Wildlife-vehicle collisions: The disproportionate risk of injury faced by motorcyclists.

We applied a generalized linear mixed-effects model to determine the factors leading to injuries from wildlife-vehicle collisions (WVCs). We used the Police database representing WVCs which took place on the Czech road network between 2009 and 2022. The majority of WVCs in Czechia are with roe deer, followed by wild boar, i.e., both relatively small ungulates. Less than 2 % of these encounters ends with an injury to the motor vehicle occupants. We found that the probability of sustaining injury was systematically higher for motorcyclists than for car occupants. The odds of sustaining an injury during WVC were roughly 1600 times higher for motorcyclists than for car occupants. When applying an evading manoeuvre, the odds of sustaining an injury were approximately 68 times higher for car occupants while only 2.3 times higher for motorcyclists compared to a direct hit to an animal. The lack of helmets (for motorcyclists) and missing seat belts (for car occupants) were additional factors which made the outcomes worse for WVCs. While the acceptance of a direct hit (preceded by braking) seems to be a reasonable strategy for car drivers, WVC awareness (including maintaining a lower speed during critical times and places) should be raised among motorcyclists as both manoeuvres are almost comparably dangerous for them.

An effect of canopy bridges on monkey-vehicle collision hotspots: Spatial and spatiotemporal analyses.

Almost one-quarter of primate species are reported to be involved in vehicle collisions. To mitigate these collisions, canopy bridges are used though their effectiveness is not broadly substantiated. We studied bridge impact on 23 years of vehicle collisions (2000-2022: N = 765) with colobus (Colobus angolensis palliatus), Sykes' (Cercopithecus mitis albogularis), and vervet (Chlorocebus pygerythrus hilgerti) monkeys in Diani, Kenya. Along a 9 km road, collisions did not decrease over the study duration, although bridges increased from 8 to 30. Using the kernel density estimation plus (KDE+) method, collisions appeared highly concentrated at some locations. These concentrations, called hotspots, represent hazardous road segments, though the hotspots for all three species overlapped for only 3% of the road length. We then inspected the collision hotspots over time, using the spatiotemporal extension of the KDE+ method. We compared hotspot presence in the 3 years before and after bridge installation to determine if bridges mitigated these hotspots. Hotspots disappeared for ~60% of bridges postinstallation, suggesting that bridges effectively reduce some collisions. However, of the bridges installed in locations that were not hotspots, 13% had hotspots emerge. Surprisingly, regardless of preinstallation hotspot occurrence, almost one-fifth of bridges had postinstallation hotspots. To understand the extent to which bridges mitigate collisions, other factors need consideration, including species attributes and crossing behavior, and road features and vehicle volume. We used the novel analytical method because it best suited our data set, given the challenges of determining the bridge impact zone and the low collision frequency.

COVID-19 related travel restrictions prevented numerous wildlife deaths on roads: A comparative analysis of results from 11 countries.

Millions of wild animals are killed annually on roads worldwide. During spring 2020, the volume of road traffic was reduced globally as a consequence of the COVID-19 pandemic. We gathered data on wildlife-vehicle collisions (WVC) from Czechia, Estonia, Finland, Hungary, Israel, Norway, Slovenia, Spain, Sweden, and for Scotland and England within the United Kingdom. In all studied countries WVC statistics tend to be dominated by large mammals (various deer species and wild boar), while information on smaller mammals as well as birds are less well recorded. The expected number of WVC for 2020 was predicted on the basis of 2015-2019 WVC time series representing expected WVC numbers under normal traffic conditions. Then, the forecasted and reported WVC data were compared. The results indicate varying levels of WVC decrease between countries during the COVID-19 related traffic flow reduction (CRTR). While no significant change was determined in Sweden, where the state-wide response to COVID-19 was the least intensive, a decrease as marked as 37.4% was identified in Estonia. The greatest WVC decrease, more than 40%, was determined during the first weeks of CRTR for Estonia, Spain, Israel, and Czechia. Measures taken during spring 2020 allowed the survival of large numbers of wild animals which would have been killed under normal traffic conditions. The significant effects of even just a few weeks of reduced traffic, help to highlight the negative impacts of roads on wildlife mortality and the need to boost global efforts of wildlife conservation, including systematic gathering of roadkill data.

The effect of wildlife carcass underreporting on KDE+ hotspots identification and importance.

