Les méthodes géostatistiques sont couramment utilisées pour la cartographie de la qualité de l'air. En France, le système PREV'AIR diffuse depuis 2003 des prévisions quotidiennes à trois jours ainsi que des cartographies de qualité de l'air à l'échelle nationale et européenne. Ces cartographies sont construites par krigeage des observations en utilisant les sorties du modèle déterministe de chimie-transport CHIMERE en dérive externe. Dans ce domaine, de nombreuses questions et développements restent en suspens.La première partie de la thèse vise à améliorer les cartographies locales, nationales et européennes des principaux polluants réglementaires. En mode analyse (réalisation de cartographies à partir des observations passées, et donc connues), on revient sur l'utilisation des variables explicatives dans le krigeage : il s'agit de voir comment ces variables peuvent être intégrées à l'estimation pour la modélisation des non stationnarités et l’élaboration de cartographies de plus haute résolution spatiale que les simulations du modèle de chimie-transport. La réflexion porte sur la modélisation de la composante déterministe du processus stochastique, mais aussi sur la covariance du résidu. Une attention particulière est portée au réseau de mesures. L'utilisation des observations PM10 pour cartographier les PM2.5, dont le réseau de mesures est moins dense, est explorée. L'estimation dans les zones peu informées est également examinée. Enfin, une réflexion est menée sur l'extension de ces méthodes au mode prévision, afin d'améliorer le système PREV'AIR, qui dissocie les composantes temporelles et spatiales de la prévision. Après une revue des méthodes spatio-temporelles d'estimation, différentes méthodes sont évaluées.Dans la deuxième partie sont présentées des approximations pragmatiques permettant d'exploiter au mieux les cartographies produites quotidiennement ou ponctuellement afin de satisfaire aux exigencesréglementaires. A l'échelle nationale, l'exploitation des sorties des cartes analysées (horaires, journalières) de PREV'AIR permet de déduire des cartes de probabilités de dépassement des seuils réglementaires. Ces travaux amènent à reconsidérer la question de la représentativité spatiale des stations de mesures. Enfin, on insiste sur la nécessité d'utiliser des méthodes d’estimation appropriées pour le calcul des surfaces en dépassement de valeurs réglementaires. / Geostatistical methods are commonly used in air quality mapping. In France, since 2003, the PREV'AIR system has been broadcasting daily three-days forecasts as well as national and European air quality maps. They are built by a kriging of the observations using the deterministic chemistry-transport model CHIMERE outputs as an external drift. However, many issues or developments remain unresolved.The first part of the thesis aims at improving the local, national and european maps of the main regulated pollutants. Regarding the analysis (mapping of the past observations), we get back to the question of the use of explanatory variables in kriging. What are the best options for a covariate-based modelling of the underlying non stationarity, that also enables downscaling the model outputs? The statistical modelling for the deterministic component of the stochastic process is investigated, as well as the modelling for the covariance of the residual. A focus is also made on the spatial sampling of the monitoring network. The use of PM10 observations to map PM2.5, whose monitoring network is less dense, is studied and a thought is also given to the estimation in far-off extrapolation. Last, we discuss how to extend these methods to the prediction problem (mapping of the future, where no observations are available). Therefore, we could improve the PREV'AIR system which currently dissociates the time and the spatial components of the underlying process. A review of spatio-temporal methods is carried out, and some of them are evaluated.In a second part, some pragmatic though justified approximations are presented to deal with regulatory requirements. At both local and national levels, how the analyses can be used to deduce probability maps of exceedances of the regulatory thresholds? Along this line, we also come back to the question of the spatial representativeness of the monitoring stations. Last, the pragmatic approximations are confronted to non-linear estimations, theoretically more convenient to deal with non-linear functions of the concentrations. We engage a discussion to show the need of considering the appropriate estimation method to compute the surface exceeding the limit value.

