High-resolution Bayesian mapping of landslide hazard with unobserved trigger event

Pesisir Selatan District is one of the areas in West Sumatra that often experiences natural disasters of landslides. Natural disasters of landslides that occur in this area often cause losses in the form of property or human lives and disruption of road access to other areas. The aims of this research were to analyze the hazard of landslides based on land characteristics, analyze the carrying capacity of the region based on landslide hazards, analyze social vulnerability, and analyze prediction of Population loss. The method used in this study is a survey method with a geomorphological approach. Data used in the form of primary data and secondary data. The technical analysis used is in the form of scoring to analyze landslide hazards and GIS used to analyze the carrying capacity of the region, vulnerability, and exposed populations. The results showed that the hazard of landslides in the study area was divided into three, namely low, medium, and high hazards. The hazard of low landslides is generally found in plain areas near the coast. This area is formed due to the results of sedimentation processes from both rivers and the sea. The hazard of moderate landslides is generally found in hilly areas, and the hazard of high landslides is generally found in hilly or mountainous areas. The hazard of landslides is caused by physical characteristics in the form of high rainfall, steep slopes to very steep slope, high rock weathering. Social vulnerability in the research area is divided into two parts, namely low and medium. Low social vulnerability is caused by the ratio of male population more than female population, ratio of vulnerable groups, ratio of poor Population, and a ratio of disabled Population. Social vulnerability is being caused by a greater number of female populations, ratios of vulnerable groups, and disabled Population. The prediction of population loss found in the Lakitan Tengah area is in Lengayang Subdistrict, which is 11,869 Population.

Comparison of Multivariate Poisson lognormal spatial and temporal crash models to identify hot spots of intersections based on crash types

BackgroundAlthough malaria decline has been observed in most sub-Saharan African countries, the disease still represents a significant public health burden in Tanzania. There are contradictions on the effect of ownership of at least one mosquito net at household on malaria mortality. This study presents a Bayesian modelling framework for the analysis of the effect of ownership of at least one mosquito net at household on malaria mortality with environmental factors as confounder variables.MethodsThe analysis used longitudinal data collected in Rufiji and Ifakara Health Demographic Surveillance System (HDSS) sites for the period of 1999–2011 and 2002–2012, respectively. Bayesian framework modelling approach using integrated nested laplace approximation (INLA) package in R software was used. The space time models were established to assess the effect of ownership of mosquito net on malaria mortality in 58 villages in the study area.ResultsThe results show that an increase of 10 % in ownership of mosquito nets at village level had an average of 5.2 % decrease inall age malaria deaths (IRR = 0.948, 95 % CI = 0.917, 0.977) in Rufiji HDSS and 12.1 % decrease in all age malaria deaths (IRR = 0.879, 95 % CI = 0.806, 0.959) in Ifakara HDSS. In children under 5 years, results show an average of 5.4 % decrease of malaria deaths (IRR = 0.946, 95 % CI = 0.909, 0.982) in Rufiji HDSS and 10 % decrease of malaria deaths (IRR = 0.899, 95 % CI = 0.816, 0.995) in Ifakara HDSS. Model comparison show that model with spatial and temporal random effects was the best fitting model compared to other models without spatial and temporal, and with spatial–temporal interaction effects.ConclusionThis modelling framework is appropriate and provides useful approaches to understanding the effect of mosquito nets for targeting malaria control intervention. Furthermore, ownership of mosquito nets at household showed a significant impact on malaria mortality.Electronic supplementary materialThe online version of this article (doi:10.1186/s12936-016-1311-9) contains supplementary material, which is available to authorized users.

