Landslide susceptibility mapping using machine learning algorithms and comparison of their performance at Abha Basin,Asir Region,Saudi Arabia

The current study aimed at evaluating the capabilities of seven advanced machine learning techniques(MLTs),including,Support Vector Machine(SVM),Random Forest(RF),Multivariate Adaptive Regression Spline(MARS),Artificial Neural Network(ANN),Quadratic Discriminant Analysis(QDA),Linear Discriminant Analysis(LDA),and Naive Bayes(NB),for landslide susceptibility modeling and comparison of their performances.Coupling machine learning algorithms with spatial data types for landslide susceptibility mapping is a vitally important issue.This study was carried out using GIS and R open source software at Abha Basin,Asir Region,Saudi Arabia.First,a total of 243 landslide locations were identified at Abha Basin to prepare the landslide inventory map using different data sources.All the landslide areas were randomly separated into two groups with a ratio of 70%for training and 30%for validating purposes.Twelve landslide-variables were generated for landslide susceptibility modeling,which include altitude,lithology,distance to faults,normalized difference vegetation index(NDVI),landuse/landcover(LULC),distance to roads,slope angle,distance to streams,profile curvature,plan curvature,slope length(LS),and slope-aspect.The area under curve(AUC-ROC)approach has been applied to evaluate,validate,and compare the MLTs performance.The results indicated that AUC values for seven MLTs range from 89.0%for QDA to 95.1%for RF.Our findings showed that the RF(AUC=95.1%)and LDA(AUC=941.7%)have produced the best performances in comparison to other MLTs.The outcome of this study and the landslide susceptibility maps would be useful for environmental protection.     

Using high-resolution satellite imagery to provide a relief priority map after earthquake

After the earthquake occurrence, collecting correct information about the extent of damage is essential for managing critical conditions and allocating limited resources. The prepared building damage maps sometimes bring about waste of time required for rescuing individuals under the rubble by wrongly conducting rescue teams toward regions with a lower rescue priority. In this research, an algorithm based on using a proposed standard at database level was developed to prioritize damaged buildings by considering five key elements of land use type, the degree of damage to buildings, the land use differentiation index, time of the highest population density in each land use, and time of disaster’s incidence. The steps of the proposed method which was implemented in the MATLAB environment include: detecting buildings on the pre- and post-event imagery, implementing texture features for each candidate building, choosing the optimal features by genetic algorithm, determining the degree of building damage in three classes of negligible damage, substantial damage, and heavy damage by using the difference between chosen features as inputs of the designed neurofuzzy inference system. Data collected from field observations were compared to the output obtained from the proposed algorithm. This comparison presented a general accuracy of 88% and Kappa coefficient of 79% in the classification of buildings into three damage classes. The proposed standard then was used for classifying damaged buildings into relief priorities of high, medium, and low. Findings revealed that the relief priority map could be a basis for correct guidance of relief and rescue teams during crucial times following earthquakes.     

Pressuremeter Modulus and Limit Pressure of Clayey Soils Using <Emphasis Type="Italic">GMDH</Emphasis>-Type Neural Network and Genetic Algorithms

Pressuremeter modulus (\(E_{M}\)) and limit pressure (\(P_{L}\)) are used for the calculation of the settlement and bearing capacity of foundation respectively. As the determination of these parameters from pressuremeter test (PMT) is relatively time-consuming and expensive, various empirical correlations have been proposed to correlate the \(E_{M}\) and \(P_{L}\) to other soil parameters. For the existing equations are incapable of estimating these PMT parameters well, in present research group method of data handling type neural network is used to estimate the \(E_{M}\) and \(P_{L}\) of clayey soils. The \(E_{M}\) and \(P_{L}\) were modeled as a function of three variables including the moisture content (\(\omega\)), plasticity index and corrected SPT blow counts (\(N_{60}\)). A database containing 51 data sets have been used for training and testing of the models. The performances of proposed models are compared with those of existing empirical equations. The results demonstrate that appreciable improvement with respect to the other correlations has been achieved. At the end, sensitivity analysis of the obtained models has been performed to study the influence of input parameters on model outputs and shows that the \(N_{60}\) is the most influential parameter on the PMT parameters.     

