Les événements extrêmes sont l'expression de la variabilité climatique naturelle. Puisque les émissions anthropiques affectent le climat mondial, il est naturel de se demander si les événements extrêmes observés récemment sont une manifestation du changement climatique. Cette thèse se propose de contribuer à la compréhension de l'influence du changement climatique anthropique sur les événements extrêmes observés, tout en évaluant si et comment cette information scientifique - et plus généralement, l'attribution d'événements extrêmes (AEE) - pourrait être utile à la société. Je propose des outils statistiques et j'utilise un ensemble d'entretiens qualitatifs pour répondre à ces questions.La partie statistique s'applique aux vagues de chaleur européennes. Je quantifie le rôle joué par la circulation atmosphérique dans l'intensité de quatre vagues de chaleur récente. Cette analyse s'appuie sur des analogues de circulations, qui identifient des jours ayant une circulation similaire à celle de l'événement étudié. Ensuite, je dissocie l'influence du changement climatique sur les processus dynamiques et non dynamiques menant aux vagues de chaleur. Je calcule des tendances sur l'occurrence de circulations favorisant les fortes chaleurs et sur la température pour une circulation fixée, pour les vagues de chaleur de 2003 en Europe de l'Ouest et de 2010 en Russie. Je trouve que la significativité des résultats dépend de l'événement étudié, ce qui montre l'intérêt de calculer des tendances pour des types de circulation atmosphérique précis.La partie épistémologique analyse les utilisations sociales potentielles de l'AEE. Je mesure comment elle pourrait informer les négociations internationales sur le climat, en particulier les pertes et préjudices, en réponse à des arguments de scientifiques dans ce sens. Je trouve que le seul rôle que l'AEE puisse jouer pour renforcer les pertes et préjudices est un rôle de sensibilisation des politiques, en marge du processus de négociations. Je compare également les motivations avancées par les scientifiques dans les entretiens avec les résultats existants sur l'utilité sociale de ce type d'information scientifique. Je montre que la pertinence sociale des résultats d'AEE est ambiguë, et qu'il y a un manque de données empiriques pour mieux comprendre comment différents acteurs s'approprient et réagissent à cette information. / Extreme events are an expression of natural climate variability. Since anthropogenic emissions affect global climate, it is natural to wonder whether recent observed extreme events are a manifestation of anthropogenic climate change. This thesis aims at contributing to the understanding of the influence of anthopogenic climate change on observed extreme events, while assessing whether and how this scientific information - and more generally, the science of extreme event attribution (EEA) - could be useful for society. I propose statistical tools to achieve the former, while relying on qualitative interviews for the latter.The statistical part focuses on European heatwaves. I quantify the role played by the atmospheric circulation in the intensity of four recent heatwaves. This analysis is based on flow analogues, which identify days with a similar circulation pattern than the event of interest. I then disentangle the influence of climate change on the dynamical and non-dynamical processes leading to heatwaves. I calculate trends in the occurrence of circulation patterns leading to high temperatures and trends in temperature for a fixed circulation pattern, applied to the 2003 Western Europe and 2010 Russia heatwaves. I find that the significance of the results depend on the event of interest, highlighting the value of calculating trends for very specific types of circulation.The epistemological part evaluates the potential social uses of extreme event attribution. I assess how it could inform international climate negotiations, more specifically loss and damage, in response to a number of claims from scientists going in this direction. I find that the only potential role EEA could play to boost the loss and damage agenda would be to raise awareness for policy makers, aside from the negotiation process itself. I also evaluate how the different motivations stated by EEA scientists in interviews fare compared to the existing evidence on social use of this type of scientific information. I show that the social relevance of EEA results is ambiguous, and that there is a lack of empirical data to better understand how different non-scientific stakeholders react and appropriate EEA information.

