Geoinformatics and Modelling of Landslide Susceptibility and Risk

The dynamic interplay of disturbance and succession in ecosystems are meaningfully explained by the occurrence of landslides. It is very difficult task to restore the landslide surface area because of the presence of high degree of spatial and temporal variability in soil stability and fertility (Walker et al. 2009). The variability of landslides and its destructive character have brought attention of many research scholars in logical and scientific understanding of the concept, mechanism, vulnerability and risk of landslides. Landslides can be defined as the movement of mass of rocks, earth materials, and debris down the slope under the influence of gravity by which nature finds its way of adjusting slope stability.

Occurrences of landslides are affected by geomorphic, tectonic and hydrologic parameters. Balason river basin of Darjeeling Himalaya exhibits these parameters which are being analyzed in regard to landslide occurrences. In the present work, we carried out an overlay analysis of various data layers such as elevation, slope, geology, geomorphology, soil, rainfall, drainage density and drainage frequency with the landslide distribution data layer to assess the probability of different classes of landslide causative factors. Frequency ratio (FR) values for each class were estimated considering both landslide affected pixels and total pixels of a class. The result showed that, those areas which were characterized by elevation of 1065–1257 m, slope of 36–89°, south facing slope, high concave area, lineament density of 1.22–2.29, 20–40 m distance from lineament, NDVI of 0.16–0.21, drainage density of 620.17–741.51, 100–150 m distance from drainage, SPI of 12.34–13.14, TWI of −19.41 to −17.59 and rainfall of 2560–2618 mm were registered with high frequency ratio and high landslide probability. In addition, the probability of landslide occurrences is also high in Darjeeling gneiss, lower hill, human settlement and fine loamy to coarse loamy soil textural area.

Morphometric analysis of any drainage basin has paid immense importance in hydrological investigation. Morphometric parameters describe the topology, the structure, the planform and the relief of basin which are being applied for the prioritization of watersheds. In the present study, an attempt has been made to prioritize sub-watersheds based on morphometric analysis in relation to slope instability. The base map of stream network were digitized from toposheets no. 78A/4, 78A/8, 78B/1, 78B/5 and 78B/6 (Scale 1:50,000) and then updated on Google earth. Arc GIS 9.3 software and MS excel-2007 were used to assess 13 morphometric parameters i.e. bifurcation ratio (Rb), length of overland flow (Lof), drainage density (Dd), stream frequency (Fs), texture ratio (Rt), drainage texture (Td), compactness coefficient (Cc), constant of channel maintenance (Ccm), shape factor/basin shape (Sf), form factor (Ff), circularity ratio (Rc), elongation ratio (Re) and relief ratio (Rr) of the Balason River basin of Darjeeling Himalaya and to prioritize subwatersheds (SW). The results showed that SW 4, SW 5 and SW 8 fall under very high priority class in respect of soil erosion and soil loss or simply instability having Rb of 3.532–4.002, Dd of 5.986–6.538, Fs of 14.934–29.447, Td of 9.765–23.158, Rt of 7.055–16.958, Lof of 0.069–0.084, Rr of 0.209–0.302, Cc of 1.265–1.554, Ccm of 0.139–0.167, Sb of 2.969–5.556, Ff of 0.180–0.337, Rc of 0.420–0.634 and Re of 0.479–0.655. Sub-watershed wise highest priority obtained by SW 5 followed by SW 8, SW 4, SW 3, SW 9, SW 1, SW 6, SW 16, SW 2, SW 17, SW 15, SW 13, SW 7, SW 10, SW 11, SW 12, SW 14, and SW 18.

The present study attempts to assess the role of basin morphometric parameters in slope instability using morphometric diversity (MD) model. Also try to find out the role of drainage parameters and relief parameters in slope failure using drainage diversity (DD) and relief diversity (RD) models respectively. For that total 14 morphometric data layers were considered. The relationship of each data layers with landslide susceptibility was judge using frequency ratio (FR) approach. Parameters like drainage density, drainage frequency, relative relief, drainage texture, junction frequency, infiltration number, ruggedness index, dissection index, elevation, slope, relief ratio and hypsometric integral were positively related with landslide potentiality while bifurcation ratio and drainage intensity negatively correlated with slope failure. The principal component analysis (PCA) based weight assign to each data layers of each model which multiplied with unidirectional reclassified data layers for each model using weighted linear combination (WLC) approach to prepare landslide susceptibility maps. The receiver operating characteristics curve showed that, the landslides prediction accuracy of the DD, RD and MD models was 71.4, 73.9 and 76.3% respectively. The FR plots of the aforesaid three models suggested that, the chance of landslide increases from very low to very high susceptibility zones.

