عنوان مقاله [English]
Probabilistic design is a new approach in the design of water resources systems. Accordingly, a framework for a flood warning system design is proposed in this study. The flood warning reliability assessment model, introduced by other authors, is then completed and presented in detail. The model is applied to the Madarsou river flood warning system in northeastern Iran, which was installed in 2005. In this case, one control cross section, three vulnerable areas, three time intervals of 30, 90 and 180 minutes for forecasting the flood crest height, two forecaster triggering stages (i.e. 2.5m and 2.7m) and three warning thresholds (i.e. 3m, 3.5m and 4m) are considered. Regardless of the warning thresholds, the longest potential lead-time is obtained for the 30 minute forecasting intervals. The potential lead-time decreases by increasing the system reliability, for higher trigger stages. The trade-off discussion between potential lead-time and system reliability helps decision makers to find out the optimum operational point based on the accepted risk. The analysis results for 30 minutes forecasting intervals in Madarsou showed that by increasing the lead time up to 9.28 hours, the number of missed flood increases from 15.10 to 17.78. Indeed, the discussion on the model reliability and the lead-time trade-off depends on the system performance and the acceptable risks for a region.