Slope movements are among the most widespread and damaging natural hazards in Hungary and worldwide. In recent decades, the occurrence and impact of landslides and related mass movements have markedly increased, a trend commonly linked to ongoing climate change. This study presents a landslide hazard assessment of climate-sensitive slope processes affecting the Danube riverside built structures and houses at the Dunaszekcső high bank, Hungary (Central Europe), focusing on the loess bluff area, where several slope failures and erosional events have been documented in recent decades. The study area is located on the steep Danube-facing slopes of the settlement high bank, composed mainly of Pleistocene loess, loess-derived paleosols, and interbedded sandy and clayey sediments. These lithologies exhibit strong variability in cohesion, permeability, and moisture sensitivity and are covered by shallow soils, resulting in high susceptibility to surface erosion, earth slides, and loess collapses. Steep slopes, locally sparse vegetation, and unfavourable slope exposure further increase landslide hazard. The applied methodology integrates detailed field mapping, geomorphological and engineering geological analysis, and evaluation of long-term and event-based precipitation data. Special attention was given to the identification of active sliding areas and the trigger mechanism. The results indicate that both short, high-intensity convective storms and prolonged rainfall events can initiate landslides. Under current and projected climatic conditions, slope failures and sediment mobilisation are expected, highlighting the urgent need for integrated landslide risk mitigation strategies. These include continuous slope monitoring, rainfall-based early-warning systems, and targeted structural and non-structural protection measures. The paper benefited from the results of GeoNetSee project “An AI/IoT-based system of GEOsensor NETworks for real-time monitoring of unStablE tErrain and artificial structures”, which is financed through the Interreg Danube Region programme, contract DRP0200783.

BackgroundCardiovascular Diseases (CVDs) are health-threatening conditions that account for high mortality in the world. Approximately 23.6 million deaths due to CVD is expected in the year 2030 worldwide. The CVD burden is more severe in developing countries, including Tanzania.ObjectivesThis study analyzed the spatial-temporal trends and determinants of cardiovascular diseases in Tanzania from 2010 to 2019.MethodsIndividual data were extracted from Jakaya Kikwete Cardiac Institute (JKCI), Mbeya Zonal Referral Hospital (MZRH), Kilimanjaro Christian Medical Centre (KCMC) and Bugando hospitals and the geographical data from TMA. The model containing spatial and temporal components was analyzed using the Bayesian hierarchical method implemented using Integrated Nested Laplace Approximation (INLA).ResultsThe results found that the incidence of CVD increased from 2010 to 2014 and decreased from 2015 to 2019. The southern highlands, lake, central and coastal zones were more likely to have CVD problems than others. It was also revealed that people aged 60–64 years OR = 1.49, females OR = 1.51, smokers OR = 1.76, alcohol drinkers OR = 1.48, and overweight OR = 1.89 were more likely to have CVD problems. Additionally, a 1oC increase in the average annual air maximum temperature was related to a 14% risk of developing CVD problems. The study revealed that the model, which included spatial and temporal random effects, was the best-predicting model.ConclusionThe study shows a decreased CVD incidence rate from 2015 to 2019. The CVD incidences occurred more in Tanzania’s coastal and lake areas between 2010 and 2019. The demographic, lifestyle and geographical risk factors were significantly associated with the CVD.

IntroductionThere is an increasing interest in small area analyses in cancer surveillance; however, technical capacity is limited and accessible analytical approaches remain to be determined. This study demonstrates an accessible approach for small area cancer risk estimation using Bayesian hierarchical models and data visualization through the smallareamapp R package.Materials and methodsIncident lung (N = 26,448), female breast (N = 28,466), cervical (N = 1,478), and colorectal (N = 25,457) cancers diagnosed among British Columbia (BC) residents between 2011 and 2018 were obtained from the BC Cancer Registry. Indirect age-standardization was used to derive age-adjusted expected counts and standardized incidence ratios (SIRs) relative to provincial rates. Moran’s I was used to assess the strength and direction of spatial autocorrelation. A modified Besag, York and Mollie model (BYM2) was used for model incidence counts to calculate posterior median relative risks (RR) by Community Health Service Areas (CHSA; N = 218), adjusting for spatial dependencies. Integrated Nested Laplace Approximation (INLA) was used for Bayesian model implementation. Areas with exceedance probabilities (above a threshold RR = 1.1) greater or equal to 80% were considered to have an elevated risk. The posterior median and 95% credible intervals (CrI) for the spatially structured effect were reported. Predictive posterior checks were conducted through predictive integral transformation values and observed versus fitted values.ResultsThe proportion of variance in the RR explained by a spatial effect ranged from 4.4% (male colorectal) to 19.2% (female breast). Lung cancer showed the greatest number of CHSAs with elevated risk (Nwomen = 50/218, Nmen = 44/218), representing 2357 total excess cases. The largest lung cancer RRs were 1.67 (95% CrI = 1.06–2.50; exceedance probability = 96%; cases = 13) among women and 2.49 (95% CrI = 2.14–2.88; exceedance probability = 100%; cases = 174) among men. Areas with small population sizes and extreme SIRs were generally smoothed towards the null (RR = 1.0).DiscussionWe present a ready-to-use approach for small area cancer risk estimation and disease mapping using BYM2 and exceedance probabilities. We developed the smallareamapp R package, which provides a user-friendly interface through an R-Shiny application, for epidemiologists and surveillance experts to examine geographic variation in risk. These methods and tools can be used to estimate risk, generate hypotheses, and examine ecologic associations while adjusting for spatial dependency.

