El impacto de la contaminación del aire es un tema crítico para el medioambiente y el clima. Una mala calidad del aire es un tema de importancia para la salud pública, especialmente en ambientes urbanos. El material particulado (PM), el ozono (O3) y el dióxido de nitrógeno (NO2) son los contaminantes más problemáticos en Europa y España. La Comisión Europea ha mostrado una gran preocupación por desarrollar técnicas que permitan incrementar el conocimiento sobre la dinámica de los contaminantes atmosféricos para asegurar el cumplimiento de la legislación y para informar a la población acerca de sus niveles. Además, la directiva europea 2008/50/CE establece la posibilidad de usar técnicas de modelización para informar sobre calidad del aire. Esta tesis doctoral está desarrollada en el marco de dos proyectos: El proyecto CALIOPE y el proyecto CICYT CGL2006-08903, ambos basados en la necesidad de desarrollar un sistema de calidad del aire que permita informar y entender los niveles de contaminación en Europa y España, con el objetivo de obtener un preciso pronóstico de la calidad del aire. Con ese propósito, el sistema de modelización CALIOPE se ha desarrollado con alta resolución espacial y temporal sobre Europa (12 km x 12 km y 15 capas, 1 hora), dominio madre, y España (4 km x 4 km y 15 capas, 1 hora), dominio anidado. CALIOPE consiste en un conjunto de modelos que tienen en cuenta la contaminación tanto antropogénica como natural. La disponibilidad del supercomputador MareNostrum, alojado en el Barcelona Supercomputer Center- Centro Nacional de Supercomputación, ha permitido trabajar a tan alta resolución. El objetivo principal de esta tesis es aumentar la confianza científica en el sistema CALIOPE, identificando sus puntos fuertes y débiles con un nivel de detalle que contribuya a establecer necesidades de mejora en el proceso de modelización. Por tanto, el presente trabajo ha evaluado espacial y temporalmente las simulaciones de calidad del aire sobre Europa y España en términos de O3, NO2, SO2, PM2.5 y PM10 en superficie sobre el año completo 2004. Para identificar el origen de las incertidumbres en la modelización del PM, su composición química ha sido también evaluada en ambos dominios. Las evaluaciones han sido realizadas sobre más de 150 estaciones de calidad del aire (más de 2 millones de datos experimentales). Además, esta tesis ha usado el sistema CALIOPE para analizar los patrones de calidad del aire sobre 2004, identificando claramente las áreas de contaminación. Las ideas más importantes que se desprenden de esta tesis son tres. Primero, las condiciones de contorno químicas basadas en un modelo global, como el LMDz-INCA2, son esenciales para modelizar el O3 troposférico sobre los dominios de estudio. Segundo, para simular la concentración de PM en el sur de Europa, tanto a escala rural como urbana, la contribución de polvo procedente del desierto del Sahara deber ser considerada debido a la proximidad al continente africano. La contribución del polvo del desierto a través del modelo BSC-DREAM8b ayuda satisfactoriamente a modelizar los picos de PM10 observados. Tercero, para ser capaz de modelizar la calidad del aire a escala urbana sobre España es esencial (1) una alta resolución espacial y temporal que permita describir fenómenos mesoescalares en áreas de topografía compleja , (2) un modelo de emisiones altamente desagregado como HERMES; (3) unos modelos que representen el estado actual del conocimiento en meteorología y química atmosférica / The impact of air pollution is a critical topic in environment and climate. Poor air quality is an important public health issue, especially in urban environments. Particulate matter (PM), tropospheric ozone (O3) and nitrogen dioxide (NO2) are the main problematic pollutants in Europe and Spain. The European Commission has shown a great concern for developing actions that allow increasing the knowledge on dynamics of atmospheric pollutants to assure the accomplishment of legislation and to inform the population about their levels. The European directive 2008/50/EC establishes the possibility of using modelling techniques to assess air quality. This Ph.D. thesis is developed in the framework of two projects: the CALIOPE project and the CGL2006-08903 CICYT project, both based on the necessity to develop an air quality modelling system that allows assessing and understanding the air pollution levels in Europe and Spain, with the aim of obtaining a precise air quality forecast. For that purpose, the