基于GIMMS AVHRR NDVI数据的三北防护林工程区植被覆盖动态变化与影响因子分析研究(英文)

本文基于1982-2006年连续25年的GIMMS AVHRR NDVI植被覆盖指数,采用了最大化NDVI均值法、与气温及降水变化的相关性和一元线性回归趋势分析法,对中国三北防护林工程区连续25年的植被覆盖时空变化特征进行了动态变化研究。结果表明:(1)近25年来,研究区植被NDVI平均值总体呈缓慢上升趋势,增速为每10年0.007;(2)研究区植被和气温、降水整体呈正相关关系,植被与降水正相关面积明显大于植被与气温正相关面积,说明降水是研究区植被生长的关键因子;(3)1982-2006年,研究区植被覆盖增加的区域主要分布在大兴安岭中、南部,小兴安岭中部,长白山东北段,燕山,辽西低山丘陵区,阿尔泰山,天山,祁连山东段,西北荒漠区东部和黄土高原丘陵沟壑区南部等;植被覆盖减少的区域主要是在大兴安岭两侧,呼伦贝尔高原西部,三江平原北部,科尔沁沙地南端,西北荒漠区南部和黄土高原丘陵沟壑区北部等。     

1982~2011年北半球中纬度秋季植被绿度变化的主要模态

利用系统聚类方法和经验正交函数分解（EOF分析）2种方法,分别提取了北半球中纬度地区1982~2011年秋季（9~11月）归一化差值植被指数（NDVI）变化的主要模态,辨识了植被绿度变化的区域差异;并采用奇异值分解（SVD）的方法,综合时间和空间2个维度上的变异信息,揭示了植被绿度变化的气候背景。结果表明,北半球中纬度秋季植被绿度变化有2种基本模态,一种是持续增加模态（模态I）,该模态广泛分布于北美大陆、亚欧大陆的北半部（大约在55oN以北）和东西两端,NDVI平均增速为0.014/10a;另一种是趋势转折模态（模态II）, NDVI先增加,后减少,转折点大致出现在1994年,该模态主要出现在亚欧大陆中部,NDVI变化的平均速率分别是0.027/10a和-0.017/10a,其中以40oE~80oE最为典型。植被绿度变化与温度变化的时空特征基本一致。模态I区域的温度变化以持续性升高为主要特征,模态II区域的温度变化则以先增加后降低为主要特征,转折年份与NDVI变化的转折年份基本一致。SVD分析的第一模态NDVI与温度的时间系数相关系数为0.82,第二模态为0.92。由此表明,植被绿度变化主导模态可能由温度变化模态所致,在区域-大洲尺度上,温度变化的区域差异导致了秋季植被绿度变化的区域差异。     

近25年三北防护林工程区土地退化及驱动力分析

以长时间序列的遥感数据NOAA/AVHRR NDVI为基础,利用Sen趋势度与Mann-Kendall分析法相结合的方式分析了三北防护林工程区1982年到2006年土地退化趋势,结合气候因子降雨,运用残差法模型评价了人类活动在工程区内土地退化中所起的作用.结果表明:区域内土地退化程度整体趋于减轻,即植被上升的区域大于下降的区域,13.00％的地区退化程度显著减轻,6.20％的地区退化程度显著加重.其中绝大多数省份的土地退化程度趋于减轻,尤以内蒙古、青海和新疆最为明显,只有甘肃省土地退化程度明显趋于加重.而人类活动对植被变化起显著正作用的占11.93％,显著负作用的为6.19％.这说明在干旱、半干旱区,由于降雨随时间的变化不显著,对植被变化的影响相对较为微弱,植被显著变化主要受人类活动的影响.     

