基于SPEI和NDVI的中国流域尺度气象干旱及植被分布时空演变

近些年中国干旱灾害频频发生且其影响不断加剧,因此探讨干旱对植被生长的影响对农业生产具有重要的意义。该文基于标准化降雨蒸散指数(standard precipitation evapotranspiration index,SPEI),应用Mann-Kendall(MK)趋势分析方法研究了流域尺度的气象干旱格局;利用新一代归一化植被指数(normalized different vegetation index,NDVI)分析了植被的生长状况;并探讨了SPEI和NDVI的相关性,结果表明:1)1982-2012年间,SPEI值在各时间尺度上都呈现为微弱下降趋势,即为微弱的干旱化;在空间上,干旱化趋势主要分布于东北、黄土高原和西南地区,而西北地区则呈现出明显的湿润化趋势;2)全国年及各季(冬季除外)NDVI序列均呈现上升趋势,空间上其下降趋势主要出现在西北内陆和东北地区部分流域,上升趋势则因季节而异;3)在年时间尺度上,SPEI与NDVI的相关分析表明,209个流域中有56个通过正相关检验(P0.05),主要分布在新疆北部、华北以及东北地区,负相关主要分布在秦岭淮河以南地区,气象干旱对植被生长的影响在干旱半干旱地区表现得更加明显。研究结果可为中国流域尺度干旱影响评估提供参考依据。

Assessing impacts of meteorological drought on vegetation at catchment scale in China based on SPEI and NDVI

Abstract: Drought is a natural disaster caused by long-time water shortage and uneven water distribution, which is one of the most common disasters around the world. Under the background of global warming, droughts have become more and more frequent and devastating, causing negative impacts on agricultural production, ecological environment and social economy. For the protection of vegetation and crops, it would be of practical significance to investigate how vegetation develops under the influence of drought which varies in time and space. This research applied the standard precipitation evaporation index (SPEI) derived from high spatial resolution dataset of precipitation and evaporation, as well as the newly published normalized difference vegetation index (NDVI) which represents the growth condition of vegetation on land. Mann-Kendall (MK) test and correlation analysis were taken to study the spatial and temporal evolution of SPEI and NDVI at catchment scale with the help of ArcGIS 10.1 software. The results of the research illustrated that: 1) The MK value of SPEI at catchment scale showed great coincidence with that of the 4189 grid points both in time and space, indicating the feasibility to investigate the response of vegetation growth under the influence of meteorological drought at catchment scale. SPEI was getting increasing slightly at different time scales during 1962-2012, indicating that China as a whole was being more humid. However, taking the time series of 1982-2012 as the research object, China was getting drier slightly at different time scales. In space, both at annual and at seasonal scale, drought trends were mainly detected in catchments in the Loess Plateau area, Northeast and Southwest China, while catchments in Northwest China presented significant humidifying tendency. 2) The NDVI value of the whole China was upgrading with time during 1982-2012, among which the NDVI in spring, growing season and whole year had a significant increasing trend. In space, some catchments of northern Xinjiang and Northeast China demonstrated a significant downgrading tendency, while the upgrading tendency was different owing to different time scales. 3) At annual scale, 56 of the 209 catchments showed significant positive correlation (P<0.05) between SPEI and NDVI, and most of them were located in north Xinjiang, and northern and northeastern China, which indicated that vegetation in arid and semiarid area was significantly affected by droughts. At the same time, negative correlation was mainly distributed in southern China (few of them passed the significant level of 0.95), indicating that droughts in southern China exerted little influence on vegetation growth. 4) Correlation analysis between seasonal NDVI and SPEI showed that in summer and autumn, catchments with positive relation mainly occurred in north area, while catchments with negative correlation mainly occurred in Southeast China and the middle and lower reaches of the Yangtze River plain. In spring and winter, negative relation came to dominate, which mainly occurred in Northeast China, the Yangtze River Basin and most catchments in Southeast China. Lag-relation analysis between seasonal NDVI and SPEI one or two seasons early showed similar correlation distribution as the result of relation analysis without lag. 5) For most of the catchments, air temperature, precipitation and evaporation exerted significant influence on NDVI of vegetation; among them, temperature played a leading role on the vegetation growth in catchments in southern China, while precipitation played a leading role on the vegetation growth in catchments in northern China. Our study on the relationship between meteorological drought and vegetation growth can provide a scientific basis for drought predication and precaution, and thus help to guide agricultural production and protect vulnerable ecosystems.