Using mPING data to generate Random Forests for Precipitation Type Forecasts

Random forests driven by mPING data are developed on environmental profiles from the RAP, NAM and GFS forecast models. These random forests are then investigated for intra-model domain dependence by applying a random forest developed for a certain forecast lead time to a different forecast lead time. Inter-model domain dependence is investigated by developing random forests for one model and applying them to a different model. Interim results are reported.