自然植被对气候变化响应的研究:综述

植被-气候关系的研究已经超越了植被地理学、植被生态学的研究范围，而成为全球变化研究的核心内容之一，从而受到地理学家、生态学家等的广泛关注，并开展了大量的研究。本文概述了过去、现在和未来自然植被对气候变化响应研究的主要进展，阐述了自然植被响应气候变化，特别是未来全球气候变化的基本的可能结果。     

近35年三北防护林体系建设工程的防风固沙效应

通过分析三北工程区生态系统防风固沙服务量变化,进而评价工程的防风固沙效应。结果表明：① 近35 a,三北工程区林地面积持续增加、草地面积持续减少,特别是半干旱风沙区和黄土高原区。植被覆盖度在前20 a持续增加,近15 a则呈现先下降而后略微增加趋势。②土壤风蚀模数持续减少,近15 a减幅远高于前20 a,沙地、草地、其它类型转林地或转草地区域的减幅尤为明显。生态系统防风固沙服务保有率呈上升态势,干旱荒漠区增幅显著。草地和沙地为工程区生态系统防风固沙服务总量贡献了71%,且转草地区域的贡献高于转林地区域。③ 考虑到植被好转同时受气候变化和人类活动的影响,以风场减弱为主的气候变化导致土壤风蚀力减弱与三北防护林等生态工程对生态系统防风固沙服务变化的贡献率分别介于85%~89%和11%~15%之间。     

基于NOAA NDVI的中国植被绿度始期变化

在对1982-1999 年NOAA/AVHRR NDVI 数据集进行缺失处理、平滑处理以及残存误差订正的基础上,利 用基于NDVI 累积频率曲线的Logistic 拟合模型计算生成了各年份全国植被绿度始期数据, 统计了所有气候植被 类型和生物气候区的平均绿度始期,并对不同空间类型植被绿度始期的年际变化趋势与显著性程度、时空变化特 征等做了统计分析。结果表明：①全国大部分地区植被绿度始期呈提前趋势,其中华北平原、贵州东北部、湖南中南 部山地丘陵区、广东西部地区以及内蒙古高原东部典型草原区等明显提前。②植被绿度始期年际波动程度高于10 天的地区主要分布在农业植被区、常绿植被区和草原区,其中河套平原、关中盆地、河南中东部、川西盆地、广东、藏 东南边缘地带以及台湾部分地区超过15 天。③除温带荒漠以外,其它所有植被类型绿度始期均为提前趋势。④从 各生物气候区空间变化来看,随着年积温降低和干旱指数增大,植被绿度始期大致呈推迟趋势;从年际变化来看, 华南、华东、华中、华北、内蒙、东北和青藏区植被绿度始期提前,陕晋和西北区植被绿度始期推迟。⑤受1982～1983 年厄尔尼诺事件影响,各空间类型植被绿度始期分别表现为推迟和提前趋势。     

中国植被覆盖度时空演变及其对气候变化和城市化的响应

植被覆盖变化不仅与气候因子密切相关,而且也受人类活动的影响。目前,从省级尺度研究中国植被时空变化特征以及定量分析气候因子结合人类活动对植被覆盖影响研究仍较少。基于Google Earth Engine(GEE)平台和2000—2020年Landsat数据及同期气候与夜间灯光数据,采用像元二分法、线性回归分析、变异系数、偏相关分析和贡献度模型等方法对中国植被覆盖度时空演变及其对气候变化和城市化的响应进行了分析。结果表明：（1）2000—2020年中国植被覆盖度以0.32%·a-1的速率增长。植被覆盖区域以高覆盖度为主,面积占研究区域的38%,总体呈现从东南至西北递减的趋势。（2）黄土高原、云南省、西藏自治区和新疆维吾尔自治区西部植被覆盖度呈现增长趋势。植被年际波动在南部比北部、东部比西部稳定。黑龙江省植被覆盖度最高,为91.7%;新疆维吾尔自治区最低,为14.4%;宁夏回族自治区植被覆盖度以0.98%·a-1的速率增长,植被得到显著改善。（3）气候因子和城市化对植被覆盖度的影响存在明显空间差异性。气温和降水量对中国北部地区植被覆盖度的影响分别为负相关和正相关,城市化主要影响经济较为发达的...     