abstract: Environmental hazards and disaster researchers have demonstrated strong associations between sociodemographic indicators, such as age and socio-economic status (SES), and hazard exposures and health outcomes for individuals and in certain communities. At the same time, behavioral health and risk communications research has examined how individual psychology influences adaptive strategies and behaviors in the face of hazards. However, at present, we do not understand the explanatory mechanisms that explain relationships between larger scale social structure, individual psychology, and specific behaviors that may attenuate or amplify risk. Extreme heat presents growing risks in a rapidly warming and urbanizing world. This dissertation examines the social and behavioral mechanisms that may explain inequitable health outcomes from exposure to concurrent extreme heat and electrical power failure in Phoenix, AZ and extreme heat in Detroit, MI. Exploratory analysis of 163 surveys in Phoenix, AZ showed that age, gender, and respondent’s racialized group identity did not relate to thermal discomfort and self-reported heat illness, which were only predicted by SES (StdB = -0.52, p < 0.01). Of the explanatory mechanisms tested in the study, only relative air conditioning intensity and thermal discomfort explained self-reported heat illness. Thermal discomfort was tested as both a mechanism and outcome measure. Content analysis of 40 semi-structured interviews in Phoenix, AZ revealed that social vulnerability was associated with an increase in perceived hazard severity (StdB = 0.44, p < 0.01), a decrease in perceived adaptation efficacy (StdB = -0.38, p = 0.02), and an indirect increase (through adaptive efficacy) in maladaptive intentions (StdB = 0.18, p = 0.01). Structural equation modeling of 244 surveys in Phoenix, AZ and Detroit, MI revealed that relationships between previous heat illness experience, perceived heat risk, and adaptive intentions were significantly moderated by adaptive capacity: high adaptive capacity households were more likely to undertake adaptive behaviors, and those decisions were more heavily influenced by risk perceptions and previous experiences. However, high adaptive capacity households had lower risk perceptions and fewer heat illness experiences than low adaptive capacity households. A better understanding of the mechanisms that produce social vulnerability can facilitate more salient risk messaging and more targeted public health interventions. For example, public health risk messaging that provides information on the efficacy of specific adaptations may be more likely to motivate self-protective action, and ultimately protect populations. / Dissertation/Thesis / Doctoral Dissertation Environmental Social Science 2019

No description available.

Meine kumulative Dissertation

This thesis has analyzed the relationship between the punitive turn and the crime policies of the Green party (Miljöpartiet de Gröna) and the Sweden democrats (Sverigedemokraterna) to answer the research question; what influence the punitive turn has had on the parties policies. The choice of method has fallen on a quantitative content-analysis with a qualitative complementarity and qualitative facilitation. From an account of the punitive turn has a word-list with recording units been created, of those recording units has a computer-search been made of the two parties most recent official documents which accounts for the parties holistic politics; MP´s Partyprogramme from 2013, and SD´s Principleprogramme from 2011. The result of the qualitative complementarity shows; a greater frequency of recording units for MP than for SD. The analysis of the qualitative facilitation-result shows; a lesser direct influence of the punitive turn for MP than for SD. The result and analysis are discussed regarding whether the research question has been answered, and no such conclusion is considered to be made by the researcher, instead are the result and analysis open for interpretation of the reader.

