Landslide susceptibility mapping using logistic regression analysis in Latyan catchment

Document Type: Research Paper


1 Young Researchers and Elite Club, Islamic Azad University, Karaj Branch, Karaj, Iran.

2 Tehran Univercity

3 university of Tehtan, Iran

4 University of Tehran, Iran


Every year, hundreds of people all over the world lose their lives due to landslides. Landslide susceptibility map describes the likelihood or possibility of new landslides occurring in an area, and therefore helping to reduce future potential damages. The main purpose of this study is to provide landslide susceptibility map using logistic regression model at Latyan watershed, north Iran. In the first stage, 208 Landslide locations were identified and mapped using extensive field surveys. 75 % of these landslides were used for training and 25 % of them for validation of the model. The mapped landslides were then georeferenced using ArcGIS 10 to provide the landslide inventory map. In the second stage, maps of factors affecting the occurrence of landslides were prepared in ArcGIS 10. Finally in the last stage, the relationships between these affecting factors and the landslide inventory map were calculated using Logistic regression algorithm. The amount of pseudo R2 (0.32) and AUC (0.85) shown the high efficiency of Logistic regression model. According to the coefficients obtained by the model, lithology is the most important factor affecting landslide occurrence (coefficient= +12.032). Most landslides (69%) are located within Ek Formation. The results indicated that 7.56% of the basin is located in high susceptibility class and 2.88% in very high susceptibility class. 


Main Subjects

Atkinsson, P.M., R. Massari, 1998. Generalized linear
     modeling of susceptibility to landsliding in the central
     appennines, Italy. Computer & Geoscience, 24; 373-

Ayalew, L., H. Yamagishi, 2005. The application of GIS
     based logistic regression for landslide susceptibility
     mapping in the Kakuda-Yahiko Mountains, Central
     Japan. Geomorphology, 65; 15–31.

Ayalew, L., H. Yamagishi, H. Marui, T. Kanno, 2005.
     Landslide in Sado Island of Japan: Part II. GIS-based
     susceptibility mapping with comparison of results
     from two methods and verifications. Enginering
     Geology, 81; 432–445.

Brenning, A., 2005. Spatial prediction models for
     landslide hazards: review, comparison and
     evaluation. Natur. Hazards and Earth System.
     Science, 5; 853–862.

Bui, D.T., B. Pradhan, O. Lofman, I. Revhaug, O.B.
     Dick, 2011. Landslide susceptibility mapping at Hoa
     Binh province (Vietnam) using an adaptive neuro-
     fuzzy inference system and GIS. Computer
     Geoscience, doi: 10.1016/ j.cageo.2011.10.031.

Clark, W.A., P.L. Hosking, 1986. Statistical methods for
     Geographers. John Wily and Sons, New York. 518pp.

Dai, F.C., C.F. Lee, 2002. Landslide characteristics and
     slope instability modeling using GIS, Lantau Island,
     Hong Kong. Geomorphology, 42; 213–228.

Dai, F.C., C.F. Lee, Z.W. Xu, 2001. Assessment of
     landslide susceptibility on the natural terrain of
     Lantau Island, Hong Kong. Environmental Geology,
     40; 381–391.

Devkota, K.C., A.D. Regmi, H.R. Pourghasemi, 2013.
     Landslide susceptibility mapping using certainty
     factor, index of entropy and logistic regression
     models in GIS and their comparison at Mugling–
     Narayanghat road section in Nepal Himalaya. Natural
     Hazards, 65; 1-31.

Dou, J., U. Paudel, T. Oguchi. 2015. Differentiation of
     shallow and deep-seated landslides using support
     vector machines: a case study of the Chuetsu area,
     Japan. Terr Atmos Ocean Sci., 26; 227–239.

Duman, T.Y., T. Can, C. Gokceoglu, H. A. Nefeslioglu,
     H. Sonmez, 2006. Application of logistic regression
     for landslide susceptibility zoning of Cekmece Area,
     Istanbul, Turkey. Environmental Geology, 51;241–

Ercanoglu, M., F.A. Temiz, 2011. Application of logistic
    regression and fuzzy operators to landslide
     susceptibility assessment in Azdavay (Kastamonu,
     Turkey). Environmental Earth Science, 64; 949–964.

Formetta, G., G. Capparelli, P. Versace, 2016.
     Evaluating performance of simplified physically
     based models for shallow landslide susceptibility.
     Hydrol. Earth Syst. Sci., 20; 4585–4603.

Garcia-Rodriguez, M.J., J.A. Malpica, B. Benito, M.
     Diaz, 2008. Susceptibility assessment of
     earthquaketriggered landslides in El Salvador using
     logistic regression. Geomorphology, 95;172–191.

