吳仁吉,康薩如拉*,張慶,2,任海娟,任婧,周俊梅,王珍,李丹,牛建明,2*(.內(nèi)蒙古大學(xué)生命科學(xué)學(xué)院,內(nèi)蒙古 呼和浩特0002;2.內(nèi)蒙古大學(xué)中美生態(tài)、能源與可持續(xù)性科學(xué)研究中心,內(nèi)蒙古 呼和浩特0002)
?
錫林河流域羊草草原植被分異的驅(qū)動(dòng)力
吳仁吉1,康薩如拉1*,張慶1,2,任海娟1,任婧1,周俊梅1,王珍1,李丹1,牛建明1,2*
(1.內(nèi)蒙古大學(xué)生命科學(xué)學(xué)院,內(nèi)蒙古 呼和浩特010021;2.內(nèi)蒙古大學(xué)中美生態(tài)、能源與可持續(xù)性科學(xué)研究中心,內(nèi)蒙古 呼和浩特010021)
植被與環(huán)境關(guān)系是群落生態(tài)學(xué)研究重要內(nèi)容,闡明氣候與放牧共同作用下草原植被分異,量化解析自然與人為干擾因子的影響具有理論與實(shí)踐價(jià)值。本研究以錫林河流域羊草草原為例,采用除趨勢(shì)典范對(duì)應(yīng)分析法(detrended canonical correspondence analysis,DCCA)和結(jié)構(gòu)方程模型(structural equation modeling,SEM),定量研究氣候、海拔、土壤及放牧因子對(duì)羊草群落分異的作用。結(jié)果表明,1)DCCA前兩個(gè)排序軸集中了67.63%的信息量,第1排序軸反映了放牧和水熱因子的作用,第2排序軸為土壤養(yǎng)分的影響;2)結(jié)構(gòu)方程模型分析發(fā)現(xiàn),羊密度對(duì)植被分異的總影響最大,其次為氣候因子;3)氣候、海拔和放牧因子對(duì)植被分異也有間接作用,主要途徑是影響土壤理化性質(zhì)。本研究明晰了放牧與氣候是主導(dǎo)錫林河流域羊草草原群落分異的關(guān)鍵因子,深入揭示了氣候因子在局地尺度所發(fā)揮的作用。在草地管理實(shí)踐中,結(jié)合氣候特征制定具體措施,有助于促進(jìn)草地資源的合理利用與保護(hù)。
錫林河流域;羊草草原;植被分異;結(jié)構(gòu)方程模型;除趨勢(shì)典范對(duì)應(yīng)分析
植被是指一個(gè)地區(qū)所有植物群落的總和,亦能夠指示一定地域的生態(tài)環(huán)境特征[1-2],研究植被與環(huán)境關(guān)系具有重要的理論與實(shí)踐價(jià)值[3-5]。植物群落的種類組成、結(jié)構(gòu)與外貌等特征不僅是大氣候作用的產(chǎn)物,也受到地形、土壤、放牧、火燒等因子的綜合影響[2]。前人普遍利用多元分析方法探討植被與環(huán)境之間的關(guān)系[5-9],其中排序和分類等方法的應(yīng)用已經(jīng)很成熟[6,10-11]。例如,李紹忠[12]對(duì)遼寧省白石砬子自然保護(hù)區(qū)內(nèi)的森林類型進(jìn)行了分類和排序,發(fā)現(xiàn)海拔與RA(reciprocal averaging)第二排序軸有相關(guān)性;張新時(shí)[13]利用DCA(detrended correspondence analysis)與TWINSPAN方法,揭示出熱量與濕度梯度對(duì)西藏阿里地區(qū)的植被分布起主導(dǎo)作用;張金屯[14]則將排序(principal component analysis,PCA)、分類(TWINSPAN)與多元回歸分析方法相結(jié)合,闡明了山西植被地帶分布與氣候的關(guān)系。然而,無(wú)論是大尺度還是局地尺度的植被分異,都是環(huán)境因子共同作用的結(jié)果,直接作用與間接作用并存[15-19]。以放牧脅迫為例,牲畜的采食和踐踏直接作用于植株個(gè)體,另一方面,踐踏和養(yǎng)分回歸改變了土壤環(huán)境,對(duì)群落演替的間接影響也十分重要[20-22]。上述傳統(tǒng)的植被環(huán)境關(guān)系分析方法對(duì)于多個(gè)原因、多個(gè)結(jié)果、測(cè)量中的誤差以及同一因子的不同路徑很難同時(shí)處理[23-25]。因此,有必要引入新的方法,更為深入地剖析與量化環(huán)境因子對(duì)植被作用的路徑關(guān)系。
結(jié)構(gòu)方程模型(structural equation modeling,SEM)綜合了因子分析、回歸分析、路徑分析等多變量統(tǒng)計(jì)方法,能夠同時(shí)探討自然系統(tǒng)內(nèi)變量間的“因果”關(guān)系強(qiáng)弱以及不同因子對(duì)同一過(guò)程直接與間接影響[26]。Grace等[27]利用SEM分析了生物多樣性、生產(chǎn)力、環(huán)境條件和干擾因子之間的關(guān)系,發(fā)現(xiàn)在小尺度上,生物多樣性對(duì)生產(chǎn)力的影響較弱。SEM在生態(tài)學(xué)領(lǐng)域中的應(yīng)用已有較多報(bào)道[26,28-31],呈現(xiàn)出強(qiáng)勁的發(fā)展勢(shì)頭。
本研究以內(nèi)蒙古錫林河流域羊草草原(Leymuschinensis)為例,基于69個(gè)樣地的野外調(diào)查數(shù)據(jù),利用除趨勢(shì)典范對(duì)應(yīng)分析(detrended canonical correspondence analysis,DCCA)與結(jié)構(gòu)方程模型(SEM)相結(jié)合的方法,試圖通過(guò)量化羊草草原不同退化演替階段群落與環(huán)境因子之間的關(guān)系,揭示驅(qū)動(dòng)羊草草原退化的主導(dǎo)因子以及路徑關(guān)系,加深對(duì)典型草原退化演替機(jī)制的認(rèn)識(shí),并為退化草地恢復(fù)與草地資源的合理利用提供科學(xué)依據(jù)。