La violence en temps de guerre prend parfois des formes extrêmes. Non seulement les belligé-rants tuent fréquemment un grand nombre de civils, mais leurs atrocités vont parfois au-delà du seul meurtre pour inclure des actes de cruauté tels que la mutilation, le viol, la torture publique ou encore les abus des corps humains. Si des cas individuels de ce type de violence se produisent probablement dans presque tous les conflits et peuvent s’expliquer par la psychopathie, le sa-disme, ou un processus de « barbarisation » inhérent à la guerre, ni la psychologie individuelle ni les conditions propres à la lutte armée ne peuvent expliquer pourquoi l’ampleur de ce type d’actes spécifiques varie d’une organisation sociale à une autre. Dans certains groupes armés, les atrocités extrêmes demeurent des cas isolés. Dans d’autres, cependant, nombre de combattants font de telles pratiques une part intégrante de leur « répertoire de violence ». Comment expliquer cette variation ? Dans cette thèse, je propose une série d’arguments théoriques permettant d’expliquer, au moins en partie, pourquoi la violence en temps de guerre prend parfois la forme d’atrocités extrêmes. Je définis les « atrocités extrêmes » comme des actes de violence caractérisés par une cruauté manifeste et publique. Je soutiens que ces types d’actes émergent souvent dans un contexte de guerre irrégulière ; des conflits caractérisés par une impor-tante asymétrie des capacités militaires, l’utilisation de tactique de guérilla, et, par voie de consé-quence, un contrôle territorial hautement fragmenté. Dans ce contexte, je montre que les atrocités extrêmes découlent souvent de deux proces-sus alternatifs. Le premier processus implique un modèle de prise de décision « du haut vers le bas, » via lequel des dirigeants organisationnels adoptent des tactiques centrées sur la violence extrême à des fins stratégiques. L’objectif premier est souvent de dissuader les civils de collabo-rer avec l’ennemi dans les territoires contestés. Le deuxième processus implique l’émergence d’atrocités extrêmes en tant que « pratique de guerre ». Ici, les techniques de cruauté émergent au niveau des unités militaires, sans découler d’ordres venus d’en haut. Je soutiens qu’une telle émergence est plus probable dans les unités militaires présentant tant un niveau élevé de cohésion sociale qu’une faible discipline. Dans des conditions de guerre irrégulière, ces unités peuvent dé-velopper des normes informelles légitimant la violence extrême comme moyen de venger les pertes au combat. Ces arguments sont développés et testés dans ma thèse à l’aide d’une variété de matériaux empiriques disséminés dans trois articles qui ont été ou seront bientôt soumis pour publication. Le premier article (chapitre 2) définit le concept d’atrocité extrême et utilise des données venant de quatre guerres civiles pour vérifier la plausibilité d’une série d’explications tirées de la littérature sur la violence envers les civils. Le deuxième article (chapitre 3) analyse l’utilisation des décapita-tions par les groupes djihadistes. Je montre que seule une minorité de ces groupes utilise ce type de violence de manière récurrente et que la variation peut être expliquée par le contexte straté-gique dans lequel les organisations mènent leurs opérations et par la nature de leurs liens transna-tionaux. Le troisième article (chapitre 4) se concentre sur l’émergence des atrocités extrêmes en tant que « pratique » au sein des forces de sécurité étatiques, en utilisant l’exemple des mutila-tions commises par les soldats américains au cours de la guerre du Vietnam. Ce cas démontre spécifiquement comment les atrocités extrêmes peuvent se généraliser malgré leur interdiction au niveau du commandement. Enfin, le chapitre 5 montre que les idées théoriques développées dans mes articles ont une application plus large, en utilisant une base de données originale sur les atro-cités extrêmes perpétrées dans les guerres civiles entre 1980 et 2011. / Wartime violence sometimes takes particularly extreme forms. Not only do belligerents frequent-ly kill large numbers of civilians, but violent atrocities sometimes go beyond killing to include acts of overt cruelty such as mutilation, rape, public torture, and the abuse of human remains. While individual instances of such violence likely occur in almost all wars, and might be ex-plained by a certain prevalence of psychopathy or sadism among combatants, or by a process of “barbarization” inherent in war, neither individual psychology nor universal wartime conditions can explain why armed actors seem to vary in the extent to which they perpetrate such violence. In some armed groups, episodes of extreme atrocity remain isolated cases. In others, in contrast, large numbers of combatants appear to adopt such practices as an established part of their “reper-toire of violence.” What explains such variation? In this dissertation, I develop and test a series of explana-tions that help account for variation in the occurrence of “extreme atrocities” within and across conflicts. I define extreme atrocities as acts of physical violence characterized