An integrated retrieval framework for multiple polarization, multiple frequency radar networks

Abstract

Radar networks form the backbone of severe weather and remote sensing in throughout most of the world. These networks provide diverse measurements of weather phenomenon, but ultimately are measuring indirect parameters rather than detecting the physics of the situation. One of the long standing goals of weather remote sensing is to relate the measurements from the various instruments to the physics that give rise to the measurements. Weather radar networks give both a better spatial coverage than single radars, as well as providing multiple looks at the environment. Newly developed radar networks have started to incorporate multiple frequencies and multiple polarizations to take advantage of attributes of different radar frequencies. Raindrops occupy different scattering regimes based on the frequency of the radar being used. Based on this, multiple radars at different wavelengths provide unique information about the microphysical characteristics of the atmosphere. Nonetheless, very little work has been conducted on fusing multiple radar measurements at heterogeneous frequencies to improve microphysical retrievals. This work presents a forward variational algorithm for multiple radar fusion that retrieves microphysical parameters from the atmosphere. The single radar case and the multiple radar case will both be addressed. Ground instrumentation will be used for verification, and the spatial and temporal variability of precipitation microphysics will be discussed.