Phenology-based classification of crop species and rotation types using fused MODIS and Landsat data: The comparison of a random-forest-based model and a decision-rule-based model

•We proposed rotation classification models based on Random Forest and decision rule.•The overall accuracy of the two classification models was 90.0 % and 89.7 %.•High spatiotemporal NDVI time series reduced statistical bias caused by mixed pixels.•Random-Forest-based model was suitable for mapping crops over many growing seasons.•Decision-rule-based model was suitable for mapping crops over one growing season.