1999—2010年中国西北地区植被覆盖对气候变化和人类活动的响应

中国西北地区气候干旱,频繁出现沙尘天气,属于生态脆弱区域,而植被变化是生态系统对气候变化响应的指示器,研究其变化对改善西北生态环境具有重要意义.本文利用1999—2010年归一化植被指数(NDVI)以及气象数据研究中国西北地区植被覆盖时空变化,以Sen趋势度结合Mann-Kendall检验、相关和偏相关分析以及残差法分析人类活动和气候变化对植被覆盖变化的影响.结果表明:西北地区植被覆盖整体呈增加趋势,但在局部地区气候干旱少雨和人类活动抑制植被生长.植被变化强度空间差异是人类活动和气候要素共同作用的结果:气温高,降水少,大部分地区植被覆盖与气候要素相关显著,并且植被变化对气温和降水的响应存在一定滞后时间;蒸发量大于降水量,人类引水灌溉弥补降水不足,使得农业植被呈增长趋势.新疆北部地区植被覆盖呈下降趋势,原因是气候干旱、沙漠化严重会抑制植被生长,人类活动频繁、城市扩建同样会破坏植被生长.     

黄土高原植被降水利用效率对植被恢复/退化的响应

基于2000-2014年MODIS NDVI数据及气象数据,运用累计降水利用效率变化差异（CRD,cumulative rain use efficiency differences）估算模型和基于地形要素降水量插值法,探讨2000-2014年黄土高原RUE（降水利用效率rain use efficiency）对植被变化的响应,以期为黄土高原生态可持续发展提供数据支撑。结果表明：黄土高原大部分地区植被覆盖得以改善,其面积约占总面积的81%,区域边缘植被覆盖退化严重。黄土高原降水利用效率RUE与累计NDVI的相关性总体表现为“东南呈正相关,西北为负相关”的空间格局,全区相关系数以正相关为主。黄土高原CRD与植被变化趋势的相关性显著,其中,植被退化背景下,植被退化程度越严重,RUE越低;植被恢复背景下,RUE受“退耕还林还草”作用显著,2000-2005年,RUE呈上升趋势,2007年后,随着退耕还林还草政策的工作重心转移,RUE呈波动变化。     

北半球春季植被NDVI对温度变化响应的区域差异

利用1982年到2000年的探路者NDVI资料，采用奇异值分解分析方法，研究北半球春季NDVI对温度变化响应的空间差异，前7对模态对总的协方差平方的解释率高达91％以上，反映出NDVI和气温的相关性非常高，第一对模态解释率达42．6％，显示北半球最显著的NDVI响应中心在西西伯利亚，其次是北美大陆，中心在其中东部，第三对及以后的模态反映的是次一次的空间特征，分析表明这些NDVI一温度的耦合模态受大尺度的大气环流系统的显著影响，9个重要的大气环流指标能解释整个北半球NDVI方差的55．6％，其中对欧洲、北美东南部，北美西北部，亚洲高纬以及东亚地区的影响最突出，因此，研究未来植被生态系统对全球变化响应的区域特征时，必须要考虑到这些环流系统的可能变化及其影响。     

陕甘宁地区植被恢复对气候变化和人类活动的响应(英文)

The "Grain for Green Project" initiated by the governments since 1999 were the dominant contributors to the vegetation restoration in the agro-pastoral transitional zone of northern China. Climate change and human activities are responsible for the improvement and degradation to a certain degree. In order to monitor the vegetation variations and clarify the causes of rehabilitation in the Shaanxi-Gansu-Ningxia Region, this paper, based on the MODIS-NDVI and climate data during the period of 2000-2009, analyzes the main charac-teristics, spatial-temporal distribution and reasons of vegetation restoration, using methods of linear regression, the Hurst Exponent, standard deviation and other methods. Results are shown as follows. (1) From 2000 to 2009, the NDVI of the study area was improved progres-sively, with a linear tendency being 0.032/10a, faster than the growth of the Three-North Shelter Forest Program (0.007/10a) from 1982 to 2006. (2) The vegetation restoration is characterized by two fast-growing periods, with an "S-shaped" increasing curve. (3) The largest proportion of the contribution to vegetation restoration was observed in the slightly improved area, followed by the moderate and the significantly improved area; the degraded area is distributed sporadically over southern part of Ningxia Hui Autonomous Region as well as eastern Dingbian of Shaanxi province, Huanxian and Zhengyuan of Gansu province. (4) Climate change and human activities are two driving forces in vegetation restoration; more-over anthropogenic factors such as "Grain for Green Project" were the main causes leading to an increasing trend of NDVI on local scale. However, its influencing mechanism remains to be further investigated. (5) The Hurst Exponent of NDVI time series shows that the vegetation restoration was sustainable. It is expected that improvement in vegetation cover will expand to the most parts of the region.     

