Forecasting landslide hazards in the vicinity of Krasnaya Polyana basing on the linear discriminatory analysis

Application of modern statistical methods (LDA) for landslide susceptibility assessment is discussed. A computer model was created to forecast if an area falls into the category of potentially landslidehazardous. The model is based on the landslides inventory carried out earlier by one of the authors in the vicinity of the Krasnaya Polyana settlement (Russia, Krasnodarskiy Krai). The sites with the highest landslide susceptibility were identified, and the landslide-prone objects and buildings were revealed.

By 2018, a lot of factual data on landslide formation has been accumulated for a number of relatively small areas around the world. There are many examples of the elaboration of mathematical models explaining the spatial patterns of landslides. The most common errors during the construction of such models are characterized. It is also shown that, at least in English-language publications, there are no examples of statistical probabilistic modeling of the landslide hazard in the territory of Russia.

The structure and the operation mode of computational algorithm implemented in the R environment are described. The characteristic durations of each computational stage and the whole algorithm using common desktop PC are presented. Two most important advantages of the algorithm are as follows: 1) the user does not decide which particular set of characteristics should be tested for predictive power, 2) «sensitivity», rather than «accuracy», is used to evaluate the model’s «quality», which is more coherent considering the incompletion of landslide processes within the territory.

The model is based on six geomorphometric parameters of terrain: the minimum normalized height, the minimum standardized height, the average «terrain view factor» in a cell, the mean and standard deviation of the «multiresolution index of valley bottom flatness», the maximum negative topographic openness. The reliability of forecast by the model for the territory under study is 73%.

1. Aditian A., Kubota T., Shinohara Y. Comparison of GIS-based landslide susceptibility models using frequency ratio, logistic regression, and artificial neural network in a tertiary region of Ambon, Indonesia. Geomorphology, 2018, vol. 318, p. 101–111.

2. Anezi V., Kostanzo D., Minina M.V., Korolev V.A., Rotilyano E. Primenenie statisticheskih neparametricheskih metodov dlya ocenki faktorov opolznevyh processov na territorii rajonov Scillato i Caltavuturo (o. Sicilia) [Application of statistical non-parametric techniques for the assessment of landslide factors within the territory of Scillato and Caltavuturo communes (Sicilia)] Academician Sergeev’s readings, vol. 15, Moscow, Peoples’ Friendship University of Russia Publ., 2013, p. 101–105. (In Russian)

3. Baoping W., Sijing W., Enzhi W., Jianmin Z. Characteristics of rapid giant landslides in China. Landslides, 2004, no. 1, p. 247–261.

4. Boehner J., Selige T. Spatial prediction of soil attributes using terrain analysis and climate regionalization. SAGA – Analysis and Modelling Applications, Goettingen, Goettinger Geographische Abhandlungen, 2006, p. 13–28.

5. Bolysov S.I., Bredikhin A.V., Eremenko E.A. Kompleksnaya melkomasshtabnaya otsenka geomorfologicheskoj bezopasnosti Rossii [Integral small-scale assessment of the geomorphologic safety of Russia]. Vestn. Mosk. un-ta, Ser. 5, Geogr., 2016, no. 2, p. 3–12. (In Russian)

6. Box G.E.P., Cox D.R. An analysis of transformations. Journal of the Royal Statistical Society, Series B, 1964, no. 26(2), p. 211–252.

7. Carrara A. Multivariate Models for Landslide Hazard Evaluation. Mathematical Geology, 1983, vol. 15, no. 3, p. 403–426.

8. Dozier J., Frew J. Rapid calculation of terrain parameters for radiation modeling from digital elevation data. IEEE Transactions on Geoscience and Remote Sensing, 1990, no. 28, p. 963–969.

9. Fawcett T. An introduction to ROC analysis. Pattern Recognition Letters, 2006, no. 27, p. 861–874.

10. Fedorenko V.S. Gornye opolzni i obvaly, ih prognoz [Landslides and rock falls, and their forecasting], Moscow, MGU Publ., 1988, 214 p. (In Russian)

11. Gallant J.C., Dowling T.I. A multiresolution index of valley bottom flatness for mapping depositional areas. Water Resources Research, 2003, no. 39(12), p. 1347–1360.

12. Gorsevski P.V., Gessler P.E., Foltz R.B., Elliot W.J. Spatial prediction of landslide hazard using logistic regression and ROC analysis. Transactions in GIS, 2006, vol. 10, no. 3, p. 395–415.

