Use of a stochastic precipitation nowcast scheme for fluvial flood forecasting and warning

Abstract

In collaboration with the Bureau of Meteorology (Melbourne, Australia), the Met Office (Joint Centre for Hydro-Meteorological Research, UK) has developed a stochastic precipitation nowcast scheme, designed to model and predict the PDF of surface rain rate and rain accumulation in space and time. Here we demonstrate the range of probabilistic products generated by the scheme, and their potential applications for fluvial flood forecasting and warning.With the aid of a hydrological model (the PDM), we consider the use of ensembles of predicted catchment rain accumulation in evaluating the range of possible river flow responses from a given catchment. When employed in conjunction with a catchment specific, cost-based decision-making model, we highlight the value of PDFs of forecast catchment rainfall accumulation and river flow as an aid to objective decision making within the flood warning process.  Crown

Development and verification of a real-time stochastic precipitation nowcasting system for urban hydrology in Belgium

“…One of the first applications of STEPS in hydrology is presented in Pierce et al (2005), who used the STEPS ensemble nowcasts to quantify the accuracy of flow predictions in a medium-sized catchment in the UK. The value of STEPS nowcasts for urban hydrology was extensively analyzed by , , Liguori and Rico-Ramirez (2013) and Xuan et al (2014).…”

Very short period quantitative precipitation forecasting

“…There are a number of approaches that are under investigation. Pierce et al (2004a) use the S-PROG system in stochastic mode to generate an ensemble of nowcast precipitation fields that are then fed into a hydrological model. This produces a suite of forecast hydrographs, which not only indicate the 'best-guess' streamflow profile but also show a range of results, giving an indication of the possible variety of flow scenarios.…”

“…This approach accounts for not only the uncertainty in motion that has been attempted before (e.g Schmid et al 2002) but also the uncertainty in the development of the precipitation field. Pierce et al (2004a), as well as Fox and Wikle (2004), attempt to determine more explicit measures of uncertainty by running multiple realizations of their nowcast schemes. Pierce et al (2004a) choose to run 100 ensemble members to obtain some measure of the range of the QPF that may be of interest to the hydrologist, while Fox and Wikle (2004) employ a statistical Gibbs Sampler to determine the number of model runs required to obtain the full distribution of nowcast outcomes and thereby produce explicit measures of variance in the same spatial field as the mean nowcast product.…”

“…Taking this approach one step further, Pierce et al (2004a) employed S-PROG in a stochastic mode and inputed the resultant ensemble QPF into a hydrological model. The experiment reported used a hydrological model of a small (142 km 2 ) catchment in England, and demonstrated how distributional streamflow forecasts can be generated from ensemble short-period QPF.…”

Providing distributed forecasts of precipitation using a Bayesian nowcast scheme

“…The inherent uncertainty in forecast products is shown by Seed (2003) and Germann and Zawadzki (2002), but, while these methods acknowledge the limits of precipitation predictability, they do not provide explicit measures of uncertainty useful to a hydrologist, yet represent the uncertainty through the maximum resolved scale of a precipitation object. Recently, Pierce et al (2004) have demonstrated the use of a very short period QPF scheme in a stochastic mode to produce an ensemble of precipitation and streamflow forecasts. This methodology illustrates the potential range of forecast outcomes without explicit error parameterization.…”