Landslide susceptibility mapping using weighted linear combination for the roshtqala district, Tajikistan

Размещено на http://www.allbest.ru/

Landslide susceptibility mapping using weighted linear combination for the roshtqala district, Tajikistan

Saidaliev Ismoil Mirzovalievich, Master student; Chen Xi, Senior engineer; Wang Wei Sheng, Senior engineer; Abdudzhaborzoda Bakhromshokh, Master student; Narzuloeva Manizha, Master student; Yogibekov Dzhovid, Master student of Xinjiang Institute of Ecology and Geography Chinese Academy of Sciences

Abstract

Investigating slope instability and susceptibility mapping is a fundamental component of management that reduces the risk of life with landslides. Susceptibility to landslides is defined as the tendency of a locality to slope damage occurs, and susceptibility is usually expressed in a cartographic way. The aim of this study is to assess the condition of landslides at the Roshtqala district in east part of Tajikistan employing Analytic Hierarchy Process and Weighted Linear Combination to produce landslide susceptibility map. To carry out this research, six landslide conditioning factors including landslide inventory map, slope, aspect, elevation, distance to streams and topographic wetness index were employed to produce the final susceptibility map.

Keywords: Roshtqala, landslide susceptibility mapping, weighted linear combination, analytical hierarchy process, GIS

Introduction

A landslide is a dangerous geomorphological phenomenon, the displacement of rock masses on a slope under the influence of its own weight and additional load due to undermining of the slope, waterlogging, seismic shocks and other processes [1]. Landslides occur on the slopes of valleys or river banks, in the mountains, on the shores of the seas, etc., the most ambitious at the bottom of the seas [2]. Most often landslides occur on slopes composed of alternating water-resistant and aquifers. The displacement of large masses of earth or rock along a slope or cliff is caused in most cases by wetting the soil with rain water so that the soil mass becomes heavier and more mobile. It may also be caused by earthquakes or destructive activities of the sea. The forces of friction, which ensure the adhesion of soils or rocks on the slopes, turn out to be less than the force of gravity, and the whole mass of rock begins to move.

Many authors have different attitudes to the concept of a landslide [3, 4] mentions that, at the minimum 90% of damage from landslides could be evaded if the problem had been known prior to the landslide incident. Analyzing disaster management is one of the vital areas of GIS based on Multi Criteria Decision Analyses(MCDA). GIS - MCDA provides mighty methods to assess and predict the landslide hazards. Landslides and slope instability pose a serious danger to human activities, as a result of economic losses[5], property damage and high running costs expenses as well as injuries or deaths[6]. These losses can be mitigated if causal relationship of events known[7]. Landslide susceptibility mapping (LSM) is a solution to understanding and predicting future hazards in order to mitigate their consequences. It is one of the study fields portraying the spatial distribution of future slope-failure susceptibility[8]. This research study is based on Analytical Hierarchy Process(AHP) [9-15] and might help the citizens, planners and engineers for decisionmaking to minimize the losses reasoned by modern and futurity landslides by prevention, mitigation and avoidance[16, 17].

In producing a susceptibility map, the straight mapping technique entails holding regions susceptible to slope failure, by comparing particular geological and geomorphological properties with those of landslide locations. GIS MCDA gives an ample accumulation of techniques and methods for organizing decision problems and designing, evaluating and prioritizing alternate decisions. [18-20] describes the synergic potentials of GIS and MCDA that observes the advantages for advancing theoretical and have done study with the employment of GIS - MCDA. Many different landslide susceptibility zonation methods such as weighted overlay, ANN, decision tree model, AHP, IVM, MCDA and physically based landslide hazard models are GIS based models for prognostication the likelihood of landslide hazards[12, 21-28]. According to a complete literature review of GIS-MCDA [19], Boolean operators and weighted linear combination procedures compose nearly 40% of all GIS-based multicriteria analysis applications.

