杜加強(qiáng),舒儉民,趙晨曦,賈爾恒·阿哈提,王麗霞,香 寶,方廣玲,劉偉玲,何 萍,*

1 中國(guó)環(huán)境科學(xué)研究院, 北京 100012 2 中國(guó)環(huán)境科學(xué)研究院環(huán)境基準(zhǔn)與風(fēng)險(xiǎn)評(píng)估國(guó)家重點(diǎn)實(shí)驗(yàn)室, 北京 100012 3 新疆環(huán)境保護(hù)科學(xué)研究院, 烏魯木齊 830011 4 環(huán)境保護(hù)部南京環(huán)境科學(xué)研究所, 南京 210042

兩代AVHRR GIMMS NDVI數(shù)據(jù)集的對(duì)比分析
——以新疆地區(qū)為例

杜加強(qiáng)1,2,舒儉民1,2,趙晨曦3,賈爾恒·阿哈提3,王麗霞4,香 寶1,2,方廣玲1,2,劉偉玲1,2,何 萍1,2,*

1 中國(guó)環(huán)境科學(xué)研究院, 北京 100012 2 中國(guó)環(huán)境科學(xué)研究院環(huán)境基準(zhǔn)與風(fēng)險(xiǎn)評(píng)估國(guó)家重點(diǎn)實(shí)驗(yàn)室, 北京 100012 3 新疆環(huán)境保護(hù)科學(xué)研究院, 烏魯木齊 830011 4 環(huán)境保護(hù)部南京環(huán)境科學(xué)研究所, 南京 210042

最新發(fā)布的1981—2012年的AVHRR GIMMS NDVI3g數(shù)據(jù)為了解區(qū)域植被的近期變化狀況提供了數(shù)據(jù)基礎(chǔ)。深入理解該版本與老版本GIMMS NDVIg(1981—2006年)之間的關(guān)系,對(duì)于使用新數(shù)據(jù)時(shí)充分利用已有老版本的研究結(jié)果具有重要意義。以我國(guó)西北干旱區(qū)的典型區(qū)域——新疆為例,研究了兩個(gè)數(shù)據(jù)集在反映生長(zhǎng)季、春季、夏季和秋季植被現(xiàn)狀,植被變化趨勢(shì)及其對(duì)氣候變化響應(yīng)方面的異同。研究結(jié)果表明:兩個(gè)數(shù)據(jù)集在描述植被活動(dòng)空間分布、變化趨勢(shì)及其與氣候的相關(guān)性方面大體相似,但在數(shù)值、動(dòng)態(tài)變化率及其對(duì)氣候變化響應(yīng)強(qiáng)度等方面存在的差異也不容忽略。NDVI3g數(shù)據(jù)生長(zhǎng)季和各季節(jié)NDVI數(shù)值多大于NDVIg,尤其是在夏季和在植被覆蓋較好的區(qū)域。區(qū)域尺度,NDVI3g所反映的植被變化趨勢(shì)更為平穩(wěn),尤其是在夏季和較長(zhǎng)的時(shí)段,這可能與像元尺度NDVI3g顯著增加范圍小于NDVIg,而顯著減少范圍多于NDVIg有關(guān)。兩個(gè)數(shù)據(jù)集對(duì)氣溫、降水量、潛在蒸散發(fā)和濕潤(rùn)指數(shù)的響應(yīng)具有大體一致的空間格局,但對(duì)氣候因子變化的敏感性存在差異,哪一個(gè)數(shù)據(jù)集更為靈敏依賴于不同的氣候因子和時(shí)段。一般規(guī)律是NDVI3g與熱量因子顯著正相關(guān)的區(qū)域小于NDVIg,而與水分因子顯著正相關(guān)的區(qū)域則大于NDVIg。利用長(zhǎng)期的生態(tài)數(shù)據(jù)集,盡快理清兩個(gè)數(shù)據(jù)集在表征植被變化之間的異同并建立兩者的轉(zhuǎn)換關(guān)系,對(duì)于合理開(kāi)展植被變化、碳平衡、生態(tài)系統(tǒng)服務(wù)功能評(píng)估等廣泛利用NDVI數(shù)據(jù)的相關(guān)研究十分重要。

GIMMS NDVI3g；GIMMS NDVIg；植被活動(dòng)；氣候變化；比較分析；新疆

植被是陸地生態(tài)系統(tǒng)最重要的組成部分,聯(lián)接了土壤圈、水圈和大氣圈的物質(zhì)循環(huán)和能量流動(dòng),在調(diào)節(jié)陸地碳平衡和氣候系統(tǒng)方面發(fā)揮了重要作用[1- 3],監(jiān)測(cè)植被動(dòng)態(tài)變化具有重要的科學(xué)價(jià)值和現(xiàn)實(shí)意義。植被覆蓋變化受到氣候變化、人類活動(dòng)的強(qiáng)烈影響[4- 9],地表植被對(duì)外界干擾的響應(yīng)已經(jīng)成為國(guó)內(nèi)外學(xué)術(shù)界研究的熱點(diǎn)。開(kāi)展大范圍的野外實(shí)地調(diào)查無(wú)疑是理論上監(jiān)測(cè)植被狀況的最優(yōu)方法[10],但該方法需要耗費(fèi)大量人力、財(cái)力,同時(shí)由點(diǎn)上的結(jié)果擴(kuò)展到面上時(shí)也可能產(chǎn)生偏差[11]。因此,在大尺度監(jiān)測(cè)植被變化的最有效的方法是基于衛(wèi)星的植被監(jiān)測(cè)[10,12],其具有時(shí)間連續(xù)、空間范圍廣、可重復(fù)、廉價(jià)等特點(diǎn)[10,13- 14]。

