Spatial prediction models for shallow landslide hazards: a comparative assessment of the efficacy of support vector machines,artificial neural networks,kernel logistic regression,and logistic model tree

Preparation of landslide susceptibility maps is considered as the first important step in landslide risk assessments, but these maps are accepted as an end product that can be used for land use planning. The main objective of this study is to explore some new state-of-the-art sophisticated machine learning techniques and introduce a framework for training and validation of shallow landslide susceptibility models by using the latest statistical methods. The Son La hydropower basin (Vietnam) was selected as a case study. First, a landslide inventory map was constructed using the historical landslide locations from two national projects in Vietnam. A total of 12 landslide conditioning factors were then constructed from various data sources. Landslide locations were randomly split into a ratio of 70:30 for training and validating the models. To choose the best subset of conditioning factors, predictive ability of the factors were assessed using the Information Gain Ratio with 10-fold cross-validation technique. Factors with null predictive ability were removed to optimize the models. Subsequently, five landslide models were built using support vector machines (SVM), multi-layer perceptron neural networks (MLP Neural Nets), radial basis function neural networks (RBF Neural Nets), kernel logistic regression (KLR), and logistic model trees (LMT). The resulting models were validated and compared using the receive operating characteristic (ROC), Kappa index, and several statistical evaluation measures. Additionally, Friedman and Wilcoxon signed-rank tests were applied to confirm significant statistical differences among the five machine learning models employed in this study. Overall, the MLP Neural Nets model has the highest prediction capability (90.2 %), followed by the SVM model (88.7 %) and the KLR model (87.9 %), the RBF Neural Nets model (87.1 %), and the LMT model (86.1 %). Results revealed that both the KLR and the LMT models showed promising methods for shallow landslide susceptibility mapping. The result from this study demonstrates the benefit of selecting the optimal machine learning techniques with proper conditioning selection method in shallow landslide susceptibility mapping.     

Multi-scenario Rockfall Hazard Assessment Using LiDAR Data and GIS

Transportation corridors that pass through mountainous or hilly areas are prone to rockfall hazard. Rockfall incidents in such areas can cause human fatalities and damage to properties in addition to transportation interruptions. In Malaysia, the North–South Expressway is the most significant expressway that operates as the backbone of the peninsula. A portion of this expressway in Jelapang was chosen as the site of rockfall hazard assessment in multiple scenarios. Light detection and ranging techniques are indispensable in capturing high-resolution digital elevation models related to geohazard studies. An airborne laser scanner was used to create a high-density point cloud of the study area. The use of 3D rockfall process modeling in combination with geographic information system (GIS) is a beneficial tool in rockfall hazard studies. In this study, a 3D rockfall model integrated into GIS was used to derive rockfall trajectories and velocity associated with them in multiple scenarios based on a range of mechanical parameter values (coefficients of restitution and friction angle). Rockfall characteristics in terms of frequency, height, and energy were determined through raster modeling. Analytic hierarchy process (AHP) was used to compute the weight of each rockfall characteristic raster that affects rockfall hazard. A spatial model that considers rockfall characteristics was conducted to produce a rockfall hazard map. Moreover, a barrier location was proposed to eliminate rockfall hazard. As a result, rockfall trajectories and their characteristics were derived. The result of AHP shows that rockfall hazard was significantly influenced by rockfall energy and then by frequency and height. The areas at risk were delineated and the hazard percentage along the expressway was observed and demonstrated. The result also shows that with increasing mechanical parameter values, the rockfall trajectories and their characteristics, and consequently rockfall hazard, were increased. In addition, the suggested barrier effectively restrained most of the rockfall trajectories and eliminated the hazard along the expressway. This study can serve not only as a guide for a comprehensive investigation of rockfall hazard but also as a reference that decision makers can use in designing a risk mitigation method. Furthermore, this study is applicable in any rockfall study, especially in situations where mechanical parameters have no specific values.     

