The agricultural pastoral ecotone of Northern China (APENC), as a fragile ecological zone, has experienced dramatic land use/land cover changes (LULCC) owing to intensive human disturbances. The understanding of the impacts of LULCC on climate, especially the diurnal temperature range (DTR), is not sufficient in this region. The spatio‐temporal processes of LULCC and its impacts on DTR were quantitatively analyzed, using the Terra/Aqua MODIS data (MYD11A2 and MCD12Q1, 2003–2013), on the premise of eliminating or weakening the effects of background climate. Results showed that the areas of croplands, forests, and waterbody increased with the annual rate of 2.84%, 0.95%, and 8.2%, respectively, whereas grasslands and bare land decreased with the annual rate of 0.69% and 7.32%, and built‐up remained quite stable during the period of 2003–2013. The annual changing rate of LULC for 2008–2013 was nearly 2.84 times the value for 2003–2008. In general, LULC contributed to a decline of DTR over the whole APENC both at monthly/seasonal timescale. Croplands, forests, and mutual transformation between croplands and grasslands tended to decrease daytime maximum temperature (T_max), resulting in a decline of DTR. The grasslands could reduce the DTR by decreasing T_max and increasing nighttime minimum temperature (T_min). By contrast, the conversion of grasslands into forests tended to increase T_max and decrease T_min, thus a rise of DTR for winter, summer, and autumn had observed. These results can assist land resource development and relevant policies making in APENC.

Estimating terrestrial water storage (TWS) with high spatial resolution is crucial for hydrological and water resource management. Comparing to traditional in-situ data measurement, observation from space borne sensor such as Gravity Recovery and Climate Experiment (GRACE) satellites is quite effective to obtain a large-scale TWS data. However, the coarse resolution of the GRACE data restricts its application at a local scale. This paper presents three novel convolutional neural network (CNN) based approaches including the Super-Resolution CNN (SRCNN), the Very Deep Super-Resolution (VDSR), and the Residual Channel Attention Networks (RCAN) to spatial downscaling of the monthly GRACE TWS products using the outputs of the Ecological Assimilation of Land and Climate Observations (EALCO) model over Canada. We also compare the performance of CNN-based methods with the empirical linear regression-based downscaling method. All comparison results were evaluated by root mean square error (RMSE) between the reconstructed GRACE TWS and the original one. RMSEs over the matched pixels are 22.3, 14.4, 18.4 and 71.6 mm of SRCNN, VDSR, RCAN and linear regression-based method respectively. Obviously, VDSR shows the best accuracy among all methods. The result shows all CNN-based super resolution methods preform much better than traditional method in spatial downscaling.