Passive and active microwave remote sensing data are analyzed to identify signatures of precipitation and vertical motion in tropical convection. A database of cloud/radiative model simulations is used to quantify the surface rain rates and latent heating profiles consistent with these signatures. At satellite footprint-scale (~10 km), rain rate and latent heating estimates are subject to significant random errors, but by averaging the estimates in space and time, random errors are substantially reduced. Bias errors have been minimized by improving the microphysics in the supporting cloud/radiative model simulations.

Remotely-sensed precipitation and latent heating distributions in the tropics are derived from Tropical Rainfall Measuring Mission (TRMM) and Special Sensor Microwave/Imager (SSM/I) sensor data. Latent heating distributions are compared to estimates derived from a simpler technique that utilizes observations of surface rain rate and stratiform rain proportion to infer heating vertical structure. Latent heating profiles derived from satellite observations of individual mesoscale convective systems are compared to coincident estimates based upon a combination of dual-doppler radar and rawinsonde data. Results of these tests will be presented at the conference.