1.1 研究區(qū)概況

圖1 樣地分布Fig.1 The distribution of samples
研究區(qū)位于內(nèi)蒙古錫林河流域。地理位置介于北緯43.41°-44.13°;東經(jīng)116.12°-117.24°之間,海拔在1000~1500 m。該地區(qū)屬溫帶大陸性氣候,年平均氣溫0.88 ℃[32]。年平均降水量338 mm。土壤以栗鈣土為主,以典型草原占優(yōu)勢(shì),羊草草原主要分布在地形較為平坦的傾斜平原、寬谷與丘間。
1.2 野外取樣
取樣時(shí)間為2014年7月末至8月初,共設(shè)置了69個(gè)樣地,均處在地形平坦的地段,兩個(gè)樣地之間距離不小于5 km(圖1)。每個(gè)樣地記錄地理和群落外貌,并設(shè)置3個(gè)1 m×1 m的樣方,記錄每個(gè)植物種的高度與株叢數(shù),并分種齊地面剪取地上部分,稱取鮮重,帶回實(shí)驗(yàn)室,用65 ℃烘箱烘干至恒重之后(約48 h)得到干重。在3個(gè)測(cè)產(chǎn)樣方內(nèi),采用土鉆法,每個(gè)樣方各鉆取0~10 cm、10~20 cm和20~30 cm三個(gè)土層的土樣,并將同一樣地的3個(gè)樣方鉆取的土樣按同一土層進(jìn)行混合,帶回實(shí)驗(yàn)室測(cè)定土壤養(yǎng)分與機(jī)械組成。
1.3 土壤樣品測(cè)試
測(cè)試的土壤養(yǎng)分包括土壤全氮(total nitrogen,TN)、全磷(total phosphorus,TP)、速效氮(available nitrogen,AN)、速效磷(available phosphorus,AP)和有機(jī)碳(organic carbon,OC),土壤機(jī)械組成包括石礫(gravel)、砂粒(sand)、黏粒(silt)以及粉粒(clay)的含量。土壤全氮含量測(cè)定采用凱氏定氮法;全磷采用硫酸-氫氧化鈉熔融-鉬銻抗比色法;速效氮(水解性氮)采用硼酸-氫氧化鈉堿解土壤-擴(kuò)散吸收法;速效磷采用碳酸氫鈉浸提-鉬銻抗比色法;有機(jī)碳采用重鉻酸鉀氧化外加熱法[33]。土壤機(jī)械組成采用土壤粒度儀(S3500,美國(guó))進(jìn)行測(cè)定。
1.4 環(huán)境因子的獲取
氣候因子包括1月平均氣溫(average temperature in January,T1)、7月平均氣溫(average temperature in July,T7)、年平均氣溫(annual average temperature,T)、年平均降水量(average annual precipitation,P)、干燥度(K),通過(guò)內(nèi)蒙古氣候因子空間模型[34]計(jì)算獲得。海拔(altitude,ALT)數(shù)據(jù)源自研究區(qū)數(shù)字高程模型(digital elevation model,DEM),數(shù)據(jù)下載于地理空間數(shù)據(jù)云(http://www.gscloud.cn/)網(wǎng)站,通過(guò)相應(yīng)樣地的地理位置讀取。氣候因子和海拔指標(biāo)的提取均采用ArcGIS 9.3。
放牧因子采用羊密度(sheep density)表示,數(shù)據(jù)下載于聯(lián)合國(guó)糧農(nóng)組織(http://data.fao.org/mapwas/)網(wǎng)站。
1.5 數(shù)據(jù)分析
統(tǒng)計(jì)原始樣方資料,以相對(duì)干重為指標(biāo)。由于一、二年生植物的種類與生物量明顯受到降水的控制,一般不參與群落分析,因此在剔除掉一、二年生植物后,構(gòu)建了69個(gè)樣地×91個(gè)物種的樣方——物種矩陣。
采用Canoco for Windows 4.5對(duì)樣方——物種矩陣、氣候因子與海拔矩陣、土壤因子矩陣、放牧因子矩陣進(jìn)行DCCA分析,獲得前兩個(gè)排序軸信息量的大小[19,34]。從而分析氣候、海拔、土壤、放牧及其交互作用與群落演替的關(guān)系。
基于DCCA前兩個(gè)排序軸得分,利用AMOS 17.0構(gòu)建氣候與海拔、土壤、放牧等因子與群落演替路徑關(guān)系的結(jié)構(gòu)方程模型。在剔除了T1、T7、AP、OC、石礫以及砂粒等不能通過(guò)模型檢驗(yàn)的因子后,采用最大似然法估計(jì)路徑系數(shù)值及參數(shù),通過(guò)結(jié)構(gòu)方程模型的擬合指數(shù)來(lái)檢驗(yàn)和評(píng)價(jià)模型的擬合程度。本研究選取絕對(duì)擬合指數(shù)、相對(duì)擬合指數(shù)和代替性指標(biāo)等3類指標(biāo),其中絕對(duì)擬合指數(shù)包括卡方自由度比x2/df(Chi square degrees of freedom)、擬合優(yōu)度指數(shù)GFI(goodness of fit index)以及簡(jiǎn)約擬合優(yōu)度指數(shù)PGFI(parsimony goodness of fit index);相對(duì)擬合指數(shù)包括規(guī)范擬合指數(shù)NFI(normed fit index)、增量擬合指數(shù)(incremental fit index)以及簡(jiǎn)約規(guī)范擬合指數(shù)PNFI(parsimony normed fit index);代替性指標(biāo)包括比較擬合指數(shù)CFI(comparative fit index)、簡(jiǎn)約比較擬合指數(shù)PCFI(parsimony comparative fit index)[23,36-37],調(diào)整參數(shù)以修正模型,定量分析諸因子對(duì)群落演替的直接與間接影響。
