Multiple cropping index of Northern China based on MODIS/EVI

Multiple cropping is an important mean for increasing regional grain output and also a crucial cropping pattern in China’s farming system. This study proposed a new method for extracting multiple cropping index (MCI) on pixel level with multi-temporal moderate-resolution imaging spectroradiometer (MODIS) enhanced vegetation index (EVI) data based on the crop phenology and decision tree (DT). The method could be divided into two steps. First of all, according to the local crop phenology, several features were put forward for discriminating the pixel-level MCI, which contained three types: fallow, single cropped and double cropped. Second, the threshold for each feature was brought up by using CART Algorithm. Finally, the multiple cropping index of 15 provinces of Northern China were extracted in 2005. Then, the result was compared with that of former researches, and it shows that DT method is more efficient.