13. Green D.M., Swets J.A. Signal detection theory and psychophysics. New York, NY: John Wiley and Sons Inc., 1966, 234 p.

14. Grus J. Data Science from Scratch. Sebastopol, CA: O’Reilly, 2015, 320 p.

15. Guo Ch., Montgomery D.R., Zhang Y., Wang K., Yang Z. Quantitative assessment of landslide susceptibility along the Xianshuihe fault zone, Tibetan Plateau, China. Geomorphology, 2015, vol. 248, p. 93–110.

16. He S., Pan P., Dai L., Wang H., Liu J. Application of kernelbased Fisher discriminant analysis to map landslide susceptibility in the Qinggan River delta, Three Gorges, China. Geomorphology, 2012, no. 171–172, p. 30–41.

17. Katok A.B., Hassel’blat B. Vvedenie v sovremennuyu teoriyu dinamicheskih system [Introduction to the modern theory of dynamic systems], Moscow, Faktorial Publ., 1999, 768 p. (In Russian)

18. King D., Bourennane H., Isambert M., Macaire J.J. Relationship of the Presence of a Non-Calcareous Clay-Loam Horizon to DEM Attributes in a Gently Sloping Area. Geoderma, 1999, no. 89(1–2), p. 95–111.

19. Korup O., Clague J.J., Hermanns R.L., Hewitt K., Strom A.L., Weidinger J.T. Giant landslides, topography and erosion. Earth and Planetary Science Letters, 2007, no. 261, p. 578–589.

20. Krestin B.M., Malneva I.V. Aktivnost opolznevogo i selevogo processov na territorii Bolshogo Sochi i ee izmeneniya v nachale XXI veka [Landslide and debris flow activity in Sochi agglomeration and its changes in the beginning of XXI century], Geoecology, Engineering Geology, Hydrogeology, Geocriology, 2015, no. 1, p. 58–66.

21. Lee S., Pradhan B. Landslide hazard mapping at Selangor, Malaysia using frequency ratio and logistic regression models. Landslides, 2007, vol. 4, no. 1. p. 33–41.

22. Milanovskij E.E. Novejshaya tektonika Kavkaza [Recent tectonics of the Caucasus], Moscow, Nedra Publ., 1968, 483 p. (In Russian)

23. Pochvenno-ekologicheskij atlas Krasnodarskogo Kraya [Soilenvironmental atlas of the Krasnodar Krai] Ed. Vidnov A.S. et al. Krasnodar, KZRZ KK Publ., 1999, 20 p. (In Russian)

24. Prima O.D.A., Echigo A., Yokoyama R., Yoshida T. Supervised landform classification of Northeast Honshu from DEM-derived thematic maps. Geomorphology, 2006, vol. 78, p. 373–386.

25. Reichenbach P., Rossi M., Malamud B.D., Mihir M., Guzzetti F. A review of statistically-based landslide susceptibility models. Earth-Science Reviews, 2018, no. 180, p. 60–91.

26. Shvarev S.V. Analiz parametrov drevnego katastroficheskogo opolznya v doline reki Psluh (Zapadnyj Kavkaz) s ispol’zovaniem dannyh lazernogo skanirovaniya [Analysis of the parameters of an ancient catastrophic landslide in the valley of the Psluh River (Western Caucasus) using the laser scanning data], Geomorfologiya, 2015, no. 4, p. 90–98. (In Russian)

27. Trigila A., Iadanza C., Esposito C., Scarascia-Mugnozza G. Comparison of logistic regression and random forests techniques for shallow landslide susceptibility assessment in Giampilieri (NE Sicily, Italy). Geomorphology, 2015, vol. 249, p. 119–136.

28. Von Ruette J., Papritz A., Lehmann P., Rickli C., Or D. Spatial statistical modeling of shallow landslides – Validating predictions for different landslide inventories and rainfall events. Geomorphology, 2011, vol. 133, no. 1–2, p. 11–22.

29. Yokoyama R., Shirasawa M., Pike R.J. Visualizing topography by openness: A new application of image processing to digital elevation models. Photogrammetric Engineering and Remote Sensing, 2002, vol. 68, p. 251–266.

30. Zhidkov M.P. Usloviya vozniknoveniya krupnyh obval’noopolznevyh yavlenij na Bol’shom Kavkaze [Factors of large landslides and rock falls in the Great Caucasus], Geomorfologiya, 2000, no. 1, p. 73–82. (In Russian)