利用MODIS fAPAR傅立叶时间序列分析研究植被光合作用活动对净辐射和降雨的响应：青藏高原个例研究

气候变化对植被动力学有非常大的影响。为了定量描述气候变化对植被的影响，文章利用MODIS fAPAR 数据和NCEP 的净辐射和降雨再分析数据对青藏高原地区气候变化对植被的影响进行了时间序列分析。研究所用的数据时间跨度为2000年至2005年。首先利用NCEP 再分析数据建立了干旱度因子的时间序列，为了与MODIS fAPAR 具有相同的时间采样间隔，由NCEP的日净辐射和日降雨量得到每8天的平均净辐射和8日降雨的和。根据一定时间间隔的净辐射与降雨量的比可以用来衡量相对于可利用水分的剩余能量，因此该比值也是干旱灾害的度量。其次，对MODIS fAPAR 的傅立叶时间序列分析提供了两个植被光合作用对干旱相应的因子，即fAPAR的年平均值及其年振幅值。在时间和空间尺度上对植被光合作用活动与干旱指数之间的关系进行了定量分析。对湿年和干年之间的响应差异进行了比较。研究表明较干地区对气候变化的响应最为显著。分析应该扩展到更长的时间跨度以便更加有效地在时间和空间尺度上评估气候变化对植被动力学的影响。

Response of Vegetation Photosynthetic Activity to Net Radiation and Rainfall: A Case Study on the Tibetan Plateau by Means of Fourier Analysis of MODIS fAPAR Time Series

Climate variability has a large impact on the vegetation dynamics. To quantify this impact in the Tibetan plateau a study was carried out using time-series of MODIS fAPAR satellite data products and NCEP net radiation and rainfall re-analysis data. The data set spanned over the years between 2000 and 2005. The NCEP data are used to construct a time series of a radiational indicator of drought: daily net radiation and rainfall data for each NCEP grid are integrated over a period of eight days to match the temporal sampling interval of MODIS data products. The ratio of net radiation over rainfall for a given period of time is a measure of excess energy relative to available water and is therefore a measure of drought hazard. Fourier analysis of time series of the MODIS fAPAR provides two indicators of the response of vegetation photosynthetic activity to drought, as measured by the indicator just described. The two indicators used in this study are the mean yearly fAPAR value and its annual amplitude. The algorithm used (HANTS) fits iteratively a Fourier series to a set of irregularly spaced observations, after elimination of outliers, such as due to cloud-contaminated observations. The relationships between photosynthetic activity of vegetation and the radiational drought hazard indicator are determined and quantified spatially and temporally. The response during the wettest respectively driest year during the period covered by available observations was compared. The drier areas prove to be the most sensitive to climate impact. The analysis should be extended over a longer period of time to obtain a more robust assessment of climate impact on vegetation dynamics, particularly as regards the response of vegetation to temporal respectively spatial variability of climate.