A hybrid clustering-fusion methodology for land subsidence estimation

In many parts of Canada, limited data are available for hydrodynamic model inputs, and the ability to generate quality flood grids through 1D, 2D or 3D methods is nonviable. In this paper, the capability of simplified flood models, which rely solely on digital terrain models (DTMs), was explored to assess the quality and speed of their results. Results were validated against historic floods in two locations. Three non-physics-based simplified conceptual flood models were tested: (1) planar method, (2) inclined plane and (3) height above nearest drainage network (HAND) model. The accuracy and performance were evaluated using three criteria: inundation extent, water depth and computation time. Findings show that the HAND model is the best predictor of inundation extent, with Probability of Detection and Critical Success Index being higher than 0.90 in both study areas. Though the preprocessing time for the HAND model is lengthy, once completed, the time to simulate flooding at a variety of water levels is rapid, making this model the most suitable choice for web-based, on-demand flood inundation mapping. Knowledge of the fit of these flood models and associated uncertainty can be helpful to emergency managers such that they can better understand exposure and vulnerability while preparing flood response plans.     

A tectono-stratigraphic record of an extensional basin: the Lower Jurassic Ab-Haji Formation of east-central Iran

The Lower Jurassic Ab-Haji Formation consists of siliciclastic strata which are widespread and superbly exposed across the Tabas and Lut blocks of east-central Iran. The formation records the geodynamic history of central Iran during the Early Jurassic in the aftermath of the main Cimmerian event (near the Triassic–Jurassic boundary) through its sedimentary facies and stratigraphic architecture and allows palaeogeographic and palaeoenvironmental reconstructions. We measured and studied three well-exposed outcrop sections and identified lithofacies and facies associations (fluvial plain, delta plain, delta front, prodelta, and shallow-marine siliciclastic shelf). The integration of all geological, stratigraphic, and sedimentological data shows a west-to-east continental-to-marine gradient within the Ab-Haji Formation. Based on thickness variations, lateral facies changes, palaeocurrent patterns, and changes in the nature of the basal contact of the Ab-Haji Formation on the Tabas and Lut blocks, we locate the fault-bounded Yazd Block in the west and the Shotori Swell at the eastern edge of the Tabas Block as provenance regions. The pattern of thickness variations, rapid east–west facies changes, and provenance is best explained by a tectonic model invoking large tilted fault blocks in an extensional basin. The basal unit shows distinct increase in grain size at the base of the Ab-Haji Formation, similar to the Shemshak Group of the Alborz Mountains (the base of the Alasht Formation) and the non-marine time-equivalent succession of the Binalud Mountains of northeastern Iran. This grain size pattern may have been caused by rapid source area uplift due to slab break-off of the subducted Iran plate in the course of the Cimmerian collision in east-central Iran.     

Spatial prediction of soil erosion susceptibility: an evaluation of the maximum entropy model

Soil erosion is considered as the most widespread form of soil degradation which causes serious environmental problems. This study investigates the performance of the maximum entropy (ME) in mapping rill erosion susceptibility in the Golgol watershed, Ilam province, Iran. To this end, ten rill erosion conditioning factors were selected to be employed in the modelling process based on an investigation of the literature. These layers are: elevation, slope percent, aspect, stream power index, topographic wetness index, distance from streams, plan curvature, lithology, land use, and soil. Then, a training dataset of rill erosion locations was used for modelling this phenomenon. The area under receiver operating characteristics curve was used for evaluating the performance of the ME model. In addition, Modified Pacific South-West Inter Agency Committee (MPSIAC) framework was applied and sediment yield was determined for different hydrological units in the study area. At last, Jackknife test was implemented to show the contribution of the factors in the modelling process. The results depicted that area under ROC curve for training and validation datasets were 0.867, and 0.794, respectively. Therefore, this conclusion can be achieved that ME worked well and could be a good tool for generating rill erosion susceptibility maps and its output could be employed for soil conservation in similar areas.     

GIS-based gully erosion susceptibility mapping: a comparison among three data-driven models and AHP knowledge-based technique

Toroud Watershed in Semnan Province, Iran is a prone area to gully erosion that causes to soil loss and land degradation. To consider the gully erosion, a comprehensive map of gully erosion susceptibility is required as useful tool for decreasing losses of soil. The purpose of this research is to generate a reliable gully erosion susceptibility map (GESM) using GIS-based models including frequency ratio (FR), weights-of-evidence (WofE), index of entropy (IOE), and their comparison to an expert knowledge-based technique, namely, Analytic Hierarchy Process (AHP). At first, 80 gully locations were identified by extensive field surveys and Google Earth images. Then, 56 (70%) gully locations were randomly selected for modeling process, and the remaining 26 (30%) gully locations were used for validation of four models. For considering geo-environmental factors, VIF and tolerance indices are used and among 18 factors, 13 factors including elevation, slope degree, slope aspect, plan curvature, distance from river, drainage density, distance from road, lithology, land use/land cover, topography wetness index (TWI), stream power index (SPI), normalized difference vegetation index (NDVI), and slope–length (LS) were selected for modeling aims. After preparing GESMs through the mentioned models, final maps divided into five classes including very low, low, moderate, high, and very high susceptibility. The receiver operating characteristic (ROC) curve and the seed cell area index (SCAI) as two validation techniques applied for assessment of the built models. The results showed that the AUC (area under the curve) in training data are 0.973 (97.3%), 0.912 (91.2%), 0.939 (93.9%), and 0.926 (92.6%) for AHP, FR, IOE, and WofE models, respectively. In contrast, the prediction rates (validating data) were 0.954 (95.4%), 0.917 (91.7), 0.925 (92.5%), and 0.921 (92.1%) for above models, respectively. Results of AUC indicated that four model have excellent accuracy in prediction of prone areas to gully erosion. In addition, the SCAI values showed that the produced maps are generally reasonable, because the high and very high susceptibility classes had very low SCAI values. The results of this research can be used in soil conservation plans in the study area.     