歸一化植被指數(shù)NDVI是公認(rèn)的陸地植被生長(zhǎng)狀況的最佳表征指標(biāo),廣泛地應(yīng)用于從全球到區(qū)域尺度的植被動(dòng)態(tài)及其對(duì)氣候變化響應(yīng)、土地退化區(qū)域識(shí)別、植被生產(chǎn)力和碳平衡等領(lǐng)域的研究之中。AVHRR GIMMS NDVI數(shù)據(jù)集具有時(shí)間序列長(zhǎng)、覆蓋范圍廣、時(shí)空可比、較強(qiáng)地植被動(dòng)態(tài)變化表征能力[15]等特點(diǎn),被證明是描述植被生長(zhǎng)動(dòng)態(tài)變化最好的數(shù)據(jù)集之一[7,16- 18],得到了非常廣泛的應(yīng)用,取得了大量的研究成果。目前,最常用的NDVI數(shù)據(jù)集是時(shí)間序列為1981—2006年的GIMMS NDVIg數(shù)據(jù)集,有關(guān)最近幾年植被變化及其與過(guò)去30年比較的研究需要擴(kuò)展GIMMS NDVIg數(shù)據(jù)集的時(shí)間序列[14,19- 22]。國(guó)內(nèi)外已經(jīng)有學(xué)者開(kāi)始進(jìn)行GIMMS NDVIg和與其他傳感器NDVI數(shù)據(jù)集的比較與數(shù)據(jù)插補(bǔ)工作[3,12,23- 30]。最近,最新版本的GIMMS NDVI數(shù)據(jù)集發(fā)布[31],被稱作GIMMS NDVI3g,時(shí)間跨度為1981—2012年,其數(shù)據(jù)處理目標(biāo)旨在提高高緯度地區(qū)的數(shù)據(jù)質(zhì)量[32- 33],以便于更適合北半球生態(tài)系統(tǒng)植被活動(dòng)變化的研究[23,34]。作為新一代的長(zhǎng)時(shí)間序列數(shù)據(jù)集,該數(shù)據(jù)集可以為地表植被30多年來(lái)的整體變化趨勢(shì)提供基礎(chǔ)數(shù)據(jù),勢(shì)必得到廣泛的應(yīng)用。然而,由于AVHRR傳感器設(shè)計(jì)之初并不是以植被研究為目的,獲取NDVI之前需要進(jìn)行一系列的校正處理工作,導(dǎo)致GIMMS NDVI數(shù)據(jù)集本質(zhì)上是動(dòng)態(tài)變化的,每一次有更新的數(shù)據(jù)加入必須要重新計(jì)算[35],這就使得即使是重疊年份1981—2006年的GIMMS NDVI3g與GIMMD NDVIg也不相同。因此,最新的GIMMS NDVI3g與過(guò)去應(yīng)用最為廣泛的GIMMS NDVIg版本之間的對(duì)比研究勢(shì)在必行,這也是連接已有研究和未來(lái)利用GIMMS NDVI3g監(jiān)測(cè)植被活動(dòng)的橋梁。目前有關(guān)兩代數(shù)據(jù)的對(duì)比研究剛剛開(kāi)始,尚未檢索到有關(guān)我國(guó)干旱區(qū)兩代數(shù)據(jù)集對(duì)比研究的相關(guān)報(bào)道。

為此,本文以中國(guó)西北干旱區(qū)的主體——新疆地區(qū)為例,利用生長(zhǎng)季(3—11月)、春季(3—5月)、夏季(6—8月)和秋季(9—11月)4個(gè)合成時(shí)段,評(píng)價(jià)兩代數(shù)據(jù)集在表征植被時(shí)空變化趨勢(shì)及其對(duì)氣候變化響應(yīng)的量級(jí)和空間模式上存在的差異和相似性,以期為未來(lái)利用GIMMS NDVI3g數(shù)據(jù)集評(píng)估干旱區(qū)植被動(dòng)態(tài)變化趨勢(shì)時(shí),如何合理利用GIMMS NDVIg的研究結(jié)果提供支撐和參考。

1 材料和方法

1.1 研究區(qū)域

新疆位于我國(guó)西北邊陲,介于73°20′—96°25′E,34°15′—49°10′N之間,總面積約1.66×106km2。境內(nèi)大致呈緯向伸展的三大山系阿爾泰山、天山和昆侖山分割著準(zhǔn)噶爾和塔里木盆地,形成了獨(dú)特的山體、盆地相間的地貌格局。山體垂直地帶性差異明顯,天山、阿爾泰山、昆侖山發(fā)育有大面積的森林和草地植被,準(zhǔn)噶爾盆地和塔里木盆地分布有典型的溫帶荒漠植被,綠洲和城市則分布在河谷平原區(qū)。由于新疆南北跨度大,形成了以天山為界,南疆、北疆自然地理狀況差異較大的格局,分別屬于溫帶大陸性干旱半干旱氣候和暖溫帶大陸性干旱氣候。區(qū)域干旱、少雨、多大風(fēng)的氣候特點(diǎn),形成了廣布的沙漠戈壁景觀,植被覆蓋率總體較低,生態(tài)系統(tǒng)較為脆弱、敏感,是研究植被變化及其對(duì)氣候變化響應(yīng)的理想?yún)^(qū)域。

1.2 數(shù)據(jù)來(lái)源與處理

GIMMS NDVIg和GIMMS NDVI3g數(shù)據(jù)集均來(lái)源于NASA戈達(dá)德航天中心,合成時(shí)段均為15 d。GIMMS NDVIg的空間分辨率為8km×8km(約為0.072°×0.072°),時(shí)間跨度是1981—2006年；GIMMS NDVI3g則分別為0.083°×0.083°,1981—2012年。

