Candan Gokceoglu

LANDSLIDE SUSCEPTIBILITY MAPPING OF THE SLOPES IN THE RESIDUAL SOILS OF THE MENGEN REGION (TURKEY) BY DETERMINISTIC STABILITY ANALYSES AND IMAGE PROCESSING TECHNIQUES

Abstract The aim of the present study is to prepare a landslide susceptibility map of a region of about 120 km2, between Gokcesu and Pazarkoy (around Mengen, NW Turkey) at approximately 10 km north of the North Anatolian Fault Zone, where frequent landslides occur. For this purpose, mechanisms of the landslides were studied by two-dimensional stability analyses together with field observations, and the parameters controlling the development of such slides were identified. Field observations indicated that the failures generally developed within the unconsolidated and/or semiconsolidated soil units in forms of rotational, successive shallow landslides within the weathered zone in Mengen, Cukurca and Sazlar formations. Although consisting of residual soils, Capak and Gokdag formations do not exhibit landslides as the natural slopes formed on these, do not exceed the critical slope angles. Statistical evaluations and distribution of the landslides on the topographical map showed that such parameters as cohesion, angle of internal friction, slope, relative height, orientation of slopes, proximity to drainage pattern, vegetation cover and proximity to major faults were the common features on the landslides. Digital images were obtained to represent all these parameters on gray scale on the SPOT image and on the digital elevation model (DEM) of the area using image processing techniques. Soil mechanics tests were carried out on 36 representative samples collected from different units, and parameters were determined for two-dimensional stability analyses basing on “sensitivity approach” and for the preparation of digital shear strength map. In order to determine the critical slope angle values for the residual soils, a series of sensitivity analyses were realized by using two-dimensional deterministic slope stability analyses techniques for varying values of cohesion, angle of internal friction and slope height along with varying saturation conditions. According to the results of the sensitivity analyses, the Mengen formation was found to be most susceptible unit to landslides, covering about 33.5% of the region studied in terms of surface area. The distribution of the critical slopes were determined by superimposing the critical slope values from sensitivity analyses on slope map of the study area. On the other hand, iso-cohesion and iso-friction maps were produced by locating the values of cohesion and internal friction angles in a geographic coordinate system such that they coincide with sample locations on the DEM and by further interpolation of the values concerned. The pixel values were evaluated in gray scale from 0 to 255, 0 representing the lowest pixel value and 255 representing the highest. Sensitivity analyses on cohesion and angle of internal friction to investigate the effects of these parameters only on stability, revealed that cohesion was effective at a rate of 70% by itself, while angle of internal friction alone controlled the stability by a rate of 30%. The iso-cohesion and iso-friction maps previously obtained were digitally combined in these rates and a “shear strength map” was prepared. The geographic setting of the study area is such that northern slopes usually receive dense precipitation. In relation to this fact, about 42% of the landslides are due north. Thus, a slope orientation map was prepared using the DEM, and slopes facing north were evaluated as being more susceptible to sliding. Proximity to the drainage pattern was another important factor in the evaluation, as streams could adversely affect the stability by either eroding the toe or saturating the slope, or both. When considered together, in conjunction with the field observations, faults and landslides showed a close association. In the area, about 88% of the landslides were detected within an area closer than 250 m to major faults, therefore, a main discontinuity map was produced using the SPOT image of the region, and “proximity to major faults” was evaluated as a parameter as most of the landslides developed in areas where the vegetation was rather sparse. A vegetation cover map was therefore obtained from the SPOT image, and the areas with denser vegetation were considered to be less susceptible to sliding with respect to the areas with less or no vegetation. Having prepared the maps accounting for the distribution of critical slopes, shear strength properties, relative height, slope angle, orientation of the slopes, vegetation cover, proximity to the drainage pattern, geographic corrections were carried on each of these, and a potential failure map was obtained for the residual soils by superimposing all these maps. Next, a classification was performed on the final map and five relative zones of susceptibility were defined. When compared with this map, all of the landslides identified in the field were found to be located in the most susceptible zone. The performance of the method used in processing the images appears to be quite high, the zones determined on the map being the zones of relative susceptibility.