Airborne particulate matter (PM) consists of solid and liquid particles suspended in air. PM causes many negative health effects when inhaled like exacerbations of asthma, chronic obstructive pulmonary disease (COPD), and premature death. The health impacts of PM depend on the physical size and chemical composition of the inhaled particles. Particles less than 2.5 micrometers can penetrate the deep lung and enter the bloodstream. Understanding the composition of PM helps study human exposures and evaluate PM sources to support control and mitigation strategies. This thesis examines PM in power plant emissions, in ambient air, and in homes, with an emphasis on characterizing hazardous metals. PM emissions from the University of Iowa power plant were examined during a transition away from coal. Biomass is a renewable fuel and when used in place of or co-fired alongside coal it directly reduces emissions of fossil CO2 and PM to the atmosphere. PM emissions were examined under two scenarios: the first was a stoker boiler that fired 100% renewable energy pellets. PM and metal emissions reduced by 59 and 80% respectively despite increases to polycyclic aromatic hydrocarbons (PAHs) that was attributed to emission of unburned carbon fuel. The second test was done in a circulating fluidized bed boiler firing 78% oat hulls, 17% coal, and 5% energy pellets. Decreases in PM, PAHs, and metal emissions were 32, 33, and 50%, respectively. Reductions in PM, PAHs, and metals when firing renewable fuels provide environmental advantages to local air quality while trying to eliminate the use of coal. Ambient air quality was examined downwind of the power plant. The levels of PM2.5 in Iowa City in 2016 ranged from 1.4-32.1 µg m-3 with an annual mean of 7.5 µg m-3. These levels are below the National Ambient Air Quality Standards (NAAQS) for 24-hours (35 µg m-3) and the annual average (12 µg m-3). On average, the analyzed metals accounted for 10.7 ± 5.3% of PM2.5. Potassium, calcium, and zinc were the most abundant metals (averaging 250 ± 10, 170 ± 40, and 11.2 ± 0.5 ng m-3, respectively) and are associated with geological sources and biomass burning. Metals associated with fossil fuel combustion such as arsenic, lead, or vanadium were observed at low levels (averaging 0.58 ± 0.01, 1.32 ± 0.03, and 0.19 ± 0.01 ng m 3, respectively). Harmful metals were at least an order of magnitude lower than the World Health Organization’s guideline concentrations in Iowa City PM2.5. Overall, the hazardous metals in ambient PM2.5 observed in Iowa City were not at concerning levels. The work in this chapter helps provide a metal speciation profile for future studies and these measurements can be used to assess future changes in PM2.5 metal concentrations. Indoor PM collected in homes of Eastern Iowa COPD patients was characterized for its metal content. COPD patients are at higher risk of developing respiratory infections, which cause acute exacerbations of COPD—the leading cause of mortality in COPD patients—and airborne PM increases risk of infection. From the 21 homes studied 6-87 mg of indoor PM was collected over 30 days in winter. Crustal metals such as magnesium, iron, and aluminum were the most concentrated in indoor PM, having mass fractions from 3000-25,000 ng mg-1. Toxic metals like vanadium and arsenic were at lower levels from 3-65 ng mg-1. Between homes the relative abundance of metals was similar, but the absolute abundance was highly variable. Analyzing indoor PM is essential since Americans on average spend 80% of their day indoors making it a significant environment for PM exposure. This chapter provides a chemical profile for different homes when studying the impact PM has on respiratory health. This thesis also provides new measurements of the chemical composition of PM at the point of emission, in ambient air, and within homes. Advancing knowledge of PM composition in different environments is vital in understanding its impacts on human health. Improvements to local air quality with reductions to PM and metal emissions were observed when firing alternative fuels. Ambient PM2.5 concentrations in Iowa City were lower than NAAQS levels and the composition of both ambient and indoor PM was evaluated. With greater understanding of PM composition, better control strategies can be studied and employed to improve local air quality.

This research uses air pollution data, information about the locations of the air quality monitoring stations, and political events data from 2016 to 2017 to investigate the possible existence and characteristics of China’s routine short-term environmental policies aimed for ambient air quality. The results of multiple difference-in-difference suggest that air quality in the urban area, where ambient air pollution levels are usually higher due to population density, is lower on diplomatic events dates, while there is limited evidence that air quality is better in the urban area on domestic events dates. Therefore, this research reveals that there are short-term policies taking place on diplomatic events dates.

