Changes in extent of open-surface water bodies in China's Yellow River Basin (2000–2020) using Google Earth Engine cloud platform

Abstract

Dynamics of open surface water is of great significance to understand the impacts of global changes and human activities on water resources. However, long-term changes of open-surface water bodies in the Yellow River Basin have remained poorly characterized. In the study, 26,681 high-quality Landsat images were processed using the Google Earth Engine (GEE) cloud platform to study the long-term changes of the open water bodies in the Yellow River Basin from 2000 to 2020. A percentile-based image synthesis method was used to collect training samples. The multi-index water body extraction rule (MIWER) was proposed to rapidly extract surface water bodies. The results show that (1) The MIWER is suitable for long-term and large-scale surface water mapping based on Landsat images; (2) The average area of permanent and seasonal water in the last 20 years is 9062.59 km² and 6918.83 km², respectively, accounting for 56.7% and 43.3% of the total surface water in the Yellow River Basin, respectively; (3) From 2000–2020, both permanent and seasonal water in the whole Yellow River Basin and sub-basins have increased. In general, the total open surface water shows an increasing trend with obvious spatial heterogeneity; (4) The changes of surface water in the Yellow River Basin are related to climate change and intensive human activities, but the influencing factors vary in different regions. The findings of this study demonstrate a novel robust, low-cost water extraction method that satisfies the requirements of terrestrial water inundation mapping and management across the Yellow River Basin. The methodology of this study is expected to be extended to global surface waters, providing valuable information for global surface water resource planning and management.