Many approaches have been developed in order to mitigate wildlife-vehicle collisions (WVC), their causes and consequences. Reliable data on the amount and location of killed animals along roads are therefore necessary. The existing WVC databases are usually, however, far from complete. This data underreporting causes problems when identifying the riskiest places along a transportation infrastructure. WVC data underreporting can distort the results of WVC hotspots determination. In this work, we simulated WVC hotspots identification and stability under various rates of WVC data underreporting. Our aim was to investigate whether WVC hotspots can be found at the original locations even when data are strongly underreported. We applied the KDE + method for WVC hotspots identification. The KDE + method also allows for hotspots ranking according to cluster strength and collective risk. These two measures were then used for detection of diminishing hotspot signals with a rising level of underreporting. We found that WVC hotspots with a greater cluster strength suffered less from underreporting whereas hotspots will lower values of both cluster strength and collective risk were not detected when underreporting in the data increased. Hotspots with a cluster strength above 0.5 were almost always detected when data underreporting remained below 50%. More than 50% of these hotspots (with cluster strength above 0.5) were detectable even when underreporting rate was between 50 and 80%. We further studied the effects of both spatial and temporal underreporting. Whereas temporal change of underreporting was not a problem in hotspots detection, spatial underreporting introduced significant errors producing both false positive and false negative results (hotspots). We conclude that both researchers and practitioners should be aware of the phenomenon of underreporting and should also try to maintain the same sampling effort of spatial reporting.

Application of statistical techniques to proportional loss data: Evaluating the predictive accuracy of physical vulnerability to hazardous hydro-meteorological events.

Knowledge about the cause of differential structural damages following the occurrence of hazardous hydro-meteorological events can inform more effective risk management and spatial planning solutions. While studies have been previously conducted to describe relationships between physical vulnerability and features about building properties, the immediate environment and event intensity proxies, several key challenges remain. In particular, observations, especially those associated with high magnitude events, and studies designed to evaluate a comprehensive range of predictive features are both limited. To build upon previous developments, we described a workflow to support the continued development and assessment of empirical, multivariate physical vulnerability functions based on predictive accuracy. Within this workflow, we evaluated several statistical approaches, namely generalized linear models and their more complex alternatives. A series of models were built 1) to explicitly consider the effects of dimension reduction, 2) to evaluate the inclusion of interaction effects between and among predictors, 3) to evaluate an ensemble prediction method for applications where data observations are sparse, 4) to describe how model results can inform about the relative importance of predictors to explain variance in expected damages and 5) to assess the predictive accuracy of the models based on prescribed metrics. The utility of the workflow was demonstrated on data with characteristics of what is commonly acquired in ex-post field assessments. The workflow and recommendations from this study aim to provide guidance to researchers and practitioners in the natural hazards community.

On reliable identification of factors influencing wildlife-vehicle collisions along roads.

Wildlife-vehicle collisions (WVCs) pose a serious global issue. Factors influencing the occurrence of WVC along roads can be divided in general into two groups: spatially random and non-random. The latter group consists of local factors which act at specific places, whereas the former group consists of globally acting factors. We analyzed 27,142 WVC records (roe deer and wild boar), which took place between 2012 and 2016 on Czech roads. Statistically significant clusters of WVCs occurrence were identified using the clustering (KDE+) approach. Local factors were consequently measured for the 75 most important clusters as cases and the same number of single WVCs outside clusters as controls, and identified by the use of odds ratio, Bayesian inference and logistic regression. Subsequently, a simulation study randomly distributing WVC in clusters into case and control groups was performed to highlight the importance of the clustering approach. All statistically significant clusters with roe deer (wild boar) contained 34% (27%) of all records related to this species. The overall length of the respective clusters covered 0.982% (0.177%) of the analyzed road network. The results suggest that the most pronounced signal identifying the statistically significant local factors is achieved when WVCs were divided according to their occurrence in clusters and outside clusters. We conclude that application of a clustering approach should precede regression modeling in order to reliably identify the local factors influencing spatially non-random occurrence of WVCs along the transportation infrastructure.

ROCA - An ArcGIS toolbox for road alignment identification and horizontal curve radii computation.

We present the ROCA (ROad Curvature Analyst) software, in the form of an ESRI ArcGIS Toolbox, intended for vector line data processing. The software segments road network data into tangents and horizontal curves. Horizontal curve radii and azimuth of tangents are then automatically computed. Simultaneously, additional frequently used road section characteristics are calculated, such as the sinuosity of a road section (detour ratio), the number of turns along an individual road section and the average cumulative angle for a road section. The identification of curves is based on the naïve Bayes classifier and users are allowed to prepare their own training data files. We applied ROCA software to secondary roads within the Czech road network (9,980 km). The data processing took less than ten minutes. Approximately 43% of the road network in question consists of 42,752 horizontal curves. The ROCA software outperforms other existing automatic methods by 26% with respect to the percentage of correctly identified curves. The segmented secondary roads within the Czech road network can be viewed on the roca.cdvgis.cz/czechia web-map application. We combined data on road geometry with road crashes database to develop the crash modification factors for horizontal curves with various radii. We determined that horizontal curves with radii of 50 m are approximately 3.7 times more hazardous than horizontal curves with radii accounting for 1000 m. ROCA software can be freely downloaded for noncommercial use from https://roca.cdvinfo.cz/ website.