2000~2014年黑龙江流域（中国）植被覆盖时空变化及其对气候变化的响应

采用MODIS/NDVI数据,利用Theil-Sen Median 趋势分析、Mann-Kendall 以及Hurst指数方法分析2000~2014年黑龙江流域（中国）植被的时空变化特征、植被变化发展趋势及可持续性特征;应用相关分析法研究了气候变化对植被生长的影响。结果表明,2000~2014年黑龙江流域（中国）植被NDVI指数呈缓慢增加趋势,山区植被覆盖增加显著,东北部平原区植被覆盖持续退化,总体上植被覆盖持续改善能力较弱。植被NDVI对气候响应的季节差异显著,且不同类型植被对气候因子的响应不一致：春季植被NDVI主要受温度影响,夏季植被NDVI主要受降水量影响,秋季林地NDVI与温度正相关、草地NDVI与降雨量正相关。     

Response of vegetation restoration to climate change and human activities in Shaanxi-Gansu- Ningxia Region

The “Grain for Green Project” initiated by the governments since 1999 were the dominant contributors to the vegetation restoration in the agro-pastoral transitional zone of northern China. Climate change and human activities are responsible for the improvement and degradation to a certain degree. In order to monitor the vegetation variations and clarify the causes of rehabilitation in the Shaanxi-Gansu-Ningxia Region, this paper, based on the MODIS-NDVI and climate data during the period of 2000-2009, analyzes the main characteristics, spatial-temporal distribution and reasons of vegetation restoration, using methods of linear regression, the Hurst Exponent, standard deviation and other methods. Results are shown as follows. (1) From 2000 to 2009, the NDVI of the study area was improved progressively, with a linear tendency being 0.032/10a, faster than the growth of the Three-North Shelter Forest Program (0.007/10a) from 1982 to 2006. (2) The vegetation restoration is characterized by two fast-growing periods, with an “S-shaped” increasing curve. (3) The largest proportion of the contribution to vegetation restoration was observed in the slightly improved area, followed by the moderate and the significantly improved area; the degraded area is distributed sporadically over southern part of Ningxia Hui Autonomous Region as well as eastern Dingbian of Shaanxi province, Huanxian and Zhengyuan of Gansu province. (4) Climate change and human activities are two driving forces in vegetation restoration; moreover anthropogenic factors such as “Grain for Green Project” were the main causes leading to an increasing trend of NDVI on local scale. However, its influencing mechanism remains to be further investigated. (5) The Hurst Exponent of NDVI time series shows that the vegetation restoration was sustainable. It is expected that improvement in vegetation cover will expand to the most parts of the region.     

气候变化和生态建设对秦岭—淮河南北植被动态的影响

论文基于2000-2019年秦岭—淮河南北MODIS-NDVI植被覆盖数据,对中国南北过渡带植被时空变化进行分析,并探讨植被动态变化驱动因素.结果 表明:①在趋势变化上,2000-2019年秦岭—淮河南北植被显著恢复.其中,秦巴山区植被恢复面积占比最高,其次是巫山山区和关中平原;植被退化区面积占比仅为6.4％,主要分布...     

