The objective of this study is to develop a technique to use lightning obse
rvations for estimating convective rainfall. A framework for rainfall estim
ation is developed in which key elements are 1) the rainfall-lightning rati
o, that is, the convective rainfall mass per cloud-to-ground lightning flas
h; 2) the spatial distribution of rainfall relative to flash locations; and
3) the temporal distribution of rainfall relative to the time of lightning
occurrence. These three elements are examined through a study of 22 summer
thunderstorms in the domain covered by the Melbourne, Florida, WSR-88D rad
ar during August of 1992 and 1993. The analyses are carried out by combinin
g lightning observations from the National Lightning Detection Network with
storm parameters computed from 3D reflectivity observations using the Thun
derstorm Identification Tracking and Nowcasting storm-tracking and analysis
algorithms. The effect of the prevailing convective regime on the variabil
ity of lightning-rainfall relationships is investigated. The rainfall estim
ation procedure is implemented and tested for a thunderstorm that occurred
on 20 August 1992. Striking similarities in the spatial distribution of rai
nfall estimates are observed for the rainfall maps derived from lightning o
bservations and those derived from WSR-88D reflectivity observations. Rainf
all estimates derived from lightning observations are of potential use for
short-term prediction of flash floods, especially in regions of poor radar
coverage; Potential uses of this method also include correction of radar-es
timated rainfall for range effects.