The present study is dealt with the application of Information value model (IVM), landslide nominal risk factor mode (LNRFM), fuzzy logic approach (FLA) and statistical index model (SIM) and the preparation of landslide susceptibility maps of the Balason river basin of Darjeeling Himalaya using various geomorphic, hydrologic, and tectonic attributes such as elevation, slope, aspect, curvature, geology, geomorphology, soil, distance to lineament, lineament density, drainage density, distance to drainage, stream power index (SPI), topographic wetness index (TWI), land use and land cover (LULC) and NDVU. All the landslide conditioning factors were being processed in GIS platform. The prepared landslide susceptibility maps were also validated using ROC curve which stated that fuzzy logic approach is best suited for developing landslide susceptibility zonation map of the Balason river basin of Darjeeling Himalaya.

The present study is associated with the implication of weight of evidence model and certainty factor model to prepare landslide susceptibility maps of the Balason river basin of Darjeeling Himalaya using data layers of elevation, slope, aspect, curvature, geology, geomorphology, soil, distance to lineament, lineament density, drainage density, distance to drainage, stream power index (SPI), topographic wetness index (TWI), land use and land cover (LULC) and NDVU in ARC GIS 10.1. The developed landslide susceptibility map was classified in five i.e. very low, low, moderate, high and very high landslide susceptibility. The prepared landslide susceptibility maps were also validated using ROC curve which stated that certainty factor mode is best suited for developing landslide susceptibility zonation map of the Balason river basin of Darjeeling Himalaya.

The present study is dealt with the preparation of landslide susceptibility map of the Balason river basin of Darjeeling Himalaya with the help of GIS tools machine learning model i.e. support vector machine (SVM) and artificial neural network model (ANNM). Fifteen landslide causative factors i.e. slope, aspect, curvature, elevation, geology, geomorphology, soil, distance to drainage, drainage density, distance to lineaments, lineament density, land use and land cover, stream power index (SPI), topographic wetness index (TWI) and rainfall were considered to produce the landslide susceptibility zonation maps. To generate all these factors map topographical maps, geological map, geomorphological map, soil map, satellite imageries, and google earth images were processed and constructed into a spatial data base using GIS and image processing techniques. SVM classification algorithm was performed for each factor by using the RBF kernel to prepare landslide susceptibility map. And, the back-propagation method was also applied to estimate factor’s weight and the landslide hazard indices were derived with the help of trained back-propagation weights using ANN model. Then, the landslide susceptibility zonation map of the Balason river basin was made using GIS tool and classified into five i.e. very low, low, moderate, high and very low landslide susceptibility. To validate the prepared landslide susceptibility map landslide inventory was used and accuracy result was obtained after processing ROC curve.

The present study attempts to assess geo-spatial distribution of landslide susceptibility in the Balason river basin of Darjeeling Himalaya using clustering of various factors i.e. geomorphological factors, lithological factors group, hydrological factors, triggering factor, protective factor and anthropogenic factor. The geomorphological factors, lithological factors group, hydrological factors, triggering factor, protective factor and anthropogenic factor were being integrated with the help of geomorphological factor group model, lithological factor group model, hydrological factor group model, triggering factor group model, protective factor group model and anthropogenic factor group model. To prepare data layers such as elevation, slope, aspect, curvature, geology, geomorphology, soil, drainage density, distance to drainage, lineament density, distance to lineament, stream power index, topographic wetness index, NDVI and LULC Google earth, topographical maps, SRTM DEM, satellite images were processed properly in GIS environment.

Since ancient time, the Darjeeling Himalaya has been subjected to landslide hazards and similar phenomena. The Balason river basin is situated in this region and suffered several types of large and small landslides. At the time of pre-independence landslides took place mostly in the areas where the setup of the physical environment was weak, less strong and more disturbed geological formations, rugged and high slope areas, low vegetation cover areas etc. These natural geo-environmental parameters play an important role in slope instability (Hays in Facing geologic and hydrologic hazards: earth-science considerations. U.S. Geological Survey, Reston, Virginia, 1981; Sarkar and Kanungo in Nat Hazard Manag 1–6, 2010). In recent times, the severity and the frequency of the landslide phenomena have been increased significantly.

Among the various natural hazards, slope failure is the most widespread and damaging hazard (De Smedt in Slope stability analysis using GIS on a regional scale: a case study of Narayanghat Mungling highway section, Nepal, 2005). A sudden failure of the slope is caused by sliding, rolling, falling or slumping. When failure occurs, material is transported down slope until a stable slope condition is re-established. The Darjeeling Himalayan terrain is very high susceptible to slope instability due to a complex geological structure and complex interaction among various processes acting upon the steep mountain southern escarpment slopes. In Darjeeling, the spatial extents of landslides are increasing day by day and causing severe damage to lives and properties. The Balason river basin is not an exception to it.