Floods are common in most parts of India particularly during the monsoon season. Flood control and mitigation involves flood plain mapping, zoning, insuring and warning systems as floods cannot be avoided. These measures minimize damage and loss in terms of life and property. In this regard real-time data based decision supports systems (DSS) are crucial for effective flood management.
Inflow forecasting at any given point along the river is fundamental to flood management. This forecast depends on hydro-meteorological forecast on short or long terms basis. Here meteorological forecast is treated as an input for flood forecasting, by translating precipitation (or rainfall) to runoff via deterministic hydrological models. The advance time of forecast (or lead time) is crucial in forecasting. In general shorter time forecast are more reliable.
Also short time inflow-forecasting based on real-time data collected using an automatic data acquisition system (DAS) and Grid data through Remote Sensing are reasonably accurate. The DAS and the hydrological models (rainfall runoff and hydrodynamic) constitute the main components with DSS as the interface. The DSS updates the hydrological model (data assimilation) for each time step and executes the model to forecast flows for the lead time in advance and predicts scenarios of flood levels, discharges, storages (in case of reservoirs) and possible inundations downstream to enable the decision making process by the DM (decision maker). The DM can thus activate the government machinery fairly in advance. Longer time steps in inlfow forecasting are generally used for planning purposes and are less accurate.
Real time DSS are currently being evolved in the BBMB (Bhakra Beas Management Board) for managing multi-reservoirs in the Sutlej system.
/articles/need-real-time-dss-flood-management