by the public dis-play of overt cruelty, and argue that the occurrence of such violence is closely connected to the context of irregular warfare, that is, of warfare characterized by pronounced asymmetry in mili-tary capabilities and fragmented territorial control. Within this context, I show that there are two common pathways towards extreme atrocity. The first involves a process of “top down” decision making, whereby organizational leaders adopt extreme forms of violence for strategic ends. Among the most important of these is the imperative to deter civilian collaboration with the ene-my in contested territories. Because of the terror they inspire, extreme atrocities can usefully serve this purpose, at least under certain conditions. A second pathway involves the “bottom up” emergence of extreme atrocities among rank-and-file combatants as an unordered “practice of war.” Such emergence, I argue, is more likely in military units with high levels of social cohesion but low levels of discipline. Under conditions of irregular warfare, such units can develop infor-mal norms that endorse extreme violence as a means of avenging combat losses. These arguments are developed and tested in my dissertation using a variety of different empirical material, most of which is presented in three articles that have been or are soon to be submitted for publication. The first article (Chapter 2) defines the concept of “extreme atrocity” and uses violence data from four civil wars to probe the plausibility of a series of explanations of such violence derived from the literature on civilian victimization. The second article (Chapter 3) further develops the idea that irregular warfare creates strategic incentives for the top-down adoption of extreme violence, focusing specifically on the use beheadings by jihadist groups. I show that variation in the use of beheadings among jihadist groups can be explained by a combi-nation of local strategic context and transnational ties. The third article (Chapter 4) focuses on the emergence of extreme atrocity as a “practice” among state security forces, using the example of mutilations perpetrated by American soldiers during the Vietnam war to show how extreme forms of violence can become widespread despite being unambiguously prohibited by military policy. Finally, Chapter 5 shows that the theoretical ideas developed in my articles have broader application. Using original data on extreme atrocities perpetrated in civil wars between 1980 and 2011, I show that the patterns in perpetration of such violence by state security forces and rebel groups are consistent with the theories of top-down adoption and bottom-up emergence of ex-treme atrocity described above. This thesis contributes to our understanding of wartime violence by explicitly theorizing a hitherto neglected dimension of violence, and developing and testing explanations that can ac-count for variation in its occurrence at multiple different levels.

In many environmental and climate applications, extreme data are spatial by nature, and hence statistics of spatial extremes is currently an important and active area of research dedicated to developing innovative and flexible statistical models that determine the location, intensity, and magnitude of extreme events. In particular, the development of flexible sub-asymptotic models is in trend due to their flexibility in modeling spatial high threshold exceedances in larger spatial dimensions and with little or no effects on the choice of threshold, which is complicated with classical extreme value processes, such as Pareto processes. In this thesis, we develop new flexible sub-asymptotic extreme value models for modeling spatial and spatio-temporal extremes that are combined with carefully designed gradient-based Markov chain Monte Carlo (MCMC) sampling schemes and that can be exploited to address important scientific questions related to risk assessment in a wide range of environmental applications. The methodological developments are centered around two distinct themes, namely (i) sub-asymptotic Bayesian models for extremes; and (ii) flexible marked point process models with sub-asymptotic marks. In the first part, we develop several types of new flexible models for light-tailed and heavy-tailed data, which extend a hierarchical representation of the classical generalized Pareto (GP) limit for threshold exceedances. Spatial dependence is modeled through latent processes. We study the theoretical properties of our new methodology and demonstrate it by simulation and applications to precipitation extremes in both Germany and Spain. In the second part, we construct new marked point process models, where interest mostly lies in the extremes of the mark distribution. Our proposed joint models exploit intrinsic CAR priors to capture the spatial effects in landslide counts and sizes, while the mark distribution is assumed to take various parametric forms. We demonstrate that having a sub-asymptotic distribution for landslide sizes provides extra flexibility to accurately capture small to large and especially extreme, devastating landslides.

