7124

456 دكتري

دكتري

خاكشناسي

اصفهان: دانشگاه صنعتي اصفهان، دانشكده كشاورزي

1391

بيست و يك،288ص.: مصور﴿رنگي﴾،جدول،رنگي

ص.ع.به فارسي و انگليسي

محمدعلي حاج عباسي، شمس ا... ايوبي

احمد جلاليان، مجيد افيوني، فريد شيخ الاسلام

1/8/91

محمدرضا مصدقي، بهزاد قرباني، محسن شكل آبادي

مهندسي كشاورزي

ID456 دكتري

به فارسي و انگليسي: قابل رويت در نسخه ديجيتالي

Modelling of Soil Erosion Hazard in the Bazoft Watershed using Fuzzy Logic Algorithm SWAT Model and Genetic Algorithm Fuzzy Clustering Asghar Besalatpour E mail address a besalatpour@ag iut ac ir Date of submission July 20 2012 Department of Soil Science College of Agriculture Isfahan University of Technology Isfahan 84156 83111 Iran Mohammad Ali Hajabbasi hajabbas@cc iut ac irShamsollah Ayoubi ayoubi@cc iut ac irAhmad Jalalian jalalian@cc iut ac irMajid Afyuni afyuni@cc iut ac irFarid Sheikholeslam sheikh@cc iut ac irCollege Graduate Program Coordinator Ahmad Riasi Prof Assoc Prof and Prof of Soil Science respectively Department of Soil Science College ofAgriculture Isfahan University of Technology Isfahan Iran Assoc Prof Department of Electronic and Computer Science Isfahan University of Technology Isfahan Iran AbstractFuzzy logic algorithm soil and water assessment tool SWAT and genetic algorithm fuzzy clustering models were used to identify and prioritize the critical sub basins in ahighly mountainous watershed with imprecise and uncertain data Bazoft watershed southwestern Iran In the fuzzy logic algorithm model three important landscapefeatures related to soil erosion hazard including vegetation cover slope and soilerodibility factor were used The potential use of generalized likelihood uncertaintyestimation GLUE sequential uncertainty fitting SUFI 2 program and particle swarmoptimization PSO algorithms for calibration and validation of SWAT model wasevaluated After determining the proper algorithm the calibrated and validated SWATmodel was used for the prediction of soil erosion hazard in the study area For thegenetic algorithm fuzzy clustering model the mean weighted of 12 easily availableparameters affecting soil loss from the study area were used for modeling of soil erosionhazard Results showed that a large part of the watershed around 91428 ha waspredicted to be endangered of a high or very high erosion risk using the Fuzzy logicmodel Use of the PSO algorithm to calibrate the SWAT model resulted in bettersimulation results comparing to use of the GLUE and SUFI 2 algorithms The sub basins with high soil erosion hazard based on the SWAT model were accounted forabout 84 of soil loss from the study area According to the results of the geneticalgorithm fuzzy clustering model 12 sub basins out of 55 sub basins in the watershedhad a very high erosion hazard which cover about 29 5 of the study area around63879 ha The sub basins S7 S4 S8 S9 and S25 had respectively the highest erosionhazard and were accounted as critical sub basins in the Bazoft watershed forimplementation of soil and watershed conservation programs to reduce soil loss fromthe study area Keywords SWAT model PSO algorithm Fuzzy logic Genetic algorithm Fuzzyclustering Soil erosion

محمدعلي حاج عباسي، شمس ا... ايوبي

احمد جلاليان، مجيد افيوني، فريد شيخ الاسلام

محمدرضا مصدقي، بهزاد قرباني، محسن شكل آبادي