tRIBS model
Short-term Radar Rainfall Nowcasting


MIT Lincoln Laboratory Storm Growth and Decay Tracker (GDST)

This research project involved the application of an existing short term rainfall extrapolation technique to a radar data set obtained from WSI (Weather Services Inc) for the NEXRAD system over the continental United States. The current model was developed by MIT Lincoln Laboratory for the express purpose of forecasting windfields and reflectivity in the near vicinity of airports (Wolfson et al, 1999). It consists of applying a spatial filter to a NEXRAD reflectivity field and determining the wind field through a cross correlation technique at the larger, synoptic scale. The software has been extensively tested by MIT Lincoln Laboratory for the purpose outlined above.

As part of our hydrometeorological forecasting efforts, we wish to obtain an accurate forecast of rainfall at the watershed scale for the forecast lead time of 0-3 hours. Nowcasting is a viable tool at this temporal scale, capable of providing quantitative forecasts that can be used as input to a distributed hydrologic model. The spatial extent of the WSI NEXRAD data set also provides the necessary spatial coverage for our hydrologic modeling purposes and even opens the door to more extended prediction from the radar rainfall mosaic in the continental US. Used in conjunction with the tRIBS distributed hydrologic model, the precipitation forecast product can improve our ability to forecast various types of flooding events in watersheds.