Assimilation of Doppler weather radar observations in a mesoscale model for the prediction of rainfall associated with mesoscale convective systems

Obtaining an accurate initial state is recognized as one of the biggest challenges in accurate model prediction of convective events. This work is the first attempt in utilizing the India Meteorological Department (IMD) Doppler radar data in a numerical model for the prediction of mesoscale convective complexes around Chennai and Kolkata. Three strong convective events both over Chennai and Kolkata have been considered for the present study. The simulation experiments have been carried out using fifth-generation Pennsylvania State University-National Center for Atmospheric Research (PSU-NCAR) mesoscale model (MM5) version 3.5.6. The variational data assimilation approach is one of the most promising tools available for directly assimilating the mesoscale observations in order to improve the initial state. The horizontal wind derived from the DWR has been used alongwith other conventional and non-conventional data in the assimilation system. The preliminary results from the three dimensional variational (3DVAR) experiments are encouraging. The simulated rainfall has also been compared with that derived from the Tropical Rainfall Measuring Mission (TRMM) satellite. The encouraging result from this study can be the basis for further investigation of the direct assimilation of radar reflectivity data in 3DVAR system. The present study indicates that Doppler radar data assimilation improves the initial field and enhances the Quantitative Precipitation Forecasting (QPF) skill.     

Landslide hazard mapping at Selangor, Malaysia using frequency ratio and logistic regression models

The aim of this study is to evaluate the landslide hazards at Selangor area, Malaysia, using Geographic Information System (GIS) and Remote Sensing. Landslide locations of the study area were identified from aerial photograph interpretation and field survey. Topographical maps, geological data, and satellite images were collected, processed, and constructed into a spatial database in a GIS platform. The factors chosen that influence landslide occurrence were: slope, aspect, curvature, distance from drainage, lithology, distance from lineaments, land cover, vegetation index, and precipitation distribution. Landslide hazardous areas were analyzed and mapped using the landslide-occurrence factors by frequency ratio and logistic regression models. The results of the analysis were verified using the landslide location data and compared with probability model. The comparison results showed that the frequency ratio model (accuracy is 93.04%) is better in prediction than logistic regression (accuracy is 90.34%) model.     

Probabilistic landslide hazards and risk mapping on Penang Island, Malaysia

This paper deals with landslide hazards and risk analysis of Penang Island, Malaysia using Geographic Information System (GIS) and remote sensing data. Landslide locations in the study area were identified from interpretations of aerial photographs and field surveys. Topographical/geological data and satellite images were collected and processed using GIS and image processing tools. There are ten landslide inducing parameters which are considered for landslide hazard analysis. These parameters are topographic slope, aspect, curvature and distance from drainage, all derived from the topographic database; geology and distance from lineament, derived from the geologic database; landuse from Landsat satellite images; soil from the soil database; precipitation amount, derived from the rainfall database; and the vegetation index value from SPOT satellite images. Landslide susceptibility was analyzed using landslide-occurrence factors employing the probability-frequency ratio model. The results of the analysis were verified using the landslide location data and compared with the probabilistic model. The accuracy observed was 80.03%. The qualitative landslide hazard analysis was carried out using the frequency ratio model through the map overlay analysis in GIS environment. The accuracy of hazard map was 86.41%. Further, risk analysis was done by studying the landslide hazard map and damageable objects at risk. This information could be used to estimate the risk to population, property and existing infrastructure like transportation network.     

Effect of seasonal and cyclonic winds on internal tides over the Bay of Bengal

The influence of seasonal and cyclonic winds is studied on the characteristics of internal waves (IWs) over the western Bay of Bengal (BoB) by using MITgcm model. As the BoB experiences reversal of seasonal winds and also tropical cyclones during pre-monsoon and post-monsoon months, its effect is seen through the computation of spectral estimates of the IWs. It is seen that the peak estimate is associated with the semidiurnal frequency at all the depths and is found higher in May compared to November. This is attributed to the presence of shallow mixed layer depth and deep thermocline due to the upwelling favorable winds. The computation of isopycnal displacement infers that the internal tides are present from 40 to 120 m depth in case of upwelling favorable winds of May, whereas, the presence of internal tides is restricted between 90 and 120 m for the downwelling favorable winds of November. During May, the available potential energy is also seen in a narrow coastal stretch, whilst it is absent in November. During the Hudhud cyclone period of October 7–14, 2014, it is noticed from the spectral estimates that the IWs of tidal frequency are replaced by inertial frequency with a periodicity of about 2 days as a consequence of strong cyclonic winds. The progressive vector diagram shows the mean current is initially westward up to October 17, 2014 and then northeastward with well-defined clockwise circulation. The maximum radius of inertial oscillation of 15 km is observed. After the cyclone ceases, the estimate associated with inertial frequency slowly diminishes and enhances the estimates related to internal tides. The simulations also suggest that the internal tides are absent for about 6 weeks as a response of the cyclonic winds.

