Comment on "A Stochastic Model Relating Rainfall Intensity to Raindrop Processes" by J. A. Smith and R. D. De Veaux

Smith and De Veaux [1994] develop a stochastic model for rainfall rates based on approximate expressions for the special case when raindrop diameters follow a lognormal distribution (their equations (3)-(5)). It will be shown here that some minor changes to their rainfall rate model yield the following advantages: (1) the assumption of a lognormal distribution of raindrop diameters is no longer required; (2) approximate expressions for rainfall rates are avoided and can be replaced with simpler exact expressions; and (3) the normal distribution can be used to obtain confidence limits for rainfall rates when large numbers of droplets are involved. It is also noted that Smith and De Veaux's variance expression for log rainfall rate appears to underestimate the true variance associated with measured rates.