Prioritizing flood producing potential in ungauged watersheds using the AHP-VIKOR method (Case study: Haji-Bakhtiar Watershed, Ilam)

Extended abstract 1- Introduction Flood is one of the most complex and destructive natural phenomena that cause significant damage to agriculture, fisheries, housing, and infrastructure and significantly affects social and economic activities (Chang et al., 2008). The relationship between geomorphological and hydrological parameters makes it possible to predict floods in gauged basins and generalize predictions to similar ungauged basins by creating relationships between the geomorphologically similar basins (Jain & Sinha, 2003). Prioritization of areas for flood control projects is a fundamental decision that must be confirmed by studying the areachr('39')s physical, social, and economic conditions and determining the effectiveness of the plans (Djrodjetive & Bruck, 1989).  Sub-basins with critical conditions or are close to main rivers, or public facilities (reservoir dams, diversion dams, etc.) prioritize carrying out rehabilitation projects. The region being mountainous, lack of proper past and present management of pastures, very steep slopes and slope instability, relatively severe soil erosion, the existence of rock outcrops with inadequate vegetation, and human interventions are the factors that cause flood problems in the Haji-Bakhtiar watershed. In the current study, the sub-basins in the Haji Bakhtiar watershed will be prioritized based on the factors mentioned above. To this end, two techniques are combined: AHP and VIKOR.    2- Methodology In order to meet the purpose of the research, the following steps are carried out. - Studying the theoretical foundations by searching library documents and using expert opinions to identify the most important indicators of flood potential in sub-basins - Collecting and preparing the required climate data and digital layers - Calculating the weights of the indicators by using Analytic Hierarchy Process (AHP) - Prioritizing (15) sub-basins in terms of flood potential using the VIKOR method and preparing a sub-basin potential prioritization map in two cases of 25 and 50 year return period rainfall - Classifying flood potential of sub-basins in 4 classes (very high, high, medium, low) based on the VIKOR method.   3- Results  In the AHP method, important flood-producing indices were determined using the Delphi technique (consulting with experts) and 15 questionnaires. The results showed that the indicators affecting flood potential have different weights, based on expertschr('39') judgments, and the compatibility ratio in all questionnaires is less than 0.1. Among the indicators, the area index (final weight 0.44) has the most significant effect on flood potential, and the average slope index (final weight 0.096) has the least effect on flood potential. The range of changes in the VIKOR index of options, based on the AHP-VIKOR method, varies from 0.032 to 0.994. The results of prioritizing flooding potential of sub-basins, based on two rainfall intensities with 25 and 50 return periods, showed that in both cases, sub-basin H12-2 (Palkaneh village) with VIKOR index of 0.994 has the highest priority (Maximum flooding potential) and H9 sub-basin with VIKOR index of 0.032 has the lowest priority (minimum flooding potential). After prioritization in Arc GIS 10.3 environment based on the VIKOR index in two cases of rainfall intensity, the flooding potential of the sub-basins were classified into four classes. The sub-basin flooding potential map results at both return periods showed that, for both rainfall intensities, about 31% of the area had medium flood potential, 53% high flood potential, and 16% of the area had very high flooding potential. Out of 15 sub-basins, seven sub-basins are in moderate flooding condition, seven sub-basins are in high flooding condition, and one sub-basin is in very high flooding condition.   4- Discussion & Conclusions  The results obtained from the AHP method in the present study are based on aggregating the data collected by a questionnaire, which professors and watershed management experts complete. The results show that, among the four indicators affecting the flooding potential of sub-basins, the area index has the most significant impact on the flooding potential of the study area. This finding is consistent with Dehghani et al. (2013) and Ozturk & Batuk (2011). The rainfall intensity index is the next most important factor affecting the flood potential of the study area, which is consistent with the results of Karam and Derakhshan (2012); Dehghani et al. (2013); Yahaya (2008); and Ozturk & Batuk (2011). The curve number index is the third important index, which corresponds to Khosroshahi and Saghafian (2003); Soleimani Sardo et al., (2013). Finally, the fourth index is the average slope index; this result is consistent with the results of Karam and Derakhshan (2012); Dehghani et al. (2013), and Yahaya (2008); Fernandez & Lutz (2010). After calculating the weights of the indicators using the AHP method, the VIKOR method (Asghari Saraskanrood et al., 2015; Khalghi, 2002) was used to prioritize the flood potential sub-basins. The flood classification potential of sub-basins shows the high potential of the region in terms of flood generation. The lands with very high risk need watershed management measures such as preventing soil erosion and destruction, reducing water flow rate, increasing flood concentration-time, creating opportunities for water infiltration in the subsoil and recharge the aquifers, and cultivating suitable plants for the conditions. The geography of the slopes and rangelands restoration should be done to change what makes the region vulnerable to its strengths.