Improved flood susceptibility mapping using a best first decision tree integrated with ensemble learning techniques

Improving the accuracy of flood prediction and mapping is crucial for reducing damage resulting from flood events. In this study, we proposed and validated three ensemble models based on the Best First Decision Tree (BFT) and the Bagging (Bagging-BFT), Decorate (Bagging-BFT), and Random Subspace (RSS-BFT) ensemble learning techniques for an improved prediction of flood susceptibility in a spatially-explicit manner. A total number of 126 historical flood events from the Nghe An Province (Vietnam) were connected to a set of 10 flood influencing factors (slope, elevation, aspect, curvature, river density, distance from rivers, flow direction, geology, soil, and land use) for generating the training and validation datasets. The models were validated via several performance metrics that demonstrated the capability of all three ensemble models in elucidating the underlying pattern of flood occurrences within the research area and predicting the probability of future flood events. Based on the Area Under the receiver operating characteristic Curve (AUC), the ensemble Decorate-BFT model that achieved an AUC value of 0.989 was identified as the superior model over the RSS-BFT (AUC = 0.982) and Bagging-BFT (AUC = 0.967) models. A comparison between the performance of the models and the models previously reported in the literature confirmed that our ensemble models provided a reliable estimate of flood susceptibilities and their resulting susceptibility maps are trustful for flood early warning systems as well as development of mitigation plans.     

Investigating the impacts of typhoon-induced floods on the agriculture in the central region of Vietnam by using hydrological models and satellite data

Flooding associated with landing tropical cyclones (TCs) is one of the major natural hazards in the coastal region of Vietnam. Annually, approximately 5 or 6 TCs make landfall in Vietnam, bringing heavy rains and inducing flooding, particularly to the central coastal region because of its topography and geographic configuration. This study focuses on the modelling of typhoon-induced floods that have resulted in widespread damage to agriculture over the central Thua Thien Hue Province of Vietnam by coupling two well-known hydrological models, KINEROS2 and HEC-RAS (Daniel et al. in Open Hydrol J 5(1), 2011), and using GSMaP (Global Satellite Mapping of Precipitation) data as the satellite rainfall input. Landsat imagery and GIS are also used for mapping and analysing the inundated areas. The discharge and water level from the KINEROS2 and HEC-RAS models displayed acceptable results for the floods modelled from three selected typhoons; both the Nash–Sutcliffe simulation efficiency coefficient (NSE) and the coefficient of determination (R²) were greater than 0.6. The simulated inundation maps of these typhoon-induced floods were compared with those extracted from the Landsat imagery to assess consistency. The result revealed a similar spatial extension of the inundated agricultural areas. This information, together with the forecasted TC movements and associated rainfalls, will be helpful to plan methods for mitigating potential typhoon-induced flooding and damage, particularly damage to agricultural regions.     

Correcting the distortion of apparent resistivity pseudosection produced by 3D effect of the embankment geometry

Two-dimensional electrical resistivity tomography data acquired on the crest of the embankment are distorted by the 3D effect of the embankment geometry, reservoir water level and abutment. The distortion affects seriously the final solution of the 2D inverse problem. By comparing the apparent resistivity pseudosections from a 3D and 2D electrical resistivity model of the embankment, the distortion degrees of the apparent resistivity pseudosections along the axis on the crest were estimated for the cases of reservoir and which does not contain water. The obtained results indicate that the distortion degree acquired in the case of a reservoir that contains water was much less than that in the case of the reservoir that does not contain water. In the case of reservoir that contains water, the apparent resistivity pseudosections of the P–P and ED–ED arrays had the largest distortion degree and of D–D, W–S and P–D arrays had the smallest distortion degree. In the case of the reservoir that does not contain water, the apparent resistivity pseudosection of P–P array had the smallest distortion among all arrays. Through modeling investigation, a correction process to reduce the distortion of the apparent resistivity pseudosection was proposed. The correction process was tested in the embankment model, and two field works were carried out in the To Lich River in Hanoi and Khuan Cat embankment in Lang Son province, Vietnam. It is possible to bring the distortion of the apparent resistivity pseudosection down to 2.8–13.9%, depending on the type of electrode arrays and the type of reservoirs, containing or does not contain water. The distortion correction of the apparent resistivity pseudosection is strongly recommended before doing the 2D inverse interpretation.     

