Stochastic Flood Forecasting System : The Middle River Vistula Case Study

This book presents the novel formulation and development of a Stochastic Flood Forecasting System, using the Middle River Vistula basin in Poland as a case study. The system has a modular structure, including models describing the rainfall-runoff and snow-melt processes for tributary catchments and the transformation of a flood wave within the reach. The sensitivity and uncertainty analysis of the elements of the study system are performed at both the calibration and verification stages. The spatial and temporal variability of catchment land use and river flow regime based on analytical studies and measurements is presented. A lumped parameter approximation to the distributed modelling of river flow is developed for the purpose of flow forecasting. Control System based emulators (Hammerstein-Wiener models) are applied to on-line data assimilation. Medium-range probabilistic weather forecasts (ECMWF) and on-line observations of temperature, precipitation and water levels are used to prolong the forecast lead time. The potential end-users will also benefit from a description of social vulnerability to natural hazards in the study area.

This book presents the novel formulation and development of a Stochastic Flood Forecasting System, using the Middle River Vistula basin in Poland as a case study. The system has a modular structure, including models describing the rainfall-runoff and snow-melt processes for tributary catchments and the transformation of a flood wave within the reach. The sensitivity and uncertainty analysis of the elements of the study system are performed at both the calibration and verification stages. The spatial and temporal variability of catchment land use and river flow regime based on analytical studies and measurements is presented. A lumped parameter approximation to the distributed modelling of river flow is developed for the purpose of flow forecasting. Control System based emulators (Hammerstein-Wiener models) are applied to on-line data assimilation. Medium-range probabilistic weather forecasts (ECMWF) and on-line observations of temperature, precipitation and water levels are used to prolong the forecast lead time. The potential end-users will also benefit from a description of social vulnerability to natural hazards in the study area.