Prediction of strength parameters of sedimentary rocks using artificial neural networks and regression analysis

Accurate laboratory measurement of geo-engineering properties of intact rock including uniaxial compressive strength (UCS) and modulus of elasticity (E) involves high costs and a substantial amount of time. For this reason, it is of great necessity to develop some relationships and models for estimating these parameters in rock engineering. The present study was conducted to forecast UCS and E in the sedimentary rocks using artificial neural networks (ANNs) and multivariable regression analysis (MLR). For this purpose, a total of 196 rock samples from four rock types (i.e., sandstone, conglomerate, limestone, and marl) were cored and subjected to comprehensive laboratory tests. To develop the predictive models, physical properties of studied rocks such as P wave velocity (V_p), dry density (γ_d), porosity, and water absorption (A_b) were considered as model inputs, while UCS and E were the output parameters. We evaluated the performance of MLR and ANN models by calculating correlation coefficient (R), mean absolute error (MAE), and root-mean-square error (RMSE) indices. The comparison of the obtained results revealed that ANN outperforms MLR when predicting the UCS and E.     

Large scale mass movements triggered by the Kashmir earthquake 2005, Pakistan

The SPOT image analysis in Muzaffarabad Azad Kashmir,northwest Himalayas,Pakistan reveals that the Kashmir earthquake 2005 triggered a number of coseismic mass movements along the hanging wall block of the Muzaffarabad Fault.The Neelidandi and Langarpura rock falls have been identified as two major reactivated mass movements with an estimated volume of 3.1 × 106m3and 5.76 × 106m3,respectively.The Neelidandi and Langarpura mass movements were initiated during earthquake in the direction of northwest-southeast extension and northeastsouthwest directed thrusting,respectively.The Neelidandi rock fall occurred in sheared cherty dolomites and limestones of the Cambrian Muzaffarabad Formation,whereas the Langarpura rock fall occurred in alternating clays,shales,claystones,siltstones and sandstones of the Miocene Murree Formation.These rock units along the fault are highly fractured and jointed.The geotechnical maps and geological longitudinal profiles show the relationship between the geometrical characteristics and mechanism of these mass movements.Their characteristics were analyzed according to the role of topographic,seismic,geological and tectonic factors.The steep topography,sheared rocks,lithology,coseismic uplift and strong ground shaking of the hanging wall block along Muzaffarabad Fault facilitated the gravity collapse of these mass movements.     

Landslide Susceptibility Mapping along Bhalubang–Shiwapur Area of Mid-Western Nepal Using Frequency Ratio and Conditional Probability Models

Roads constructed in fragile Siwaliks are prone to large number of instabilities. Bhalubang–Shiwapur section of Mahendra Highway lying in Western Nepal is one of them. To understand the landslide causative factor and to predict future occurrence of the landslides, landslide susceptibility mapping(LSM) of this region was carried out using frequency ratio(FR) and weights-of-evidence(W of E) models. These models are easy to apply and give good results. For this, landslide inventory map of the area was prepared based on the aerial photo interpretation, from previously published/unpublished reposts, and detailed field survey using GPS. About 332 landslides were identified and mapped, among which 226(70%) were randomly selected for model training and the remaining 106(30%) were used for validation purpose. A spatial database was constructed from topographic, geological, and land cover maps. The reclassified maps based on the weight values of frequency ratio and weights-of-evidence were applied to get final susceptibility maps. The resultant landslide susceptibility maps were verified andcompared with the training data, as well as with the validation data. From the analysis, it is seen that both the models were equally capable of predicting landslide susceptibility of the region(W of E model(success rate = 83.39%, prediction rate = 79.59%); FR model(success rate = 83.31%, prediction rate = 78.58%)). In addition, it was observed that the distance from highway and lithology, followed by distance from drainage, slope curvature, and slope gradient played major role in the formation of landsides. The landslide susceptibility maps thus produced can serve as basic tools for planners and engineers to carry out further development works in this landslide prone area.