An evaluation of odor repellent effectiveness in prevention of wildlife-vehicle collisions.

Wildlife-vehicle collisions (WVC) amount to 11% of all registered traffic crashes in the Czech Republic causing, apart from numerous deaths and serious injuries to animals, property damage and injuries to car passengers. Odor repellents have the potential to lower the overall number of WVC and allow animals to cross roads at the same time. We tested the effectiveness of odor repellent preparation in prevention of WVC. 18 places were selected on the Czech road network where WVC were concentrated on the basis of traffic crash data. Control sections on the same road segments were also delimited in order to keep the traffic intensities constant. We applied a Before-After-Control-Impact (BACI) study design to control not only the effect of the measures but also the expected natural variations in wildlife populations over time. Data were compared before and after odor repellent installations. Wildlife carcass gathering was carried out during the spring and autumn. We also used the police crash database to supplement carcass data when no field works were carried out. 201 killed mammals (roe deer and wild boars) were identified in total over 47 months. We applied a Bayesian approach as only a limited numbers of WVC were available. A WVC decrease between 26 - 43% can be expected on the treated road sections. These numbers are, however, up to three-times lower than those claimed by producers of odor preparations.

Manner of Death of Older People with Regard to Blood Alcohol Concentration.

OBJECTIVE: Alcohol abuse is related to a wide variety of negative health outcomes including mortality in older people. Alcohol abuse in older people is characterised by certain specific features uncommon in general adult population. The main objective of this study was to analyse the autopsy protocols of deceased older people in relation to blood alcohol concentration (BAC), sex, age, and manner of death. As a positive BAC, >0.20 g/kg was accepted. METHODS: The sample consists of 1,012 deceased older people (i.e. aged 65 years and over) selected out of 2,377 autopsied subjects in the period from 2003­2013. Subjects included into the sample were chosen via the proportional sampling method. Data (BAC, sex, age, and manner of death) was recorded in a single structured protocol. Data was evaluated statistically (Kolmogorov-Smirnov two-sample test, Wilcoxon two-sample test, risk ratio). RESULTS: Among older people, there has been a statistically significant correlation of natural death with sex (men died earlier) and with increased BAC (people with positive BAC died earlier). In case of violent death there is a difference in the types of accidents in older people with positive BAC (>0.2 g/kg) and with negative BAC (≤0.2 g/kg). Drowning is more common in older people with positive BAC. CONCLUSIONS: Health campaigns in Europe and the Czech Republic aimed at reducing alcohol consumption mainly deal with young people. Alcohol abuse has an impact on premature mortality even in older people. As shown by this study, older people with positive BAC die significantly earlier.

Circumstances and causes of fatal cycling crashes in the Czech Republic.

OBJECTIVES: The circumstances and causes of death of 129 cyclists registered in the Olomouc and the Zlín regions, the Czech Republic, between 2005 and 2013 were the subject of this study. METHODS: We analyzed the autopsy reports, where the principal cause of death was stated, and obtained a detailed description of the circumstances recorded by the police officers. RESULTS: Eighty-three cases (64.3% of the set) were collisions involving a motor vehicle. The driver was the guilty party in 57 cases (68.7%) and the cyclist in the remaining 26 cases (31.3%). The most frequent cause of the crash was connected with right of way (29 cases). Cars were involved in 52 cases; heavy vehicles, including buses, in 26 cases; and motorcycles in 5 cases. Single-vehicle crashes consisted of 43 (33.3%) cases. We divided this group into 3 subgroups based on whether the particular case could be attributed to a cyclist having lost control of the bicycle (31 cases) or to other particular causes. Sixty-eight cases (52.7%) of fatal outcomes were directly linked to intracranial injuries. Multiple injuries were the principal cause of death in 19 cases (14.7%), followed by hemorrhagic traumatic shock (12 cases, 9.3%). Seventy-two (55.8%) cyclists died immediately after the crash and 23 (17.8%) cyclists died within a day of the accident. CONCLUSIONS: Trucks were more dangerous to cyclists than cars at intersections, whereas cars were more dangerous on straight sections. The most important pattern was identified as a motor vehicle hitting a cyclist from behind on a straight road section. We identified a strong underestimation of natural death as a cause of cycling fatalities in the official police reports.

Identification of hazardous road locations of traffic accidents by means of kernel density estimation and cluster significance evaluation.

This paper proposes a procedure which evaluates clusters of traffic accident and organizes them according to their significance. The standard kernel density estimation was extended by statistical significance testing of the resulting clusters of the traffic accidents. This allowed us to identify the most important clusters within each section. They represent places where the kernel density function exceeds the significance level corresponding to the 95th percentile level, which is estimated using the Monte Carlo simulations. To show only the most important clusters within a set of sections, we introduced the cluster strength and cluster stability evaluation procedures. The method was applied in the Southern Moravia Region of the Czech Republic.