A common practice in the analysis of the spatial distribution of neglected tropical diseases is to assume that in silent areas (no reports) there are no cases. However, when the problem is underreporting, it risks reinforcing the neglect of areas that should be a priority. Instead of this assumption, one can predict the number of cases in silent areas using epidemiologic and spatial dependence information. The present study exemplifies this approach, using the integrated nested Laplace approximation in Bayesian spatial models that relate social vulnerability and the number of reported cases of feline (zoonotic) sporotrichosis in census tracts (CTs) of the municipality of Guarulhos. In addition to predictions for silent CTs, we assigned a priority index to all CTs. The results showed a more problematic epidemiologic situation, compared to the scenario in which it is assumed that there are no cases in silent CTs. To iteratively validate the index predictions and calibrate the degree of confidence assigned to the predictions, one can compare the distribution of the priority indices of silent CTs with the distribution of cases identified through active surveillance in a sample of silent CTs.

Chapter 3 - Morphometric conditions underpinning the spatial and temporal dynamics of landslide hazards on the volcanics of Mt. Elgon, Eastern Uganda

This paper studies the relationship between the student abilities in the second year of high school and the infrastructural endowment in all Italian municipalities, using spatial Bayesian modelling. Municipal student scores are obtained by averaging standardised and spatially homogeneous indicators of student outcomes provided by the INVALSI Institute for two subjects: Italian and mathematics. Given the nature of the data, we employ a multilevel regression model assuming a bivariate intrinsic conditionally autoregressive (ICAR) latent effect to explain the spatial variability and account for the correlation between the two subjects. Bayesian model estimation is obtained using the integrated nested Laplace approximation (INLA), implemented in the package. We find that along with a significant association with the current state of school infrastructure and facilities, spatially structured latent effects are still necessary to explain the different student outcomes across municipalities.

This paper adopts a Bayesian spatial random effect modelling approach to analyse the risk of domestic burglary in Cambridgeshire, England, at the census output area level (OA). The model, in the form of Binomial spatial logistic regression, integrates offence and offender based theories and takes into account unknown local risk factors (represented as unexplained spatial autocorrelation in the model). A score of ‘proximity to offenders’ was calibrated for each OA based on the number of likely offenders in the county, the OAs they reside, and their proximities. Our results indicate that areas that have a score higher than the average score were at higher risks of being burgled. Household occupied by non-couple and economically inactivity are positively associated confounders. Household occupied by owner is a negatively associated confounder. These confounders diminish the effect of high score of proximity to offenders, which, however, remains positively associated with the risk of burglary. Bayesian spatial random effect modelling, which adds to the traditional (non-spatial) regression model a spatial random effect term, stabilizes estimated risks and remarkably improves model fit and causation inference. Mapping the results of spatial random effect reveals locations of high risk of burglary after controlling for offender and socioeconomic factors. Limitations of the study and strategies to deter burglaries based on the results of spatial random effect modelling are discussed.