Gorsevski, P.V., P.E. Gessler, R.B.  Foltz, W.J. Elliot,
     2006. Spatial prediction of landslide hazard using
     logistic regression and ROC analysis. Trans GIS,
     10; 395–415.

Gritzner, M. L., W.A. Marcus, R. Aspinall, S.G. Custer,
     2001. Assessing landslide potential using GIS, soil
     wetness modeling and topographic attributes, Payette
     River, Idaho. Geomorphology, 37; 149–165.

Iranian Landslide working party, 2007. Iranian
     landslides list, Forest, Rangeland and Watershed
     Association, Iran, 60p.

Jaafari, A., A. Najafi, H.R. Pourghasemi, J. Rezaeian, A.
     Sattarian, 2014. GIS-based frequency ratio and index
     of entropy models for landslide susceptibility
     assessment in the Caspian forest, northern Iran. Int. J.
     Environ. Sci. Technol., 11; 909–926.

Komac, M., 2006. A landslide susceptibility model using
     the analytical hierarchy process method and
    multivariate statistics in perialpine Slovenia.
     Geomorphology, 74;17–28.

Lee, S., 2004. Application of likelihood ratio and logistic
     regression models to landslide susceptibility mapping
     using GIS. Environmental Management, 34; 223-232. 

Lee, S., 2005. Application of logistic regression model
     and its validation for landslide susceptibility mapping
     using GIS and remote sensing data. International
     Journal of Remote Senssing, 26;1477–1491.

Lee, S., J. Choi, K. Min, 2004. Probabilistic landslide
     hazard mapping using GIS and remote sensing data at
     Boun, Korea. International Journal of Remote
     Sensing, 25; 2037–2052.

Lee, S., K. Min, 2001. Statistical analysis of landslide
     susceptibility at Yongin, Korea. Environmental
     Geology, 40;1095–1113.

Lee, S., T. Sambath, 2006. Landslide susceptibility
     mapping in the Damrei Romel area, Cambodia using
     frequency ratio and logistic regression models.
     Environmental Geology, 50; 847–855.

Nefeslioglu, H.A., C. Gokceoglu, H. Sonmez, 2008. An
     assessment on the use of logistic regression and
     artificial neural networks with different sampling
     strategies for the preparation of landslide
     susceptibility maps. Engineering Geology, 97; 171–

Oh, H.J., B. Pradhan, 2011. Application of a neuro-fuzzy
     model to landslide susceptibility mapping for shallow
     landslides in tropical hilly area. Computer
     Geoscience, 37; 1264–1276.

Pourghasemi, H.R., B. Pradhan, C. Gokceoglu, M.
     Mohammadi, H.R. Moradi, 2012a. Application of
     weights-of-evidence and certainty factor models and
     their comparison in landslide susceptibility mapping
     at Haraz watershed, Iran. Arabian Journal of
     Geoscience, doi: 10.1007/s12517-012-0532-7.

Pourghasemi, H.R., B. Pradhan, C. Gokceoglu, K.
     Deylami Moezzi, 2012b. A comparative assessment
     of prediction capabilities of Dempster-Shafer and
     Weightsof-evidence models in landslide
     susceptibility mapping using GIS. Geomantic Natural
     Hazards Risk, doi: 10.1080/19475705.2012.662915.

Pourghasemi, H.R., B. Pradhan, C. Gokceoglu, 2012c.
     Application of fuzzy logic and analytical hierarchy
     process (AHP) to landslide susceptibility mapping at
     Haraz watershed, Iran. Natural Hazards, 63; 965–996.

Pourghasemi, H.R., B.M. Mohammady, B. Pradhan,
     2012d. Landslide susceptibility mapping using index
     of entropy and conditional probability models in GIS:
     Safarood Basin, Iran. Catena, 97; 71–84.

Pourghasemi, H.R., B.M. Mohammady, S.F. Aghda,
     2013a. Landslide susceptibility mapping by binary
     logistic regression, analytical hierarchy process, and
     statistical index models and assessment of their
     performances. Natural hazards, 69; 749–779.

Pourghasemi, H.R., B. Pradhan, C. Gokceoglu, K.D.
     Moezzi, 2013b. A comparative assessment of
     prediction capabilities of Dempster–Shafer and
     weights-of-evidence models in landslide
     susceptibility mapping using GIS. Geomatics Nat
     Hazards Risk, 4;93–118.

Pourghasemi, H.R., B. Pradhan, C. Gokceoglu, M.
     Mohammady, H.R. Moradi, 2013c. Application of
     weights-of-evidence and certainty factor models and
     their comparison in landslide susceptibility mapping
     at Haraz watershed, Iran. Arabian Journal of
     Geosciences, 6; 2351–2365.

