Heikki Pohjola, FMI
Precipitation in cold climates, like that in Finland, is quite shallow, in winter frequently only 2-5 km high. It has been found that the vertical profile of reflectivity (VPR) is the dominating factor in the accuracy of measurements of precipitation at ranges of 50-250 km from a radar. The large bias of 2-20 dB, often observed in gauge-radar comparisons, is fully originated from the sampling difference between the actual reflectivity (precipitation) at the ground level and in the contribution volume aloft.
We have implemented a real time VPR correction scheme of the 500 m PseudoCAPPI reflectivity to represent reflectivity at ground level. The correction is calculated to all ranges (0-250 km) in a network of 7 C-band Doppler radars. The basic principle is comparison of the beam smoothed reflectivity aloft to the reflectivity at ground level. Both of them are estimated from the measured VPRs close to the radars (0-40 km) applying the known elevation angles, PseudoCAPPI height, elevation angle of the horizon and beamwidth. The following factors are taken into account:
1 Diagnosis of precipitation profiles and elimination of clutter from the measured individual VPRs.
2 The periods with no precipitation in the VPR measurement range apply climatological profiles, adjusted to the daily freezing level height.
3 Small scale profile variations are smoothed by applying 6-hourly time-integrated VPR corrections.
4 If the VPR correction is very large (shallow precipitation at longer ranges) or the uncorrected dBZ is already quite high, the magnitude of the correction is attenuated to avoid unrealistically high precipitation intensities e.g. in cases of embedded convection.
5 If the bright band hits the ground, the ground level reflectivity (the reference value in the VPR correction) will be adjusted to avoid too large spurious VPR corrections at longer ranges.
6 At each network composite pixel the magnitude of the VPR correction is a weighted mean of the corrections derived at each radar site. The weights are determined according to the distances to each radar.
Layers of elevated precipitation, not reaching the ground, or areas of significant evaporation are problematic for any correction based on an observed VPR close to a radar. The representativity of an instantaneous time-space interpolated freezing level height at a radar site is not necessarily very good as the operational radiosoundings are performed only in three sounding stations in Finland every 12 hours.
Verification of the method is performed by comparing the profile corrected dBZ values at longer ranges (140 - 250 km) from a radar (radar A) to the dBZ values at the same location but measured with an other radar (radar B), which is close to the location (5 - 50 km). The dBZ measurements from radar B should have an insignificant bias due to the VPR as the actual measurement height is the same as the nominal CAPPI height of 500 m at short ranges.
References:
Koistinen, J. and Pohjola H., 2002: Operational vertical reflectivity profile correction in radar network composities in Finland. Preprints, 2nd European Radar Conference, ERAD, 2002