2.1 DCCA排序
錫林河流域羊草草原植物群落與氣候、海拔、土壤及放牧因子的DCCA排序見(jiàn)圖2,排序軸與諸因子相關(guān)分析結(jié)果見(jiàn)表1。DCCA第一排序軸占總信息量的38.75%,第二排序軸占總信息量的28.86%,前兩個(gè)排序軸合計(jì)達(dá)到總信息量的67.63%,能夠代表樣本總體[38-42]。從圖2可以看出,研究區(qū)域羊草草原被劃分為4種群落類型,即羊草+黃囊苔草(Carexkorshinskyi)+克氏針茅(Stipakrylovii)群落(Ⅰ)、羊草+克氏針茅+糙隱子草(Cleistogenessquarrosa)群落(Ⅱ)、羊草+糙隱子草+大針茅(S.grandis)群落(Ⅲ)、羊草+冷蒿(Artemisiafrigida)+糙隱子草群落(Ⅳ)。雖然各類型之間的界限不明顯,但對(duì)群落演替仍有所體現(xiàn),DCCA1反映了由羊草+大針茅群落(Ⅲ)向羊草+克氏針茅群落(Ⅰ與Ⅱ)的退化, DCCA2則與土壤基質(zhì)變化有關(guān),羊草+冷蒿群落(Ⅳ)得分較高。進(jìn)一步由表1可知,與第1排序軸存在極顯著相關(guān)性的因子包括7月平均溫度、年平均溫度、年平均降水量、干燥度指數(shù)、海拔以及羊密度因子,且羊密度因子相關(guān)系數(shù)最高;所有環(huán)境因子與第二軸均沒(méi)有顯著相關(guān)性。可見(jiàn)環(huán)境因子對(duì)群落演替解釋的貢獻(xiàn)率集中在第一排序軸,羊密度因子是主導(dǎo)因子。

表1 DCCA排序軸與環(huán)境因子的相關(guān)系數(shù)Table 1 Correlation coefficients of DCCA axes with environmental factors
**表示P<0.01;*表示P<0.05。 ** indicateP<0.01; * indicateP<0.05.T1:1月平均氣溫 Average temperature in January;T7:7月平均氣溫Average temperature in July;T:年平均氣溫 Annual mean temperature;P:年平均降水量 Annual mean precipitation;K:干燥度 Drought index;ALT:海拔Altitude;TN:全氮 Total nitrogen;AN:速效氮 Available nitrogen;TP:全磷 Total phosphorus;AP:速效磷 Available phosphorus;OC:有機(jī)碳 Organic carbon.下同 The same below.

圖2 樣地DCCA排序Fig.2 The result of DCCA ordination Ⅰ:羊草 L. chinensis+黃囊苔草 Carex korshinskyi+克氏針茅 Stipa krylovii;Ⅱ:羊草 L. chinensis+克氏針茅 S. krylovii+糙隱子草Cleistogenes squarrosa;Ⅲ:羊草 L. chinensis+糙隱子草 C. squarrosa+大針茅 S. grandis;Ⅳ:羊草 L. chinensis+冷蒿 Artemisia frigida+糙隱子草 C. squarrosa.
2.2 結(jié)構(gòu)方程模型的構(gòu)建與檢驗(yàn)
錫林河流域羊草草原植被與環(huán)境因子間相互關(guān)系的結(jié)構(gòu)方程模型見(jiàn)圖3。檢驗(yàn)(表2)表明,該結(jié)構(gòu)方程模型比較合理,能夠支持初始的目標(biāo),較好地反映錫林河流域羊草草原植被特性與環(huán)境因子之間相互關(guān)系。
2.3 群落演替驅(qū)動(dòng)力解析
通過(guò)圖3可知,預(yù)設(shè)模型中氣候、海拔、土壤以及放牧因子對(duì)羊草草原群落分異均有直接影響,其中氣候因子的直接影響路徑系數(shù)最高(0.70),放牧因子次之(-0.57)。然而,分析氣候、土壤以及放牧因子對(duì)植被的標(biāo)準(zhǔn)化總影響、標(biāo)準(zhǔn)化直接影響以及標(biāo)準(zhǔn)化間接影響路徑系數(shù)(表3)發(fā)現(xiàn),放牧因子的總影響路徑系數(shù)最高。另一方面,氣候、海拔以及放牧因子通過(guò)作用于土壤對(duì)羊草草原群落演替有著間接影響,其中海拔的間接影響路徑系數(shù)最高,放牧因子最低。

表2 模型擬合指數(shù)Table 2 Fitting index of SEM model

圖3 環(huán)境因子與植被特性的結(jié)構(gòu)方程模型Fig.3 Structural equation model of environmental factors and vegetation characteristics 箭頭表示因子之間的“依賴”關(guān)系,路徑系數(shù)表示兩個(gè)相關(guān)因子之間直接“依賴”關(guān)系的強(qiáng)弱。The arrow indicates the factor between the “dependency” relationship; the path coefficients indicate the strength of the correlation between the two factors directly “dependency” relationship.

