Precipitation Quantification Informed by Polarimetric Phased Array Radar Process Observations

Microphysical observations associated with clouds, convection, and precipitation are constrained by the capabilities of the instruments used to measure them. Phased array radar (PAR) technology offers unprecedented spatial sampling to derive fingerprints of microphysical processes, e.g., in the vertical. This work explores an approach for process classification using uniquely derived polarimetric PAR returns and a machine learning framework for improved quantitative precipitation estimation primed by process information. Probabilistic microphysical characteristics and precipitation rates estimates that account for uncertainty and errors are derived using the developed classification and retrieval framework and applied on PAR observations and bin microphysics modeled scenarios.