Thursday, 27 October 2005: 4:30 PM
Alvarado ABC (Hotel Albuquerque at Old Town)
Accurate calibration of radars is essential if rainfall is to be derived from conventional reflectivity observations, Z. If the polarisation parameter differential reflectivity, ZDR, is to be used as well as Z, then the precision of the Z calibration may be the limiting factor in accuracy of the derived rain rates. It has been suggested that the ratio of KDP/Z in rain is a well-defined function of ZDR, and that calibration of Z to 0.5dB can be achieved by computing the theoretical phase shift along a ray obtained by integrating the predicted KDP from the observed Z and ZDR. Differences between observed and predicted phase shifts are attributed to Z calibration errors, and are subsequently minimised by scaling observations of Z. This method has the advantage of avoiding the use of noisy KDP observations.
We present an analysis of C-band observations suggesting that accurate and consistent calibration can be achieved by limiting the data set to rays with small enough phase shifts so that attenuation of Z and ZDR is negligible. Rays where the consistency breaks down can be identified as containing hail. An alternative approach proposed in the literature is to assume that the radar has been independently calibrated, and use the consistency to find the appropriate drop shape model to be used on a particular day. Our results do not support this approach but instead suggest that the drop shape model does not change significantly with time.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner