Forecasting convection initiation (CI) has been a difficult problem in the Atmospheric Sciences today. There are several means (via radar and satellite) which can attempt to nowcast CI within the 0-1 hour timeframe. Forecasting CI within the 1-6 hour timeframe remains difficult, due to the frequent inability of numerical weather prediction models to �spin up� the important gradients in a timely fashion, and because simple time extrapolation fail beyond ~2 hours when attempting to predict convective initiation.

We will present a new product that will help with the forecasting of air mass thunderstorms in the 1-6 hour time-frame by identifying potential sources of updrafts near differential heating boundaries. The product will be easily generated daily, and in a very timely fashion, so that it could be used as a tool by operational forecasters to assess regions of preferred CI each day.

The hypothesis that is being tested via this product is that under synoptically �calm� conditions, when surface winds are generally <5 m/s and baroclinicity is weak, thermal circulations will form along differential heating gradients, similar to �inland sea-breezes�. In the past, these thermals have been termed �Nonclassical Mesoscale Circulations�, or simply, NCMCs (Segal and Arritt, 1992). Given enough atmospheric instability, the location of these circulations will act as source regions for updrafts and, eventually, deep convection. Data sets of remotely-sensed vegetation from MODIS Level 3, 16-day composites, recent rainfall estimates observed from radar and NOAA/NWS, and GOES-derived insolation are used to create differential heating indicies that will help identify these source regions.