表3 各因子影響植被特性的標(biāo)準(zhǔn)化路徑系數(shù)Table 3 Standardized path coefficients of environmental factors affecting vegetation characteristics
一般地,氣候是大尺度控制草原植被分異的主導(dǎo)因子,在局地尺度上,決定草原植物群落組成與結(jié)構(gòu)的因子多以土壤以及放牧等人為干擾為主,特別是由于放牧導(dǎo)致的退化演替[19,43-45]。氣候因子對(duì)植被特性的路徑系數(shù)比較高(表3),表明氣候?qū)﹀a林河流域羊草草原植物群落分異具有重要影響,因此有必要進(jìn)一步審視草原植被—環(huán)境關(guān)系中氣候的地位與作用。事實(shí)上,研究區(qū)域典型草原植物群落的組成與結(jié)構(gòu)對(duì)氣候因子改變的響應(yīng)可能是十分敏感的,地形導(dǎo)致的植被分異就是例證之一。眾所周知,地形使得水熱發(fā)生了再分配,大的地形地貌變化甚至產(chǎn)生了植被地帶的空間替代[46]。盡管錫林河流域地形起伏較小,但對(duì)于典型草原分布的影響也是明顯的,丘陵的陰坡、陽(yáng)坡與丘間發(fā)育了不同的群落類型[31,47-48]。不同的草地利用方式及強(qiáng)度與地形效應(yīng)相結(jié)合,導(dǎo)致土壤特征發(fā)生改變[31],也調(diào)控著植物群落的構(gòu)建。并且,驅(qū)動(dòng)研究區(qū)域氣候因子發(fā)生變化的不僅僅是地形。由于錫林河流域地處大興安嶺南段余脈的北坡,由南向北,不僅海拔逐漸下降,同時(shí)年平均降水量與年平均氣溫也呈現(xiàn)出梯度變化的趨勢(shì)[49-50],年平均降水量由410 mm減少到260 mm,年平均氣溫0.5 ℃升高到2.1 ℃[51-53]。綜上所述,基于典型草原植被對(duì)氣候因子敏感性的認(rèn)知,結(jié)合研究區(qū)域氣候變異性的實(shí)際,有理由認(rèn)為氣候因子對(duì)羊草草原植物群落分異具有主導(dǎo)作用。
本研究表明,放牧是驅(qū)動(dòng)羊草草原群落分異的最主要因子,這與該地區(qū)放牧活動(dòng)頻繁與劇烈的實(shí)際相符合[51],也與前人研究的主流結(jié)論相一致[20,54-55]。在放牧過(guò)程中,草食動(dòng)物通過(guò)采食和踐踏途徑直接作用于草地,強(qiáng)度放牧增加了適口性差與有毒有害牧草,降低了優(yōu)勢(shì)植物的地位,從而改變了群落的組成與結(jié)構(gòu)[21,55-57];另一方面,踐踏與養(yǎng)分回歸過(guò)程顯著影響土壤有機(jī)質(zhì)含量、土壤水分及土壤pH值等,其對(duì)草地植被的間接影響甚至超過(guò)了直接影響[54,58-59]。本研究量化了放牧因子對(duì)羊草草原群落的直接影響與間接影響,但是間接影響非常微弱,特別是放牧對(duì)土壤的直接影響路徑系數(shù)僅為0.04。顯然,其結(jié)果不僅與前人的很多研究結(jié)論不符[20-21,59],而且也確實(shí)低估了間接作用。究其原因,我們認(rèn)為可能與使用羊密度數(shù)據(jù)代替放牧強(qiáng)度有關(guān)。羊密度數(shù)據(jù)是依據(jù)草畜供需關(guān)系的狀態(tài)參量計(jì)算而來(lái)(http://data.fao.org/mapwas/),因此草地地上生物量的高低對(duì)于該指標(biāo)的大小有決定性作用。在錫林河流域,草地地上生物量不僅代表了食草動(dòng)物的采食,更是植被特性的表征,因此,本研究中放牧因子對(duì)植被特性的直接影響實(shí)質(zhì)上包含了植被本身的信息,一方面增加了數(shù)據(jù)冗余,削弱了放牧本身的獨(dú)立性與作用;另一方面,掩蓋了放牧對(duì)土壤的直接影響。在今后的研究中,迫切需要采用能夠精準(zhǔn)量化放牧作用的指標(biāo)。
結(jié)構(gòu)方程模型在21世紀(jì)初開(kāi)始在我國(guó)的經(jīng)濟(jì)學(xué)、教育學(xué)研究中得到應(yīng)用,而在生態(tài)學(xué)中的應(yīng)用剛剛起步[25-26]。在本研究中,結(jié)構(gòu)方程模型不僅定量描述了氣候因子、海拔、土壤因子與放牧因子對(duì)羊草草原植被特性的直接影響,也辨析了間接影響路徑,比較準(zhǔn)確地揭示了環(huán)境因子對(duì)羊草草原群落分異的復(fù)雜作用,因此可以成為深入探討植被環(huán)境關(guān)系的有效手段。然而,本研究也發(fā)現(xiàn)一些應(yīng)用問(wèn)題,主要表現(xiàn)為模型的擬合指數(shù)沒(méi)有達(dá)到理論上的最佳擬合標(biāo)準(zhǔn),其原因可能是樣本量相對(duì)較少。在經(jīng)濟(jì)學(xué)、教育學(xué)和心理學(xué)等領(lǐng)域,樣本量達(dá)到幾百甚至上千,提升了通過(guò)率及擬合矯正效果[23-24],而生態(tài)學(xué)研究中獲取如此多的野外觀測(cè)樣本難度極大。為此,在具體計(jì)算時(shí),刪除了不能通過(guò)模型檢驗(yàn)或者顯著性低的因子,使得觀測(cè)變量與樣本量的比值盡可能降低,適當(dāng)?shù)貜浹a(bǔ)了因樣地?cái)?shù)目不足產(chǎn)生的缺陷。
References:
[1] Cheng Z H, Zhang J T. Impacts of tousist development on vegetation in Tianlong mountains. Scientia Geographica Sinica, 2000, 2: 8.
