Merging precipitation data from rain gauge-, microwave link-, and radar measurements for incorporation into a fully coupled data assimilation framework

点击翻译按钮获取中文摘要

Project P3 aims at an improved representation of the spatiotemporal distribution of precipitation based on sparse and/or inaccurate observations. We will further develop merging methods for different observation types including rain gauges, commercial microwave links, and weather radars, and we will use our merged precipitation product as super-observations for data assimilation in the fully coupled terrestrial Systems Modeling Platform (TerrSysMP). The merging methodology encompasses a copula-based reconstruction of high-resolution precipitation fields from different combinations of observation types. Subsequently, different approaches for incorporating the merged precipitation product within the data assimilation framework will be implemented and evaluated. Approaches will range from replacement of the model precipitation by the precipitation product for the land components of the coupled model (insertion), to weakly coupled data assimilation by restricting assimilation to the atmosphere component, and to a fully coupled data assimilation. The results will be evaluated for both the analyses and predictions with respect to the minimization of errors in the simulated (for both the weakly and strongly coupled approach) and predicted spatiotemporal distribution of precipitation. Here, also the mutual relation of precipitation with other hydro-meteorological variables is considered, as well as computational costs.