Quantitative Precipitation Estimation (QPE) by exploiting the potential of advanced radar observations and data assimilation

点击翻译按钮获取中文摘要

The overarching objective of this project is to improve hydrological modeling on the mesoscale by providing quantitative ground precipitation estimates (QPE) in an adequate spatiotemporal resolution. Target scale is 1.000 km² in 1000 m / 1 hour resolution. To this end, we will investigate and combine two QPE approaches: The first is to combine convection-permitting atmospheric modeling and new observations using advanced data assimilation techniques (Variational techniques, Ensemble Kalman Filters and combinations thereof). Key observations are C-band radar reflectivity, radial velocity, and polarization, Micro Rain Radar (MRR) vertical profiles of reflectivity, rain rate, and drop size distribution (DSD), as well as distrometer DSD. Particularly, we are exploring the value of MRR observations and of C-Band radar polarimetric measurements. High-resolution DA is a promising tool for QPE as it takes full advantage of the 3D information content of radar data. The physically consistent 3D thermodynamic field produced by the model allows quality control of observational data and acts as a dynamic-physical interpolator at unobserved locations. The second approach is QPE based on advanced observations, consisting of a synergy of radar reflectivity and polarization measurements, as well as MRR, distrometer, and raingauge observations using advanced geostatistical interpolation and simulation techniques. We will also explore ways to produce hybrid combinations of modeland observation-based QPE fields. Test area is the Alzette catchment in Luxembourg (1090 km²), with the focus area of the research unit, Attert (288 km²), nested within.