Multi-scale tracer and full-chemistry simulations with the STEM atmospheric chemistry model have been used to analyze the effects of both transported and local production of pollutants on California air quality during the ARCTAS-CARB experiment conducted in June 2008. During this summer experiment, simulated and observed O3 transport patterns from the coast to inland northern California are shown to vary based on the meteorological conditions and the oceanic O3 profiles, which are strongly episodically affected by Asian inflows. During a specific period, high coastal O3 air-masses at altitudes ~2-4 km can be transported inland and can significantly influence the surface O3 20-30 hours later over northern Sacramento valley, and the southern California can be indirectly affected by in-state transport. The model performance was improved by using LBCs downscaled from RAQMS global model that assimilated satellite data, as well as the LBC based on NASA DC-8 airborne observations during the experiment. The effects of SOx in these transported Asian air-masses over California were relatively less strong than O3 and its precursors, and the local emissions acted as the major contributor to the elevated sulfur concentrations below 5 km. Several SOx emission inventories (EI) were compared and the simulated SOx were validated with various observational datasets, with special focus on three regions - South Coast, San Francisco and Central Valley. The resolutions and the spatial and/or temporal variations of SOx emissions in all EIs remain to be further improved. Both terrestrial and maritime emissions were found important to coastal SOx distributions. Their percentile contributions to the coastal SOx spatial distributions for the experiment week were estimated, and the absolute contributions during flight periods were quantified. The California-Mexico atmosphere interaction majorly occurred between two sister city pairs-the coastal city pair (San Diego-Tijuana) was heavily affected by both California and Mexico maritime emissions through both directions. The effect was extended to the inland city pair (Mexicali-Calexico) by westerly winds, where was shortly impacted by southerly winds from Mexico, and limited areas over southeast of California were also affected.

Thesis (M.S.)--University of Nevada, Reno, 2006. / "May 2006." Includes bibliographical references (leaves 40-42). Online version available on the World Wide Web.

Fine and coarse particles were collected for eight weeks during the summer of 1991, at a coastal site (Yaquina Head) and a non-industrial site (Corvallis) in Western Oregon to characterize the aerosol composition and evaluate whether the sites are appropriate for sampling "background" marine air. Concentrations of up to 11 species (S0₄²⁻, NO₃⁻, Cl⁻, Na, Fe, Ni, Pb, Cr, Co, Sb, and CH₃SO₃H) for 95 samples were determined using four chemical analysis techniques. The influences of seasalt and soil dust were identified by analyzing concentrations of Na and Fe in the aerosol samples. Relative elemental composition in fine and coarse fractions indicated that the aerosol composition at Yaquina Head was greatly affected by seasalt. "Seasalt" enrichment factors (relative to Na) indicated that seasalt is the only source of Cl⁻ and SO₄²⁻ in coarse particles at Yaquina Head. In contrast, the seasalt influence was relatively weak at the Corvallis site. "Crustal" enrichment factors suggested that soil dust was not a major source of Na⁺, Cl⁻, or S0₄²⁻ at either site. A simple conceptual model that relies on meteorological conditions was used to identify sampling periods with long range transport from either marine or continental areas as well as local influences. This model suggested that during 61% of the experiment period the aerosols were advected from marine areas. At Yaquina Head, 52% of the sampling periods are associated with the "clean" background air (marine air with no local influences). Thus, Yaquina Head represents a useful location for collecting marine background air from the Pacific Ocean. The chemical composition of the marine background air collected at the Yaquina Head site is similar to that for other remote sites around the world. At Corvallis, "clean" marine background air can occasionally (21%) be collected even though Corvallis is located 64 km from the ocean. / Graduation date: 1993