A fuzzy model to predict the uniaxial compressive strength and the modulus of elasticity of a problematic rock

Abstract Although the uniaxial compressive strength and modulus of elasticity of intact rocks are highly important parameters for rock engineering and engineering geology projects, the necessary core samples cannot always be obtained from weak, highly fractured, thinly bedded, or block-in-matrix rocks. For this reason, the predictive models are often employed for the indirect estimation of mechanical parameters. However, to obtain the realistic values is very important for a predictive model. In this study, some predictive models using regression analysis and fuzzy inference system have been developed for the greywackes cropping out in the city of Ankara and its close vicinity. For this purpose, a series of rock mechanics tests were applied and the relevant intact rock parameters were obtained. Following the tests, descriptive statistical studies on the parameters, regression analyses and construction of fuzzy inference system studies were carried out. While meaningful relationships were not obtained from the simple regression analyses, both multiple regression analyses and the fuzzy inference system exhibited good predictive performance. In addition to the coefficient of correlation, the values account for (VAF) and the root mean square error indices were also calculated to check the prediction performance of the obtained models. The VAF and root mean square error indices were calculated as 41.49% and 15.62 for the uniaxial compressive strengths obtained from the multiple regression model; 64.02% and 8.85 for the modulus of elasticity values obtained from the multiple regression model; 81.24% and 13.06 for uniaxial compressive strengths obtained from the fuzzy inference system; and 78.64% and 6.87 for the modulus of elasticity values obtained from the fuzzy inference system. As a result, these indices revealed that the prediction performances of the fuzzy model are higher than those of multiple regression equations.

A fuzzy triangular chart to predict the uniaxial compressive strength of the Ankara agglomerates from their petrographic composition

Abstract High-quality core samples are necessary for the laboratory uniaxial compressive strength determinations. However, such core samples cannot always be obtained from weak, thinly bedded and block-in-matrix rocks, particularly from agglomerates and conglomerates. For this reason, the development of predictive models for the mechanical properties of rocks, mechanical indices or petrographical characteristics seems to be an attractive study area in rock engineering. Predictive models, generally, include simple and multivariate regression techniques, fuzzy logic and neural network approaches. In the present study, a fuzzy triangular chart for the prediction of uniaxial compressive strength of the Ankara agglomerates from their petrographical composition is suggested. A simple image classification method is used to determine the percentages of constituents of the agglomerate core samples. The Ankara agglomerates are mainly composed of tuff which is a cementing material, and pink and black andesite blocks ranging from few millimetres to about a meter. The classification chart developed in this study for the Ankara agglomerates includes 25 sub-triangle characterizing different petrographical composition expressed by if–then fuzzy rules. Based on the petrographical composition and uniaxial compressive strength values, a total of 15 membership function graphs were produced using if–then rules. Employing the membership functions and triangular petrographical composition chart, a fuzzy triangular chart for the prediction of uniaxial compressive strength of the agglomerates was obtained. To control performance of prediction capacity of the triangle, the variance accounts for (VAF) and the root mean square error (RMSE) indices were calculated as 96.76% and 9.37, respectively. It is noted that the fuzzy triangular chart exhibited a very high prediction capacity.

Manifestation of an adaptive neuro-fuzzy model on landslide susceptibility mapping: Klang valley, Malaysia

The purpose of the present paper is to manifest the results of the neuro-fuzzy model using remote sensing data and GIS for landslide susceptibility analysis in a part of the Klang Valley areas i Malaysia. Landslide locations in the study area were identified by interpreting aerial photographs and satellite images, supported by extensive field surveys. SPOT 5 satellite imagery was used to map vegetation index. Maps of topography, lineaments, NDVI and land cover were constructed from the spatial datasets. Seven landslide conditioning factors such as altitude, slope angle, plan curvature, distance from drainage, soil type, distance from faults and NDVI were extracted from the spatial database. These factors were analyzed using a neuro-fuzzy model (adaptive neuro-fuzzy inference system, ANFIS) to construct the landslide susceptibility maps. During the model development works, total 5 landslide susceptibility models were obtained by using ANFIS results. For verification, the results of the analyses were then compared with the field-verified landslide locations. Additionally, the ROC curves for all landslide susceptibility models were drawn and the area under curve values was calculated. Landslide locations were used to validate results of the landslide susceptibility map and the verification results showed 98% accuracy for the model 5 employing all parameters produced in the present study as the landslide conditioning factors. The validation results showed sufficient agreement between the obtained susceptibility map and the existing data on landslide areas. Qualitatively, the model yields reasonable results which can be used for preliminary landuse planning purposes. As a conclusion, the ANFIS is a very useful tool for regional landslide susceptibility assessments.

