Road crashes pose a significant public health issue worldwide, often leading to severe injuries and fatalities. This dissertation embarks on a comprehensive examination of the factors affecting road crash severity, with a special focus on older drivers and the unique challenges introduced by the COVID-19 pandemic. Utilizing a dataset from Virginia, USA, the research integrates advanced statistical methods and machine learning techniques to dissect this critical issue from multiple angles. The initial study within the dissertation employs multilevel ordinal logistic regression to assess crash severity among older drivers, revealing the complex interplay of various factors such as crash type, road attributes, and driver behavior. It highlights the increased risk of severe crashes associated with head-on collisions, driver distraction or impairment, and the non-use of seat belts, specifically affecting older drivers. These findings are pivotal in understanding the unique vulnerabilities of this demographic on the road. Furthermore, the dissertation explores the efficacy of both parametric and non-parametric machine learning models in predicting crash severity. It emphasizes the innovative use of synthetic resampling techniques, particularly random over-sampling examples (ROSE) and synthetic minority over-sampling technique (SMOTE), to address class imbalances. This methodological advancement not only improves the accuracy of crash severity predictions for severe crashes but also offers a comprehensive understanding of diverse factors, including environmental and roadway characteristics. Additionally, the dissertation examines the influence of the COVID-19 pandemic on road safety, revealing a paradoxical decrease in overall traffic crashes accompanied by an increase in the rate of severe injuries. This finding underscores the pandemic's transformative effect on driving behaviors and patterns, heightening risks for vulnerable road users like pedestrians and cyclists. The study calls for adaptable road safety strategies responsive to global challenges and societal shifts. Collectively, the studies within this dissertation contribute substantially to transportation safety research. They demonstrate the complex nature of factors influencing crash severity and the efficacy of tailored approaches in addressing these challenges. The integration of advanced statistical methods with machine learning techniques offers a profound understanding of crash dynamics and sets a new benchmark for future research in transportation safety. This dissertation underscores the evolving challenges in road safety, especially amidst demographic shifts and global crises, and advocates for adaptive, evidence-based strategies to enhance road safety for all, particularly vulnerable groups like the older drivers. / Doctor of Philosophy / Road crashes are a major concern worldwide, often leading to serious injuries and loss of life. This dissertation delves into the critical issue of road crash severity, with a special focus on older drivers and the challenges brought about by the COVID-19 pandemic. Drawing on data from Virginia, USA, the research combines cutting-edge statistical methods and machine learning to shed light on this pressing matter. One important part of the research focuses on older drivers. It uses advanced analysis to find out why crashes involving this group might be more serious. The study discovered that situations like head-on collisions, driver distraction or impairment, and not wearing seat belts greatly increase the risk for older drivers. Understanding these risks is crucial in identifying the special needs of older drivers on the road. Then, the study explores the power of machine learning in predicting crash severity. Here, the research stands out by using innovative techniques to balance out the data, leading to more accurate predictions. This part of the study not only improves our understanding of what leads to severe crashes but also highlights how different environmental and road factors play a role. Following this, the research looks at how the COVID-19 pandemic has impacted road safety. Interestingly, while the overall number of crashes went down during the pandemic, the rate of severe injuries in the crashes that occurred increased. This suggests that the pandemic changed driving behaviors, posing increased risks especially to pedestrians and cyclists. In summary, this dissertation offers valuable insights into the complex factors affecting road crash severity. It underscores the importance of using advanced analysis techniques to understand these dynamics better, especially in the face of demographic changes and global challenges like the pandemic. The findings are not just academically significant; they provide practical guidance for policymakers and road safety experts to develop strategies that make roads safer for everyone, particularly older drivers.

