Investigation of hydrological drought in the Eastern Black Sea Region

In this study, daily data of 7 stations selected among 41 observation stations located in the Eastern Black Sea region and belonging to the State Hydraulic Works (DSI) were used. Station data covers the years 1980 to 2015. It has been determined that there are missing data on an annual basis at the stations and these years are given in a table. In the cases determined for the 7 stations used in the study, the daily flow values were completed with linear regression and then the 7-day minimum flow values were calculated. By using the calculated daily flow values, the flow continuity lines were calculated and shown separately for the 7 stations. Q90 intervals were given and it was determined that all of the Q90 flow rates at the stations were low. Then, the 7-day minimum flow values are calculated and shown. Statistical values and L-Moment parameters have been calculated by using the 7-day minimum current values of the stations. The flow frequency lines drawn for each station are shown. Different return intervals were determined for each station used, and distributions with two (Lognormal, Weibull and Exponential) and three parameters (Lognormal, Weibull, Pearson Type 3, Logpearson Type 3, General Extreme and General Logistic) were used to estimate low current values. The parameters of these distributions were calculated. Then, minimum current estimations are made for the determined turning intervals and shown. As a result, it was determined that the lowest flow estimates belong to the exponential distribution function. Determined estimation values were compared with the flow frequency values within the year. By applying L-Moments, K-S and PPCC tests on the determined station data, it is aimed to determine the most appropriate distributions for the data we have. L-moment test was applied to the stations and the results were given. Accordingly, 14.29% LN2, 85.71% W2 in two-parameter distributions and 100% conformity with GL distribution in three-parameter distributions were determined. In the application of the K-S test, the W3 distribution has a 28.6% agreement. In the PPCC test application, it was determined that the LN distribution was 85.71% and the LN3 distribution was 71.43% compatible. Finally, Mann-Kendall trend analysis was applied in the study and the results of the analysis were shown. As a result, it was determined that there was a trend at station D22A084 and it was seen that this trend was decreasing over time.