[2] Zhang X S. The potential evapotranspiration(pe)index for vegetation and vegetation-climatic classification(1)-an introduction of main metliods and pep program. Acta Phytocologica et Geobotanica Sinica, 1989, (1): 1-9. 張新時(shí). 植被的pe(可能蒸散)指標(biāo)與植被-氣候分類(一)——幾種主要方法與pep程序介紹. 植物生態(tài)學(xué)與地植物學(xué)學(xué)報(bào), 1989, (1): 1-9.
[3] Cowles H C. The Ecological Relations of the Vegetation on the Sand Dunes of Lake Michigan[M]. Chicago: University of Chicago Press, 1899.
[4] Guisan A, Zimmermann N E. Predictive habitat distribution models in ecology. Ecological Modelling, 2000, 135(2): 147-186.
[5] He M, Zheng J, Li X,etal. Environmental factors affecting vegetation composition in the Alxa Plateau, China. Journal of Arid Environments, 2007, 69(3): 473-489.
[6] Niu J M, Hu H. The review of study on relationship between vegetation and environments in China. Acta Scientiarum Naturalium Universitatis NeiMongol, 2000, (1): 76-80. 牛建明, 呼和. 我國(guó)植被與環(huán)境關(guān)系研究進(jìn)展. 內(nèi)蒙古大學(xué)學(xué)報(bào)(自然科學(xué)版), 2000, (1): 76-80.
[7] Woodward F I. Climate and Plant Distribution[M]. Cambridge: Cambridge University Press, 1987.
[8] Box E O. Plant functional types and climate at the global scale. Journal of Vegetation Science, 1996, 7(3): 309-320.
[9] Ellenberg H. Vegetation Ecology of Central Europe[M]. Cambridge: Cambridge University Press, 1988.
[10] Zhang J T. Fuzzy set ordination and its application. Acta Ecologica Sinica, 1992, (4): 325-331. 張金屯. 模糊數(shù)學(xué)排序及其應(yīng)用. 生態(tài)學(xué)報(bào), 1992, (4): 325-331.
[11] Zhang J T. Analysis of relationshlls between vegetation and its environmental variables.II.Cca and dcca constrained ordinations. Journal of Shanxi University: Natural Science Edition, 1992, (3): 292-298. 張金屯. 植被與環(huán)境關(guān)系的分析ⅱ: CCA和DCCA限定排序. 山西大學(xué)學(xué)報(bào): 自然科學(xué)版, 1992, (3): 292-298.
[12] Li S Z. 0rdination and classifieation of forest communities at Baishilazi natural reserve in Liaoning. Chinese Journal of Ecology, 1985, (4): 13-15, 20. 李紹忠. 遼寧省白石磖子自然保護(hù)區(qū)森林的排序和分類. 生態(tài)學(xué)雜志, 1985, (4): 13-15, 20.
[13] Zhang X S. Indirect gradient analysis,quantitative classificaion and environmental interpretation of plant communities in Ngari,Xizang(tibet). Acta Phytocologica et Geobotanica Sinica, 1991, (2): 101-113. 張新時(shí). 西藏阿里植物群落的間接梯度分析、數(shù)量分類與環(huán)境解釋. 植物生態(tài)學(xué)與地植物學(xué)學(xué)報(bào), 1991, (2): 101-113.
[14] Zhang J T. Canonical principal component analysis and its application to the study of vegetation-climate relationships in Shanxi, China. Acta Geographica Sinica, 1998, (3): 66-73. 張金屯. 典范主分量分析及其在山西植被與氣候關(guān)系分析中的應(yīng)用. 地理學(xué)報(bào), 1998, (3): 66-73.
[15] Burke A. Classification and ordination of plant communities of the Naukluft Mountains, Namibia. Journal of Vegetation Science, 2001, 12(1): 53-60.
[16] Miao S, Carstenn S, Nungesser M. Real World Ecology[M]. New York: Springer, 2009.
[17] Clark J S. Models for Ecological Data: An Introduction[M]. New Jersey, USA: Princeton University Press Princeton, 2007.
