The ESA CCI projects will deliver the next generation of satellite derived geophysical parameters, with quantified uncertainties that will allow each parameter to be assessed against requirements from the Global Climate Observing System (GCOS) for Essential Climate Variables (ECV) and the Climate Modelling Community (CMC), represented within the CCI programme by the Climate Modelling User Group (CMUG). A critical step in the acceptance of the CCI products by the GCOS and CMC communities is providing confidence in the quality of each CCI product and its uncertainties through validation against independent data such as ground based reference measurements or alternate estimates from other projects and sensors.
The below figure shows a flowchart of the validation activities for Level 2 algorithm development and final ECV production.
Validation and/or product assessment is a multi-faceted undertaking because there is no reference data set that can be considered as a "truth" for the EO based surface soil moisture retrievals. In-situ soil moisture sensors installed near the soil surface come closest to what can be considered as truth.
However, even for these measurements there are two major problems that need to be considered: Firstly, the thin surface layer (0.5-5 cm) sensed by the microwave sensors is directly exposed to the atmosphere and the sun, while sensors need to be installed in a more protected deeper layer (typically at 5 cm depth). Therefore, the satellite measurements represent a more variable soil layer than the in-situ measurement. Secondly, there is a significant scaling problem as the in-situ measurements only represent a very small area (a few dm²) while the satellite senses an area in the order of hundreds of km². Therefore, even a truly extensive in-situ data base as now e.g. assembled in the International Soil Moisture Network, cannot give a final answer to what the accuracy of the ECV soil moisture products is. Instead, additional references and advanced mathematical tools such as triple collocation or data assimilation are necessary to capture the various aspects necessary for a complete error characterisation and to narrow down the uncertainty range of the error estimates.
Therefore, within the Soil Moisture CCI a number of different validation activities will be performed:
Point-wise validation with in-situ measurements
Comparison to global model simulations
Data assimilation experiments at catchment scale
Assessment of scaling errors with SAR-based retrievals
Validation over boreal and sub-arctic environments
A product validation plan (PVP) is currently in preparation. The first version is planned to be available in May 2012, providing a description of the methods and designs that will be applied in the calibration/validation of the CCI soil moisture products.
- No upcoming events available