Global Soil Moisture Dataset From a Multitask Model (GSM3)

Description

GSM3 (Global Soil Moisture from a Multitask Model, v1.0) is a daily, 9 km global soil moisture dataset spanning 2015–2020. It was generated using a multitask deep learning model that simultaneously learns from multiple soil moisture networks worldwide, enabling robust generalization across diverse climatic and land cover conditions. 

The dataset and methodology are described in detail in the companion paper below. If you use this dataset, please cite the paper:

Liu, J., Hughes, D., Rahmani, F., Lawson, K., and Shen, C.: Evaluating a global soil moisture dataset from a multitask model (GSM3 v1.0) with potential applications for crop threats, Geoscientific Model Development, 16, 1553–1567, https://doi.org/10.5194/gmd-16-1553-2023, 2023.

Interactive Viewer

Explore the dataset in your browser (no download required): https://jayhydro.github.io/gsm3-viewer/

Data Specification

Shortname: GSM3
Longname: Global Soil Moisture Dataset From a Multitask Model
Version: 1.0
Format: GeoTIFF
Spatial Coverage: Global
Temporal Coverage: 2015-01-01 to  2020-12-31
File Size: ~10.3 MB per file
Data Resolution
Spatial: 9-km
Temporal: Daily
Coordinate Reference System (CRS): EPSG:6933 - WGS 84 / NSIDC EASE-Grid 2.0 Global