@article {
author = {},
title = {Assessment of Analysis Network Process and Logistic Regression in the Investigation of Landslide Potential in the Axis Range and Reservoir Dam
(Case Study: Ghalea Chai Dam)},
journal = {Quantitative Geomorphological Research},
volume = {5},
number = {3},
pages = {67-80},
year = {2018},
publisher = {},
issn = {22519424},
eissn = {},
doi = {},
abstract = {Introduction
Management of natural disasters needful local information set able for perils and decrease of process them. At this contex the indication of the potential landslides in the area which due to geographical condition and the construction of human is prone to landslides is necessary.Ghalae chai dam is one of these areas.Therefor, the purporse that is followed in this study is that, the Motions and instabilities hillside and affective factors in it be known to deleterious effect of that be prevented in the field of the natural resources and in the other parts of construction of development and economic and areas with high potential risk is identification and be zonied. Therefore the main purpose of this research, is to assessment of Analytic Network Process and Logistic Regression methodi n determine the landslide prone areas in the range Axis and Reservoir of Ghalea Chai Ajabshir.
Methodology
To evaluate the potential of landslides in the dam area in this study to evaluate the efficiency of Analytic network process and logistic regression methods were investigated. ANP model building requires the definition of elements and their assignment to clusters, and a definition of their relationships (i.e., the connections between them indicating the flow of influence between the elements). Like AHP, ANP is founded on ratioscale measurement and pairwise comparisons of elements to derive priorities of selected alternatives. In addition, relations among criteria and sub-criteria are included in evaluations, allowing dependencies both within a cluster (inner dependence) and between clusters (outer dependence) (Saaty, 2001). Pairwise comparison is now done, both for weighting the clusters (i.e., criteria) and for estimating the direction and importance of influences between elements, numerically pictured as ratio scales in a so-called supermatrix. Network analytic was used to examines the potential of landslides for the first time in Iran and for runing it the Super Desition and ArcGis software was used. However logistic regression is a multivariate statistical technique predictor for the dependent variable that is modes of zero and one with lack of occurrence or the event (eg landslide) are. In this method, regression relation the variables not to linear and logistic curve is S-shaped. In The models are estimates in the range of zero to one that the numbers are nearly zero, indicating less likelihood of preparing and numbers are nearly one indicates higher risk. In the models was runed by using the Edrisi and ArcGis software.
Results and discussion
In the Analytic Network Process method, for runing a three-layer model consists of, goal, criteria, Alternatives was designed. After runing paired comparisons between elements and clusters the priority of the danger classes based on their significance was determined and the coefficients of the factors showed that the land use factor has the maximum effect in occurrence of the landslides if the area.The zoning of map were classified in five classes of very high to very low risk class. In the logistic regression method the eveletion factor became known as the most influential factor in the occurrence of the landslides of the area .For Testing Accuracy of the model , three-Statistical indexs were used that includs Chi Square , Pseudo R Square, ROC. The Value of Chi Square is 1657/0673 and value of Pseudo R Square is 0/5677 and the ROC value is 0/9726 that showed accuracy of the model. And finaly two listed methods were compared with each other and the best model was selected and introduced.
The Result:
The results showed that among eight factors effective factors, Land use, Elevation and Aspect have a greater role in the occurrence of landslides. Also comparison of the degree of fit between landslide distribution and zoning maps of the mentioned method showed that, the analytic network process with 67/33% fitness has the better performance than the logistic regression in Identiting risk areas in the study area. Thus, the two statistical methods used in this study was obtained from the logistic regression model and the ANP model (Eq. 2) selection to as the best model was introduced. Also, according the results of landslide hazard zonation in the study area using the two methods mentioned 39/19% of the total area of landslides is very high risk.},
keywords = {Ghalea Chai dam,Analytic Network Process (ANP),logistic regression,Landsat Satellite,Landslide},
title_fa = {ارزیابی روشهای تحلیل شبکه ( ANP ) و رگرسیون لجستیک در بررسی پتانسیل وقوع زمینلغزش در محدوده محور و مخزن سد، مطالعه موردی: سد قلعه چای},
abstract_fa = {ارزیابی پتانسیل وقوع زمینلغزش در منطقهای که به دلیل وضعیت جغرافیایی و ساختوسازهای انسانی مستعد لغزش میباشد ضروری مینماید. سد مخزنی قلعه چای عجبشیر، یکی از این نوع نواحی میباشد. در این مطالعه، جهت بررسی پتانسیل وقوع زمینلغزش، روشهای تحلیل شبکه ( ANP ) و رگرسیون لجستیک مورد ارزیابی قرار گرفت . جهت این مطالعه از تصویر TM، 2011 ماهواره لندست استفاده شد. فاکتورهای مؤثر وقوع زمینلغزش ( شیب، جهت دامنه، لیتولوژی، کاربری زمین، فاصله از گسل، فاصله از رودخانه، فاصله از جاده، طبقات ارتفاعی ) در محیط GIS آماده و سپس با لایه پراکنش زمینلغزشها قطع دادهشده و نقشههای پهنهبندی خطر زمینلغزش در روشهای فوق تولید شد. نتایج نشان داد که فرآیند تحلیل شبکه نسبت به روش رگرسیون لجستیک عملکرد بهتری را در بررسی پتانسیل وقوع زمینلغزش در منطقه موردمطالعه دارد همچنین تفسیر ضرایب نشان داد که ، کاربری اراضی، طبقات ارتفاعی، جهت دامنه، نقش مهمی در وقوع زمینلغزش دارند. و با استفاده از نقشه پیشبینی احتمال وقوع زمینلغزش، منطقه به پنج گروه حساسیت تقسیم: بسیار پایین، پایین، متوسط، بالا، بسیار بالا.},
keywords_fa = {سد قلعه چای,تحلیل شبکه (ANP),رگرسیون لجستیک,ماهواره لندست,زمین لغزش},
url = {http://www.geomorphologyjournal.ir/article_78054.html},
eprint = {http://www.geomorphologyjournal.ir/article_78054_4d1e52fbe6c433f381c4b3e6245d2656.pdf}
}