植被指数是反映地表植被覆盖状况的重要参数,分析气候因子与植被指数间的相互关系有助于揭示气候变化对植被的影响,然而当前研究有两种分析植被指数与气候因子关系的方法,分别为分析植被指数与生长季内和生长季间气候因子的关系,然而这两种法差异如何,何种方法更为合适需要进一步分析。利用2000年-2009年生长季的MODIS的归一化植被指数NDVI(Normalized Difference Vegetation Index)数据集和藏北那曲地区3个气象站逐月气象资料(月平均气温、≥0℃活动积温和月降水量),分析比较了生长季内和生长季间气候因子对植被生长影响的差异,并分析了两种方法的优劣。结果表明:(1)生长季内植被NDVI与同期气温和降水量均呈高度正相关,生长季内时滞时间尺度为1个月时,植被NDVI对月平均气温及降水响应均最为强烈。(2)生长季间NDVI与同期降水量相关性并不明显,气候因子的滞后效应在生长季间也较弱。(3)生长季内和生长季间植被NDVI与气候因子的关系所得出的结论有一定差异性,可能是因为两方面的原因:生长季内植被NDVI与水热因子的高相关性与中国季风季候造成的高温多雨出现在夏季有关,而生长季内高水热条件与高植被指数对应的多年重复必然造伪的高相关系数,但这种相关性不一定能真实反映植被与水热条件的关系,而生长季间水热等气候因子与植被指数年际变化相关性分析不存在水热与高植被指数同期问题,更能真实反映气候因子年际变化对植被的影响。

Vegetation index is an important parameter that reflects the status of vegetation in an area. Analyzing the relationship between climatic factors and the vegetation index is helpful to fully understand the impact of climate change on vegetation. However, some conclusions on the relationship between the vegetation index and climatic factors are inconsistent across various time scales. Thus, this issue is addressed in the present study to enhance our understanding of the relationship between vegetation and climatic factors.
With the use of the Normalized Difference Vegetation Index (NDVI) data of moderate resolution imaging spectroradiometer (MODIS) during the growing seasons from 2000 to 2009, the monthly climatic factors (i.e., mean air temperature, accumulated temperature above 0℃, and monthly precipitation) of three observations in the northern Tibetan Naqu were combined, and the within-growing-season and cross-growing-season correlations between the NDVI and the climatic factors were analyzed. First, we preprocessed the data. To eliminate the interference of human factors, especially the urban buildings in the near site, we obtained the NDVI values outside the radius of 25km around the meteorological station. Second, we calculated the correlation coefficient between the NDVI and the monthly mean air temperature. Similarly, the correlation coefficient between the mean air temperature for the month ahead and NDVI was also calculated using the NDVI (4-9 months) and the monthly mean temperature series (3-8 months). The same process is applied to the two months ahead and the other factors. Third, we calculated the correlation coefficient between the NDVI and the mean air temperature of the month. Similarly, the correlation coefficient of the mean air temperature for the month ahead and the NDVI for April was calculated using the NDVI for April and the mean air temperature for March. The same process is applied to the two months ahead and the other factors.
The within-growing-season correlations between the NDVI and the temperature and precipitation factors were highly and positively significant, and the lag effects of the climatic factors on NDVI were most obvious for the one-month lag. By contrast, the inter-growing-season correlation between NDVI and precipitation was not significant, and the lag effect was much weaker than the within-growing-season lag effect. Therefore, the correlations between the NDVI and climatic factors vary between the within-growing-season and the inter-growing-season. Such a variation can be attributed to two aspects:the within-growing-season correlation fully considered the synchronization of the rainfall and temperature, whereas the inter-growing-season did not; the difference in sample sizes resulted in different results.
In this paper, the relationship between NDVI and climatic factors is discussed at different time scales. Results show differences in some aspects. At present, most of the studies are based on the relationship between vegetation changes and climate factors in the growing season. The analysis of the relationship between vegetation development and climatic factors are more scientific and persuasive. In conclusion, much more attention should be paid to the different approaches to obtain the various correlations between NDVI and climatic factors. spects:the within-growing-season correlation fully considered the synchronization of the rainfall and temperature, whereas the inter-growing-season did not; the difference in sample sizes resulted in different results.