Using ensemble reforecasts to generate flood thresholds for improved global flood forecasting

Ervin Zsoter,Elisabeth Stephens,Christel Prudhomme,Hannah Cloke,Florian Pappenberger

doi:10.1111/jfr3.12658

Abstract

AbstractGlobal flood forecasting systems rely on predefining flood thresholds to highlight potential upcoming flood events. Existing methods for flood threshold definition are often based on reanalysis datasets using a single threshold across all forecast lead times, such as in the Global Flood Awareness System. This leads to inconsistencies between how the extreme flood events are represented in the flood thresholds and the ensemble forecasts. This paper explores the potential benefits of using river flow ensemble reforecasts to generate flood thresholds that can deliver improved reliability and skill, increasing the confidence in the forecasts for humanitarian and civil protection partners. The choice of dataset and methods used to sample annual maxima in the threshold computation, both for reanalysis and reforecast, is analysed in terms of threshold magnitude, forecast reliability, and skill for different flood severity levels and lead times. The variability of threshold magnitudes, when estimated from the different annual maxima samples, can be extremely large, as can the subsequent impact on forecast skill. Reanalysis‐based thresholds should only be used for the first few days, after which ensemble‐reforecast‐based thresholds, that vary with forecast lead time and can account for the forecast bias trends, provide more reliable and skilful flood forecasts.

Highlights

Flood forecasting systems use meteorological data and hydrological modelling to deliver forecasts of river discharge and other hydrological variables such as inundation or soil moisture
Comparing the number of events forecasted with the number of events identified in the benchmark set (i.e., GloFAS-ERA5 river discharge reanalysis which is the nearest equivalent to the “observations”), expressed as percentage occurrence frequency
Ensemble reforecasts for day 1 to day 30 lead times, over 20 years (1997–2016), with 104 forecasts in each year, flood thresholds computed by fitting an extreme value distribution on the 20 annual maxima, for 5,665 global catchments and 4 return periods (50, 20, 10, and 5% annual exceedance probability (AEP))

Summary

Introduction

Flood forecasting systems use meteorological data and hydrological modelling to deliver forecasts of river discharge and other hydrological variables such as inundation or soil moisture. In the Global Flood Awareness System of the Copernicus Emergency Management Service (GloFAS; Alfieri et al, 2013, Hirpa et al, 2018), the severity of the predicted flood is defined according to a set of three thresholds, as shown in Figure 1 for the example of tropical cyclone Idai in Mozambique in March 2019. These thresholds are computed from a 40-year long river discharge reanalysis (Harrigan et al, 2020).

Methods

Results

Discussion

Conclusion