Landslide Susceptibility Mapping by Neuro-Fuzzy Approach in a Landslide-Prone Area (Cameron Highlands, Malaysia)

This paper presents the results of the neuro-fuzzy model using remote-sensing data and geographic information system for landslide susceptibility analysis in a part of the Cameron Highlands areas in Malaysia. Landslide locations in the study area were identified by interpreting aerial photographs and satellite images, supported by extensive field surveys. Landsat TM satellite imagery was used to map the vegetation index. Maps of the topography, lineaments, Normalized Difference Vegetation Index (NDVI), and land cover were constructed from the spatial data sets. Eight landslide conditioning factors such as altitude, slope gradient, curvature, distance from the drainage, distance from the road, lithology, distance from the faults, and NDVI were extracted from the spatial database. These factors were analyzed using a neuro-fuzzy model adaptive neuro-fuzzy inference system to produce the landslide susceptibility maps. During the model development works, a total of five landslide susceptibility models were constructed. For verification, the results of the analyses were then compared with the field-verified landslide locations. Additionally, the receiver operating characteristic curves for all landslide susceptibility models were drawn, and the area under curve values were calculated. Landslide locations were used to validate the results of the landslide susceptibility map, and the verification results showed a 97% accuracy for model 5, employing all parameters produced in the present study as the landslide conditioning factors. The validation results showed a sufficient agreement between the obtained susceptibility map and the existing data on the landslide areas. Qualitatively, the model yields reasonable results, which can be used for preliminary land-use planning purposes.

Assessment of Landslide Susceptibility by Decision Trees in the Metropolitan Area of Istanbul, Turkey

The main purpose of the present study is to investigate the possible application of decision tree in landslide susceptibility assessment. The study area having a surface area of 174.8 km2 locates at the northern coast of the Sea of Marmara and western part of Istanbul metropolitan area. When applying data mining and extracting decision tree, geological formations, altitude, slope, plan curvature, profile curvature, heat load and stream power index parameters are taken into consideration as landslide conditioning factors. Using the predicted values, the landslide susceptibility map of the study area is produced. The AUC value of the produced landslide susceptibility map has been obtained as 89.6%. According to the results of the AUC evaluation, the produced map has exhibited a good enough performance.

Application of weights-of-evidence and certainty factor models and their comparison in landslide susceptibility mapping at Haraz watershed, Iran

The main goal of this study was to investigate the application of the weights-of-evidence and certainty factor approaches for producing landslide susceptibility maps of a landslide-prone area (Haraz) in Iran. For this purpose, the input layers of the landslide conditioning factors were prepared in the first stage. The landslide conditioning factors considered for the study area were slope gradient, slope aspect, altitude, lithology, land use, distance from streams, distance from roads, distance from faults, topographic wetness index, stream power index, stream transport index and plan curvature. For validation of the produced landslide susceptibility maps, the results of the analyses were compared with the field-verified landslide locations. Additionally, the receiver operating characteristic curves for all the landslide susceptibility models were constructed and the areas under the curves were calculated. The landslide locations were used to validate results of the landslide susceptibility maps. The verification results showed that the weights-of-evidence model (79.87%) performed better than certainty factor (72.02%) model with a standard error of 0.0663 and 0.0756, respectively. According to the results of the area under curve evaluation, the map produced by weights-of-evidence exhibits satisfactory properties.