Study area and dataset

Pamir is situated in the crowing area of largest Asia mountains - Hindu Kush, Karakoram, Kun - Lun and Tien Shan.The northern and eastern borders of this area are Zaalay and Sarykol ridges, and southern and south-west boundaries are the valleys of the rivers Panj and Pamir[29]. The Darvoz Ridge is the northwest boundary of the Pamir. Roshtqala District is a district in the east part of Tajikistan, in the south-western part of Pamir. It stretches along the Shakhdara River basin among Shughnon range to the north and the Shakhdara range to the south. According to Census 2010, the population of Roshtqala district is 24,400. In Roshtqala district, the elevation increases from 2122 to 6676m above the sea level in Pamir Mountains.

Mass movement, rock falls and landslides are common in the Roshtqala district.

The landslides inventory database for Roshtqala district shows 10 landslide incidents were recorded by GPS in field surveys. To produce a landslide susceptibility map of current area slope, aspect, elevation, topographic wetness index and distance to streams were used and compared with landslide inventory map for efficiency of research.

Figure 1. Location of study area

Quaternary system

In the quaternary system the glacial, alluvial, proluvial, deluvial and lacustrine deposits are found.

Proluvial deposits occupy the valleys of small tributaries and from the debris cones. Deluvial sediments form different screes, deluvial cover on the slope, and others. The recent deposits that were formed as a result of landslides and rock falls as well as debris flows form the Usoy dam body. Additionally, the big blocks of the sandstone and marble are found on the Usoy dam, and they mix with detrital rocks.

The zone of the South-Western Pamir occupies the northern slopes of the North Alichursky and Rushansky ridges, and its northern boundary is the Rushano-Pshart deep fault. Two Rushan and Alichur-Gurumdinskaya sub-zones are distinguished there.

All necessary geometric thematic editing was done on the original datasets and a topology was created, vector layers were transformed into raster format with 30 m resolution, and the spatial datasets were utilized in Arc GIS. Standardization was performed on the criteria. Standardization is vital phase in MCDA approach. [30] characterized a process that transforms and resizes the original criteria into equivalent units. This method is an annex of the binary logic class, which enables the explanation of sets outside of keen limits and permits elements to be assigned to a particular set. A fuzzy set is materially a set, where members have degrees of membership ranging between 0 and 1, as opposed to a binary class set in which each element must have a membership degree of either 0 or 1 [31]. In landslide analysis for Roshtqala district, the factors have been presented as separate GIS dataset layers with memberships of different potential classes and were subsequently standardized using the maximum eigenvectors approach on a 0-1 scale.

While carrying out this study six thematic maps including landslide inventory map, slope, aspect, elevation, topographic wetness index(TWI) and distance to streams were employed to produce landslide susceptibility map.

Slope map vital parameter as it set ups the subsurface flow velocity of terrain and shows the gradient of slopes in degrees. Aspect factor relates to various weathering, exposure to sunlight and drying winds, soil moisture and shows the direction of slopes in degrees from 0-360°. Elevation has a high influence on landslide occurrence. TWI is extensively used while preparing susceptibility maps. Using a depressionless DEM equation sinks in formation image are removed to eliminate weather index. Distance to streams was derived from DEM using applicable formula in Arc GIS.

Methodology

The selected set of criteria should adequately reflect the decision-making environment and contribute to the achievement of the ultimate goal[32]. There isn't any guideline for selecting condition factors that influence landslides for susceptibility mapping and it depends on presence of factor datas.

To prepare landslide susceptibility maps, various methods like fuzzy logic, statistic method and AHP is used. AHP method is used to define weights of influencing factors for landslide susceptibility. AHP is a matrix based pair-wise comparison of effect of every factors that influences land sliding. A spacious explanation about AHP is described by [15, 33]. To compare importance of factors to each other, every factor is rated against every other factors by giving a kinsman prevalent value between 1 and 9 as shown in Table 1.

Table 1. Scale of pairwise comparison[15].

In matrix based pair wise comparison the factors in horizontal axis are compared to vertical axis with value between 1 and 9. Factor weights are showed in Table 2.

Table 2. Factor weights

The Weighted linear combination method is one of the mostly used technique in GIS-MCDA[18]. The WLC is a method that is customized in several GIS and is relevant for the flexible combination of maps. WLC is a hybrid between qualitative and quantitative methods. This method presents the simplest procedure, first the criterion scores needs to be standardized and the weights need to be computed to create a single score of evaluation[34].

