9071

8416

كارشناسي ارشد

محيط زيست

اصفهان: دانشگاه صنعتي اصفهان، دانشكده منابع طبيعي

1392

ده،113ص.: مصور،جدول،نمودار

ص.ع.به فارسي و انگليسي

محمودرضا همامي

حسين بشري

8/4/93

مصطفي تركش اصفهاني، علي لطفي

مهندسي منابع طبيعي

ID8416

111 Modelling habitat suitability for Asiatic cheetah Acinonyx jubatus venaticus using Bayesian belief networks Farhad Moeinizade Email address f moeinizade@in iut ac ir Department Of Natural Resources Isfahan University of Technology Isfahan Iran Degree M Sc language Persian Mahmoud reza Hemami Email address mrhemami@cc iut ac irAbstractHistorically Asiatic cheetah Acinonyx jubatus venaticus was distributed in vast areas ofAsia and was found from Sinai through Iran to India Today the distribution of Asiaticcheetah is only limited to Iran with estimated numbers of less than 70 As a consequence thistaxon is categorised as Critically Endangered CR in the IUCN red list To save the remnantpopulation of Asiatic cheetah in Iran Conservation of Asiatic Cheetah Project CACP wasinitiated more than a decade ago by UN Successful conservation of threatened speciesrequires comprehensive understanding of their ecology and their interaction withenvironmental factors Bayesian Belief Network BBN is a kind of graphical model thatincorporates probabilistic relationships among variables and integrate different sources ofknowledge The study area was located in central Iran where containing the most importantAsiatic cheetah habitat An influence diagram was developed construct a cause and effectrelationship between the most important habitat variables affecting the species This diagramwas then converted to an initial BBN by populating the model by semi subjectiveprobabilities obtained from Cheetah experts Quantifying the variables relationship wasperformed in Netica 3 12 software The behavior of the model was tested using sensitivityanalysis Model accuracy was evaluated by application of hypothetical data as well as 146actual presence point and 30 absence points of cheetah in the 5 protected areas The modelcorrectly predicted 97 2 of the presence points and 100 of the absence points of thespecies The studied variables were mapped by Arc Map 9 3 software A habitat suitabilitymap for Asiatic cheetah in central Iran was then produced by overlaying suitability classifiedenvironmental variables used for building the model Habitat security prey availability andpoaching were recognized as the main variables determining habitat suitability of Asiaticcheetah Siyah Kuh National Park the northern part of Bafgh Protected Area and thesouthern part of Dare Anjir Wildlife Refuge were recognized as the most suitable habitats The developed model is a valuable tool for predicting cheetah habitat suitability which canassist in adaptive management of the species Keywords cheetah Habitat suitability modeling Bayesian Belief Networks Sensitivityanalysis

محمودرضا همامي

حسين بشري

مصطفي تركش اصفهاني، علي لطفي