Using Different Fuzzy Logic Methods to Optimize DRASTIC Model, Case Study: Tabriz Plain Aquifer

With respect to the concentration of population، agricultural activities and industrial manufactures in Tabriz Plain area، vulnerability assessment of the plain aquifer is very useful for development، management، decision making for land use and preventing groundwater contamination. In this research، vulnerability of the Tabriz plain complex aquifer to contamination was considered by DRASTIC model in GIS media and different fuzzy logic methods adapted to optimize the model. The DRASTIC model uses seven environmental parameters effective on assessment of aquifers vulnerability such as Depth to groundwater level، net Recharge، Aquifer media، Soil media، Topography، Impact of vadose zone، and hydraulic Conductivity، which created as seven layers in ArcGIS media and weighted، rated and compiled، finally groundwater contamination vulnerability map was prepared and the DRASTIC index for the study area calculated between 40 to 126. The vulnerability map created by DRASTIC is compared to nitrate data and the results indicated a relative correlation between the nitrate level and vulnerability index. The Sugeno، Mamdani and Larsen fuzzy logic methods were adapted for optimizing the weights of the DRASTIC parameters. For this purpose، the DRASTIC parameters as input data and the vulnerability index as output data were defined for fuzzy models and nitrate concentration data were divided in two categories for training and test steps. The output of model in training step was corrected by related nitrate concentration، and after model training، the output of model in test step was verified by nitrate concentration. The results obtained from different fuzzy models show that the Sugeno fuzzy logic model is an applicable and useful method for optimizing DRASTIC model. According to the final model results، the eastern part of the area، inside the Tabriz city limits، has the maximum potential rate for contamination.