氣象數(shù)據(jù)來(lái)源于中國(guó)氣象科學(xué)數(shù)據(jù)共享服務(wù)網(wǎng),包含53個(gè)站點(diǎn)的月氣溫、月降水量數(shù)據(jù)(氣象站點(diǎn)分布見(jiàn)圖1)。1∶1000000矢量化植被類型圖來(lái)自中國(guó)植被類型圖[36],將新疆地區(qū)的植被分為森林、灌叢、草地、耕地、荒漠和無(wú)植被6大類。

兩種NDVI數(shù)據(jù)經(jīng)過(guò)子集提取、圖像鑲嵌、裁剪、數(shù)據(jù)格式轉(zhuǎn)換、投影轉(zhuǎn)換及質(zhì)量檢驗(yàn)等預(yù)處理過(guò)程,形成新疆GIMMS NDVIg和GIMMS NDVI3g數(shù)據(jù)集。采用最大值合成方法得到月尺度GIMMS NDVI數(shù)據(jù),以進(jìn)一步去除云的影響,并減少月內(nèi)物候循環(huán)的影響[35]。參考相關(guān)研究[3,7- 8],采用0.05的NDVI值作為的植被閾值,排除非植被因素的影響。

1.3 研究方法

數(shù)值比較[15,29- 30]、相關(guān)性分析[17- 18]、趨勢(shì)一致性分析[15,35]以及NDVI對(duì)氣候變化響應(yīng)[32]被認(rèn)為是評(píng)價(jià)NDVI數(shù)據(jù)集之間一致性的有效方法。本文從NDVI數(shù)值、變化趨勢(shì)及其與氣候變化的相關(guān)性3個(gè)方面,來(lái)評(píng)價(jià)重疊年份(1982—2006年)GIMMS NDVIg和GIMMS NDVI3g數(shù)據(jù)集的一致性,以期在靜態(tài)、動(dòng)態(tài)、空間格局、對(duì)外部干擾反映等多方面綜合反映兩個(gè)數(shù)據(jù)集的相似性。為了更好地利用兩代數(shù)據(jù)集的優(yōu)勢(shì)與特點(diǎn),以及方便利用已有NDVIg成果和未來(lái)NDVI3g數(shù)據(jù)處理,保留兩代數(shù)據(jù)集各自的分辨率。

NDVI的變化趨勢(shì)采用其與年份的最小二乘法回歸進(jìn)行分析,得到回歸方程的斜率(Slope)和Pearson相關(guān)系數(shù),分別用來(lái)表示植被生長(zhǎng)的變化速率和變化趨勢(shì)。定義(SlopeGIMMS NDVI3g-SlopeGIMMS NDVIg)/ SlopeGIMMS NDVIg×100為兩個(gè)數(shù)據(jù)集變化趨勢(shì)的差值比率(ratio of difference between NDVI3g slope and NDVIg slope, RDNS)。NDVI對(duì)氣候變化的響應(yīng),采用NDVI與同期氣候要素的相關(guān)性來(lái)表征。氣象數(shù)據(jù)采用Kriging方法插值到GIMMS NDVIg和GIMMS NDVI3g數(shù)據(jù)集的空間分辨率。為了更好地表征植被與干濕條件之間的關(guān)系,計(jì)算了蒸散發(fā)和濕潤(rùn)指數(shù)。蒸散發(fā)是唯一一個(gè)即出現(xiàn)在在水量平衡方程又出現(xiàn)在地表能量平衡方程中的要素[37- 38],與生態(tài)系統(tǒng)水分利用密切相關(guān)[8,38],尤其是在干旱和半干旱地區(qū)。參考相關(guān)研究[8],研究區(qū)域潛在蒸散發(fā)(ET)采用Thornthwaite方法計(jì)算,濕潤(rùn)指數(shù)(HI)采用降水量與潛在蒸散發(fā)之比計(jì)算。

時(shí)段不同植被變化趨勢(shì)不同,時(shí)段長(zhǎng)度也可能會(huì)對(duì)結(jié)果產(chǎn)生影響。因此,為深入探討兩個(gè)數(shù)據(jù)集所反映的新疆植被變化及其對(duì)氣候變化響應(yīng)的動(dòng)態(tài)過(guò)程,分別在1982—1999年、1982—2000年、…、1982—2006年8個(gè)時(shí)段計(jì)算NDVI變化趨勢(shì)及其與氣候因子的相關(guān)性。顯著增加、顯著減少區(qū)域面積在8個(gè)時(shí)段的變化趨勢(shì)、強(qiáng)度,采用其與1999—2006年年份的Pearson相關(guān)系數(shù)、斜率來(lái)計(jì)算。根據(jù)顯著性檢驗(yàn)結(jié)果,將變化趨勢(shì)分為如下3個(gè)等級(jí)：極顯著(P<0.01)；顯著(P<0.05)；不顯著(P>0.05)。

2 結(jié)果

2.1 數(shù)值差異與相關(guān)性

生長(zhǎng)季多年平均NDVI3g和NDVIg數(shù)據(jù)的數(shù)值分布顯示(圖1),≤0.20區(qū)間NDVIg的分布區(qū)域均大于NDVI3g,在大于0.20區(qū)間則相反。兩者之間逐像元的差值結(jié)果顯示(圖1),66%的研究區(qū)域NDVI3g均大于NDVIg,生長(zhǎng)季多年平均的平均偏差為0.0230,且植被覆蓋度較高的天山南北、阿爾泰山以及塔里木盆地西北邊緣、西南邊緣偏差較大。

圖1 NDVI3g與NDVIg生長(zhǎng)季多年平均值分布與差值Fig.1 Distribution and differences between NDVIg and NDVI in growing season

分別統(tǒng)計(jì)1982—2006年植被區(qū)域NDVI3g和NDVIg生長(zhǎng)季、各季節(jié)的區(qū)域平均年NDVI值,并計(jì)算相關(guān)系數(shù)、平均偏差、均方根誤差和相對(duì)偏差(表1,圖2)。各季節(jié)兩者之間的相關(guān)性均達(dá)到了0.01的顯著性水平,秋季相關(guān)性最強(qiáng),春季最弱；數(shù)值差異方面,植被區(qū)域NDVI3g數(shù)值比NDVIg高19%—28%,秋季相對(duì)差異最小,夏季較大。

