Since the distant past, various extreme events, especially floods and droughts, have always threatened security and human civilization. In this regard, human being has investigated the behaviors of these phenomena in order to deal with and reduce the impacts of extreme events from the distant past. However, what has recently received more attention is the combined effects of one or more limiting phenomena in relation to each other. Nowadays, with the development of knowledge, it is clear that extreme events can have different intensity and impacts under the influence of conditions or the effects of other phenomena. Therefore, in the present study, the relationship between flood and drought phenomena and eva‎luating their seasonal behavior in thirteen watersheds of the fourth and third degree in the Karkheh basin was investigated, a new hydrological variable was also defined, which to some extent reflects the contrasts of these two events are limit on each other. The defined hydrological variable is the time interval, in months, between the peak flow of the flood and the severe drought before it, if any. For this purpose, in the present study, the Standard Precipitation Index (SPI) was used to quantify and determine the severe droughts, and the date of the peak discharge of floods were used to determine the time of occurrence of floods. Since the spatial scale of the study is a watershed, flood peaks and the scale of the SPI index is the meteorological station, to convert the drought scale from a station scale to a basin scale, the rainfall of the rain gauge (about 100 stations) is averaged by the area of ​​their Thyssen polygons in each watershed. The weight has been taken to calculate the SPI index so that the SPI index represents the drought conditions in each basin. Frequency analysis process is developed to check the significance of the defined hydrological variable. Since the time interval between the peak of the flood and the peak of the previous drought has a discrete and counting inherence, for frequency analysis and choose the best distribution function to fit it, eight well-known discrete probability distribution functions including Poisson (PO), Generalized Poisson (GPO) , double Poisson (DPO), inverse Poisson-Gaussian (PIG), geometric (GEOM), logarithmic (LG) and negative binomial of the first and second type (NBI and NBII) were tested by statistical and graphical goodness of fit tests. In general, the results of the research showed that in most of the small and fourth degree basins, except some unusual observations, a common pattern and connection between the two phenomena of flood and drought can be observed. As the intensity of annual droughts increases, the corresponding peak flood discharges will decrease up to a numerical value of SPI= -1, and after that, with the continuation of droughts (SPI<-1), the peak flood discharges abruptly increase with an exponential and quadratic pattern of growth. While in large basins and the third-degree basins, different patterns of conflict between floods and droughts have been observed. For example, the results showed that in large basins that have an area of ​​more than 30,000 km2, with the decline in the value of the SPI index (extreme droughts), the peak floods will also decrease continuously. Investigations on the timing of floods and droughts showed that droughts do not have a specific seasonal behavior in Karkheh region. While the analysis of the seasonal behavior of floods showed well that this phenomenon has a seasonal pattern in Karkheh region. Thus, the highest frequency of peak floods in the studied basins has occurred in April, March, May and December, respectively. The results of the frequency analysis also indicate that the probability distribution functions of DPO and NBI have the most appropriate fits to the time interval between the peak floods and the previous intense drought.

عليرضا گوهري , سعيد پورمنافي

عليرضا گوهري , سعيد پورمنافي