MODIS植被指数监测农业干旱的适宜性评价

MODIS传感器提供的短波红外光谱波段为农业干旱遥感监测带来了新的机遇，因为它对植被水分十分敏感。本文选择中国东北松嫩平原为研究区，旱田为农业干旱的监测目标。基于2001—2010年的8天合成MODIS产品（MOD09A1），分别计算了四种基于可见光和近红外光谱的植被绿度指数和四种基于近红外和短波红外光谱的植被水分指数，并以多尺度标准化降水指标(SPI)为判别植被指数农业干旱敏感性的标准，利用一种气象站点与象元配对关联的方法计算了不同植被指数与多尺度SPI的皮尔逊相关系数。研究表明，在农业干旱监测敏感性方面，MODIS植被水分指数（NDII6和NDII7）明显好于植被绿度指数。其中NDII7的表现最为出色，研究证实了MODIS短波红外光谱在监测农业干旱方面的潜力，为今后相关研究提供了新的见解。

Evaluating the Utility of MODIS Vegetation Index for Monitoring Agricultural Drought

The exclusive shortwave bands provided by MODIS sensors offer new opportunities for agricultural drought monitoring, since they are very sensitive to vegetation moisture. In the present work, we selected Songnen Plain in Northeast China as study area aiming at monitoring agricultural drought of dry farmland here. Four types of vegetation water indices and vegetation greenness indices were calculated from the 8-day composite MODIS product (MOD09A1) in vegetation growing season between 2001 and 2010, respectively. Multi-scale standardized precipitation index (SPI) derived from precipitation data of weather stations was used as reference data to estimate drought sensitivity of various vegetation indices, and a pixel-to-weather station paired correlation approach was used to calculate the Pearson correlation coefficient between vegetation index and SPIs. The result indicated that vegetation water indices established by near infrared and shortwave infrared bands outperformed vegetation greenness indices based on visible and near infrared bands. Of these indices, NDII7 performs the best with highest correlation coefficients across all SPIs. The authors’ results demonstrated the potential of MODIS shortwave spectral bands in monitoring agricultural drought, and this provides new insights to future research.