GLASS SM: Global 1-km Daily Surface Soil Moisture from 2000 to 2020

A 21-year global daily surface soil moisture product at a 1-kilometer resolution, representing the volumetric water content of the uppermost soil layer (0–5 cm). It was generated by Yufang Zhang of Wuhan University using an ensemble machine learning model that fused GLASS satellite products and ERA5-Land reanalysis data. Validation results show a median unbiased RMSE of 0.052 m³/m³ and an R of 0.74.

Use Cases

• Modeling water, carbon, and energy exchanges between land and the atmosphere based on soil moisture data.
• Validating hydrological and climate models using long-term, high-resolution soil moisture estimates.
• Monitoring agricultural drought and soil conditions based on daily global surface soil moisture values.
• Analyzing spatial and temporal trends in an essential climate variable over a two-decade period.

Strengths

• 21 years of daily data from 2000 to 2020 provides a long-term record.
• High 1-kilometer spatial resolution offers detailed global coverage.
• Validation metrics include a reported median unbiased RMSE of 0.052 m³/m³ and R of 0.74.
• Product generation fused multiple satellite-derived and reanalysis datasets.

Limitations

• Data represents only the surface layer (0–5 cm), not deeper soil profiles.
• Column-level documentation is absent; field semantics must be inferred after download.
• The description does not specify file size or total data volume, which may affect handling.

Provenance

Source

Global Land Surface Satellite (GLASS) products suite, Wuhan University.

Collection Method

Generated using an ensemble machine learning model from GLASS albedo, LST, LAI products, ERA5-Land reanalysis, and auxiliary datasets.

Time Range

2000 to 2020

Freshness

Covers data up to 2020; last update date for the dataset product is unknown.

Geography

Global coverage.