Formation process of two massive dams following rainfall-induced deep-seated rapid landslide failures in the Kii Peninsula of Japan

Extreme heavy rainfall due to Typhoon Talas on September 2–4, 2011 in the Kii Peninsula, Japan, triggered numerous floods and landslides. This study investigates the mechanism and the entire process of rainfall-induced deep-seated landslides forming two massive dams in the Kuridaira and Akatani valleys, respectively. The mechanism of the rapid deep-seated landslides is examined through a series of laboratory experiments on samples from sliding surfaces by using undrained high-stress dynamic-loading ring-shear apparatus. The test results indicate that the failure of samples is triggered by excess pore water pressure generation under a shear displacement from 2 to 7 mm with a pore pressure ratio ranging from 0.33 to 0.37. The rapid movement of landslides is mainly attributed to high mobility due to the liquefaction behavior of both sandstone-rich and shale samples. Geomorphic settings and landslide mobility are major contributing factors to the dam formation. Additionally, shear displacement control tests show that a certain amount of shear displacement between 2 and 7 mm along the sliding surfaces of the gravitationally deformed slopes might have led to the failures. Importantly, computer simulation with LS-RAPID software using input parameters obtained from physical experiments is employed to interpret the entire formation process of the abovementioned two landslide dams. The simulation results are examined in accordance with the observed on-site geomorphic features and recorded data to explain the possibility of sliding processes. The results further point out that local failures are initiated from the lower middle part of the landslide bodies where the geological boundary exists. This condition most probably influences the landslide initiation in the two case studies. This research is therefore helpful for hazard assessment of slopes that are susceptible to deep-seated landslides and other sequential processes in areas with geology and geomorphology similar to that of the Kii Peninsula.     

Subduction initiation of Indochina and South China blocks: insight from the forearc ophiolitic peridotites of the Song Ma Suture Zone in Vietnam

The Song Ma region, which is located in the northwestern Vietnam represents the zone of amalgamation between Indochina and South China blocks. Numerous scattered ultramafic rocks occur in this region in association with Early to Middle Palaeozoic greenschists and paragneisses, and all these rocks were subjected to hydrous metamorphism and deformation. Here, we present new field data, mineral chemistry and geochemistry from a suite of hydrated peridotites within the Song Ma region and discuss the tectonic significances of the region. We also combine the available data within the Song Ma region and Indochina–South China blocks to discuss the tectonic evolution of the subduction zone. Based on the results, we suggest that the peridotites from the Song Ma are mantle residues that suffered a high degree of partial melting in a forearc tectonic setting. The present data together with the available data within the Song Ma region and the Indochina and South China blocks clearly represent a southward directed Middle Palaeozoic subduction system. The Middle Palaeozoic subduction and accretion events mark the evolutionary history along an active convergent margin between the Indochina and South China blocks, possibly related to the amalgamation of the Pangaea supercontinent. Copyright © 2015 John Wiley & Sons, Ltd.     

Groundwater Potential Mapping Using GIS-Based Hybrid Artificial Intelligence Methods

Groundwater is one of the major valuable water resources for the use of communities, agriculture, and industries. In the present study, we have developed three novel hybrid artificial intelligence (AI) models which is a combination of modified RealAdaBoost (MRAB), bagging (BA), and rotation forest (RF) ensembles with functional tree (FT) base classifier for the groundwater potential mapping (GPM) in the basaltic terrain at DakLak province, Highland Centre, Vietnam. Based on the literature survey, these proposed hybrid AI models are new and have not been used in the GPM of an area. Geospatial techniques were used and geo-hydrological data of 130 groundwater wells and 12 topographical and geo-environmental factors were used in the model studies. One-R Attribute Evaluation feature selection method was used for the selection of relevant input parameters for the development of AI models. The performance of these models was evaluated using various statistical measures including area under the receiver operation curve (AUC). Results indicated that though all the hybrid models developed in this study enhanced the goodness-of-fit and prediction accuracy, but MRAB-FT (AUC = 0.742) model outperformed RF-FT (AUC = 0.736), BA-FT (AUC = 0.714), and single FT (AUC = 0.674) models. Therefore, the MRAB-FT model can be considered as a promising AI hybrid technique for the accurate GPM. Accurate mapping of the groundwater potential zones will help in adequately recharging the aquifer for optimum use of groundwater resources by maintaining the balance between consumption and exploitation.     

