Surface water is a highly dynamical object on the earth’s surface. At present, satellite remote sensing is the most effective way to accurately depict the temporal and spatial variation characteristics of surface water on a large scale. In this study, a region-adaptive random forest algorithm is designed on the Google Earth Engine (GEE) for automatic surface water mapping by using data from multi-sensors such as Landsat 7 ETM+, Landsat 8 OLI, and Sentinel-1 SAR images as source data, and China as a case study region. The visual comparison of the mapping results with the original images under different landform areas shows that the extracted water body boundary is consistent with the water range in the image. The cross-validation with the JRC GSW validation samples shows a very high precision that the average producer’s accuracy and average user’s accuracy of water is 0.933 and 0.998, respectively. The average overall accuracy and average kappa is 0.966 and 0.931, respectively. The independent verification results of lakes with different areas also prove the high accuracy for our method, with a maximum average error of 3.299%. These results show that the method is an ideal way for large-scale surface water mapping with a high spatial–temporal resolution.

The moderate spatial resolution and high temporal resolution of MODIS imagery make it an ideal resource for time series surface water monitoring and mapping. We used MODIS MOD09Q1 surface reflectance archive images to create an Inland Surface Water Dataset in China (ISWDC), which maps water bodies larger than 0.0625 km2 within the land mass of China for the period 2000–2016, with 8 d temporal and 250 m spatial resolution. We assessed the accuracy of the ISWDC by comparing it with the national land cover derived surface water data and global surface water (GSW) data. The results show that the ISWDC is closely correlated with the national reference data with coefficient of determination (R2) greater than 0.99 in 2000, 2005, and 2010, while the ISWDC possesses very good consistency, very similar change dynamics, and similar spatial patterns in different regions with the GSW dataset. The ISWDC dataset can be used for studies on the inter-annual and seasonal variation of the surface water systems. It can also be used as reference data for verification of the other surface water dataset and as an input parameter for regional and global hydro-climatic models. The ISWDC data are available at: https://doi.org/10.5281/zenodo.2616035.

With changing climates globally, we see changes in not just average conditions, but also in extreme events, and such events require special attention due to their unpredictable yet significant impact on native biotas. One such event is the formation of a landscape scar at Zonag Lake caused by a climate change‐induced outburst flooding event that occurred on 15 September 2011. During the winter, the scar region became a new birthplace for sandstorms, and since the flooding, remote sensing monitoring shows that between 2011 and 2020, there were 285 sandstorm days (between November and March), relative to none prior. The outburst flooding event and consequential sandstorms threaten the key lambing area of the Tibetan antelope (Chiru), affect the water balance of the Zonag Lake and downstream lakes, and may even impact on the flow in the Yangtze River. Active human intervention may be needed to repair this new desert spit and reverse the slew of consequences which may otherwise lead to significant population declines in one of the major Chiru breeding grounds due to the progressive loss of vegetation productivity across their main breeding area.

Protecting and restoring the degraded arid lakes are globally urgent issues. We document a potential recovery of the dried salt-lake, Lop Nur called "the Sea of Death" which is located at the terminus of the largest inland basin in China, the Tarim River Basin. The changes and relationship of surface water with climate parameters and groundwater in the basin over the last 30 years are analyzed, by using satellite remote sensing and land data assimilation products. We find that with increased surface water in the basin, the groundwater level in Lop Nur began to show an obvious positive response in 2015; and the rate of decline of the groundwater level is slowing down. We argue that after a balance is achieved between regional groundwater recharge and evapotranspiration, the Lop Nur ecosystem will gradually recover. This study shows an encouraging case for the protection and restoration of degraded lakes in dryland regions around the world.