Contrasting vegetation changes in dry and humid regions of the Tibetan Plateau over recent decades

An overall greening over the Tibetan Plateau(TP) in recent decades has been established through analyses of remotely sensed Normalized Difference Vegetation Index(NDVI), though the regional pattern of the changes and associated drivers remain to be explored. This study used a satellite Leaf Area Index(LAI) dataset(the GLASS LAI dataset) and examined vegetation changes in humid and arid regions of the TP during 1982–2012. Based on distributions of the major vegetation types, the TP was divided roughly into a humid southeastern region dominated by meadow and a dry northwestern region covered mainly by steppe. It was found that the dividing line between the two regions corresponded well with the lines of mean annual precipitation of 400 mm and the mean LAI of 0.3. LAI=0.3 was subsequently used as a threshold for investigating vegetation type changes at the interanual and decadal time scales: if LAI increased from less than 0.3 to greater than0.3 from one time period to the next, it was regarded as a change from steppe to meadow, and vice versa. The analysis shows that changes in vegetation types occurred primarily around the dividing line of the two regions, with clear growth(reduction) of the area covered by meadow(steppe), in consistency with the findings from using another independent satellite product. Surface air temperature and precipitation(diurnal temperature range) appeared to contribute positively(negatively) to this change though climate variables displayed varying correlation with LAI for different time periods and different regions.     

过去50年气候变化下中国潜在植被NPP的脆弱性评价

借助动态植被模型IBIS,首先模拟了过去50年（1961-2010年）气候变化下中国潜在植被NPP的动态变化,然后采用IPCC第五次评估报告选定的标准气候态时段（1986-2005年）平均气候状态作为“标准年气候”,并将该气候条件下的潜在植被NPP作为评价基准。通过与基准进行比较,计算每一年潜在植被NPP的波动情况,进而评价该年的气候条件是否使潜在植被“不适应”以及“不适应”的程度,最后根据过去50年的“不适应”次数和程度综合判断气候变化下潜在植被NPP的脆弱性。评价结果显示：在过去50年的气候变化下,天山以南的暖温带荒漠生态系统、北方温带草原生态系统以及青藏高原西部的高寒草原生态系统更容易受到气候变化的不利影响,NPP呈现出较高的脆弱性;而大部分以森林为主的生态系统则不容易受到气候变化的影响,NPP脆弱性较低,其中以常绿阔叶林和针叶林为主的生态系统NPP脆弱性更低。此外,天山以北的温带荒漠生态系统以及青藏高原中部和东部的高寒草原草甸生态系统NPP也呈现出较低的脆弱性。     

NPP vulnerability of the potential vegetation of China to climate change in the past and future

Using the Integrated Biosphere Simulator, a dynamic vegetation model, this study initially simulated the net primary productivity (NPP) dynamics of China’s potential vegetation in the past 55 years (1961–2015) and in the future 35 years (2016–2050). Then, taking the NPP of the potential vegetation in average climate conditions during 1986–2005 as the basis for evaluation, this study examined whether the potential vegetation adapts to climate change or not. Meanwhile, the degree of inadaptability was evaluated. Finally, the NPP vulnerability of the potential vegetation was evaluated by synthesizing the frequency and degrees of inadaptability to climate change. In the past 55 years, the NPP of desert ecosystems in the south of the Tianshan Mountains and grassland ecosystems in the north of China and in western Tibetan Plateau was prone to the effect of climate change. The NPP of most forest ecosystems was not prone to the influence of climate change. The low NPP vulnerability to climate change of the evergreen broad-leaved and coniferous forests was observed. Furthermore, the NPP of the desert ecosystems in the north of the Tianshan Mountains and grassland ecosystems in the central and eastern Tibetan Plateau also had low vulnerability to climate change. In the next 35 years, the NPP vulnerability to climate change would reduce the forest–steppe in the Songliao Plain, the deciduous broad-leaved forests in the warm temperate zone, and the alpine steppe in the central and western Tibetan Plateau. The NPP vulnerability would significantly increase of the temperate desert in the Junggar Basin and the alpine desert in the Kunlun Mountains. The NPP vulnerability of the subtropical evergreen broad-leaved forests would also increase. The area of the regions with increased vulnerability would account for 27.5% of China.