Bivariate Approaches Research Articles

Application of bivariate approaches for flood susceptibility mapping: a district level study in Eastern India

The main objective of this present study is to prepare flood susceptibility mapping using bivariate approaches considering parameters such as rainfall, geomorphology, geology, soil, elevation, ground slope, curvature of the land, flow directions, stream density, distance from road, distance from river, land use land cover, topographic wetness index, normalized differentiate vegetation index and water index. Bivariate methods like Shannon's Entropy Index (SEI) and Weight of Evidence (WofE-IV) models have been used in conjunction with remote sensing and geographical information system and it was found that 17.77% (SEI) and 11.20% (WofE-IV) area of the district categorized as highly susceptible to flood especially found in the adjoining area of Tangon, Punarbhaba, and Atreyee rivers. Around 26–35% of the total area is safe from flood especially located in the Tapan and Banshihari block. The area under curve values (SEI = 0.954, WofE-IV = 0.913) shows that both models are performing well for flood susceptibility mapping.

Predictors Of Length Of Care For Covid-19 Patients Based On Patient Characteristics At The North Aceh Cut Mutia Hospital In 2020-2021

According to the data released by the Indonesian Ministry of Health on November 22, 2021, Indonesia has confirmed that COVID-19 cases have reached 4,253,598 cases and 143,744 deaths, occupying the first position in the highest cases among countries in ASEAN region. One of the areas confirmed by COVID-19 is Aceh. The purpose of this study was to find out predictors of the length of care for covid-19 patients based on the characteristics of COVID-19 patients at Cut Mutia Hospital, North Aceh. This type of research is cross-sectional with secondary data. The population was all patients treated for COVID-19, both recovered and died in the COVID-19 treatment room at Cut Mutia Hospital, North Aceh, in 2020-2021, totaling 467 people with a sampling technique using proportions with a total of 200 respondents. The data used were secondary data. The data were analyzed using bivariate and multivariate approaches. The research found that there were no mean differences of length of stay between Covid-19 patients with low and higher education (p value= 0.39), with clinical symptoms and without clinical symptoms ( p value= 0.686), with comorbidities and without comorbidities (p value = 0.006), female and male ( p value= 0.038). The results obtained that the predicting variables of length of stay of the Covid-19 patients at Cut Meutia Hospital, North Aceh in 2020-2021, were gender (p value= 0.038, p< 0.05) and comorbidities (p value= 0.006, p< 0.05). It is suggested that Cut Mutia Hospital could conduct more intensive counseling and health promotion to achieve shorter length of care.

Frequency dependence between oil futures and international stock markets and the role of gold, bonds, and uncertainty indices: Evidence from partial and multivariate wavelet approaches

This paper examines the frequency dynamic co-movements between crude oil prices and stock market returns of three developed economies (Canada, Japan, and the USA) and the emerging BRICS (Brazil, Russia, India, China, and South Africa) economies by considering four global factors (U.S. treasury bills, S&P volatility index, gold price, and U.S. EPU index). Using bivariate and multivariate wavelet approaches, the results show evidence of time-frequency co-movements between the considered markets at medium and low frequencies. Besides, the results reveal that the co-movement is intensified during global financial crisis and COVID-19 pandemic periods, confirming recoupling hypothesis. The risk analysis reveals dependence and persistence of co-movements, and aggravation of portfolio risk in the BRICS economies and across markets during bouts of afflictions. These findings should encourage the relevant national and transnational policy makers to consider these co-movements which vary over time and in duration when setting up regulations that deem to enhance the market efficiency.

