Optimally leveraging radar observations to provide information on rain microphysical processes

Description

Current bulk microphysics schemes exhibit deficiencies that are due in part to their simplified representation of a complex natural state, and in part due to a fundamental lack of understanding of microphysical processes. These schemes also employ numerous ad-hoc assumptions, parameter choices, and process rate formulations that make both rigorous constraint by observations and quantitative estimates of uncertainty difficult or impossible. On the other hand, observations such as polarimetric radar provide direct information about cloud and precipitation properties but not the microphysical process rates themselves. To address this gap, we have developed a novel probabilistic parameterization framework called the Bayesian Observationally-constrained Statistical-physical Scheme (BOSS). BOSS combines existing, though limited, process level microphysical knowledge with flexible process rate formulations and parameters trained to observations through Bayesian inference. Using a raindrop size distribution (DSD) normalization method that relates DSD moments to one another via generalized power series, generalized multivariate power expressions are derived for the microphysical process rates as functions of a set of prognostic DSD moments in BOSS. The approach is flexible and can utilize any number and combination of prognostic moments and any number of terms in the process rate formulations. This allows for systematic quantification of both parametric and structural uncertainty associated with the process rate formulations, which is not possible using traditional schemes. Here we use a Monte Carlo Markov Chain sampler within the BOSS framework to constrain process rates and parameters directly with “synthetic” polarimetric radar observations. The synthetic observations are generated by a bin microphysics rainshaft model run over a wide range of conditions. We explore the information content that is gained from different observed polarimetric radar variables (ZH, ZDR, KDP) with respect to various modeled warm rain microphysical processes (evaporation, collision/coalescence, and drop breakup). The constraint of BOSS and information content gained by polarimetric radar is investigated for different configurations of BOSS, that is, with different combinations and numbers of the prognostic DSD moments (two-moment and three-moment). We show that prognosing the third and sixth moments rather than the traditional choice of zeroth and third moments in two-moment bulk