In this formula, S shows the final score, Wi represents the weight of the criterion i, and Ui defines the standardized criterion score.

Figure 2. Landslide Influencing Factors (a)Aspect (b)Distance to streams (c)Elevation (d)Slope (e) Landslide Inventory Map (f)TWI

Results and Discussion

GIS based Multi Criteria Decision Analyses weighted linear combination with five conditioning factors such as slope, aspect, elevation, topographic wetness index and distance to streams were used to prepare landslide susceptibility map. The final map showing the spatial dispensation of the landslide susceptibility index was classified into five categories, very low susceptibility, low susceptibility, moderate susceptibility, high susceptibility and very high susceptibility. To examine the efficiency of final map 10 observed landslide locations was overlaid with final map.

That shows that 6 of past landslides occurred in very high susceptibility zones, where 3 landslides occurred in high susceptibility zones and 1 of past landslides occurred in moderate susceptibility zone as shown in Figure 3.

Figure 3.Landslide Susceptibility Map with overlaid observer landslides

mapping slope landslide

Conclusion

To analyze the landslide susceptibility map in this study five thematic layers were used. Applying AHP the weights of each factor were obtained and by implementing Weighted Overlay Method the landslide susceptibility index was calculated for every pixel. And the final phase was the creating and reclassifying landslide susceptibility map into five categories. To test the performance of final susceptibility map we compared it with major landslides events occurred in past and the predicated map showed the satisfactory results. The produced susceptibility map might be useful to decision makers for choosing the appropriate locations for future planning.

mapping slope landslide

References

1. Abella E.A.C., Multi-scale landslide risk assessment in Cuba. Utrecht University, 2008. 154.

2. Intarawichian N., S. Dasananda, Analytical hierarchy process for landslide susceptibility mapping in lower mae chaem watershed, northern thailand. Suranaree Journal of Science & Technology, 2010. 17(3).

3. Varnes D.J., Landslide Hazard Zonation: a review of principles and practice; UNESCO. UNESCO, Natural Hazards, No.3, 1984.

4. Brabb E. Proposal for worldwide landslide hazard maps. in Seventh International Conference and Field Workshop on Landslides in Czech and Slovak Republics. 1993.

5. Mirshakar Odinaev, Y.D.G., Alieva S, Ismoilov Kh, Bakhtiyorov Zulfiyor, Saidaliev Ismoil, The evolution process and economic growth on the example of the industry of the Republic of Tajikistan. wschodnioeuropejskie czasopismo naukowe, 2018(6-2(34)): p. 4-9.

6. Das I., et al., Landslide susceptibility assessment using logistic regression and its comparison with a rock mass classification system, along a road section in the northern Himalayas (India). Geomorphology, 2010. 114(4): p. 627-637.

7. Intarawichian N., S. Dasananda, Analytical Hierarchy Process for Landslide Susceptibility Mapping in Lower Mae Chaem. 2010. 17(3): p. 1-16.

8. Zhu L. and J.-f. Huang, GIS-based logistic regression method for landslide susceptibility mapping in regional scale. Journal of Zhejiang University-Science A, 2006. 7(12): p. 2007-2017.

9. Ge Y., et al., A comparison of five methods in landslide susceptibility assessment : a case study from the 330-kV transmission line in Gansu Region , China. Environmental Earth Sciences, 2018. 0(0): p. 0.

10. Ma F., et al., Application ot analytical hierarchy process and least-squares method for landslide susceptibility assessment along the Zhong-Wu natural gas pipeline,China. 2014: p. 481--492.

11. Millet I., T.L. Saaty, On the relativity of relative measures-accommodating both rank preservation and rank reversals in the AHP. European Journal of Operational Research, 2000. 121(1): p. 205-212.

12. Mondal S., R. Maiti, Landslide Susceptibility Analysis of Shiv-Khola Watershed, Darjiling: A Remote Sensing & GIS Based Analytical Hierarchy Process (AHP). Journal of the Indian Society of Remote Sensing, 2012. 40(3): p. 483-496.