Influence of the spatial configuration of available data on hydraulic conductivity estimates for a geostatistical-Kalman filter method

Abstract Accurate hydraulic conductivity estimates are vital for groundwater evaluation. Usually, interpolations of hydraulic conductivity data are needed to obtain spatial estimates over larger areas, but the results present a high uncertainty which can be reduced by adding a secondary variable in the estimation. In this paper, the influence of the number and spatial configuration of hydraulic conductivity (K) and hydraulic head (HH) data on the estimation of K is evaluated using univariate and bivariate geostatistical-Kalman filter approaches (similar to kriging and cokriging, respectively). A synthetic case based on a transient groundwater flow model is used to generate different numbers, spatial arrays, and data. With these data, variogram models for the univariate and bivariate cases were fitted and used to calculate the corresponding covariance matrices for the Kalman filter. The results show that K estimates are more reliable when HH data is added than when only K is used, independently of the number and distribution of the data, since there is a better agreement between the calculated errors and estimate error variances. HH data provides valuable information only where K is not sampled. This evaluation could support the design of optimal sampling strategies to obtain reliable K estimates.

Frequency spillovers and portfolio risk implications between Sukuk, Islamic stock and emerging stock markets

This study examines the multiscale spillovers between five important emerging stock markets namely, Brazil, Russia, India, China, and South Africa (BRICS) and both Dow Jones Islamic stock market index (DJIM) and Dow Jones Sukuk index (DJ Sukuk) using bivariate and multivariate wavelet approaches. The results show evidence of strong time-scale co-movements between conventional and Islamic stock markets at different frequencies. Moreover, the pure contagion is evident at the short-term whereas fundamental contagion appears in the long run. Sukuk show relatively less integration with the conventional stock markets of BRICS at high and medium frequencies. DJIM and DJ Sukuk provide risk diversification opportunities for BRICS stock market volatility. The diversification benefits in terms of lower portfolio VaR is available in the high frequency scale or at the short-term. In long-run, the Islamic equity market become fundamentally integrated with the BRICS stock markets, thus, reduces hedging gains.

Comovement and instability in cryptocurrency markets

We analyze the extent of comovement between daily price returns of nine major cryptocurrencies from April 2013 to February 2021. We assess its evolution using bivariate and multivariate modeling approaches, and detect pronounced time variation. Generally, comovement is initially low and positive, but increases substantially between early 2017 and late 2018 and stays high until the end of the sample. We then adopt a right-tail version of the Augmented Dickey–Fuller unit root test to identify periods of mildly explosive behavior (statistical instability) in the Network Value to Transactions (NVT) ratio of seven cryptocurrencies. The NVT ratio is a novel measure of the dollar value of cryptocurrency transaction activity relative to network value. We show evidence of significant mild explosiveness in all seven cryptocurrencies. In 2018 and in the second half of 2019, several cryptocurrencies experience often simultaneous instability. Instability is a dominant feature of these markets.

WHERE ARE WE IN SPACE: EXAMINING TEACHERS’ ICT COMPETENCY IN TEACHING LEARNERS IN SPECIAL SCHOOLS IN KENYA

<p>Information Communication Technology (ICT) plays a central role in facilitating education in the 21st century. For successful integration of ICT in teaching and learning teachers play a key role. As such, there is a need for competent teachers in applying ICT skills to teach and facilitate learning. This study sought to examine teachers’ ICT competency in implementing Digital Literacy Programme (DLP) in public primary special schools in Nairobi County ii Kenya. The study used a semi-structured questionnaire tool to collect data from 99 teachers and 41 head teachers from public primary special schools in Nairobi County. Mixed-Methods Research (MMR) where both Statistical Methods and thematic analysis approaches were used to analyse the data. Statistical Methods used include descriptive analysis such as frequencies and percentages and bivariate analysis approaches including Chi-square tests of hypothesis regarding the relationship between teacher’s ICT competency and implementation of DLP. The study found that the role played by DLP in building teacher ICT skill capacity is 28.4% lower in special schools compared to regular schools. As a result, teachers of children with disabilities and special educational needs have limited relevant ICT skills such as assistive technology to support teaching and learning. The study also found that whilst there may be a significant proportion of teachers skilled with ICT skills, lack of access to digital devices and internet connectivity makes it difficult for them to demonstrate their competency and apply the same to the teaching of children with disabilities. Based on the results of this study, recommendations for policy enhancement and improvement of practice have been made.</p><p> </p><p><strong> Article visualizations:</strong></p><p><img src="/-counters-/edu_01/0978/a.php" alt="Hit counter" /></p>

