A Bayesian Quantitative Precipitation Nowcast Scheme

Abstract

Very short-period quantitative precipitation forecast (QPF) or nowcast schemes provide deterministic output that fails to convey explicit measures of the uncertainty in the forecast. Presented here is a forecast methodology based upon a Bayesian hierarchical model that produces a QPF product for a 1-h period along with an associated estimated forecast error field. The precipitation forecast quality is comparable to other nowcast schemes and the uncertainty measures increase the utility of the methodology by allowing forecasters to judge the trustworthiness of the products.