<p>Tropical mountainous regions are often identified as landslide hotspots with particularly vulnerable populations. Whereas both natural (e.g., rainfall, lithology) and anthropogenic (e.g., deforestation, mining) factors are expected to be involved in landslide occurrence, the relative importance of such factors remains poorly documented. The objective of this study is to understand the role played by natural and anthropogenic factors in the occurrence of landslides in the Rift Flank west of Lake Kivu in DR Congo. First, we inventoried 2856 landslides of various types using Google Earth imagery, high resolution topographic data, historical aerial photographs from1950’sand intense field surveys. We made a distinction between old and recent deep-seated landslides, shallow landslides and mining-related landslides. The difference in susceptibility patterns and in size distributions between old and recent deep-seated landslides indicates that natural factors contributing to their occurrence were either different or changed over time. The shallow landslides are all recent. The susceptibility analysis shows that their regional pattern is mainly controlled by forest dynamics and the presence of roads. The occurrence of shallow landslides in forest areas, i.e. in natural environments, strongly increases with slope gradient. In areas with similar topographic conditions where deforestation has occurred since the 1950’s, shallow landslides are more frequent, but of smaller size. This size reduction is attributed to the decrease of regolith cohesion due to forest loss, which allows a smaller minimum critical area for landsliding. In areas that were already deforested in 1950’s, shallow landslides are less frequent, larger, and occur on less steep slopes. This suggests a combined role between regolith availability and soil management practices that influence erosion and water infiltration. Landslides associated with mining activities are larger than shallow landslides but smaller than the recent deep-seated instabilities. They are not well predicted by the susceptibility models, showing that they respond to environmental factors that are not present under natural conditions. This research highlights the importance of human activities on the occurrence of landslides and the need to consider this context when studying hillslope instability patterns in regions under anthropic pressure.</p><p> </p><p>Keywords: landslide processes, multi-temporal inventory, historical deforestation, mining and quarrying, susceptibility assessment, Africa</p>

Abstract. We present the results of the application of a recently proposed model to determine landslide hazard. The model predicts where landslides will occur, how frequently they will occur, and how large they will be in a given area. For the Collazzone area, in the central Italian Apennines, we prepared a multi-temporal inventory map through the interpretation of multiple sets of aerial photographs taken between 1941 and 1997 and field surveys conducted in the period between 1998 and 2004. We then partitioned the 79 square kilometres study area into 894 slope units, and obtained the probability of spatial occurrence of landslides by discriminant analysis of thematic variables, including morphology, lithology, structure and land use. For each slope unit, we computed the expected landslide recurrence by dividing the total number of landslide events inventoried in the terrain unit by the time span of the investigated period. Assuming landslide recurrence was constant, and adopting a Poisson probability model, we determined the exceedance probability of having one or more landslides in each slope unit, for different periods. We obtained the probability of landslide size, a proxy for landslide magnitude, by analysing the frequency-area statistics of landslides, obtained from the multi-temporal inventory map. Lastly, assuming independence, we determined landslide hazard for each slope unit as the joint probability of landslide size, of landslide temporal occurrence, and of landslide spatial occurrence.

<p>A better detection of landslide occurrence is critical for disaster prevention and mitigation, and a standing pursuit owing to increasing and widespread impact of slope failures on human activities and natural environment in a changing world. However, the detection of rainfall-induced landslide is limited in some areas by data scarcity and method applicability. In this study, we proposed distributed rainfall thresholds within homogeneous slope units, by considering the interaction of landslide-influencing geo-environmental conditions and landslide-triggering rainfall variables. Homogeneous slope units are extracted based on detailed terrain analysis. Various landforms are identified and used to obtain slope units with homogeneous slope traits. The concept behind the distributed rainfall threshold models is that rainfall threshold for landslide occurrence varies with geo-environmental conditions such as slope gradient. Thus, a link can be established between landslide-influencing geo-environmental conditions and landslide-triggering rainfall variables. We used elevation, slope, plan and profile curvature, mean annual precipitation and temperature, soil texture and land cover as independent variables. Rainfall duration and cumulated rainfall of landslide-triggering rainfall events are automatically calculated and used, the former as one of independent variables, and the latter as the dependent variable. A support vector regression (SVR) and a multiple linear regression (MLR) method are used. The error and correlation coefficient measurement indicate a better performance of SVR method. Compared with grid units, the model scores high accuracy for slope units. The models are implemented at a regional scale (Guangdong, China). The SVR model in slope units ran with error of 0.16 mm and correlation coefficient of 0.93.</p>

