earth online
  • News
  • Ensuring soil moisture data qu...

Ensuring soil moisture data quality with reference measurements

10 Jan 2024

ESA is taking steps to verify the accuracy of soil moisture estimation with the Fiducial Reference Measurements for Soil Moisture (FRM4SM) project and a new version of the Quality Assurance for Soil Moisture (QA4SM) service.

While Earth Observation (EO) data are vital for monitoring and understanding environmental change, their real value can only be reached if we can trust their accuracy.

The ESA-led FRM4SM project aims to define procedures and protocols to qualify in situ Soil Moisture (SM) measurements – Fiducial Reference Measurements – so that they can validate the quality of soil moisture products over the entire duration of a satellite mission.

As part of this project, ESA supports the QA4SM service, an online validation service that validates SM products against in situ and model reference data. A new version of the QA4SM platform has been released in early 2024, with a user-friendly facelift and new helpful video tutorials to guide users through the validation process.

QA4SM facelift
QA4SM facelift

Learn more about these initiatives in this Q&A with the FRM4SM project team: Raffaele Crapolicchio, ESA’s Data Quality Manager for the Soil Moisture and Ocean Salinity (SMOS) mission; Alexander Boresch, project lead at AWST in Austria; and Alexander Gruber, science lead at TU Wien in Austria.

 

ESA: How does ESA address the challenge to quantify uncertainties in satellite data products?

Raffaele Crapolicchio: To assess the uncertainties of EO data products, we need ground reference measurements. However, obtaining a reliable ground “truth” to compare satellite measurements against is not easy - not only can data quality vary greatly across ecoregions, it is also time consuming and costly to obtain enough ground samples to cover each region. Sometimes this is not practically feasible, due to inaccessibility or harsh environmental conditions.
Fortunately, vast amounts of in situ data already exist due to the worldwide interest in Earth system variables and ESA efforts to set up campaigns for satellite data validation. ESA is also focused on making collective use of these data to address challenging aspects, such as the differences in sensor technology, sampling protocols, data quality and uncertainties.

Raffaele Crapolicchio, ESA’s Data Quality Manager for the Soil Moisture and Ocean Salinity (SMOS) mission
Raffaele Crapolicchio, ESA’s Data Quality Manager for the Soil Moisture and Ocean Salinity (SMOS) mission

 

Alexander Gruber, Science Lead for FRM4SM project at TU Wien in Austria
Alexander Gruber, Science Lead for FRM4SM project at TU Wien in Austria

ESA: Are there any recognised international standards to collect in situ data for satellite validation?

Alexander Gruber: Not for the collection because this varies greatly depending on the variable in question. What does exist are high-level requirements of what makes in situ measurements suitable for satellite product validation. They have been laid out by the Quality Assurance for Earth Observation (QA4EO) framework, which was established by the Committee of Earth Observation Satellites (CEOS). These guidelines define so-called “fiducial reference measurements” (FRMs) as in situ measurements that are independent from the satellite retrieval, have documented and maintained uncertainty budgets, and are established using SI-traceable metrology standards. FRMs are used to validate satellite products following certain FRM protocols, but for soil moisture, such protocols and metrological standards have yet to be defined.

 

ESA: How does ESA address the FRM concept for soil moisture data products?

Raffaele Crapolicchio: ESA has initiated several FRM activities to foster the estab-lishment of such independent, fully characterised, and traceable ground meas-urements for the different Earth system variables. One such activity is the Fiducial Reference Measurements for Soil Moisture (FRM4SM) project, which has made considerable progress in building trust in satellite soil moisture data products.

 

ESA: How is the FRM4SM project set-up?

Alexander Boresch: FRM4SM is a joint project of the Angewandte Wissenschaft, Software und Technologie GmbH (AWST, Austria), Technische Universität Wien (TU Wien, Austria), and Centre d ́Etudes Spatiales de la BIOsphère (CESBIO, France). Its cornerstones are the International Soil Moisture Network (ISMN) database and the Quality Assurance for Soil Moisture (QA4SM) online validation platform.

 

ESA: What is the International Soil Moisture Network?

Alexander Gruber: The ISMN is the largest database of in situ soil moisture measurements existing today and the most important data source for satellite soil moisture validation. It provides harmonised and quality-controlled in situ data from more than 3000 measurement stations via an online data portal.

 

Alexander Boresch, project lead at AWST in Austria
Alexander Boresch, project lead at AWST in Austria

 

ESA: What is QA4SM and what are the platform updates since January 2024?

Alexander Boresch: QA4SM is an online validation platform that allows users to validate and compare satellite (but also other) soil moisture products, with ground- and model-based reference data sets, following the validation good practice guidelines that have been developed by the community that are endorsed by the CEOS working group on calibration and validation [1] [2].

In the new January 2024 version, we have given the platform a facelift to make it more intuitive and user friendly. We have also added video tutorials to guide the users through the validation process.

 

 

 

ESA: Tell us about research and development activities within FRM4SM?

Alexander Gruber: FRM4SM is developing guidelines and protocols to establish SI-traceable uncertainty budgets for satellite soil moisture products in collaboration with the community. We also aim to provide recommendations for the installation, operation, and maintenance of fiducial reference measurement networks for soil moisture. Lastly, to demonstrate the utility of FRMs for soil moisture, several case studies based on SMOS soil moisture estimations are carried out as part of the project. These case studies aim to investigate the sensitivity of the SMOS validation metrics to different features, such as the probe's representativeness for the satellite footprint, the spatial and temporal collocation strategy, or the land surface composition within a satellite footprint.

SMOS data products
SMOS data products

 

ESA: How can the scientific community support the FRM concept?

Alexander Gruber: Building trust on decisions based on Earth observation data requires global, coordinated efforts and great amounts of fiducial reference data. We encourage scientists, users, and data providers to consult and share the recommendations laid out by the FRM4SM project, CEOS, and QA4EO, as well as to adhere to FAIR (findable, accessible, interoperable, and reproducible) principles whenever possible. We also invite anyone collecting in situ soil moisture data to consider sharing them with the ISMN.
Finally, in the face of climate change, we stress the need for establishing stable, long-term fiducial reference networks that can serve as a benchmark for the entire lifetime of current EO satellites and their successive generations, in as many ecoregions as possible. We hope that FRM4SM inspires future efforts for other Earth system variables and aim for a community-driven framework that builds trust in EO data to maximise their benefits for society.

 

References

  1. Gruber et al. (2020). “Validation practices for satellite soil moisture retrievals: What are (the) errors?” Remote Sensing of Environment, 244, 111806.
  2. Montzka et al. (2020). “Soil Moisture Product Validation Good Practices Protocol Version 1.0.” In: C. Montzka et al: Good Practices for Satellite Derived Land Product Validation (p. 123), Land Product Validation Subgroup (WGCV/CEOS).
Tweet