13. Moradi M., M.H. Bazyar, Z. Mohammadi, GIS-Based Landslide Susceptibility Mapping by AHP Method , A Case Study , Dena City , Iran. 2012. 2(7): p. 6715-6723.

14. Noorollahi Y., et al., Landslide modelling and susceptibility mapping using AHP and fuzzy approaches. 2018. 2(2): p. 137-148.

15. Saaty T.L., Decision making with the analytic hierarchy process. International Journal of Services Sciences, 2008. 1(1): p. 83-98.

16. Feizizadeh B., et al., Landslide susceptibility mapping for the Urmia Lake basin, Iran: a multi-criteria evaluation approach using GIS. International Journal of Environmental Research, 2013. 7(2): p. 319-336.

17. Feizizadeh B., T. Blaschke, Landslide risk assessment based on GIS multicriteria evaluation: a case study in Bostan-Abad County, Iran. Journal of earth science and engineering, 2011. 1(1): p. 66-77.

18. Malczewski J., GIS and multicriteria decision analysis. 1999: John Wiley & Sons.

19. Malczewski J., GIS-based multicriteria decision analysis: a survey of the literature. International journal of geographical information science, 2006. 20(7): p. 703-726.

20. Malczewski J. and C. Rinner, Exploring multicriteria decision strategies in GIS with linguistic quantifiers: A case study of residential quality evaluation. Journal of Geographical Systems, 2005. 7(2): p. 249-268.

21. Chang, K. and J. Liu. Landslide features interpreted by neural network method using a high-resolution satellite image and digital topographic data. in GeoImagery Bridging Continents; Proc. of 20th ISPRS Congress, Istanbul, 12-23 July 2004. 2004.

22. Saraf A., J.D. Das, V. Rawat, Satellite Based Detection of Early Occurring and Co-Seismic landslides. 2009.

23. Yeon Y.K., J.G. Han, K.H. Ryu, Landslide susceptibility mapping in Injae, Korea, using a decision tree. Engineering Geology, 2010. 116(3-4): p. 274-283.

24. Pradhan B., S. Lee, Landslide risk analysis using artificial neural network model focussing on different training sites. International Journal of Physical Sciences, 2009. 4(1): p. 1-15.

25. Kavzoglu T., E.K. Sahin, I. Colkesen, Landslide susceptibility mapping using GIS-based multi-criteria decision analysis, support vector machines, and logistic regression. Landslides, 2014. 11(3): p. 425-439.

26. Akgun A., A comparison of landslide susceptibility maps produced by logistic regression, multi-criteria decision, and likelihood ratio methods: a case study at Izmir, Turkey. Landslides, 2012. 9(1): p. 93106.

27. Ayalew L., H. Yamagishi, The application of GIS-based logistic regression for landslide susceptibility mapping in the Kakuda-Yahiko Mountains, Central Japan. Geomorphology, 2005. 65(1-2): p. 15-31.

28. Ma F., et al., Application of analytical hierarchy process and least-squares method for landslide susceptibility assessment along the Zhong-Wu natural gas pipeline, China. Landslides, 2013. 10(4): p. 481-492.

29. Lukk A.A., et al., Earthquake focal mechanisms, deformation state, and seismotectonics of the Pamir-Tien Shan region, Central Asia. Journal of Geophysical Research: Solid Earth, 1995. 100(B10): p. 20321-20343.

30. Gorsevski P.V., et al., Integrating multi-criteria evaluation techniques with geographic information systems for landfill site selection: a case study using ordered weighted average. Waste management, 2012. 32(2): p. 287-296.

31. Malczewski J., GIS-based land-use suitability analysis: a critical overview. Progress in planning, 2004. 62(1): p. 3-65.

32. Prakash T. Land suitability analysis for agricultural crops: a fuzzy multicriteria decision making approach. 2003. ITC.

33. Vargas F.Z., An Approach to Slope Length and Features Calculating Using Dem and Gis . 1994: p. 1261--1265.

34. Voogd H., Multicriteria evaluation for urban and regional planning. Vol. 207. 1983: Pion London.

Размещено на Allbest.ru