<title>Abstract</title> Species distribution models (SDMs) are vital tools in ecology and conservation. The integration of increasingly available citizen science data with planned survey data offers a significant opportunity to enhance species distribution estimates. While integrated SDMs often combine presence-only and abundance data, the interdependence between the conditional distributions of these outcomes remains to be elucidated. This study proposes a Bayesian spatial fusion modelling framework to jointly analyse presence-only and abundance data for the African baobab in Benin. The aim was to understand and map the spatial variation in the species’ distribution. We briefly reviewed process-based models for count and point process data and explored various data fusion strategies using Integrated Nested Laplace Approximations (INLA) and Stochastic Partial Differential Equations (SPDE) for inference. The results revealed a heterogeneous baobab distribution across Benin, characterised by a spatial autocorrelation range of 34.4 km (95% Bayesian credible interval, BCI = 27.59-42.52). Key drivers of this distribution include environmental factors such as annual temperature, rainfall of the driest month, soil texture (silt/clay fractions), and slope. A spatial fusion model incorporating shared latent components and common covariates' effects demonstrated the highest performance level, surpassing alternative fusion approaches. The model achieved the highest mean composite scores for the Area Under the ROC Curve (AUC) (0.85±0.02), accuracy (0.77±0.02), and True Skill Statistics (TSS) (0.66±0.05). The shared component model has the capacity to explain datasets' interdependence, estimate covariate effects missed by separate models, and enhance prediction precision. Despite relying on the assumption of an identically shared spatial signal across target responses, this research underscores the potential of spatial fusion modelling for integrating disparate data sources. The findings contribute to advancing SDM inference, particularly in data-limited contexts, and have wider applicability to spatial regression problems involving multisource outcomes.

BackgroundThe population-based mammography screening program (MSP) was implemented by the end of 2005 in Germany, and all women between 50 and 69 years are actively invited to a free biennial screening examination. However, despite the expected benefits, the overall participation rates range only between 50 and 55 %. There is also increasing evidence that belonging to a vulnerable population, such as ethnic minorities or low income groups, is associated with a decreased likelihood of participating in screening programs. This study aimed to analyze in more detail the intra-urban variation of MSP uptake at the neighborhood level (i.e. statistical districts) for the city of Dortmund in northwest Germany and to identify demographic and socioeconomic risk factors that contribute to non-response to screening invitations.MethodsThe numbers of participants by statistical district were aggregated over the three periods 2007/2008, 2009/2010, and 2011/2012. Participation rates were calculated as numbers of participants per female resident population averaged over each 2-year period. Bayesian hierarchical spatial models extended with a temporal and spatio-temporal interaction effect were used to analyze the participation rates applying integrated nested Laplace approximations (INLA). The model included explanatory covariates taken from the atlas of social structure of Dortmund.ResultsGenerally, participation rates rose for all districts over the time periods. However, participation was persistently lowest in the inner city of Dortmund. Multivariable regression analysis showed that migrant status and long-term unemployment were associated with significant increases of non-attendance in the MSP.ConclusionLow income groups and immigrant populations are clustered in the inner city of Dortmund and the observed spatial pattern of persistently low participation in the city center is likely linked to the underlying socioeconomic gradient. This corresponds with the findings of the ecological regression analysis manifesting socioeconomically deprived neighborhoods as risk factors for low attendance in the MSP. Spatio-temporal surveillance of participation in cancer screening programs may be used to identify spatial inequalities in screening uptake and plan spatially focused interventions.