[18] Swanson F, Kratz T, Caine N,etal. Landform effects on ecosystem patterns and processes. BioScience, 1988, 38(2): 92-98.
[19] Shen Z, Zhang X. The spatial pattern and topographic interpretation of the forest vegetation at Dalaoling region in the three gorges. Acta Botanica Sinica, 2000, 42(10): 1089-1095.
[20] Wang R Z, Li J D. The influence of grazing on the aneurolepidium chinense grassland in Song-nen plain. Pratacultural Science, 1992, (2): 11-14. 王仁忠, 李建東. 放牧對(duì)松嫩平原羊草草地影響的研究. 草業(yè)科學(xué), 1992, (2): 11-14.
[21] Li Y H, Wang S P. Response of plant and plant community to different stocking rates. Grassland of China, 1999, (3): 12-20. 李永宏, 汪詩(shī)平. 放牧對(duì)草原植物的影響. 中國(guó)草地, 1999, (3): 12-20.
[22] He G Y, Sun H Z, Shi X M,etal. Soil properties of Tibetan Plateau alpine wetland affected by grazing and season. Acta Prataculturae Sinica, 2015, 24(4): 12-20. 何貴永, 孫浩智, 史小明, 等. 青藏高原高寒濕地不同季節(jié)土壤理化性質(zhì)對(duì)放牧模式的響應(yīng). 草業(yè)學(xué)報(bào), 2015, 24(4): 12-20.
[23] Cheng K M. Features and application of structural equation modeling. Statistics and Decision, 2006, (10): 22-25. 程開(kāi)明. 結(jié)構(gòu)方程模型的特點(diǎn)及應(yīng)用. 統(tǒng)計(jì)與決策, 2006, (10): 22-25.
[24] Zhou T, Lu Y B. Structural equation modeling and its applications in empirical analyses. Industrial Engineering and Management, 2006, (5): 99-102. 周濤, 魯耀斌. 結(jié)構(gòu)方程模型及其在實(shí)證分析中的應(yīng)用. 工業(yè)工程與管理, 2006, (5): 99-102.
[25] Li H, Wang J K, Pei J B,etal. Equilibrium relationships of soil organic carbon in the main croplands of Northeast China based on structural equation modeling. Acta Ecologica Sinica, 2015, (2): 517-525. 李慧, 汪景寬, 裴久渤, 等. 基于結(jié)構(gòu)方程模型的東北地區(qū)主要旱田土壤有機(jī)碳平衡關(guān)系研究. 生態(tài)學(xué)報(bào), 2015, (2): 517-525.
[26] Wang Y S, Chu C J. A brief introduction of structural equation model and its application in ecology. Chinese Journal of Plant Ecology, 2011, (3): 337-344. 王酉石, 儲(chǔ)誠(chéng)進(jìn). 結(jié)構(gòu)方程模型及其在生態(tài)學(xué)中的應(yīng)用. 植物生態(tài)學(xué)報(bào), 2011, (3): 337-344.
[27] Grace J B, Michael Anderson T, Smith M D,etal. Does species diversity limit productivity in natural grassland communities. Ecology Letters, 2007, 10(8): 680-689.
[28] Grace J B, Allain L, Allen C. Factors associated with plant species richness in a coastal tall-grass prairie. Journal of Vegetation Science, 2000, 11(3): 443-452.
[29] Iriondo J M, Albert M J, Escudero A. Structural equation modelling: An alternative for assessing causal relationships in threatened plant populations. Biological Conservation, 2003, 113(3): 367-377.
[30] Lamb E G, Cahill Jr J F. When competition does not matter: Grassland diversity and community composition. The American Naturalist, 2008, 171(6): 777-787.
[31] Laughlin D C, Abella S R, Covington W W,etal. Species richness and soil properties in pinus ponderosa forests: A structural equation modeling analysis. Journal of Vegetation Science, 2007, 18(2): 231-242.
[32] Barthold F, Wiesmeier M, Breuer L,etal. Land use and climate control the spatial distribution of soil types in the grasslands of Inner Mongolia. Journal of Arid Environments, 2013, 88: 194-205.
[33] Nanjing Agricultural University. Soil Agricultural Chemistry Analysis[M]. 2nd edn. Beijing: Agriculture Press, 1981. 南京農(nóng)業(yè)大學(xué). 土壤農(nóng)化分析[M]. 第二版. 北京: 農(nóng)業(yè)出版社, 1981.
[34] Niu J M. Relationship between main vegetation types and climatic factors in Inner Mongolia. Chinese Journal of Applied Ecology, 2000, (1): 47-52. 牛建明. 內(nèi)蒙古主要植被類型與氣候因子關(guān)系的研究. 應(yīng)用生態(tài)學(xué)報(bào), 2000, (1): 47-52.
[35] Xin X P, Gao Q, Li Z Q,etal. Partitioning the spatial and environmental variations of plant community structure of alkaline grassland on Songnen plain. Acta Botanica Sinica, 1999, (7): 102-108. 辛?xí)云? 高瓊, 李鎮(zhèn)清, 等. 松嫩平原堿化草地植物群落分布的空間和環(huán)境因素分析. 植物學(xué)報(bào), 1999, (7): 102-108.
[36] Su X, Wang J J, Guo M C,etal. Coupling relationship of agricultural eco-economic system in Wuqi county based on structural equation model. Chinese Journal of Applied Ecology, 2010, (4): 937-944. 蘇鑫, 王繼軍, 郭滿才, 等. 基于結(jié)構(gòu)方程模型的吳起縣農(nóng)業(yè)生態(tài)經(jīng)濟(jì)系統(tǒng)耦合關(guān)系. 應(yīng)用生態(tài)學(xué)報(bào), 2010, (4): 937-944.
