Evaluating the impact of a time-evolving constellation on multi-platform satellite based daily precipitation estimates
Abstract
Satellite based precipitation climate data records (CDRs) have recently emerged and provide new observational sources to characterize of the changing nature of global precipitation. These products rely on the use of passive microwave instruments. At the daily scale, these CDRs are prone to performance sensitivity resulting from the availability of microwave observations. As the configuration of the microwave sounders and imagers fleet evolves over time, adding new satellites and instruments or losing old platforms, the climate-oriented performances of the CDRs are likely impacted. In this study, this effect is quantified using a prototype constellation-based quasi-global precipitation product algorithm and data-denial experiments. The constellation change has a small impact of the long-term average climatology both in terms of mean and distribution recalling the resilience of the climatology of such a multi-platform product to the fluctuations of the amount of available input data. The interannual variability on the other hand is more impacted. More large rainfall amounts are relatively more perturbed than the lower rain daily accumulation with anomalies up to 30% for some configurations. The method to correct for the artefact is detailed and while some aspects of the computations are product-specific, the major outcome of this study should apply to various similar products as well.
Origin : Publisher files allowed on an open archive