Pourghasemi, H.R., M. Beheshtirad, B. Pradhan, 2014.
     A comparative assessment of prediction capabilities
     of modified analytical hierarchy process (M-AHP)
     and Mamdani fuzzy logic models using Netcad-GIS
     for forest fire susceptibility mapping. Geomatics
     Natural Hazards and Risk,

Pradhan, B., 2010a. Application of an advanced fuzzy
     logic model for landslide susceptibility analysis;
     International Journal of Computer and Intel Systems,
     3; 370–381.

Pradhan, B., 2010b. Landslide susceptibility mapping of
     a catchment area using frequency ratio, fuzzy logic
     and multivariate logistic regression approaches. J
     Indian Soc Rem Sen, 38; 301–320.

Regmi, A.D., K.C. Devkota,  K. Yoshida, B. Pradhan,
     H.R. Pourghasemi, T. Kumamoto, A. Akgun, 2014.
     Application of frequency ratio, statistical index, and
     weights-of-evidence models and their comparison in
     landslide susceptibility mapping in Central Nepal
     Himalaya. Arab J Geosci, 7; 725–742.

Regmi, N.R., J.R. Giardino, J.D. Vitek, 2010. Modeling
     susceptibility to landslides using the weight of
     evidence approach: Western Colorado, USA.
     Geomorphology, 115; 172–187.

Regmi, A.D., K. Yoshida, H. Nagata, A. Man, S.
     Pradhan, B. Pradhan, H.R.  Pourghasemi, 2013. The
     relationship between geology and rock weathering on
     the rock instability along Mugling–Narayanghat road
     corridor, Central Nepal Himalaya. Nat Hazards, 66;

Soto, J., J.P. Galve, J.A. Palenzuela, J.M. Azañón,
     J. Tamay, C. Irigaray, 2017. A multi-method
     approach for the characterization of landslides in an
     intramontane basin in the Andes (Loja, Ecuador).
     Landslide, 96; 1-19. doi:10.1007/s10346-017-0830-y.

Van Den Eeckhaut, M., T. Vanwalleghem, J. Poesen, G.
     Govers, G. Verstraeten, L. Vandekerckhove, 2006.
     Prediction of landslide susceptibility using rare
     events logistic regression: A case study in the
     Flemish Ardennes (Belgium). Geomorphology, 76;

Yalcin, A., 2005. An investigation on Ardesen (Rize)
     region on the basis of landslide susceptibility. Ph.D.
     thesis, Karadeniz Technical University, Trabzon,

Yalcin, A., 2008. GIS-based landslide susceptibility
     mapping using analytical hierarchy process and
     bivariate statistics in Ardesen (Turkey): comparisons
     of results and confirmations. Catena, 72; 1–12.

Yalcin, A., F. Bulut, 2007. Landslide susceptibility
     mapping using GIS and digital photogrammetric
     techniques: A case study from Ardesen (NE-Turkey).
     Natural Hazards, 41; 201–226.

Yalcin, A., S. Reis, A.C. Aydinoglu, T. Yomralioglu,
     2011. A GIS-based comparative study offrequency
     ratio, analytical hierarchy process, bivariate statistics
     and logistics regression methods for landslide
     susceptibility mapping in Trabzon, NE Turkey.
     Catena, 85; 274–287

Yang, Z.Y., H.R. Pourghasemi, B. Pradhan, T.C. Chen,
     Y.H. Lee, 2014. An index to describe the earthquake
     effect on subsequent landslides in Central Taiwan.
     Arab J Geosci, DOI 10.1007/s12517-014-1435-6.

Yesilnacar, E., T. Topal, 2005. Landslide susceptibility
     mapping: A comparison of logistic regression and
     neural networks methods in a medium scale study,
     Hendek region (Turkey). Engineering Geology, 79;

Youssef, A.M., B. Pradhan, H.R. Pourghasemi, S.
     Abdullahi, 2015. Landslide susceptibility assessment
     at Wadi Jawrah Basin, Jizan region, Saudi Arabia
     using two bivariate models in GIS. Geosci J,
     doi:10.1007/ s12303-014-0065-z

Zare, M., H.R. Pourghasemi, M. Vafakhah, B. Pradhan,
     2013. Landslide susceptibility mapping at Vaz
     Watershed (Iran) using an artificial neural network
     model: a comparison between multilayer perceptron
     (MLP) and radial basic function (RBF) algorithms.
     Arab J Geosci, 6; 2873–2888.

Zhang, K., X. Xue, Y. Hong, J.J. Gourley, N. Lu, Z.
    Wan, Z. Hong, R. Wooten, 2016. iCRESTRIGRS: a 

     coupled modeling system for cascading flood–
     landslide disaster forecasting. Hydrol. Earth Syst. Sci,
     20; 5035–5048.

Zhu, C., X. Wang, 2009. Landslide susceptibility
     mapping: a comparison of information and weights-
     of evidence methods in Three Gorges Area.
     International Conference on Environmental Science
     and Information Application Technology, 342–346.