[37] Zhang Y L, Zhang J, Zhang H L,etal. Impact of culture and natural disasters on residents’ behaviors toward ecoenvironmental conservation: Sichuan province case studies. Acta Ecologica Sinica, 2014, (17): 5103-5113. 張玉玲, 張捷, 張宏磊, 等. 文化與自然災(zāi)害對(duì)四川居民保護(hù)旅游地生態(tài)環(huán)境行為的影響. 生態(tài)學(xué)報(bào), 2014, (17): 5103-5113.
[38] Yue Y M, Wang K L, Zhang W,etal. Relationships between soil and environment in peak-cluster depression areas of Karst region based on canonical correspondence analysis. Enviromental Science, 2008, (5): 1400-1405. 岳躍民, 王克林, 張偉, 等. 基于典范對(duì)應(yīng)分析的喀斯特峰叢洼地土壤-環(huán)境關(guān)系研究. 環(huán)境科學(xué), 2008, (5): 1400-1405.
[39] Song T Q, Peng W X, Zeng F P,etal. Spatial pattern of forest communities and environmental interpretation in Mulun national nature reserve, Karst cluster-peak depression region. Chinese Journal of Plant Ecology, 2010, (3): 298-308. 宋同清, 彭晚霞, 曾馥平, 等. 木論喀斯特峰叢洼地森林群落空間格局及環(huán)境解釋. 植物生態(tài)學(xué)報(bào), 2010, (3): 298-308.
[40] Ter Braak C J. Canonical correspondence analysis: A new eigenvector technique for multivariate direct gradient analysis. Ecology, 1986, 67(5): 1167-1179.
[41] Ter Braak C J. Canonical community ordination. Part i: Basic theory and linear methods. Ecoscience, 1994, (1): 127-140.
[42] Zhang Q, Niu J M, Alexander B,etal. Ecological analysis and classification ofStipabrevifloracommunities in the Inner Mongolia region: The role environment factors. Acta Prataculturae Sinica, 2012, 21(1): 83-92. 張慶, 牛建明, Alexander B, 等. 內(nèi)蒙古短花針茅群落數(shù)量分類及環(huán)境解釋. 草業(yè)學(xué)報(bào), 2012, 21(1): 83-92.
[43] Jiao J Y, Ma X H, Bai W J,etal. Correspondence analysis of vegetation communities and soil environmental factors on abandoned cropland on hilly-gullied loess plateau. Acta Pedologica Sinica, 2005, 42(5): 752.
[44] Woodward F, Mckee I. Vegetation and Climate. Environment International, 1991, 17(6): 535-546.
[45] Niu J M, Li B. Multivariate analysis of vegetation and ecological factors on the Ordos plateau,Inner Mongolia. Acta Ecologica Sinica, 1992, (2): 105-112. 牛建明, 李博. 鄂爾多斯高原植被與生態(tài)因子的多元分析. 生態(tài)學(xué)報(bào), 1992, (2): 105-112.
[46] Li G Y, Jiang C H, Cheng T,etal. Spatial-temporal variation of vegetation phenology and their relationships with vegetation degradation in a Qinghai lake watershed. Acta Prataculturae Sinica, 2016, 25(1): 22-32. 李廣泳, 姜翠紅, 程滔, 等. 青海湖流域植被物候格局時(shí)空動(dòng)態(tài)變化及其與植被退化的關(guān)系. 草業(yè)學(xué)報(bào), 2016, 25(1): 22-32.
[47] Kang S, Ma W, Li F Y,etal. Functional redundancy instead of species redundancy determines community stability in a typical steppe of Inner Mongolia. PloS One, 2015, 10(12): e0145605.
[48] Xie Y, Sha Z. Quantitative analysis of driving factors of grassland degradation: A case study in Xilin river basin, Inner Mongolia. The Scientific World Journal, 2012, doi: 10.1100/2012/169724.
[49] Schaffrath D, Vetter S, Bernhofer C. Spatial precipitation and evapotranspiration in the typical steppe of Inner Mongolia, China-a model based approach using Modis data. Journal of Arid Environments, 2013, 88: 184-193.
[50] Tong C, Feng X, Zhang Y M,etal. Soil seed banks in different grazing exclusion restoring succession stages in the Xiligole degraded steppe. Acta Ecologica Sinica, 2008, (5): 1991-2002. 仝川, 馮秀, 張遠(yuǎn)鳴, 等. 錫林郭勒退化草原不同禁牧恢復(fù)演替階段土壤種子庫(kù)比較. 生態(tài)學(xué)報(bào), 2008, (5): 1991-2002.
[51] Li B. Inner Mongolia Vegetation[M]. Beijing: Science Press, 1989. 李博. 內(nèi)蒙古植被[M]. 北京: 科學(xué)出版社, 1989.
[52] Gong X Y, Chen Q, Lin S,etal. Tradeoffs between nitrogen-and water-use efficiency in dominant species of the semiarid steppe of Inner Mongolia. Plant and Soil, 2011, 340(1-2): 227-238.
