Preimesberger et al. (2025) ESA CCI SM GAPFILLED Long-term Climate Data Record of Surface Soil Moisture from merged multi-satellite observations

Identification

• Journal: TU Wien Research Data
• Year: 2025
• Date: 2025-11-27
• Authors: Preimesberger, Wolfgang, Stradiotti, Pietro, Dorigo, Wouter Arnoud
• DOI: 10.48436/c0wbr-xf278

Research Groups

• TU Wien (Vienna University of Technology), Department of Geodesy and Geoinformation.

Short Summary

This study presents a global, gap-free long-term climate data record (CDR) of surface soil moisture spanning 1979 to 2024, derived from 19 different satellite sensors. The dataset utilizes a modified Discrete Cosine Transform Penalized Least Squares (DCT-PLS) algorithm to provide daily, spatially continuous estimates without relying on ancillary model-based variables.

Objective

• To create an independent, univariate, and gap-free global satellite-based climate data record of surface soil moisture that overcomes data fragmentation caused by limited satellite overpasses, frozen soils, dense vegetation, and radio frequency interference.

Study Configuration

• Spatial Scale: Global coverage on a $0.25^\circ \times 0.25^\circ$ (approximately $25\text{ km} \times 25\text{ km}$) horizontal grid (WGS84).
• Temporal Scale: Daily resolution covering the period from 1979 to 2024.

Methodology and Data

• Models used:
  • DCT-PLS (Discrete Cosine Transform Penalized Least Squares): A modified version of the Garcia (2010) algorithm used for interpolation and smoothing of gridded data.
  • Linear Interpolation: Applied specifically over periods of potentially frozen soils.
  • GLDAS Noah: Used as a reference reanalysis to calibrate uncertainty models for gap-filling performance.
• Data sources:
  • ESA CCI SM v9.2 COMBINED: Harmonized microwave observations from 19 satellites.
  • ERA5: Soil temperature data used to identify frozen soil conditions (temperature $< 0\text{ }^\circ\text{C}$).
  • GLDAS Noah reanalysis: Used for the calibration of uncertainty estimates.

Main Results

• Gap-Free Record: Successful generation of a continuous daily record of volumetric surface soil moisture ($m^3/m^3$) for the top $0\text{--}5\text{ cm}$ soil layer.
• Uncertainty Estimation: Provision of per-pixel uncertainty estimates that account for both the original satellite observation error and the error introduced by the gap-filling process, parameterized by gap size and local vegetation density.
• Methodological Independence: Demonstration that a univariate statistical approach (DCT-PLS) can effectively fill global satellite data gaps without the need for external land surface model drivers, preserving the observational nature of the CDR.
• Temporal Consistency: The product maintains consistency across the transition from early, data-sparse periods (1970s/80s) to the data-rich era of the last decade.

Contributions

• Univariate Gap-Filling: Provides the first global, long-term satellite soil moisture product that is gap-filled using only the original observational record, ensuring independence from model-based soil moisture estimates.
• Machine Learning Readiness: The gap-free nature of the product makes it directly compatible with machine learning architectures and hydrological models that require continuous spatio-temporal inputs.
• Enhanced Climate Monitoring: Facilitates the study of climate variability, drought applications, and land-atmosphere interactions by removing the "masking" effect of missing data in critical regions.

Funding

• European Space Agency (ESA) Climate Change Initiative (CCI) Plus Soil Moisture Project (CCN 4 to ESRIN Contract No: 4000126684/19/I-NB "ESA CCI+ Phase 1 New R&D on CCI ECVS Soil Moisture").

Citation

@article{Preimesberger2025ESA,
  author = {Preimesberger, Wolfgang and Stradiotti, Pietro and Dorigo, Wouter Arnoud},
  title = {ESA CCI SM GAPFILLED Long-term Climate Data Record of Surface Soil Moisture from merged multi-satellite observations},
  journal = {TU Wien Research Data},
  year = {2025},
  doi = {10.48436/c0wbr-xf278},
  url = {https://doi.org/10.48436/c0wbr-xf278}
}

Generated by BiblioAssistant using gemini-3-flash-preview (Google API)