Equally green? Understanding the distribution of urban green infrastructure across student demographics in four public school districts in North Carolina, USA

Green infrastructure (GI) provides a suite of ecosystem services that are widely recognized as critical to health, well-being, and sustainability on an urbanizing planet. However, the distribution of GI across urban landscapes is frequently uneven, resulting in unequal delivery of these services to low-income residents or those belonging to underserved racial/ethnic identities. While GI distribution has been identified as unequal across municipalities, we investigated whether this was true in public schoolyards within and among urban school districts. We examined schoolyards in four metropolitan areas of diverse socio-economic and demographic compositions in North Carolina, USA to determine if they provided equal exposure to GI, then compared whether this was true of the broader urban landscape. We first classified the land cover of elementary schoolyards and their neighborhoods, then used bivariate and multivariate approaches to analyze the relationships between GI (i.e. tree canopy cover and total GI) and the socioeconomic status and race/ethnicity of the schools and surrounding neighborhoods, respectively. We found that the extent of tree canopy cover and total GI in schoolyards was unrelated to the socioeconomic status and the race/ethnicity of students across the four school districts. In contrast, neighborhoods with lower socioeconomic status and larger populations of underserved race/ethnicity residents had less tree canopy cover and total GI. Although total GI was more evenly distributed in schoolyards, the extent of tree canopy cover and total GI in schoolyards was lower than that in the neighborhoods. This suggests opportunities for school districts to expand GI in schoolyards, leveraging their potential to increase ecosystem services to all children, from increased educational opportunities to improved mental, physical, and environmental well-being.

Electric motor scooter sharing: the potential of ESS in regard to sustainable mobility and city development and the derivation of policy recommendations

Electric motor scooter sharing (ESS) services have been popping up in numerous European cities throughout the last few years. Their marketing targets a young demographic, the sharing service is advertised as both fun and sustainable. Electric motor scooters can indeed be a sustainable mode of transport, for instance, when they contribute to (locally) emission-free transportation depending on the underlying electricity mix (see Weiss et al. 2012). And as we know from former studies, the availability of a city wide sharing system and its usage can decrease the will to buy a private vehicle (see the study from Firnkorn and Müller 2015). Thus, sharing systems got the potential to support multimodal mobility behavior. Besides, when shared, ESS provide relatively space efficient mobility. To evaluate the potential as a sustainable mobility solution, the analysis of mode shift effects created by such a service are a key aspect. Do electric scooter trips mostly replace trips formerly taken by cars with internal combustion engines (ICE)? Or do they replace other trips and possibly create additional traffic? This leads to the main research questions of the paper:•    How does ESS contribute to sustainable urban transportation?•    Which policy regulations or governance solutions are needed to further strengenth the advantages and counterbalance the weaknesses of such a sharing service?The paper does focus on an ESS service run within Munich/Germany. It is based on three types of data: Firstly, automatically collected data concerning customers’ renting behavior allows conclusions on the average customer trip numbers, frequency and lenghts (1). Secondly, a customer survey of the same ESS provides insights on the customers’ trip patterns and mode shift behavior. Besides, it highlights their motivation to use shared electric motor scooters. Via descriptive statistics, the datasets is analyzed using uni- and bivariate approaches. The results are completed by the third type of data consisting of interviews with ESS users on user experiences. (2)The paper features the main results of this study, addressing both ESS strenghts and challenges. On the one hand, it becomes clear to which extent ESS contributes to improving urban mobility, e. g. by replacing trips formerly taken by ICE vehicles and by furthering multimodality (like electrically assisted micromobility options). Furthermore it shows how ESS can improve the image of electric mobility. On the other hand, the paper addresses challenging mode shifts: Which role does the replacement of trips formerly taken by public transport or non-motorized modes of transportation play? And how relevant are trips taken to simply cruise around, thus creating traffic that would have not existed without ESS?Based on the results of the survey and in reflection with city planners, policy recommendations were derivated. Which approaches, instruments or frameworks have to be developed to optimize the sustainable potential of ESS and to minimize risks? The paper concludes with a set of recommendations and push and pull categories in different fields of action such as infrastructural planning, financial instruments, marketing and communication.(1) The data includes almost 490,000 trips.(2) The data was collected as part of the CIVITAS ECCENTRIC research project.The qualitative results refer to the users and their motivations, whereas the analysis of the quantitative data set focuses on substitution effects, the sustainability of ESS on the mobilities system with the aim to make overarching policy recommendations. https://civitas.eu/about The qualitative data was analysed and published in September 2021. https://journals.qucosa.de/jmv/article/view/71

