MOD09A1

该数据集名为MOD09A1，用于评估DGIN模型，其中包含了来自MODIS的时空数据，这些数据反映了澳大利亚东部的植被情况以及格陵兰岛的雪盖情况。所有数值都经过重新缩放，乘以了100的因子。DGIN模型训练时使用了不同空间足迹（50公里×50公里和100公里×100公里）以及不同时间长度（80天）的序列。在澳大利亚，数据集的方差适中（约为240），而在格陵兰岛，方差较高（约为1900）。该任务的目的是进行时空插值。

The dataset named MOD09A1 is utilized for evaluating the DGIN model. It contains spatiotemporal data from MODIS, which reflects the vegetation conditions over eastern Australia and snow cover over Greenland. All numerical values have been rescaled by a factor of 100. The DGIN model was trained using sequences with varying spatial footprints (50 km × 50 km and 100 km × 100 km) and different temporal durations (80 days). In Australia, the dataset has moderate variance (approximately 240), while in Greenland, the variance is relatively high (approximately 1900). The objective of this task is spatiotemporal interpolation.