This thesis gives an example of assessing the risk of a financial portfolio with international assets, where the assets may be of different classes, by the use of Monte Carlo simulation and Extreme Value Theory. The simulation uses univariate modelling, models of the assets’ returns as stochastic processes, as well as vine copulas to create dependency between the variables. For the asset returns a modified version of a discretized Merton jump diffusion model was used. The risk assessment used Extreme Value Theory to calculate Value at Risk and Expected Shortfall of the simulated portfolio. However, the resulting return distribution, and the risk assessment thereof, was not entirely satisfactory due to unreasonably large values ascertained. / Denna uppsats ger ett exempel på hur riskbedömning av finanisella portföljer med internationella tillgångar av olika tillgångsslag genom Monte Carlo simulering och extremvärdesteori. Simuleringen använder univariat modelling, modeller för tillgångarnas avkastningar som stokastiska processer, såväl som vine-copulas för att skapa ett beroende mellan tillgångarna. Tillgångarnas avkastningar modellerades med en modifierad version av en diskretiserad Merton-jump-diffusion model. Riskbedömningen använde extremvärdesteori för att beräkna Value-at-Risk och Expected-Shortfall. Dock blev den resulterande avkastningsfördelningen och riskbedömningen därav inte helt tillfredsällande på grund av att orimligt stora värden erhölls.

I takt med en ökande individualism i samhället ställs nya krav på dagens organisationer, där medbestämmande och personlig utveckling önskas från medarbetare. Det kan dock innebära en paradox för militära organisationer när deras kärna så tydligt grundar sig i kollektivistiska värden och en anda av laget före jaget. Samtidigt finns höga förväntningar på militär personal att kunna påverka sig själva att agera och pusha sig igenom utmanande uppgifter, vilket är särskilt påtagligt i extrema miljöer som internationella insatser. Den här studien har undersökt om självledarskap är ett verktyg som militär personal använt för att kunna prestera under sina insatser. Trots att studier visat på positiva effekter av självledarskap i påfrestande förhållanden, har få studier undersökt huruvida det kan vara effektfullt i militära kontexter. Det kan samtidigt visa på att självledarskap inte nått militära organisationer än. Därmed har studien även undersökt synen på självledarskap hos Försvarsmaktens personal. Genom att ha intervjuat sex militärer som genomfört sin insats i Mali har studien kunnat visa på att flera strategier av självledarskap används i extrema miljöer. Den avvikelse som kunde urskiljas var att självmålsättning och självbelöning inte är lika synligt som övriga. Det kan delvis förklaras utifrån kollektivistiska värderingar där personalen redan motiveras av och känner stor tillhörighet till organisatoriska mål, men även att extrema miljöer medför hög arbetsbelastning och lite utrymme för återhämtning.

