Probabilistic Forecast of Mountain Rivers Runoff in Dagestan and Armenia

Abstract

Abstract The method of forecasting based on a probabilistic approach has been considered. A system of differential equations for initial moments of the probability density distribution of daily water flow rates is used for the forecast. The forecast values of the first three moments, which characterize daily average value, variation of the water flow rate within the day, and deviation of the average value from the modal one, are the solution to the system of equations. The method is applied to forecast the runoff of spring floods on mountain rivers – the Samur River (Dagestan) and Arpa River (Armenia). The catchments of mountain rivers are characterized by more complex processes of river flow formation. It has been established that the first initial moment is forecast more reliably than the second and third. This approach is very promising for obtaining forecast probability density curves that allow determining the provided values of water flow rates (modules, and runoff layers) for daily solution. Forecasting the probabilistic characteristics of water flow rates allows assessing and preventing the risks of flooding of settlements, hydraulic structures, bridge crossings on roads and railways.