1000m resolution water production service dataset for the Loess Plateau, 1992-2020

This dataset is based on the VIC (Variable Infiltration Capacity) hydrological model, integrating multi-source remote sensing surface parameters, meteorological reanalysis data, and soil data to construct a spatiotemporal continuous dataset of 1000 m resolution water production services on the Loess Plateau from 1992 to 2020. The research area covers five major watersheds on the Loess Plateau (33 ° 39 ′ 0 ″ -42 ° 0 ′ 0 ″ N, 98 ° 31 ′ 12 ″ -113 ° 38 ′ 24 ″ E), with a total area of approximately 623000 km ². The dataset quantifies the spatiotemporal pattern of water production services and hot areas of supply-demand imbalance before and after the implementation of the Grain for Green Project through a gridded supply-demand ratio algorithm. The data processing was carried out using ArcGIS spatial analysis and Fortran programming, combined with the measured runoff data from Huaxian Hydrological Station for model validation. The Nash efficiency coefficient (Ens) reached 0.744 during the calibration period (1992-2006), increased to 0.872 during the validation period (2007-2020), and the relative error (Er) was less than 3.4%. The dataset contains 29 years of water production service raster data (in *. tif format) with a data volume of 7.52 MB, featuring long-term continuity, spatial clarity, and process traceability. The dataset can be used for ecological engineering hydrological effect assessment, vegetation restoration and water resource carrying capacity analysis, vulnerability prediction of water production services under climate change scenarios, and optimization of national spatial ecological restoration planning, providing high-precision data support for ecological governance in the Yellow River Basin and adaptive management of human land systems in semi-arid regions worldwide.