Optimized Polarimetric Radar Relations for Snow Estimation

Radar snow quantitative precipitation estimation (QPE) is challenging because of the uncertainties originating from the natural variability and diversity among snow growth habits, particle size distributions, and snow densities. However, owing to the plethora of additional information compared to conventional radars, polarimetric measurements can significantly reduce biases and errors in radar snow QPE. Recently introduced generalized polarimetric bi-variate power-law relations for snowfall rate estimation based on the joint use of radar reflectivity Z and specific differential phase Kdp (S(Kdp, Z); Bukovčić et al., 2020) that depend on the particle aspect ratio (ar) and width of the canting angle distribution (σ), are updated to take into account the degree of riming (frim) parameter. The relations are optimized using the retrievals of frim, particle orientation, and shape parameters (Fo and Fs – originating from σ and ar), and the gamma size distribution shape parameter μ from the ASOS stations measurements. Optimized polarimetric relations for snowfall rate estimation are tested with the S-band WSR-88D data against standard S(Z) and in situ measurements for several heavy snow events. Polarimetric estimates exhibit smaller biases in comparison to S(Z) close to the surface when the low-altitude Kdp measurement is reliable (Kdp > 0.03-0.05 deg/km). S(Kdp, Z) usually performs poorly in heavily aggregated snow associated with weak or negative Kdp. One venue for improvement is to replace S(Kdp, Z) with S(Z) if the latter (or some other) relations work better for a given situation.