vai al contenuto della pagina vai al menu di navigazione
 

Real-time flood forecasting systems

Keywords: flood forecasting, real-time updating, precipitation spatial field, confidence bands
Fig.1 Real-time upodating of streamflow forecasts through simulation error modelling (Author: E. Toth).
Fig. 2 Precipitation fields (time-resolution 15 mins, spatial resolution 1x1 km2) captured by the Doppler radar (Author: E. Toth).
Fig.3 Observed and real-time simulated streamflow and relative forecast confidence bands (Author: E. Toth).

Two crucial components of real-time flood forecasting systems are the estimation and forecasting of the meteorological forcing and the use of the last streamflow observations, that allow to update the rainfall-runoff model simulations and to derive an uncertainty assessment of the issued forecast. Such aspects have been analysed in the following research activities:

a) Integrated use of time-series analyisis techniques and determistic rainfall-runoff models for flood forecasting: along with traditional linear stochastic models, non-linear time-series models have been applied, that is Artificial Neural Networks (ANN) and the “nearest neighbours” method, which is a non-parametric regression methodology; such techniques are applied for forecasting the short-term future rainfall to be used as real-time input to the rainfall-runoff model and for updating the discharge predictions provided by the model (see Fig. 1).

b) Estimates and nowcasting of rainfall fields through remote-sensing techniques: the performances of an integrated flood forecasting system, based upon the use of Meteosat satellite derived rainfall maps and of a distributed rainfall- runoff model were first analysed, comparing both the input fields and the obtained streamflow forecasts. A second research topic is the evaluation of system analysis techniques for obtaining short-term (of the order of a few hours) quantitative precipitation forecasting, based on radar images (see Fig. 2).

Another important issue in flood forecasting is the analysis of the reliability and uncertainty of the streamflow forecasts. A technique for assessing the uncertainty of rainfall-runoff simulations was proposed, that makes use of a meta- Gaussian approach in order to estimate the probability distribution of the model error conditioned by the simulated river flow. The proposed technique is applied to real-world case studies, for which the confidence limits of simulated river flows are derived and compared with the actual hydrometric observations (see Fig. 3).

Main publications

Abrahart, R.J., Anctil, F., Coulibaly, P., Dawson, C.W., Mount, N.J., See, L.M., Shamseldin, A.Y., Solomatine, D.P., Toth, E., Wilby, R.L., Two decades of anarchy? Emerging themes and outstanding challenges for neural network river forecasting, Progress in Physical Geography, doi:10.1177/0309133312444943, 36(4), 480–513, 2012.

Brath A, Montanari, A., Toth, E., On the use of neural networks and non-parametric methods for improving real-time flood forecasting obtained through conceptual hydrologic models. Hydrology and Earth System Sciences. vol. 6, pp. 627-640 ISSN: 1027-5606, 2002.

Montanari L., Montanari, A., Toth, E., A comparison and uncertainty assessment of system analysis techniques for short-term quantitative precipitation nowcasting based on radar images. Journal of Geophysical Research. Atmospheres. vol. 111 ISSN: 0148-0227, 2006.

Montanari, A., Brath, A., A stochastic approach for assessing the uncertainty of rainfall-runoff simulations, Water Resources Research, Vol. 40, W01106, doi:10.1029/2003WR002540, 2004.

Montanari, A., What do we mean by 'uncertainty'? The need for a consistent wording about uncertainty assessment in hydrology, Hydrological Processes, Vol. 21, 841-845, 2007.

Toth E, Data-driven streamflow simulation: the influence of the temporal resolution of input and output variables. In: "Hydroinformatics in practice: computational intelligence and technological developments in water applications" (Eds. R. Abrahart, L. See, D. Solomatine), ISBN: 3540798803, BERLIN, Springer-Verlag, GERMANY, 2008.

Toth E., Brath, A., Montanari, A., Comparison of short-term rainfall prediction models for real-time flood forecasting. Journal of Hydrology. vol. 238, pp. 132-147 ISSN: 0022-1694, 2000.

Toth E., Brath, A., Montanari, A., Real-time flood forecasting via combined use of conceptual and stochastic models. Physics and Chemistry of the Earth Part B-Hydrology Oceans And Atmosphere. vol. 24(7), pp. 793-798 ISSN: 1464-1909, 1999.

Toth, E., Classification of hydro-meteorological conditions and multiple artificial neural networks for streamflow forecasting, Hydrology and Earth System Sciences, 13, 1555–1566, 2009.

Toth, E., Estimation of flood warning runoff thresholds in ungauged basins with asymmetric error functions, Hydrol. Earth Syst. Sci., 20, 2383-2394, doi:10.5194/hess-20-2383-2016, 2016.

Research projects

Research contracts (from 1999 on) with the Regional Civil Protection Agency of Regione Emilia- Romagna.

Italian Research Project of National Relevance 2006, “Advanced techniques for estimating the magnitude and forecasting extreme hydrological events, with uncertainty analysis (SPIE)”, financed by the Italian Ministry of University and Research (MIUR).