):198-219.
[107]Smith M J, Kay W R, Edward D, et al. AusRivAS: using macroinvertebrates to assess ecological condition of rivers in Western Australia [J]. Freshwater Biology, 1999,41(2):269-282.
[108]Davis J, Horwitz P, Norris R, et al. Are river bioassessment methods using macroinvertebrates applicable to wetlands? [J].Hydrobiologia, 2006,572(1):115-128.
[109]Joy M K, Death R G. Predictive modelling of freshwater fish as a biomonitoring tool in New Zealand [J]. Freshwater Biology,2002,47(11):2261-2275.
[110]Chessman B, Growns I, Currey J, et al. Predicting diatom communities at the genus level for the rapid biological assessment of rivers [J]. Freshwater Biology, 1999,41(2):317-331.
[111]Aguiar F C, Feio M J, Ferreira M T. Choosing the best method for stream bioassessment using macrophyte communities: indices and predictive models [J]. Ecological Indicators, 2011,11(2):379-388.
[112]Mendes T, Calapez A R, Elias C L, et al. Comparing alternatives for combining invertebrate and diatom assessment in stream quality classification [J]. Marine and Freshwater Research, 2014,65(7):612-623.
[113]Davies N M, Norris R H, Thoms M C. Prediction and assessment of local stream habitat features using large-scale catchment characteristics [J]. Freshwater Biology, 2000,45(3):343-369.
[114]唐 濤,蔡慶華,劉建康.河流生態(tài)系統(tǒng)健康及其評(píng)價(jià) [J]. 應(yīng)用生態(tài)學(xué)報(bào), 2002,13(9):1191-1194.
[115]楊文慧,嚴(yán)忠民,吳建華.河流健康評(píng)價(jià)的研究進(jìn)展 [J]. 河海大學(xué)學(xué)報(bào)(自然科學(xué)版), 2005,33(6):5-9.
[116]吳阿娜,楊 凱,車 越,等.河流健康狀況的表征及其評(píng)價(jià) [J].水科學(xué)進(jìn)展, 2005,16(4):602-608.
[117]戴紀(jì)翠,倪晉仁.底棲動(dòng)物在水生生態(tài)系統(tǒng)健康評(píng)價(jià)中的作用分析 [J]. 生態(tài)環(huán)境, 2008,17(6):2107-2111.
[118]郭向楠,張曉冰,馬 濤.河流健康評(píng)估的研究與應(yīng)用進(jìn)展研究[J]. 環(huán)境科學(xué)與管理, 2013,38(10):170-174.
[119]鄧曉軍,許有鵬,翟祿新,等.城市河流健康評(píng)價(jià)指標(biāo)體系構(gòu)建及其應(yīng)用 [J]. 生態(tài)學(xué)報(bào), 2014,34(4):993-1001.
[120]張 杰,蔡德所,曹艷霞,等.評(píng)價(jià)漓江健康的 RIVPACS預(yù)測(cè)模型研究 [J]. 湖泊科學(xué), 2011,23(1):73-79.
[121]陳 凱,劉 祥,陳求穩(wěn),等.應(yīng)用 O/E 模型評(píng)價(jià)淮河流域典型水體底棲動(dòng)物完整性健康的研究 [J]. 環(huán)境科學(xué)學(xué)報(bào), 2016,36(7):2677-2686.
[122]孟 偉,秦延文,鄭丙輝,等.流域水質(zhì)目標(biāo)管理技術(shù)研究(Ⅲ)——水環(huán)境流域監(jiān)控技術(shù)研究 [J]. 環(huán)境科學(xué)研究, 2008,21(1):9-16.
[123]孟 偉,劉征濤,張 楠,等.流域水質(zhì)目標(biāo)管理技術(shù)研究(Ⅱ)——水環(huán)境基準(zhǔn)、標(biāo)準(zhǔn)與總量控制 [J]. 環(huán)境科學(xué)研究, 2008,21(1):1-8.