Determining the changes in flood potential caused by drought periods in the Dehak watershed of South Khorasan province

Introduction
Climatic changes and the occurrence of long-term drought have greatly affected the vegetation pattern of watersheds, which results in the change of runoff coefficient and flood potential. The phenomenon of flooding in Iran is more than caused by the occurrence of heavy rains, it is the effect of disturbing the natural balance and geographical conditions of the region so that the occurrence of ordinary rains also causes floods.
 
Materials and Methods
In order to investigate the flood changes and prioritize the sub-basins of the Dehak watershed in South Khorasan province based on flood potential under the influence of drought periods, 30 years of annual rainfall statistics were used and the SPI drought index was determined. Digital elevation model maps, soil hydrological groups, and Landsat 5 and 7 satellite images for the years 1990, 2000, and 2009 were prepared and the Normalized Difference Vegetation Index (NDVI) was calculated with the help of ENVI 4.8 software and maps of vegetation and land use status were prepared. With Arc GIS 9.3 software, curve number maps from the integration of soil hydrological group maps, land use, and vegetation status, based on the Soil Conservation Service method, and with the SCS method, the participation rate of each of the sub-basins in the output flood of the entire area is determined, and by repeating the individual removal of each of the sub-areas, the sub-areas were prioritized based on flood potential.
 
Results and Discussion
The results of classifying Landsat images using the maximum likelihood method showed that the overall accuracy and the Kappa coefficient obtained from the estimation of accuracy in the NDVI map in 1990, obtained from Landsat images were equal to 73.2 and 0.68%, respectively, in the map obtained in 2000 equal to 75.2 and 0.71% and in the map obtained in 2009 it is 79.8 and 0.84 respectively. The results of the correlation of the percentage of the area of the NDVI for the three years 1990, 2000, and 2009 using SPSS software And the analysis of variance test (One-Way ANOVA) and Duncan's subset and also a significant level of 95% showed that between the three NDVI indices for the three years 1990, 2000 and 2009, the normalized difference index of vegetation cover had a high correlation (0.91) at the level There was a significant value of 0.05. Based on the standardized precipitation index, from 1998 to 2010, with the exception of 4 years, the rest were determined as drought periods in the region. The kappa coefficient obtained from the estimation of the accuracy of the NDVI map was 0.84, which was the most accurate in monitoring vegetation changes. The set of the slope, soil, and geological factors has led to placing 86.6% of the area in hydrological group C, which by definition has a high ability to produce runoff. The weighted average curve number (CN) of the Dehak watershed has changed from 62.35 in the wet year of 1990 to 65.04 and 63.50 in 2000 and 2009, which were dry years. Examining the significance of the simulated values of runoff height and flood peak flow using the analysis of variance test showed that there is a high correlation (0.71) at a significant level of 0.5 between the values of the runoff height corresponding to the three years 1990, 2000 and 2009 and there was a high correlation (0.68) between the values of the peak flow of the flood corresponding to the three years, the studied index. The peak flood discharge with a 5-year return period increased from 7.89 m3/s in 1990 to 13.67 m3/s in 2000, which was affected by drought, which is equivalent to 74.87%. This increase for the peak discharge of 200 years was equal to 21.64% so its amount increased from 93.68 m3/s in 1990 to 112.42 m3/s in 2000. The investigation of the runoff curve number of the Dehak watershed in 1990, i.e. before the drought period, showed that it was 62.35 in 1990 and 65.04 in 2000, and 63.50 in 2009, so it can be concluded that the period of Drought caused an increase in the number of runoff curves in Dehek watershed. The investigation of the height of the flood runoff of the Dehak watershed in 1990, that is, before the drought period, was determined to be 1.8 mm, and this ratio changed from 3.23 mm in 2000 to 2.8 mm in 2009, so it can be concluded that the occurrence of the period Drought has caused an increase in the height of the runoff in the Dehak watershed. Vegetation has less effect in the relative reduction of terrible floods with a high return period. In the study of the effect of changes in vegetation cover on the peak discharge and flood volume, it is also observed that the peak discharge of the flood is more sensitive to land use changes. The results of the flood discharge in the land use scenario (continuing the process of vegetation destruction) also showed that in case of further destruction of forests and pastures in the area and development of agricultural lands, the peak flood discharge will increase by 35 and 24% with a return period of 5 and 100 years, respectively. Found. This means that vegetation alone plays a limited role in controlling large floods with high return periods. The high weighted average of the curve number in the whole area in different years indicates the high risk of flooding in the Dehak watershed.
 
Conclusion
By prioritizing flood potential, it was found that among the 7 sub-areas, F3 sub-area (CN) has the highest value with 66.89 and is the most flood-prone sub-area, which is due to the existence of phyllite formations, Marne and also the abundance of sloping surfaces are in it, and sub-areas F4 and F5 are placed in the next level of flooding, which should be considered in management and executive planning. The priority map of the sub-basins in terms of flood potential shows that in the southeast margin overlooking the heights of the basin, there are factors such as high slope and rocky outcrops without vegetation cover next to the land use, the risk of flooding is very high and therefore the priority of any watershed engineering operation. In this area, it is more priority than other points.