The characteristics of long-term spatiotemporal changes in land surface water bodies in China
YIN Xiaolan
NIU Zhenguo
KE Yinghai
ZHOU Demin
Abstract：The dynamic change of land surface water bodies is one of the key indicators reflecting variations in water resources.Accurately understanding such dynamics is of great significance for studying the water cycle of the Earth system and its responses to climate change and human activities.However,existing data products on land surface water body distribution are limited in terms of spatiotemporal resolution,coverage,and continuity.To address these issues,a high-precision land surface water body distribution model was constructed by integrating multi-source geospatial data and applying machine learning algorithms,resulting in a monthly 30 m resolution water body dataset covering China from 1985 to 2021.Using third-level basins as the analysis unit and employing time series decomposition methods,the spatial distribution patterns and temporal evolution trends of land surface water bodies across China were analyzed.The results show that:(1)The land surface water body distribution model developed through the fusion of multi-source geospatial data and machine learning effectively compensates for the limitations of remote sensing-based water body monitoring.(2)The constructed monthly dynamic dataset from 1985 to 2021 demonstrates good spatiotemporal continuity.The area of land surface water bodies in China ranged between 116,900 km2 and 157,600 km2,showing an overall declining trend.The decline was more pronounced before 2003,followed by a stabilization and slight rebound thereafter.There is a clear spatial heterogeneity in the seasonal variation of water body areas,with an overall increasing trend.The seasonal difference becomes more prominent from the southeastern coastal areas(slope<-0.005)to the northwestern inland regions(slope>0.2).(3)Human activities,such as urban construction and reservoir development,are the primary driving factors of seasonal changes in water bodies,accounting for 70%of the variation.The average contribution of human factors across basins is 56.26%.This study provides valuable data support for research on water resources management and land surface water cycles in China under the influence of global climate change and human activities.
Keywords：land surface water bodiesmachine learningreconstruction of water body distributionspatiotemporal changes of water bodiesdriving factors
Publication Date：2025-08-12
Online Publishing Date：2025-08-22(First online date of this platform, not the publication date of the document)
Pages：11( 27-37 )