I världen idag pågår en urbanisering, vilket innebär att fler människor flyttar in till städerna. Det innebär att fler bostäder måste byggas för att uppfylla de nya behoven, och detta görs ofta genom förtätning av redan exploaterade områden. Vid förtätning av bostadsområden ökar ofta andelen hårdgjorda ytor. En hårdgjord yta är en icke permeabel yta där dagvatten inte kan infiltrera ner i marken utan istället bildar ytavrinning. Vattnet som avrinner färdas mot lågpunkter i terrängen vilka riskerar att översvämmas. Klimatförändringar väntas leda till häftigare väder, bland annat i form av skyfall. Kraftigare regn i kombination med större andel hårdgjorda ytor väntas öka risken för pluviala översvämningar. För att undvika pluviala översvämningar krävs strategier för att hantera städers dagvatten. Det existerande ledningsnätet är högt belastat och kombineras med hållbara dagvattenlösningar för att minska avrinningen. Grönytor ses ofta som goda infiltrationsytor, men en osäkerhet råder kring hur effektiva olika typer av grönytor är. Det är därför av intresse att undersöka hur goda infiltrationsytor urbana grönområden är och hur stor betydelse de har vid skyfall för att minimera pluviala översvämningar. Syftet med examensarbetet är att undersöka infiltrationskapaciteten för grönytor på ett antal olika platser i Stockholm. Syftet är vidare att undersöka hur resultaten från fältförsöken kan användas i det hydrauliska modelleringsprogrammet MIKE 21 för att återspegla det verkliga infiltrationsförloppet och därmed få en god bild av hur stora översvämningsriskerna är för olika områden. Totalt utfördes 13 separata mätningar i två grönområden i Stockholm. Vid 11 av mätningarna användes en dubbelringsinfiltrometer och vid två av mätningarna användes en enkelringsinfiltrometer. Mätningarna utfördes under 0,5-2 timmar beroende på vattentillgång. Infiltrationsförsöken visade att det finns en stor variation i infiltrationskapacitet, även inom mycket små områden. De visade också att det finns en tendens till högre infiltrationskapacitet för mindre kompakterade grönytor. Kornstorleksfördelningen och vattenhalten skiljde sig inte nämnvärt mellan de två områdena och dessa två parametrar kunde inte kopplas till någon skillnad i infiltrationskapacitet för de två undersökta områdena. Resultaten från simuleringarna i MIKE 21 visade att vilka värden som anges för infiltrationskapaciteten är av större betydelse än på vilket sätt dessa anges. Resultaten visade också att parametrar såsom vattenhalt och porositet hade en inverkan på infiltrationsförloppet men infiltrationszonens mäktighet hade liten inverkan på resultaten. Sammanfattningsvis kan sägas att det finns en stor variation i infiltrationskapacitet för grönytor och den osäkerheten påverkar resultaten vid modellering av översvämningsrisker i MIKE 21. / The ongoing urbanization in the world today means that more people are moving into the cities and therefore more housing is required. When building in cities there is a tendency for an increase in impermeable surfaces. An impermeable surface is defined as a surface where no water can infiltrate into the subsurface soil and instead there is an increase in surface runoff. The water flows through the terrain towards low-lying areas, which are at risk for flooding. Climate changes are expected to result in more extreme weather such as extreme rain. An increase in extreme rain in combination with more impermeable surfaces will increase the risk for pluvial flooding. To avoid pluvial flooding different strategies is required to cope with the urban stormwater. The traditional stormwater systems are usually put under high stress and sustainable stormwater management needs to be implemented to decrease the surface runoff in urban areas. Green areas are often thought to be good infiltration surfaces but there is a big uncertainty in regards to exactly how effective different green areas is for infiltration purposes. There is an interest to investigate how good the infiltration capacity is for urban green areas to map and to mitigate pluvial flooding. The aim for this master thesis is to investigate the infiltration capacity through field measurements for two different green areas in the city of Stockholm, Sweden. Furthermore, the aim is to investigate how the results from the field measurements can be implemented in the hydraulic modelling software MIKE 21 to represent the real infiltration pattern in order to map the risk for pluvial flooding for different areas. A total of 13 measurements were conducted in two green areas around Stockholm, using a double ring infiltrometer. For two of the measurements a single ring infiltrometer was used. The measurements were conducted during 0.5-2 h depending on the water accessibility. The field measurements showed that there is a large variability in infiltration capacity, even within very small areas. The measurement showed that there was a tendency for higher infiltration rates for less compacted soil. The grain size distribution showed little impact on the infiltration rate, and so did the water content. The simulations in MIKE 21 showed that the magnitude of the infiltration rate is of greater importance than the way it is implemented in MIKE 21. The results also showed that parameters such as water content and porosity had an effect on the infiltrated volume, but the depth of the infiltration zone had little impact on the results. In conclusion, there is a large variability in infiltration capacity for green areas and this uncertainty does affect the results when modelling the risk for pluvial flooding in MIKE 21.