Testing a new Radar QPE methodology for winter events with a low melting layer

The RY precipitation product of the German Weather Service (DWD) is severely affected by the presence of the low melting layer and frequently shows circular features of enhanced precipitation around the radar sites during the winter time. 
The radars tend to be installed at relatively high terrain and to scan at elevations at a minimum of 0.5° in order to avoid beam blockage and ground clutter. In doing so two problems arise:

1) the difference between the ground and the radar beams becomes a problem especially at large distances from the radar, and consequently precipitation processes in the lowest layers are not observed.
2) the radar beam often reaches the melting layer and may even cross it where it is sampling the snow above.As a result problems arise when deriving surface QPE from the radar: regions of enhanced QPE in ring shapes around the radar sites, and underestimation of the precipitation beyond the melting layer.

A new methodology (PVPR - Polarimetric Vertical Profile of Reflectivity) developed by Ryzhkov et al. 2022 is tested here for which the radar reflectivity (Z_H) is reconstructed to correct for the effect of the melting layer and snow beyond. In this methodology the melting layer is detected independently for each azimuth based on the values of Z_H and ρ_HV (cross-correlation coefficient between horizontal and vertically polarized radar waves). In particular the range bin at which the melting layer was reached is recorded (mlb_r). The strength of the melting layer (ML_S) is defined based on how much the value of ρ_HV  dropped within the melting layer. The values of ML_r and ML_S at a specific elevation are considered sufficient to characterize the melting layer, and are then compared with lookuptables which were generated by simulations of the melting layer effect on the radar beam. A correction factor is then applied based on the lookuptables to the Z_H profile within and beyond the melting layer. Visually the result shows a smoother field of reflectivity without the obvious bright band and decreased values associated with snow at farther ranges.

In this study the PVPR methodology was used to correct Z_H which in turn was used to calculate rain rates and rain accumulations in a few winter events in Germany.  The results show a strong improvement in the quality of the QPE when compared to rain gauges. The quality of the resulting QPE depends on the event and on the location of the radar. More specifically, the quality decreases when the melting layer is very low, at heights comparable to the radar height, and when the difference between the beam and the surface increases. These problems will be analyzed and potential solutions will be tested in order to improve the quality of the rainfall product.

Ryzhkov, Alexander, Pengfei Zhang, Petar Bukovčić, Jian Zhang, and Stephen Cocks. 2022. "Polarimetric Radar Quantitative Precipitation Estimation" Remote Sensing 14, no. 7: 1695. https://doi.org/10.3390/rs14071695