En contexte tempéré, les études sur l'instabilité spatio-temporelle de la sensibilité des essences forestières au climat sont rares et souvent conduites à des échelles locales et régionales ; de telles approches ne permettant pas d'obtenir une vision globale de la réponse à l'environnement et à ses variations. La fusion de jeux de données dendrochronologiques (plus de 4500 arbres carottés) a permis d'analyser le comportement de croissance de sept essences européennes tempérées majeures (Quercus petraea, Fagus sylvatica, Abies alba, Picea abies, Pinus sylvestris, Pinus nigra, Pinus uncinata) dans des contextes climatiques variés (océanique à subalpin) et sur l'ensemble du 20ème siècle. Ce travail a également permis de préciser dans quelles mesures les conditions écologiques locales modulaient cette sensibilité au climat. Les relations cerne-climat ont été évaluées par le calcul de fonctions de corrélation. Quelque soit l'essence et le contexte écologique, la sécheresse estivale est le principal facteur limitant la croissance radiale (mais non l'unique), suivie par la sécheresse de l'automne précédent et enfin le froid hivernal. La variabilité spatiale de la réponse dépend plus fortement de la pluviométrie que des températures, une pluviométrie faible conduisant à une sensibilité plus forte au froid hivernal et aux sécheresses estivale et automnale. Ce comportement général est modulé par les conditions écologiques locales, avec une sensibilité à la sécheresse moindre sur sol profond. Les différences interspécifiques s'expriment principalement hors saison de végétation (novembre à mars), même si les corrélations sont rarement significatives. La croissance des résineux est généralement stimulée par des fins d'hiver chauds (février à avril), alors que celle des feuillus est corrélée négativement aux températures et positivement aux précipitations en décembre et janvier. Ces différences entre essences s'avèrent plutôt stables le long des gradients climatiques. Enfin, l'analyse temporelle révèle de fortes instabilités des relations cerne-climat au cours du siècle dernier. Le sens et l'ampleur de ces variations sont homogènes le long des gradients écologiques, mais en revanche peu synchrones avec les instabilités climatiques (automne, hiver, printemps) ou écophysiologiquement peu logiques (été). Cette faible cohérence entre tendances climatiques et instabilité de la sensibilité au climat pourrait s'expliquer par l'absence d'une contrainte climatique de croissance unique en contexte tempéré, où la largeur de cerne est sous le double contrôle du froid hivernal et du stress hydrique estival (et automnal). Elle pourrait également provenir de phénomènes non climatiques, tels que l'effet biologique lié au vieillissement ou l'évolution progressive des pratiques de gestion forestière. Les analyses des variations spatio-temporelles de sensibilité au climat questionnent également sur la précision des relations cerne-climat, estimée le plus souvent au travers du calcul des fonctions de corrélation. En effet, les comparaisons inter-région, inter-site et inter-période des réponses révèlent souvent des variations de corrélations dont les grandeurs pourraient être de l'ordre de la précision liée à l'échantillon considéré. Nous proposons ici de quantifier l'effet de la taille (nombre d'arbres carottés) et des caractéristiques de l'échantillon (nombre de placettes, nombre d'arbres par placette, statuts sociaux couverts) sur la qualité de l'estimation du signal environnemental contenu dans la chronologie moyenne et des fonctions de corrélation. Cette analyse a permis également de préciser dans quelles mesures les différences (1) de traits fonctionnels entre espèces et (2) de contextes climatiques (plus ou moins limitants) modulent cet effet « échantillon ». [...] Suite et fin du résumé dans la thèse. / In temperate conditions, studies dealing with spatio-temporal instabilities in climate sensitivity of forest tree species are scarce and often led at local and regional scales, which prevents from drawing global responses to the environment and its variations. The dendrochronological dataset merging (more than 4500 cored trees) allowed analyzing the growth pattern of seven major European species (Quercus petraea, Fagus sylvatica, Abies alba, Picea abies, Pinus sylvestris, Pinus nigra, Pinus uncinata) in various climatic contexts (oceanic to subalpine) and over the 20th century. This thesis also investigated the climate sensitivity modulation by local ecological conditions. Climate-growth relationships were studied through the calculation of correlation functions. Regardless of the species and the ecological context, summer drought is the main growth limiting factor (but not the unique one), followed by previous autumn drought and winter frost. Spatial variability in response to climate depends more heavily on pluviometry than temperature, decreasing amount of precipitation leading to increasing sensitivities to summer and previous autumn droughts and also winter frost. This general pattern is modulated the local ecological conditions, with especially a lower sensitivity to drought on deep soils. Species-specific responses to climate are mainly evidenced out of the growing season (November to March), even if correlations are rarely significant. The growth of conifers is generally enhanced by warm late winters (February to April), while that of broadleaves is negatively correlated to temperatures and positively to precipitation in December and January. These between-species differences turn out to be stable along the climatic gradients. Lastly, the temporal analysis evidences strong climate-growth relationships instabilities over the last century. The way and the magnitude of these variations are rather homogenous along the ecological gradients, but display low synchronicity with climatic instabilities (autumn, winter and spring) and are ecophysiologically difficult to explain (summer). Such incoherencies between climatic trends and climate sensitivity trends could be related to absence of a single growth limiting factor under temperate context, since tree-ring is under the control of both winter frost and summer (and autumn) drought. They could also result from non-climatic phenomenon, such as the biological the age-related biological effect or progressive changes in forest management. The analyses of spatio-temporal variations in sensitivity climate question on the precision of the climate-growth relationships, most of the time estimated with correlation functions. Indeed, inter-plot, inter-region and inter-period comparisons of responses often highlight differences in correlations which could be of the same magnitude than that of the precision related to the investigated sample. We thus propose to quantify the effect of the sample size (number of cored trees) and characteristics (number of plots, number of trees per plot, sampled social statuses) on the accuracy of the estimation of both the environmental signal estimation contained in the growth chronology and the correlation functions. This analysis also investigates the modulation of such effects by the species-specific functional traits and the strength of the environmental growth limitation. [...] Last and final summary in the thesis.

No description available.