Public perception of climate change and disaster preparedness: Evidence from the Philippines

The Philippines is highly susceptible to both geophysical and climate-related disasters. This article explores Filipinos knowledge and perception of climate change and their association with what action Filipinos take to prepare for rapid onset natural hazards such as typhoons. Data for this study were collected from a nationally representative random survey of 5,184 adults conducted between March and April of 2017. Filipinos self-report relatively low levels of knowledge of climate change and cited increased temperatures, shifts in seasons, and heavier rains as the most likely consequences. Levels of disaster preparedness in the Philippines differ widely by region. Although most Filipinos perceive that natural hazards are a risk to them, only a third of Filipinos undertake measures to prepare for disasters. Filipinos who perceive climate-related changes directly impacting their households report taking greater action to prepare for disasters. Filipinos who believe they have been directly impacted by climate-related changes are also more likely to prepare for disasters, take planning actions, and undertake material actions to prepare, such as dwelling improvements. Other factors associated with disaster preparedness include gender, membership in an association, wealth, risk perception, and prior exposure to and losses due to disasters. The findings imply that, while posing different challenges and requiring different responses, adaptation to climate change and disaster preparedness are inherently associated and potentially mutually reinforcing. Policies and programs would arguably benefit from a more unified intervention framework that links climate change adaptation and disaster preparedness.     

The occurrence of cyanobacteria and microcystins in the Hoan Kiem Lake and the Nui Coc reservoir (North Vietnam)

The occurrence of increasing blooms of toxic cyanobacteria in freshwaters has received much attention due to the ability of many cyanobacteria to produce potent cyanotoxins. In this paper, the occurrence of dominant cyanobacteria and the concentration of microcystins (MCs) analysis were investigated monthly from July 2008 to April 2009 in the Hoan Kiem Lake and from February to April 2009 in the Nui Coc reservoir. Concentrations of intracellular MCs from water, bloom samples, and isolated strains were quantified by using high performance liquid chromatography (HPLC). During the study period, the microscopic examination of the phytoplankton samples showed the dominance of the genus Microcystis in the water environment of the Hoan Kiem Lake and the Nui Coc reservoir. The toxin analysis by HPLC demonstrated the presence of two MC variants: MC-LR and MC-RR in water samples. Total concentrations of the toxins in filtered samples from surface water ranged from non-detected to 0.91 μg L^?1 at Nui Coc reservoir and they ranged from 2.1 to 46.0 μg L^?1 at Hoan Kiem Lake. The results of the HPLC analysis confirmed the production of MCs in bloom samples (ranged from 115.9 to 184.6 μg L^?1 in the Hoan Kiem Lake and from 726.5 to 1116 μg L^?1 in the Nui Coc reservoir) and isolated strains of Anabaena sp. and Microcystis with the concentration of MC ranging from 152 to 396.2 μg g^?1 dry mass, respectively.     

Toward exploratory analysis of diversity unified across fields of study: an information visualization approach

The study of the diversity of multivariate objects shares common characteristics and goals across disciplines, including ecology and organizational management. Nevertheless, subject-matter experts have adopted somewhat separate diversity concepts and analysis techniques, limiting the potential for sharing and comparing across disciplines. Moreover, while large and complex diversity data may benefit from exploratory data analysis, most of the existing techniques emphasize confirmatory analysis based on statistical metrics and models. This work aims to bridge these gaps. First, by cross comparing the analyses of species diversity, microbial diversity, and workgroup diversity, we introduce a framework of diversity concerns aligned across the three areas. The alignment framework is validated and refined by feedback from subject-matter experts. Then, guided by the framework and theoretical information visualization and visual analytics principles (as distinguished from scientific visualization), we propose a unified taxonomy of common analytical tasks for exploration of diversity.     

Speciation and Mobility of Selected Trace Metals (As, Cu, Mn, Pb and Zn) in Sediment with Depth in Cam River-Mouth, Haiphong, Vietnam

The speciation and mobility of some selected trace metals (As, Cu, Mn, Pb and Zn) in sediments with depth was investigated in the Cam River-mouth (Vietnam) by collecting sediment cores and analysing porewater and sediment composition, complemented with single (ammonium-EDTA) and sequential (BCR 3-step) extractions and mineralogical analysis (XRD). All trace metals show overall decreasing trends with depth in porewater as a result of anthropogenic input in upper sediment layers. High porewater concentrations of As, Mn and Pb in oxic and suboxic sediment layers may result in groundwater pollution. Sediment-bound Pb and Mn dominate in the reducible and the acid-soluble fraction, respectively, while Cu and Zn distribute rather evenly between four extracted fractions. The porewater metal speciation, as predicted by a geochemical model Visual MINTEQ version 3.0, indicates that the toxicity of Cu, Mn, Pb and Zn (presented by the proportions of free metal ions) decreases with depth, while the toxicity of As increases when As(III) becomes more abundant. The dissolved concentrations of trace metals are not only controlled by the precipitation/dissolution of discrete hydroxide/oxide, carbonate and phosphate minerals, but also by sorption processes on major sorbents (i.e. As on goethite, and Mn and Zn on calcite and dolomite). Sulphide minerals do not show any control even in the anoxic zone most likely because of the low concentration of sulphur.