Suicidal Behaviors in a Nationally Representative Sample of School-Going Adolescents Aged 12\u201317 Years in Eswatini

The burgeoning body of evidence suggests that the aggregated 12-month prevalence estimates of suicidal behaviors (ideation, planning, and attempt) are relatively higher among in-school adolescents in Africa, although country-specific data on the phenomena are still inadequate from the continent. We sought to estimate the 12-month prevalence and identify some of the correlates of suicide behaviors among school-going adolescents in Eswatini. We analyzed the nationally representative data from the 2013 Eswatini World Health Organization Global School-based Student Health Survey, using univariate, bivariate, and multivariate statistical approaches. Of the 2,513 analytical samples, 17.0% (95% confidence interval [CI]: 15.4–18.4%) reported suicidal ideation, 21.0% (95% CI: 19.3–22.6%) made a suicide plan, and 15.5% (95% CI: 14.1–16.9%) attempted suicide during the previous 12 months. The final adjusted logistic models indicated health risk behavior (i.e., marijuana use), adverse interpersonal factors within the family (e.g., intrusion of privacy by parents), and school contextual factors (e.g., bullying victimization, physical fights) to be associated with increased odds of suicidal behaviors among females. However, predominantly, school-related interpersonal factors (e.g., bullying victimization, physical fights) showed strong associations with increased odds of suicidal behaviors among males. Generally, parental monitoring, parental understanding, and social support at school were associated with reduced odds of suicidal behaviors. The multi-level nature of our findings underscores the need for multi-contextual and multi-sectoral intervention and prevention programs and policy approaches targeted at mitigating the onset of suicidal ideation and possible transition to suicidal planning, attempt, and potential death by suicide in this young population.

Assessing stress variations in children during the strange situation procedure: comparison of three widely used respiratory sinus arrhythmia estimation methods

Objective. The respiratory sinus arrhythmia (RSA) is a well-known marker of vagal activity that can be exploited to measure stress changes. RSA is usually estimated from heart rate variability (HRV). This study aims to compare the RSA obtained with three widely adopted methods showing their strengths and potential pitfalls. Approach. The three methods are tested on 69 healthy preschoolers undergoing a stressful protocol, the strange situation procedure (SSP). We compare the RSA estimated by the Porges method, the univariate autoregressive (AR) spectral analysis of the HRV signal, and the bivariate AR spectral analysis of HRV and respirogram signals. We examine RSA differences detected across the SSP episodes and correlation between the estimates provided by each method. Main results. The Porges and the bivariate AR approaches both detected significant differences (i.e. stress variations) in the RSA measured across the SSP. However, the latter method showed higher sensitivity to stress changes induced by the procedure, with the mean RSA variation between baseline and first separation from the mother (the most stressful condition) being significantly different among methods: Porges, −17.5%; univariate AR, −18.3%; bivariate AR, −23.7%. Moreover, the performances of the Porges algorithm were found strictly dependent on the applied preprocessing. Significance. Our findings confirm the bivariate AR analysis of the HRV and respiratory signals as a robust stress assessment tool that does not require any population-specific preprocessing of the signals and warn about using RSA estimates that neglect breath information in more natural experiments, such as those involving children, in which respiratory frequency changes are extremely likely.