表1 區(qū)域尺度各季節(jié)NDVI3g與NDVIg的關(guān)系

2.2 變化趨勢(shì)的一致性

區(qū)域尺度,盡管兩個(gè)數(shù)據(jù)集NDVI的變化趨勢(shì)十分相似(圖2,表2),兩者在8個(gè)時(shí)段均多呈顯著增加趨勢(shì)；除夏季所有時(shí)段NDVI3g變化量小于NDVIg(約低10%—43%),生長(zhǎng)季、春季和秋季均呈前幾個(gè)時(shí)段NDVI3g變化量大于NDVIg,而后幾個(gè)時(shí)段NDVIg大于NDVI3g,春季趨勢(shì)變化幅度差異最大,達(dá)到-34%—55%,秋季最小為-15%—13%。兩個(gè)數(shù)據(jù)集夏季變化趨勢(shì)的差異呈隨時(shí)段延長(zhǎng)而增加趨勢(shì),生長(zhǎng)季、春季和秋季多呈隨時(shí)段延長(zhǎng)先減小后增加趨勢(shì)。

像元尺度,兩者生長(zhǎng)季、春夏秋3個(gè)季節(jié)NDVI變化趨勢(shì)的空間分布大致較為相似(圖3),但不同時(shí)段呈顯著變化的區(qū)域大小明顯不同(表3)。夏季、秋季NDVI3g呈增加、顯著增加趨勢(shì)的范圍8個(gè)時(shí)段均小于NDVIg,面積差值的多時(shí)段平均值范圍為-6.31% —3.15%；生長(zhǎng)季和春季也多小于NDVIg,面積差值的多時(shí)段平均值范圍為-6.91%—1.13%；而NDVI3g呈顯著減少趨勢(shì)的區(qū)域則多大于NDVIg,面積差值的多時(shí)段平均值范圍為0.96%—3.08%。兩個(gè)數(shù)據(jù)集除生長(zhǎng)季顯著增加、春季顯著增加區(qū)域的差值呈先減少后增加趨勢(shì)外,生長(zhǎng)季、夏季和秋季兩者呈增加、顯著增加和顯著減小區(qū)域面積差值均呈隨時(shí)段延長(zhǎng)而明顯快速擴(kuò)大。

2.3 對(duì)氣候變化響應(yīng)的差異

區(qū)域平均尺度,兩個(gè)NDVI數(shù)據(jù)集與同期氣溫、降水量、ET和HI的相關(guān)性見(jiàn)表4。兩個(gè)數(shù)據(jù)集在生長(zhǎng)季和各季節(jié)對(duì)不同氣候因子變化響應(yīng)的差異基本一致。除秋季NDVI3g與降水量和HI的相關(guān)系數(shù)絕對(duì)值在多數(shù)時(shí)段大于NDVIg以外,NDVIg與氣候因子的相關(guān)性總體高于NDVI3g,僅在少數(shù)時(shí)段NDVI3g與氣候因子的相關(guān)性高于NDVIg。

圖2 兩個(gè)數(shù)據(jù)集1982—2006年的NDVI變化趨勢(shì)Fig.2 NDVI dynamics of two datasets during 1982—2006

圖3 兩個(gè)數(shù)據(jù)集NDVI變化量與趨勢(shì)顯著性Fig.3 Spatial distribution of NDVI trends over Xinjiang with two datasets

季節(jié)Seasons指標(biāo)Indicators1982—19991982—20001982—20011982—20021982—20031982—20041982—20051982—2006生長(zhǎng)季NDVI3g0．0010??0．0010??0．0009??0．0009??0．0007??0．0006??0．0005??0．0004?GrowingseasonNDVIg0．0010??0．0009??0．0009??0．0009??0．0008??0．0007??0．0007??0．0006??RDNS/%1．216．493．70-2．96-10．46-16．67-21．69-33．02春季SpringNDVI3g0．0006?0．0008??0．0008??0．0007??0．00050．00040．00030．0002NDVIg0．00050．00050．0006?0．0006?0．00050．00040．00040．0004RDNS/%40．0555．1427．7216．244．81-7．98-21．16-33．56夏季SummerNDVI3g0．0014??0．0013??0．0011??0．0010??0．0008??0．0007?0．0006?0．0005NDVIg0．0016??0．0014??0．0013??0．0012??0．0011??0．0010??0．0010??0．0009??RDNS/%-12．06-9．60-13．67-20．62-26．98-30．87-35．71-43．30秋季AutumnNDVI3g0．0011??0．0009??0．0009??0．0009??0．0008??0．0007??0．0006??0．0005?NDVIg0．0010??0．0008??0．0008??0．0008??0．0008??0．0007??0．0006??0．0006??RDNS/%4．364．4813．239．264．71-0．57-1．64-14．67

**代表顯著性水平小于0.01,*代表小于0.05; RDNS: NDVI變化趨勢(shì)的差值比率

表3 兩個(gè)數(shù)據(jù)集各季節(jié)8個(gè)時(shí)段不同變化趨勢(shì)的面積比例(%)

P為相關(guān)系數(shù)為正值區(qū)域面積比例，SN、SP分別為顯著水平達(dá)到0.05的負(fù)值和正值區(qū)域面積比例

圖4 兩個(gè)數(shù)據(jù)集生長(zhǎng)季NDVI與氣候因子的相關(guān)性Fig.4 Spatial distribution of correlations between NDVI and climatic factors