CALIOPE air quality modelling system has been developed with high spatial and temporal resolution over Europe (12 km x 12 km, 1 h), as a mother domain; and Spain (4 km x 4 km, 1 h), as the nested domain. The CALIOPE system consists in a set of models that take into account both anthropogenic and natural pollution. The availability of the MareNostrum supercomputer, held in Barcelona Supercomputing Center- Centro Nacional de Supercomputación, has allowed such configuration of the CALIOPE system. The main objective of the present Ph.D. thesis is to increase the scientific confidence on the CALIOPE system, identifying skills and weakness with a degree of detail that contributes to establish necessities of improvements in the modelling process. Therefore, the present work has spatially and temporally evaluated CALIOPE air quality simulations over Europe and Spain in terms of O3, NO2, SO2, PM2.5, PM10 concentrations over the full year 2004. In order to identify the origin of uncertainties in PM modelling, PM chemical composition has been also evaluated in both target domains. Evaluations have been performed across more than 150 air quality-monitoring stations and over more than 2 million of experimental data. Furthermore, this Ph.D. thesis has used the CALIOPE system to assess air quality pattern over the year 2004, identifying clearly the areas of air pollution. There are three major thrusts of the present Ph.D. thesis. First, chemical boundary condition based on a global model, such as LMDz-INCA2, becomes essential to model O3 background concentrations in the target domains. Second, to simulate PM concentration in southern Europe, both regional and urban scales, the contribution of dust from the Saharan desert should be taken into account, since that region is frequently affected by dust outbreaks due to its proximity to the African continent. The contribution of desert dust through the BSC-DREAM8b helps to satisfactory model the observed episodic PM10 concentration peaks. Even more, the contribution of sea-salt aerosol is especially important over coastal areas. Third, to be able to model the air quality in urban scale over Spain it is essential (1) a high spatial (4 km x 4 km and 15 layers) and temporal (1h) resolution that allows describing mesoscale phenomena in very complex terrains; (2) a high disaggregated emission model to describe the sources, such as HERMES; and (3) an state-of-the-science meteorological and chemical models. This Ph.D. thesis has demonstrated that CALIOPE system applied over Europe and Spain is a useful tool which may contribute to (1) forecast air pollution in urban/suburban areas with a pervasive influence of anthropogenic emissions on a local scale and over very complex terrains and meteorology patterns; (2) assess about air pollution, discriminating between anthropogenic and natural episodes; and (3) manage air pollution, by means of modification of urban strategies or requirements of the legislation.

Diploma thesis deals with the issue of terrrorism in relation to soft targets and subsequent assessment of the selected soft target in terms of possible terrorist attack. The objective is primary school Sirotkova in Brno. The assessment criteria was evacuation of the school as a whole and its individual parts. A risk analysis was performed for the object to determine real threats in which there were clasified the probability and extent of impacts of potential attacks. The most likely scenarios of the attack were identified, based on this analysis. These scenarios were subsequently implemented into an advanced numerical 3D model, where the evacuation was simulated within these scenarios. In the framework of the school assessment, the work focused on comparing three different approaches to the evacuation process. The evacuation process was compared according to parameters of experimental data obtained from the practise of evacuation of primary school with evacuation processes set acording to the parameters of fire standards CSN and international SFPE fire safety standards. The results of the individual simulations were processed with a statistically determined probability percentile. The outcome of the assessment was to determine the probable evacuation time of the school and its individual parts. Based on these results, there was performed a proposal to optimize the evacuation process of the school, which made the process more efective and shortened the evacuation time to minimalize the impact of a potential terrorist attack.