Evaluation of spatial probability of landslides using bivariate and multivariate approaches in the Goriganga valley, Kumaun Himalaya, India

Evaluation of spatial probability of landslides using bivariate and multivariate approaches in the Goriganga valley, Kumaun Himalaya, India

Multivariate approach in analyzing medical data with correlated multiple outcomes: An exploration using ACCORD trial data

BackgroundIn clinical setting, to answer a research question, very often more than one outcome variables are used which are statistically correlated. Univariate analysis approach, which is commonly used in such context, violates the assumption of independence for correlated variables, while multivariate approach could give more robust and precise clinical decision by accounting this correlation. This paper is a demonstration of the change in statistical decision in multivariate approach compared to univariate approach in analysing medical data with multiple correlated outcome variables. Methodology: The data used in this paper was the ACCORD trial data and the variables were Systolic blood pressure (SBP) and Diastolic blood pressure (DBP) which are correlated. The condition of bivariate normality was checked after removing the outliers. We compared the differences in means of SBP and DBP separately between the groups at different follow up time points by univariate approach using unpaired ‘t’ test and thereafter together by bivariate approach using Hotelling ‘T2’. Further the role of varied levels of correlation between SBP and DBP coupled with the variations in sample size in changing the statistical inference has been investigated. Robustness of these results were demonstrated through 1000 samples of simulated data from Accord trial data. ResultsThe statistical decision regarding difference in means of either SBP or DBP between the two groups at baseline, 12th, 36th and 84th month in both univariate and bivariate approaches were similar. At baseline and 84th month the hypotheses were accepted and at 12th and 36th month the hypotheses were rejected with both the approaches. However at 60th and 72nd month the univariate and multivariate approach results contradicted each other which reflect the significance of correlation in modifying the inference. The analysis results further indicated that varied levels of correlation and sample size too modify the inference in bivariate approach and contradict univariate results. ConclusionStudies with multiple correlated outcomes, the multivariate approach which accounts this statistical correlation has potential to change statistical decisions and provide more precise results. Hence multivariate approach is recommended in such situations.

Traceability of the Norway Lobster Nephrops norvegicus in UK Shelf Seas: A Stable Isotope Approach

Ensuring sustainability of the world fisheries is a key conservation and economic objective. Traceability of seafood from the final sale back to the point of harvest is an important aspect, supporting both fishery management and consumer protection. Stable isotope–based geolocation can be applied to trace the spatial origin of seafood, drawing on comparisons between the isotopic compositions of the product and those of a reference dataset from known spatial locations. This study tests the extent to which stable isotope–based geolocation can be applied to identify catch location of the Norwegian lobster Nephrops norvegicus. Carbon, nitrogen, and sulfur isoscapes across UK shelf seas are used as the reference dataset and test the accuracy of assignment estimates using a variety of bivariate and multivariate stable isotope geolocation approaches. Two alternative Bayesian inversions, one balanced and one weighted, are applied to the outcomes of the statistical models to determine the most accurate methods of assignment. Of all the methods trialed, the multivariate approach using carbon, nitrogen, and sulfur isoscape data produced the most accurate assignments, with c. 60% of samples from each site correctly assigned among six possible fishery origins. Weighted Bayesian approaches resulted in more correct assignments to highly fished sites, but at a cost of reduced correct assignments to sites of low fishing activity. Processed Nephrops samples obtained from supermarkets were assigned to potential fishery location, with results indicating the majority were captured in the west of Scotland. The isoscape methods explored can be calibrated to any marine feeding organism and provide a useful tool for more efficient management of marine stocks.

