Abstract
The relationships between radar reflectivity factor Z, rainfall rate R, and rainfall kinetic energy flux E were analyzed based on a multiyear raindrop spectra dataset recorded by a Joss–Waldvogel disdrometer in the Goodwin Creek research watershed in northern Mississippi. Particular attention was given to the climatological variability of the relationships and the uncertainty by which one rainfall parameter may be estimated from another. Substantial variability for the coefficients of a power-law relationship Y = A_{b}X^{b} between two rainfall parameters Y and X (where Y and X may stand for any paired combination of Z, R, and E) was found. The variability of the exponent b, however, was small enough to support approaches of climatologically fixed exponents to simplify radar rainfall estimation procedures. The multiplicative factor A_{b} should typically be adjusted on a storm basis. The uncertainty of the estimation of one rainfall parameter from another, being a function of the difference in weighting of the drop size by the two parameters and the variability of raindrop spectra, was found to be approximately 50% for the Z–R relation, 40% for the E–R relation, and 25% for the Z–E relation. For extreme precipitation intensities (R ≥ 100 mm h^{−1}), this drop spectra–based uncertainty reduced to approximately 20% for all three relationships. The results exhibited significant sensitivity to the choice of method applied to determine the relationship between two rainfall parameters. Appreciable sensitivity of the relationship between rainfall parameters (i.e., power-law coefficients and drop spectra–based uncertainty) to the number of raindrops registered per 1-min drop spectrum was also found.
Corresponding author address: Dr. Matthias Steiner, Department of Civil and Environmental Engineering, Princeton University, Princeton, NJ 08544.