像元尺度,兩個(gè)NDVI數(shù)據(jù)集與同期氣溫、降水量、ET和HI相關(guān)性的空間格局(圖4)表明,總體上,兩個(gè)數(shù)據(jù)集在反映植被對(duì)氣候變化響應(yīng)方面基本一致,NDVI與氣候因子相關(guān)性的空間格局較為相似,顯著、極顯

表4 兩個(gè)數(shù)據(jù)集與4種氣候因子的相關(guān)系數(shù)

①、②、③、④分別為生長(zhǎng)季、春季、夏季和秋季

表5 兩個(gè)數(shù)據(jù)集與4種氣候變量不同相關(guān)性的區(qū)域比例(%)

Tem和Pre分別為氣溫temperature和降水量precipitation

著相關(guān)區(qū)域的分布、規(guī)模大體一致(表4)。縱觀8個(gè)時(shí)段的計(jì)算結(jié)果(表5),與NDVIg相比,生長(zhǎng)季與各季節(jié)NDVI3g與氣溫、ET顯著正相關(guān)的區(qū)域相對(duì)較小,顯著負(fù)相關(guān)的區(qū)域范圍則在多數(shù)時(shí)段相對(duì)較大。生長(zhǎng)季和各季節(jié)NDVI3g與降水量、HI呈顯著正相關(guān)的區(qū)域范圍在多數(shù)時(shí)段大于NDVIg,而在春季和秋季NDVI3g與降水量和HI呈顯著負(fù)相關(guān)的區(qū)域范圍在各時(shí)段均小于NDVIg,生長(zhǎng)季和夏季則在多數(shù)時(shí)段大于NDVIg。隨著時(shí)段延長(zhǎng),兩個(gè)數(shù)據(jù)集與氣候相關(guān)性位于同一顯著性水平的面積差距多有擴(kuò)大趨勢(shì)。

3 討論和結(jié)論

本文的研究結(jié)果與北半球高緯度地區(qū)研究得出的NDVI3g的變化量是NDVIg兩倍[32]的結(jié)果不盡相同,新疆地區(qū)NDVI3g與NDVIg在表征植被活動(dòng)動(dòng)態(tài)變化趨勢(shì)的量級(jí)方面,變化量相對(duì)大小無(wú)固定模式,哪一個(gè)數(shù)據(jù)集的動(dòng)態(tài)變化趨勢(shì)更強(qiáng),依賴于時(shí)段和季節(jié)。與北美地區(qū)的研究結(jié)果[10]一致,NDVI值較小的區(qū)間(約<0.2)NDVIg大于NDVI3g,較大的區(qū)間相反,表明在植被低覆蓋度區(qū)域NDVIg可能高估了植被NDVI值,而在植被高覆蓋區(qū)域則低估了植被NDVI值。值的說(shuō)明的是,NDVI3g和NDVIg哪一個(gè)更符合新疆的實(shí)際情況,則需要采用其他衛(wèi)星數(shù)據(jù)或大量的地面長(zhǎng)期觀測(cè)數(shù)據(jù)集、渦度通量塔觀測(cè)數(shù)據(jù)集等來(lái)評(píng)判,超出了本文的研究范疇。

NDVI3g與反映水分狀況的氣候因子(降水量、HI)正相關(guān)性多強(qiáng)于NDVIg,而與反映熱量的氣溫、ET則多呈相反規(guī)律,表明不同時(shí)段兩個(gè)數(shù)據(jù)集對(duì)不同氣候因子變化的敏感性存在差異；不同季節(jié)之間,兩個(gè)數(shù)據(jù)對(duì)各種氣候因子變化的響應(yīng)量級(jí)也不盡相同。

NDVI長(zhǎng)期數(shù)據(jù)為研究植被活動(dòng)提供了所必需的關(guān)鍵歷史視角[28]。但若想獲得任何有意義的陸表植被監(jiān)測(cè),連續(xù)的、相互校準(zhǔn)的植被指數(shù)長(zhǎng)期時(shí)間序列是關(guān)鍵需求[10,15,17,28]。NDVI3g和NDVIg之間的數(shù)值差異有可能對(duì)利用NDVI數(shù)據(jù)集的各種研究結(jié)果產(chǎn)生影響。比如,生長(zhǎng)季和夏季兩個(gè)數(shù)據(jù)集的不同,有可能對(duì)基于NDVI的植被生產(chǎn)力估算、碳匯潛力分析、農(nóng)作物估產(chǎn)、生態(tài)系統(tǒng)服務(wù)功能評(píng)估等產(chǎn)生較大影響；尤其是植被覆蓋更高的地區(qū)和生長(zhǎng)旺盛的夏季,較大的NDVI數(shù)值差異,可能對(duì)結(jié)果產(chǎn)生實(shí)質(zhì)性的影響。特別是NDVI被廣泛用于評(píng)估碳匯,NDVI數(shù)據(jù)集之間的差異可能影響國(guó)家碳賬戶的平衡,從而進(jìn)一步影響與碳排放相關(guān)國(guó)際履約和國(guó)際談判。春季、秋季NDVI的不同,則可能影響動(dòng)植物物候變化的研究結(jié)論,從而對(duì)預(yù)報(bào)農(nóng)事、指導(dǎo)農(nóng)牧業(yè)和林業(yè)生產(chǎn)、指示病蟲(chóng)害、引種和選種等方面產(chǎn)生誤導(dǎo)。

本文比較了最新的GIMMS NDVI3g和老版本GIMMS NDVIg在監(jiān)測(cè)植被動(dòng)態(tài)變化及其對(duì)氣候變化響應(yīng)方面的異同,主要發(fā)現(xiàn)歸納如下：