Landslide hazard, vulnerability, and risk assessment (HVRA), Mussoorie township, lesser himalaya, India

In the present study, landslide hazard, vulnerability and the risk assessment of the Himalayan township Mussoorie, located in the lower part of the Lesser Himalaya has been undertaken. The area famous for tourism constitutes > 7000 buildings with ~ 30,000 habitations. Four bivariate statistical approaches viz. Weight of evidence (WoE), Frequency ratio (FR), Yule Coefficient (YC), and Information Value (InV) were used for landslide hazard assessment, and all these approaches with 75%—80% success and predication rate are found to be acceptable for landslide hazard mapping. The analyses indicate that the Nagar Palika Parisad ward, Library ward, Happy valley ward and Bhadraj ward exhibit the maximum area falling under the high and very high landslide hazard zones, whereas Landaur ward, Jalkii ward and Indra Colony ward exhibit a greater part of the area in the low and very low hazard zones. On the basis of the six elements at risk present in the study area viz. settlement, crop land, water body, roads, barren and degraded land, and dense forest, the ~ 40% of the area is very high and high vulnerable, ~ 11% moderate, and ~ 49% is low and very low vulnerable. High and very high vulnerable zones are located in the west and central portion mainly because of higher habitation and more anthropogenic activities. Finally, risk map prepared intersecting hazard and vulnerability maps exhibits that Nagar Palika Parisad ward has the largest part (~ 41%) of the area with 37% buildings falling in the high and very high risk zones, whereas the Landour ward is the safest ward having ~ 83% area in the low and very low risk. Overall ~ 1604 buildings with habitation of ~ 8,000 persons in the township, are prone to high and very high landslide risk. The results of this study may be utilized for further development and planning in the area.

Suicidal behaviours among deaf adolescents in Ghana: a cross-sectional study.

BackgroundA growing global concern is that suicide research has paid little attention to young people with disabilities, particularly, in low- and middle-income countries (LAMICs). We aimed to estimate the 12-month prevalence of suicidal ideation and attempt and describe some associations among deaf adolescents in Ghana.MethodsThis is a cross-sectional anonymous self-report survey involving a nationally representative random sample of 450 school-going deaf adolescents. Data analysis included bivariate and multivariable approaches.ResultsThe overall 12-month prevalence of suicidal ideation was 19·3% (95% confidence interval [CI] = 15·8–23·3) and suicidal attempt was 15·6% (95% CI = 12·3–19·2). Although alcohol use and parental divorce were strongly associated with increased odds of both suicidal ideation and attempt, high subjective mental well-being was associated with reduced odds of both suicidal ideation and attempt. Living with no parents and being a final year student were associated with suicidal ideation, while male gender was associated with suicidal attempt.ConclusionsThe prevalence of suicidal behaviours among school-going deaf adolescents in this study compares with estimates among in-school non-deaf adolescents in Ghana and other LAMICs in Africa, and also highlights the need for prevention efforts against the onset of suicidal ideation and possible transition to attempt and suicide among deaf adolescents.

Multivariate analysis of teachers’ digital information skills - The importance of available resources

Digital information skills are prerequisites for success in the information society. Previous research on 21st century skills has shown that digital skills evolve sequentially, building on digital information skills, making them the key to future skills that teachers should acquire and be able to pass on to future generations. In the research on digital skills, shortcomings of bivariate approaches have been widely identified and interest has largely shifted to multivariate methods. However, the development of effective targeted interventions still requires a deeper understanding of the social resources of individuals and the interdependencies of predictors. This study examined the extent to which sociodemographic factors and available resources predict digital information skills and which subgroups emerge as potential targets for interventions, by analysing the interdependence of the predictor variables. The empirical data used in the study consisted of 4988 Finnish teachers' questionnaire responses and performance test results. Utilising the multiple regression analysis, the study found that digital activity and age explain most of the variation in teachers' digital information skills. Their effect was found to be the inverse of each other, with abundant digital activity increasing and age reducing teachers’ mastery of these skills. Digital self-efficacy and in-service training also emerged as promising predictors, highlighting that teachers' information skills are explained more strongly by the available resources than by sociodemographic factors. At the end, the practical significance and recommendations for more targeted interventions as well as for further research are considered on the basis of the results obtained.

