Application of bivariate stochastic methods in evaluation of response time of various hydrological components with respect to rainfall, a case study, doroudzan earth dam, fars province, iran

Abstract

Hydrological processes (models or systems) have both deterministic and stochastic components. In the deterministic models, the state of the systems in time (or space) can be exactly predicted, but in the stochastic models, some random elements are involved. With stochastic analysis, it is possible to calculate time response (delay time) of hydrologic variables in any catchment with respect to rainfall, if we have time series of those variables. Knowing the delay times, we can manage the reservoir behavior more effectively. In this study, we have various hydrologic time series in the Doroudzan Dam area, Fars Province, south of Iran, which involve: mean discharge input to the reservoir, water level changes in the reservoir, water level changes in dam’s piezometers, dam seepage flow changes in downstream (4 components) and rainfall, with time interval for all time series equals to 15 days. For computing the mean delay time (lag time or effective response time) of the various hydrological components with respect to rainfall, phase and coherency curves must be calculated and plotted. It is reported that lag times for input flow to the reservoir, water level in the reservoir, the water level in piezometers (No: 1, 4, 5 and 2) and seepage in the dam embankment are: 0.6, 1.164, 1.43, 1.77, 5.32, 10.3 and 4.7 days, respectively.