[53] Bai Y F, Li L H, Wang Q B,etal. Changes in plant species diversity and productivity along gradirents of precipitation and elevation in the Xilin river basin, Inner Mongolia. Chinese Journal of Plant Ecology, 2000, (6): 667-673. 白永飛, 李凌浩, 王其兵, 等. 錫林河流域草原群落植物多樣性和初級(jí)生產(chǎn)力沿水熱梯度變化的樣帶研究. 植物生態(tài)學(xué)報(bào), 2000, (6): 667-673.
[54] Kemper J, Cowling R M, Richardson D M. Fragmentation of south African renosterveld shrublands: Effects on plant community structure and conservation implications. Biological Conservation, 1999, 90(2): 103-111.
[55] Wang W, Liu Z L, Hao D Y,etal. Research on the restoring succession of the degenerated grassland Inner Mongolia i.Basic characteristics and driving force restoration of the degenerated grassland. Chinese Journal of Plant Ecology, 1996, (5): 449-459. 王煒, 劉鐘齡, 郝敦元, 等. 內(nèi)蒙古草原退化群落恢復(fù)演替的研究——ⅰ.退化草原的基本特征與恢復(fù)演替動(dòng)力. 植物生態(tài)學(xué)報(bào), 1996, (5): 449-459.
[56] Liu Z L, Wang W, Hao D Y,etal. Discussion on Inner Mongolia grassland degradation and recovery succession mechanisms. Journal of Arid Land Resources and Environment, 2002, (1): 84-91. 劉鐘齡, 王煒, 郝敦元, 等. 內(nèi)蒙古草原退化與恢復(fù)演替機(jī)理的探討. 干旱區(qū)資源與環(huán)境, 2002, (1): 84-91.
[57] Suttie J M, Reynolds S G, Batello C. Grasslands of the World[M]. Food & Agriculture Org, 2005.
[58] Sun R Y, Li B, Zhuge Y. General Eecology[M]. Beijing: Higher Education Press, 1993. 孫濡泳, 李博, 諸葛陽(yáng). 普通生態(tài)學(xué)[M]. 北京: 高等教育出版社, 1993.
[59] Hou F J, Chang S H, Yu Y W,etal. A review on trampling by grazed livestock. Acta Ecologica Sinica, 2004, (4): 784-789. 侯扶江, 常生華, 于應(yīng)文, 等. 放牧家畜的踐踏作用研究評(píng)述. 生態(tài)學(xué)報(bào), 2004, (4): 784-789.
Driving forces ofLeymuschinensiscommunity differentiation in the Xilin River Basin
WU Ren-Ji1, KANG Saruul1*, ZHANG Qing1,2, REN Hai-Juan1, REN Jing1, ZHOU Jun-Mei1, WANG Zhen1, LI Dan1, NIU Jian-Ming1,2*
1.SchoolofLifeScience,InnerMongoliaUniversity,Hohhot010021,China; 2.Sino-USCenterforConservation,Energy,andSustainabilityScience,InnerMongoliaUniversity,Hohhot010021,China
The relationship between vegetation and the environment is a key issue in community ecology. There are theoretical and practical reasons to clarify grassland vegetation differentiation under the coupled effects of climate variables and grazing pressure, and to quantify the effects of natural and anthropogenic disturbances on vegetation processes. In this study, we quantitatively explored the effects of climate, altitude, soil, and grazing factors onLeymuschinensiscommunity differentiation in the Xilin River Basin using a detrended canonical correspondence analysis (DCCA) and structural equation modeling (SEM). Our results can be summarized as follows: (1) The first two axes of the DCCA explained more than 67.63% of the variability; the first axis captured mainly the variance caused by grazing and thermal-hydrological conditions, while the second axis captured the variance caused by soil variables; (2) The results of the SEM indicated that sheep density had the largest total impact on vegetation differentiation, followed by climate variables; (3) We detected indirect effects of climate, altitude, and grazing on vegetation differentiation, manifested mainly by changes in soil physical and chemical properties. This work clarified the key role of grazing onL.chinensiscommunity differentiation in the Xilin River Basin, and also revealed the importance of climatic factors in vegetation differentiation on the basin scale. Therefore, these findings will be useful to design grassland management strategies that take into account climate characteristics for the rational use and protection of grassland resources.
Xilin River Basin;Leymuschinensissteppe; vegetation differentiation; structural equation modeling; detrended canonical correspondence analysis
10.11686/cyxb2016184
http://cyxb.lzu.edu.cn
2016-05-03;改回日期:2016-06-28
國(guó)家重點(diǎn)基礎(chǔ)研究發(fā)展計(jì)劃(973)課題(2012CB722201),內(nèi)蒙古自治區(qū)科技重大專項(xiàng)和國(guó)家自然科學(xué)基金(31460154)資助。
吳仁吉(1991-),男,蒙古族,內(nèi)蒙古通遼人,在讀碩士。E-mail: nmgwurenji@163.com*通信作者Corresponding author. E-mail: jmniu2005@163.com, srlkang@163.com
吳仁吉, 康薩如拉, 張慶, 任海娟, 任婧, 周俊梅, 王珍, 李丹, 牛建明. 錫林河流域羊草草原植被分異的驅(qū)動(dòng)力. 草業(yè)學(xué)報(bào), 2017, 26(4): 15-23.
WU Ren-Ji, KANG Saruul, ZHANG Qing, REN Hai-Juan, REN Jing, ZHOU Jun-Mei, WANG Zhen, LI Dan, NIU Jian-Ming. Driving forces ofLeymuschinensiscommunity differentiation in the Xilin River Basin. Acta Prataculturae Sinica, 2017, 26(4): 15-23.