Increasingly it is taken as a given that satellite observations of lightning can be of great value in the study of meteorology and climate dynamics. Satellite data are especially promising for use in remote areas where ground-based data may be difficult to obtain. Space-based monitoring of lightning provides better global coverage of intense convective activity than surface-based methods and better resolution of the convective area than infrared satellite remote sensing, which cannot distinguish between cold, high cloud massses directly over a convective area and such cloud masses that have spread out at the tropopause and may extend far from the region of deep convection. Better, more precise definition of regions of deep convection are needed: 1) to improve initializations of models on several scales, and 2) to monitor seasonal and interannual variations in intense convective activity associated with the ENSO and possible changes in intense convective activity associated with global warming on long time scales. Observed changes in intense convective activity could be used to test predictions of GCMs. What are the requirements on satellite lightning observations in order that they may be used in these ways?

One issue that arises with use of optical detectors on satellites to observe lightning is the meaning of the term “flash” as it has been used to refer to various optical observations. Apparently little attention has been paid to the long-standing traditional meaning of the term in the context of cloud-to-ground lightning observed by means of electric fields, streak cameras, and video cameras. The meaning of the term when applied to lightning without a connection to ground was already ambiguous before satellite optical observations became plentiful. We examine the usage evident in recent papers and raise the issue of a definition that might promote consistency among various types of lightning observations.