     

Alongshore Sediment Transport Near Tidal Inlets of Chilika Lagoon; East Coast of India

ABSTRACT

Chilika, a lagoon along the east coast of India, is undergoing transformation due to frequent shoreline change near inlet(s). Shoreline change near inlet includes change in position and shape of inlet, inlet channel length, and spit growth/erosion. These variable features of lagoon inlet(s) critically depend on alongshore sediment transport (LST) and discharge (water and sediment) from the lagoon to the sea. The LST and the processes responsible for sand spit growth/erosion, considered as important attributes of inlet stability, are the subject matter of the present investigation and hence the study assumes importance. The study includes integration of observational and modeling framework. Observations include nearshore wave, bathymetry, beach profile, shoreline and sediment grain size of spits while numerical modeling includes simulation of the wave using MIKE 21 Spectral Wave model and LST simulation using LITtoral DRIFT. The results indicate that the predominant wave directions as S and SSE, which induces round the year south to north alongshore transport with significant seasonal variation in magnitude. The estimated LST closely matches with previous studies near Chilika inlet and for other locations along the Odisha coast. Besides temporal variability, the study reveals spatial variability in alongshore transport near Chilika inlet and considers it as one of the important attributes along with northward spit growth for inlet migration/closure/opening.     

Retrieval and validation of soil moisture from FRS-1 data set of radar imaging satellite (RISAT-1)

In this work, we estimated soil moisture (SM) by using dielectric constant properties of soil with radar backscattering coefficient (bc) from simulated annealing techniques of RISAT-1 (radar imaging satellite, based on synthetic aperture radar (SAR) technique) data. We examined the performance of simulated annealing in retrieving SM where the vegetation cover is not very high (NDVI ≈ 0.35 for wheat-dominated area on January 19, 2013).To overcome the land surface model limits on SM estimation accuracy, point measurement spatial coverage limits, and microwave remote sensing spatial-temporal sampling limits, we reduced uncertainties through a combination of these approaches. Near-surface SM measurements from the 5.35-GHz (C-band) channels of RISAT-1 were collocated against ground-truth data (collected during the flight time of RISAT-1 over the study area), to establish SAR-SM relationships for FRS-1(circular horizontal (RH) and circular vertical (RV)) data set of RISAT-1. Comparison with the limited ground-based point (total 24 points) measurements of SM content exhibited a net improvement when near-surface SM observations were assimilated. Comparison of the SM derived from the sigma naught (σ ⁰) (of RISAT-1 data set FRS-1) using the inversion algorithm with the observed measurements (using time domain reflectometry) of SM showed root mean square error of 0.24, nRMSE = 0.03, R-RMSE = 0.38, MAE = 0.63, NRMSE = 1.02, NSE = 1, d = 0.87, r ² = 0.65, and RMSE% = 12.79 for RH polarized image while RV polarized image failed each statistical test for predicting good SM with respect to the observed SM. The ability to extract additional information comes at the expense of including more measurements, especially at frequencies lower than the L-band. This approach is therefore intended for future space-borne systems.     

Evaluation of groundwater vulnerability to pollution using DRASTIC framework and GIS

Groundwater management has a prominent role in the world especially in arid and semi-arid areas which have a shortage of water, and due to this serious problem, many researchers work on that for prevention and managing the water recourses to conserve and monitor sources. DRASTIC index can be put forward for estimating of groundwater vulnerability to such pollution. The main purpose of using the groundwater vulnerability model is to map groundwater susceptibility to pollution in different areas. However, this method has been used in various areas without modification, disregarding the effects of pollution type and characteristics. Thus, this technique must be standardized and approved for Kerman plain. Vulnerability evaluation to explain areas that are more vulnerable to contamination from anthropogenic sources has become a prominent element for land use planning and tangible resource management. This contribution aims at evaluating groundwater vulnerability by applying the DRASTIC index as well as employ sensitivity analyses to evaluate the comparative prominent of the model parameters for groundwater vulnerability in Kerman plain in the southeastern part of Iran. Moreover, the potential of vulnerability to pollution is more accurately assessed by optimizing the weights of the DRASTIC parameters with the single-parameter sensitivity analysis (SPSA). The new weights were calculated. The result of the study revealed that the DRASTIC-Sensitivity analysis exhibit more efficiently than the traditional method for a nonpoint source pollution. Observation of ultimate nitrate showed the result of DRASTIC-SPSA has more accuracy. The GIS method offers an efficient environment for carrying out assessments and greater capabilities for dealing with a huge quantity of spatial data.