Understanding Radar Refractivity

A three-year effort to validate radar-derived refractivity using two WSR-88D (S-band) and four CASA (X-band) radar systems in Oklahoma is presented. Radar refractivity is compared to Oklahoma Mesonet observations. The surface observation serves as a "truth" refractivity observation, as is the case in many previous studies. A long-term statistical analysis of the validity of radar refractivity has not been presented in the literature, and is provided here. Temporal and spatial analysis of radar refractivity agree very well with surface observations. However, periods of differences in refractivity are observed. These differences have a diurnal cycle, suggesting the root cause to be related to a meteorological mechanism that can occur on a daily basis. The magnitude of refractivity difference has been documented to be in excess of 20-30 N-units in the warm season, which translates to a potential dew point temperature bias of several degrees Celsius. Several sources of refractivity differences are presented, and the theoretical magnitude of each is examined. These sources include sampling inconsistencies, Mesonet instrumental error, radar beam propagation, drift of the transmitted frequency from the magnetron-based CASA radars, and reference map representativeness. Ultimately, the main factor in determining the magnitude of the differences between the two refractivity platforms appears to be related to the vertical gradient of refractivity, due to the difference in observation height between the radar and a surface station.