(1)兩個(gè)數(shù)據(jù)集在描述植被活動(dòng)空間分布方面較為一致,但有一定的數(shù)量差異。66%的研究區(qū)域生長(zhǎng)季NDVI3g均大于NDVIg,生長(zhǎng)季NDVI3g與NDVIg的平均偏差為0.0230?？臻g上,植被覆蓋度較高的天山南北、阿爾泰山以及塔里木盆地部分邊緣地帶偏差較大；季節(jié)上,夏季兩個(gè)數(shù)據(jù)集的數(shù)值差異較大。

(2)總體上,NDVI3g與NDVIg在反映區(qū)域植被變化趨勢(shì)及其空間格局方面基本一致,但變化強(qiáng)度差異明顯。盡管在較短的時(shí)段,生長(zhǎng)季、春季和秋季NDVI3g的增加量大于NDVIg,但總體上NDVI3g的變化更為平穩(wěn),尤其是在夏季和較長(zhǎng)的時(shí)段,其年際增加量均小于NDVIg,這與像元尺度NDVI3g顯著增加范圍小于NDVIg,而顯著減少范圍多于NDVIg有關(guān)。春季NDVI3g與NDVIg變化趨勢(shì)差異最大,達(dá)到-34%—55%,秋季最小為-15%—13%。

(3)NDVI3g和NDVIg對(duì)氣溫、降水量、ET和HI的響應(yīng)具有大體一致的空間格局,但是顯著相關(guān)的區(qū)域范圍具有一定差異。一般來(lái)說(shuō),NDVI3g與表征熱量的氣溫、ET顯著正相關(guān)的區(qū)域小于NDVIg,顯著負(fù)相關(guān)的區(qū)域則相反；NDVI3g與表征水分狀況的降水量、HI顯著正相關(guān)的區(qū)域則大于NDVIg；兩個(gè)數(shù)據(jù)集與降水量、HI呈顯著負(fù)相關(guān)區(qū)域的相對(duì)大小則依賴于季節(jié)和時(shí)段。

(4)NDVI3g數(shù)據(jù)對(duì)于了解地表植被歷史狀況以及近年的趨勢(shì)具有重要作用。鑒于NDVIg得到了廣泛應(yīng)用并取得了大量研究成果,評(píng)估NDVI3g和NDVIg數(shù)據(jù)的一致性、建立兩者的轉(zhuǎn)換關(guān)系,以及利用長(zhǎng)期生態(tài)數(shù)據(jù)集評(píng)判哪一個(gè)數(shù)據(jù)集更接近實(shí)際情況,是必要而緊迫的。

[1] Piao S L, Wang X H, Ciais P, Zhu B, Wang T, Liu J. Changes in satellite-derived vegetation growth trend in temperate and boreal Eurasia from 1982 to 2006. Global Change Biology, 2011, 17(10): 3228- 3239.

[2] Peng J, Liu Z H, Liu Y H, Wu J S, Han Y N. Trend analysis of vegetation dynamics in Qinghai-Tibet Plateau using Hurst Exponent. Ecological Indicators, 2012, 14(1): 28- 39.

[3] Bao G, Qin Z H, Bao Y H, Zhou Y, Li W J, Sanjjav A. NDVI-Based long-term vegetation dynamics and its response to climatic change in the Mongolian Plateau. Remote Sensing, 2014, 6(9): 8337- 8358.

[4] Myneni R B, Keeling C D, Tucker C J, Asrar G, Nemani R R. Increased plant growth in the northern high latitudes from 1981 to 1991. Nature, 1997, 386(6626): 698- 702.

[5] Goetz S J, Bunn A G, Fiske G J, Houghton R A. Satellite-observed photosynthetic trends across boreal North America associated with climate and fire disturbance. Proceeding of the National Academy of Sciences of the United States of America, 2005, 102(38): 13521- 13525.

[6] Peng S S, Chen A P, Xu L, Cao C X, Fang J Y, Myneni R B, Pinzon J E, Tucker C J, Piao S L. Recent change of vegetation growth trend in China. Environmental Research Letters, 2011, 6(4): 044027.

[7] Nemani R R, Keeling C D, Hashimot H, Jolly W M, Piper S C, Tucker C J, Myneni R B, Running S W. Climate-Driven increases in global terrestrial net primary production from 1982 to 1999. Science, 2003, 300(5625): 1560- 1563.

[8] Zhao X, Tan K, Zhao S, Fang J. Changing climate affects vegetation growth in the arid region of the northwestern China. Journal of Arid Environments, 2011, 75(10): 946- 952.

[9] Mohammat A, Wang X H, Xu X T, Peng L Q, Yang Y, Zhang X P, Myneni R B, Piao S L. Drought and spring cooling induced recent decrease in vegetation growth in Inner Asia. Agricultural and Forest Meteorology, 2013, 178- 179: 21- 30.

[10] Scheftic W, Zeng X B, Broxton P, Brunke M. Intercomparison of seven NDVI products over the United States and Mexico. Remote Sensing, 2014, 6(2): 1057- 1084.

[11] 卓莉, 曹鑫, 陳晉, 陳仲鑫, 史培軍. 錫林郭勒草原生態(tài)恢復(fù)工程效果的評(píng)價(jià). 地理學(xué)報(bào), 2007, 62(5): 471- 480.

[12] Piao S L, Cui M D, Chen A P, Wang X H, Ciais P, Liu J, Tang Y H. Altitude and temperature dependence of change in the spring vegetation green-up date from 1982 to 2006 in the Qinghai-Xizang Plateau. Agricultural and Forest Meteorology, 2011, 151(12): 1599- 1608.

[13] Pettorelli N, Chauvenet A L M, Duffy J P, Cornforth W A, Meillere A, Baillie J E M. Tracking the effect of climate change on ecosystem functioning using protected areas: Africa as a case study. Ecological Indicators, 2012, 20: 269- 276.

[14] De Jong R, De Bruin S, Schaepman M, Dent D. Quantitative mapping of global land degradation using Earth observations. International Journal of Remote Sensing, 2011, 32(21): 6823- 6853.

[15] Beck H E, McVicar T R, van Dijk A I J M, Schellekens J, de Jeu R A M, Bruijnzeel L. Global evaluation of four AVHRR-NDVI data sets: Intercomparison and assessment against Landsat imagery. Remote Sensing of Environment, 2011, 115(10): 2547- 2563.

[16] Beck P S A, Goetz S J. Satellite observations of high northern latitude vegetation productivity changes between 1982 and 2008: ecological variability and regional differences. Environmental Research Letters, 2011, 6(4): 045501.

[17] Fensholt R, Langanke T, Rasmussen K, Reenberg A, Prince S D, Tucker C, Choles R J, Le Q B, Bondeau A, Eastman R, Epstein H, Gaughan A E, Hellden U, Mbow C, Olsson L, Paruelo J, Schweitzer C, Seaquist J, Wessels K. Greenness in semi-arid areas across the globe 1981- 2007——an Earth Observing Satellite based analysis of trends an drivers. Remote Sensing of Environment, 2012, 121: 144- 158.

[18] Alcaraz-Segura D, Liras E, Tabik S, Paruelo J, Cabello J. Evaluating the consistency of the 1982- 1999 NDVI trends in the Iberian Peninsula across four time-series derived from the AVHRR sensor: LTDR, GIMMS, FASIR, and PAL-II. Sensor, 2010, 10(2): 1291- 1314.

[19] 毛德華, 王宗明, 宋開(kāi)山, 劉殿偉, 張柏, 張素梅, 羅玲, 張春華. 東北多年凍土區(qū)植被NDVI變化及其對(duì)氣候變化和土地覆被變化的響應(yīng). 中國(guó)環(huán)境科學(xué), 2011, 31(2): 283- 292.

[20] Mao D H, Wang Z M, Luo L, Ren C Y. Integrating AVHRR and MODIS data to monitor NDVI changes and their relationships with climatic parameters in Northeast China. International Journal of Applied Earth Observation and Geoinformation, 2012, 18: 528- 536.

[21] 信忠保, 許炯心, 鄭偉. 氣候變化和人類活動(dòng)對(duì)黃土高原植被覆蓋變化的影響. 中國(guó)科學(xué) D輯: 地球科學(xué), 2007, 37(11): 1504- 1514.

[22] 丁明軍, 張鐿鋰, 劉林山, 王兆鋒. 1982- 2009年青藏高原草地覆蓋度時(shí)空變化特征. 自然資源學(xué)報(bào), 2010, 25(12): 2114- 2122.

[23] 陳燕麗, 龍步菊, 潘學(xué)標(biāo), 鐘仕全, 莫偉華. MODIS NDVI和AVHRR NDVI對(duì)草原植被變化監(jiān)測(cè)差異. 遙感學(xué)報(bào), 2011, 15(4): 831- 845.

[24] Fensholt R, Rasmussen K, Nielsen T T, Mbow C. Evaluation of earth observation based long term vegetation trends——Intercomparing NDVI time series trend analysis consistency of Sahel from AVHRR GIMMS, Terra MODIS and SPOT VGT data. Remote Sensing of Environment, 2009, 113(9): 1886- 1898.

[25] Gallo K, Ji L, Reed B, Eidenshink J, Dwyer J. Multi-platform comparisons of MODIS and AVHRR normalized difference vegetation index data. Remote Sensing of Environment, 2005, 99(3): 221- 231.

[26] 何月, 樊高峰, 張小偉, 柳苗, 高大偉. 浙江省植被NDVI動(dòng)態(tài)及其對(duì)氣候的響應(yīng). 生態(tài)學(xué)報(bào), 2012, 32(14): 4352- 4362.

[27] Tucker C J, Pinzon J E, Brown M E, Slayback D A, Pak E W, Mahoney R, Vermonte E F, El Saleous N. An extended AVHRR 8-km NDVI dataset compatible with MODIS and SPOT vegetation data. International Journal of Remote Sensing, 2005, 26(20): 4485- 4498.

[28] Brown M E, Pinzon J E, Didan K, Morisette J T, Tucker C J. Evaluation of the consistency of long-term NDVI time series derived from AVHRR, SPOT-Vegetation, SeaWiFS, MODIS, and Landsat ETM+ Sensors. IEEE Transactions on Geoscience and Remote Sensing, 2006, 44(7): 1787- 1793.

[29] Steven M D, Malthus T J, Baret F, Xu H, Chopping M J. Intercalibration of vegetation indices from different sensor systems. Remote Sensing of Environment, 2003, 88(4): 412- 422.

[30] Fontana F M A, Coops N C, Khlopenkov K V, Trishchenko A P, Riffler M, Wulder M A. Generation of a novel 1km NDVI data set over Canada, the northern United States, and Greenland based on historical AVHRR data. Remote Sensing of Environment, 2012, 121: 171- 185.

[31] Pinzon J E, Tucker C J. A non-stationary 1981- 2012 AVHRR NDVI3gtime series. Remote Sensing, 2014, 6(8): 6929- 6960.

[32] Jiang N, Zhu W Q, Zheng Z T, Chen G S, Fan D Q. A comparative analysis between GIMSS NDVIg and NDVI3g for monitoring vegetation activity change in the northern hemisphere during 1982- 2008. Remote Sensing, 2013, 5(8): 4031- 4044.

[33] Zhu Z C, Bi J, Pan Y Z, Ganguly S, Anav A, Xu L, Samanta A, Piao S L, Nemani R R, Myneni R B. Global data sets of vegetation leaf area index (LAI)3g and fraction of photosynthetically active radiation (FPAR)3g derived from global inventory modeling and mapping studies (GIMMS) normalized difference vegetation index (NDVI3g) for the period 1981 to 2011. Remote Sensing, 2013, 5(2): 927- 948.

[34] Xu L, Myneni R B, Chapin III F S, Callaghan T V, Pinzon J E, Tucker C J, Zhu Z, Bi J, Ciais P, T?mmervik H, Euskirchen E S, Forbes B C, Piao S L, Anderson B T, Ganguly S, Nemani R R, Goetz S J, Beck P S A, Bunn A G, Cao C, Stroeve J C. Temperature and vegetation seasonality diminishment over northern lands. Nature Climate Change, 2013, 3(6): 581- 586.

[35] Fensholt R, Proud S R. Evaluation of Earth Observation based global long term vegetation trends——Comparing GIMMS and MODIS global NDVI time series. Remote Sensing of Environment, 2012, 119: 131- 147.

[36] CVAEC, CAS. Vegetation Atlas of China (1:1,000,000), 2001. Beijing: Science Press, 2001.

[37] Xu C Y, Singh V P. Evaluation of three complementary relationship evapotranspiration models by water balance approach to estimate actual regional evapotranspiration in different climatic regions. Journal of Hydrology, 2005, 308(1/4): 105- 121.

[38] Gao G, Chen D L, Xu C Y, Simelton E. Trend of estimated actual evapotranspiration over China during 1960- 2002. Journal of Geophysical Research, 2007, 112(D11): D11120.

Comparison of GIMMS NDVI3g and GIMMS NDVIg for monitoring vegetation activity and its responses to climate changes in Xinjiang during 1982—2006

DU Jiaqiang1,2, SHU Jianmin1,2, ZHAO Chenxi3, JIAERHENG Ahati3, WANG Lixia4, XIANG bao1,2, FANG Guangling1,2, LIU Weiling1,2, HE Ping1,2,*

1ChineseResearchAcademyofEnvironmentalSciences,Beijing100012,China2StateKeyLaboratoryofEnvironmentalCriteriaandRiskAssessment,ChineseResearchAcademyofEnvironmentalSciences,Beijing100012,China3XinjiangAcademyofEnvironmentalProtectionScience,Urumqi830011,China4NanjingInstituteofEnvironmentalSciences,MinistryofEnvironmentalProtectionofthePeople′sRepublicofChina,Nanjing210042,China

The released third-generation NDVI datasets in 2014, GIMMS NDVI3g, provide a data basis for quantifying recent regional vegetation dynamics over a sufficiently long term. The comparison between the new and old versions (GIMMS NDVIg, from 1981 to 2006) is necessary to link previous studies with future applications of GIMMS NDVI3g in monitoring vegetation activity trends and their responses to climate change. In this study, GIMMS NDVI3g was initially compared with GIMMS NDVIg in an evaluation of spatio-temporal patterns of seasonal vegetation changes in Xinjiang Province, China, at regional and pixel scales during overlapping periods from 1982 to 2006. The influences of climate change (including temperature, precipitation, potential evapotranspiration, and humidity index) on vegetation growth were then analyzed based on GIMMS NDVI3g and GIMMS NDVIg. To better understand the relationships between GIMMS NDVI3g and GIMMS NDVIg, NDVI trends and correlations between NDVI and climatic factors were calculated over multiple nested time series from 18 to 25 starting in 1982. The results indicated that most areas showed an approximate consistency in overall changing trends and correlations with climate variables for both datasets, but differences in many aspects should not be ignored. In most pixels, numerical values of GIMMS NDVI3g were larger than those of GIMMS NDVIg in the growing season, spring, summer, and autumn, particularly in summer, and also in those areas with dense vegetation. At a regional scale, the NDVI trends of GIMMS NDVI3g were smoother than those of GIMMS NDVIg in the growing season and all seasons, particularly in summer and longer periods. At the pixel scale, areas with a significant increase in GIMMS NDVI3g were less than those in GIMMS NDVIg, whereas this was not true in those areas with a significant decrease. The spatial patterns of correlations between GIMMS NDVI3g and four climate variables were approximately similar to those between GIMMS NDVIg and the climate variables, but there were some differences in the sensitivity of both datasets to climate change. Which dataset is more sensitive depends on climate variables and periods. In general, areas with significantly positive correlations between GIMMS NDVI3g and thermal factors were fewer than those of GIMMS NDVIg, whereas positive correlations between NDVI and moisture factors were greater in GIMMS NDVI3g than in GIMMS NDVIg. Integrated other ecological datasets, it is urgent to identify the similarities and differences between the two datasets and to establish a connection between them for reasonably monitoring vegetation dynamics using NDVI datasets.

GIMMS NDVI3g; GIMMS NDVIg; vegetation; climate change; comparison; Xinjiang

國(guó)家自然科學(xué)基金資助項(xiàng)目(41001055)；國(guó)家環(huán)保公益性行業(yè)科研專項(xiàng)經(jīng)費(fèi)資助(201209027- 5)；中國(guó)環(huán)境科學(xué)研究院中央級(jí)公益性科研院所改革啟動(dòng)專項(xiàng)(2012-YSGQ-05)

2015- 04- 19;

日期:2016- 03- 03

10.5846/stxb201504190805

*通訊作者Corresponding author.E-mail: heping@craes.org.cn

杜加強(qiáng),舒儉民,趙晨曦,賈爾恒·阿哈提,王麗霞,香寶,方廣玲,劉偉玲,何萍.兩代AVHRR GIMMS NDVI數(shù)據(jù)集的對(duì)比分析——以新疆地區(qū)為例.生態(tài)學(xué)報(bào),2016,36(21):6738- 6749.

Du J Q, Shu J M, Zhao C X, Jiaerheng Ahati, Wang L X, Xiang B, Fang G L, Liu W L, He P.Comparison of GIMMS NDVI3g and GIMMS NDVIg for monitoring vegetation activity and its responses to climate changes in Xinjiang during 1982—2006.Acta Ecologica Sinica,2016,36(21):6738- 6749.