Global Kilometer-level Spatiotemporal Continuous Surface Soil Moisture Dataset (1992-1997)

Document Title: Global 1km Surface Soil Moisture Dataset (1980-2023) DocumentationVersion: 1.0Date: March 1, 2025

Dataset Overview

This dataset is a global 1km surface soil moisture product independently developed by the research team led by Prof. Zhaoliang Li and Prof. Pei Leng at the Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences. The original data source is the ESA-CCI active-passive fused soil moisture product. The dataset integrates multi-source reanalysis and assimilation data to reduce uncertainties and fill missing values in the original ESA-CCI product. Key methodologies include:

(1) Triple Collocation Analysis (TCA) for fusing ERA5, GLEAM, and GLDAS soil moisture data.

(2) Downscaling from 25 km to 1 km resolution using subgrid variability modeling with SoilGrids high-resolution soil texture data and the Van Genuchten-Mualem model.

(3) Validation via ground observations from the International Soil Moisture Network (ISMN), achieving an unbiased root mean square error (ubRMSE) of 0.051 m³/m³.

Dataset Specifications

· Name: Global Kilometer-Level Spatiotemporally Continuous Surface Soil Moisture Dataset (1980–2023)

· Format: NetCDF

· Coordinate System: WGS84 Geographic Coordinate System

· Temporal Coverage: January 1, 1980 – December 31, 2023

· Spatiotemporal Resolution: 1km/10-day (Please be advised that due to limitations regarding data upload capacity, users requiring daily data are kindly requested to contact us: [e-mail])

· Data Source: ESA-CCI Soil Moisture Product

Data Organization

· File Naming Convention: yyyymm_xxxx.nc (e.g., 198001_early.nc represents Early January 1980).

· Units: Volumetric soil moisture content (m³/m³).

· Storage Details:

o Data is stored as int16 to optimize space.

o To obtain actual values, multiply stored data by 0.001 (e.g., a pixel value of 327 corresponds to 0.327 m³/m³).