Landslide susceptibility mapping using support vector machine and GIS at the Golestan province, Iran

The main goal of this study is to produce landslide susceptibility map using GIS-based support vector machine (SVM) at Kalaleh Township area of the Golestan province, Iran. In this paper, six different types of kernel classifiers such as linear, polynomial degree of 2, polynomial degree of 3, polynomial degree of 4, radial basis function (RBF) and sigmoid were used for landslide susceptibility mapping. At the first stage of the study, landslide locations were identified by aerial photographs and field surveys, and a total of 82 landslide locations were extracted from various sources. Of this, 75% of the landslides (61 landslide locations) are used as training dataset and the rest was used as (21 landslide locations) the validation dataset. Fourteen input data layers were employed as landslide conditioning factors in the landslide susceptibility modelling. These factors are slope degree, slope aspect, altitude, plan curvature, profile curvature, tangential curvature, surface area ratio (SAR), lithology, land use, distance from faults, distance from rivers, distance from roads, topographic wetness index (TWI) and stream power index (SPI). Using these conditioning factors, landslide susceptibility indices were calculated using support vector machine by employing six types of kernel function classifiers. Subsequently, the results were plotted in ArcGIS and six landslide susceptibility maps were produced. Then, using the success rate and the prediction rate methods, the validation process was performed by comparing the existing landslide data with the six landslide susceptibility maps. The validation results showed that success rates for six types of kernel models varied from 79% to 87%. Similarly, results of prediction rates showed that RBF (85%) and polynomial degree of 3 (83%) models performed slightly better than other types of kernel (polynomial degree of 2 = 78%, sigmoid = 78%, polynomial degree of 4 = 78%, and linear = 77%) models. Based on our results, the differences in the rates (success and prediction) of the six models are not really significant. So, the produced susceptibility maps will be useful for general land-use planning.

Factors affecting the durability of selected weak and clay-bearing rocks from Turkey, with particular emphasis on the influence of the number of drying and wetting cycles

Weathering can induce a rapid change of rock material from initial rock-like properties to soil-like properties. The resistance of a rock to short-term weathering is described through a durability parameter called the slake durability index. As durability is an important engineering parameter, particularly for weak and clay-bearing rocks, it was assessed by a number of tests. The main purpose of this study is to assess the influence of the number of drying and wetting cycles and controls of mineralogical composition and strength on durability. For this purpose, 141 samples of different types of weak and clay-bearing rocks were selected from different parts of Turkey, and relationships between the above-mentioned rock characteristics were statistically investigated. The samples were subjected to multiple-cycle slake durability testing, X-ray diffraction (XRD) analysis and uniaxial compression testing. In addition, to assess the influence of mineralogical composition on durability, the mineral contents of the original material and the material passing from the drum of the slake durability apparatus after each cycle were also determined by XRD. The results indicate that the type and amount of clay minerals are the main factors influencing the variations of the slake durability index in all samples. The durability of the clay-bearing rocks studied correlates best with the amount of expandable clay minerals. A strong relationship between the uniaxial compressive strength and the fourth-cycle slake durability index is found only for the marls among the rock types studied. Assessment of gradation results of the spoil pile materials consisting of clay-bearing rocks also reveals that the increase in percentage of fines in old piles is indicative of material degradation, as is evident by multiple-cycle slaking. It is emphasized that two-cycle conventional slake durability testing did not appear to offer an acceptable indication of the durability of weak and clay-bearing rocks when compared with multiple-cyclic wetting and drying. Comments on the performance of the test are made that aim to make the testing process and interpretation of the results more reliable.

Discontinuity controlled probabilistic slope failure risk maps of the Altindag (settlement) region in Turkey

The evaluation of potential rock slope problems using stereographic projection techniques known as kinematic analysis is one of the most important parts of a slope stability investigation to be carried out in jointed rock media. In conventional stereoprojection techniques for the assessment of possible rock slope failures, the peak orientations of joints together with the slope geometry and the friction angle of the weakness planes are used. Other possible joint orientations which may be encountered in the rock media are ignored, although they belong to the group of joint peak orientations. In this study, nearly vertical jointed andesites cropped out at the Altindag settlement region in Ankara were studied in order to evaluate the relevance of this ignored discontinuity orientation data on slope stability. As a result, probabilistic risk maps for planar, toppling and wedge failures were produced using the kinematic rules and digital elevation model of the study area. The comparison of the distribution of the actual failures in the area and the probabilistic risk maps prepared for the study area revealed that all of the identified failures are found to be located in the higher risk zones on the probabilistic risk maps.