Inferring network properties from time series using transfer entropy and mutual information: Validation of multivariate versus bivariate approaches

Functional and effective networks inferred from time series are at the core of network neuroscience. Interpreting properties of these networks requires inferred network models to reflect key underlying structural features. However, even a few spurious links can severely distort network measures, posing a challenge for functional connectomes. We study the extent to which micro- and macroscopic properties of underlying networks can be inferred by algorithms based on mutual information and bivariate/multivariate transfer entropy. The validation is performed on two macaque connectomes and on synthetic networks with various topologies (regular lattice, small-world, random, scale-free, modular). Simulations are based on a neural mass model and on autoregressive dynamics (employing Gaussian estimators for direct comparison to functional connectivity and Granger causality). We find that multivariate transfer entropy captures key properties of all network structures for longer time series. Bivariate methods can achieve higher recall (sensitivity) for shorter time series but are unable to control false positives (lower specificity) as available data increases. This leads to overestimated clustering, small-world, and rich-club coefficients, underestimated shortest path lengths and hub centrality, and fattened degree distribution tails. Caution should therefore be used when interpreting network properties of functional connectomes obtained via correlation or pairwise statistical dependence measures, rather than more holistic (yet data-hungry) multivariate models.

Safety analysis of unsignalized intersections: a bivariate extreme value approach

ABSTRACT Application of extreme value theory (EVT) to road safety analysis is gaining interest, thanks to its ability to produce quick and reliable safety evaluations without the use of crash data. Traditionally applied to single collision types and single extreme variables (i.e. surrogate measures of safety), EVT can be further exploited to simultaneously model multiple collision types, with the use of multiple extreme variables. In this paper two bivariate EVT approaches are applied for the safety evaluation of a three-leg unsignalized intersection, considering: (i) two conflict points and a single surrogate measure of safety; (ii) two surrogate measures of safety collected in a single conflict point. Each bivariate analysis was applied with two EVT methods: Component-wise Maxima (CM) and Excesses Over a Threshold (EOT). Bivariate models produced good results, especially with the EOT method, and were able to significantly improve the univariate benchmark results when the two estimation datasets were correlated.

Landslide susceptibility modelling using integrated evidential belief function based logistic regression method: A study from Southern Western Ghats, India

The present study aims to develop a spatially integrated evidential belief function based logistic regression model (EBF-LR) for landslide susceptibility mapping in a naturally sloping terrain of Southern Western Ghats in Kerala, India. For this, a landslide inventory map of 83 previous landslides was prepared using satellite imageries and verified in field check. Thereafter the landslide inventory was randomly divided into 70%–30% basis for model training and testing. Twelve landslide conditioning factors viz., lithology, land use/land cover, NDVI, slope angle, slope aspect, profile curvature, distance to stream, distance to roads, distance to lineaments, soil texture, topographic wetness index and average annual rainfall were considered for landslide susceptibility modelling. The resultant susceptibility maps were validated using receiver operating characteristics curve (ROC) with area under the curve (AUC) value, sensitivity, specificity, kappa index, mean absolute error (MAE) and root mean square error (RMSE). Analysis shows that the integrated EBF-LR model outweigh other conventional bivariate and multivariate approaches with a ROC-AUC value of 0.935, Kappa score 0.719, sensitivity 0.885, specificity 0.833 and least RMSE of 0.456 in the validation stage. The study also reveals that anthropogenic disturbances have a significant role on landslide initiation in the study area. Considering the ROC- AUC, MAE, RMSE and other validation measures, the accuracy of the proposed landslide susceptibility model is satisfactory and can be used for future land use planning and landslide mitigation in the study area.