An area of very high CI probailites (90%) caught my eye at thresholds around 90%.  Sure enough, convection did indeed form around the time of the highest probabilities, so the product worked well in highlighting the area of potential convection.  However, convection quickly died out after reaching the dBZ thresholds to blank out the CI algorithm out (30-40 dBz).  So while it proved well in identifying an area of potential storm development, in this case, the storms simply failed to materialize into anything more then a few light showers/sprinkles (although I realize the algorithm is not meant to diagnose storm sustainability).

– Deitsch