Rainfall depth-duration simulation curve and effect of the height in Tehran province

Approximation of precipitation and design of a rain-gauge system with minimum initial cost and robustness against measurement uncertainties requires finding the relationship between various stations equipped with udometers. On the other hand, in areas where the issues of erosion and sediment exist, it seems necessary to know the amount of rainfall. Among the possible solutions, increasing the number of rain gauges can be adopted which is expensive. In the present study, rainfall data of 16 base stations by Ministry of Power recording precipitation data of Tehran Province on a common date (1622 rainfalls) were extracted. In each measurement, a station was chosen as base, the rainfalls of other stations were ignored, so the rainfall of each of them were reconstructed according to base stations, which were finally compared with observed rainfalls. Then, the curve of reconstructed and observed rainfalls was transformed into a dimensionless form, and rainfall percentage in the three quarterlies and the surface under each curve were calculated. Eventually by figuring out the error percentage of them, the frequency of error percentage in anticipating the rainfall amount and the three quarterlies for both curves were calculated. In order to determine the ratio of sensitivity of the model, distance index with statistical estimation of percentage of variation coefficient was used. Results demonstrated that there is no elevation effect within the 900 to 1800 m from sea level, but, there is a jump in the amount of error that this situation reduces gradually up to 2000 m height.