Analyzing the Historical Rainfall Observations and Examining the Flood Event of Hopa, Turkey

Abstract: Hopa a city of Artvin province in northeast of Turkey witnessed a devastating flood last year. The analysis of the event is conducted in this study by analyzing the rainfall time series data. Several distribution functions are fitted to the rainfall data and the best fitting function which identified by several tests used for constructing intensity - duration – frequency (IDF) curve. A formula representing the IDF curve developed using non-linear least square regression. The returning periods of the recorded rainfall intensities on the day of the event are obtained using the developed formula. Gumbel found as the best fitting function and used for IDF construction. The developed formula gives a tight agreement with a correlation 0.994 between the observed and predicted intensities. The highest return period calculated for the event observations using the formula is 211 years which belongs to the 4 hours’ storm duration. The return period of 24 hours’ duration which is the duration of the rainfall caused the flood is 41 years. The intensities of the flood event found smaller than the intensities of the observations recorded in 1988 for the durations ? 4 hours and higher for the durations > 4 hours.