Atmospheric formaldehyde (H2CO) is an intermediate product common to the degradation of many volatile organic compounds and therefore it is a central component of the tropospheric chemistry. While the global formaldehyde background is due to methane oxidation, emissions of non-methane volatile organic compounds (NMVOCs) from biogenic, biomass burning and anthropogenic continental sources result in important and localised enhancements of the H2CO concentration. Recent spaceborne nadir sensors provide an opportunity to quantify the abundance of tropospheric formaldehyde at the global scale, and thereby to improve our knowledge of NMVOC emissions. This is essential for a better understanding of the processes that control the production and the evolution of tropospheric ozone, a key actor in air quality and climate change, but also of the hydroxyl radical OH, the main cleansing agent of our troposphere. For this reason, H2CO satellite observations are increasingly used in combination with tropospheric chemistry transport models to constrain NMVOC emission inventories in so-called top-down inversion approaches. Such inverse modelling applications require well characterised satellite data products consistently retrieved over long time periods. This work reports on global observations of formaldehyde columns retrieved from the successive solar backscatter nadir sensors GOME, SCIAMACHY and GOME-2, respectively launched in 1995, 2002 and 2006. The retrieval procedure is based on the differential optical absorption spectroscopy technique (DOAS). Formaldehyde concentrations integrated along the mean atmospheric optical path are derived from the recorded spectra in the UV region, and further converted to vertical columns by means of calculated air mass factors. These are obtained from radiative transfer simulations, accounting for cloud coverage, surface properties and best-guess H2CO profiles, the latter being derived from the IMAGES chemistry transport model. A key task of the thesis has consisted in the optimisation of the H2CO retrieval settings from multiple sensors, taking into account the instrumental specificities of each sounder. As a result of these efforts, a homogeneous dataset of formaldehyde columns covering the period from 1996 to 2010 has been created. This comes with a comprehensive error budget that treats errors related to the spectral fit of the columns as well as those associated to the air mass factor evaluation. The time series of the GOME, SCIAMACHY and GOME-2 H2CO observations is shown to be consistent and stable over time. In addition, GOME-2 brings a significant reduction of the noise on spatiotemporally averaged observations, leading to a better identification of the emission sources. Our dataset is used to study the regional formaldehyde distribution, as well as its seasonal and interannual variations, principally related to temperature changes and fire events, but also to anthropogenic activities. Moreover, building on the quality of our 15-year time series, we present the first analysis of long-term changes in the H2CO columns. Positive trends, in the range of 1.5 to 4% yr-1, are found in Asia, more particularly in Eastern China and India, and are related to the known increase of anthropogenic NMVOC emissions in these regions. Finally, our dataset has been extensively used in several studies, in particular by the BIRA-IASB modelling team to constrain NMVOC emission fluxes. The results demonstrate the high potential of satellite data as top-down constraint for biogenic and biomass burning NMVOC emission inventories, especially in Tropical ecosystems, in Southeastern Asia, and in Southeastern US. / Le formaldéhyde (H2CO) joue un rôle central dans la chimie de la troposphère en tant que produit intermédiaire commun à la dégradation chimique de la plupart des composés organiques volatils dans l’atmosphère. L’oxydation du méthane est responsable de plus de la moitié de la concentration moyenne globale du formaldéhyde. Sur les continents en revanche, les hydrocarbures non-méthaniques (NMVOCs) émis par la végétation, les feux de biomasse et les activités humaines, augmentent de façon significative et localisée la concentration de H2CO. Les récents senseurs satellitaires à visée nadir offrent la possibilité de quantifier à l’échelle globale l’abondance du formaldéhyde dans la troposphère et de ce fait, d’améliorer notre connaissance des émissions de NMVOCs. Ceci est essentiel à la compréhension des mécanismes contrôlant la production et l’évolution de l’ozone troposphérique, élément clé pour la qualité de l’air et les changements climatiques, mais aussi du composé hydroxyle OH, le principal agent nettoyant de notre troposphère. C’est pourquoi, une méthode de plus en plus répandue pour améliorer les inventaires d’émissions des NMVOCs consiste en l’utilisation d’observations satellitaires de H2CO en combinaison avec un modèle de chimie et de transport troposphérique, dans une approche appelée modélisation inverse. Ce genre d’application demande des produits satellitaires bien caractérisés et dérivés de façon cohérente sur de longues périodes de temps. Le travail présenté dans ce manuscrit porte sur l’inversion des colonnes de formaldéhyde à partir de spectres de la radiation solaire rétrodiffusée par l’atmosphère terrestre, mesurés par les senseurs GOME, SCIAMACHY et GOME-2, lancés successivement en 1995, 2002 et 2006. La méthode d’inversion est basée sur la spectroscopie d’absorption optique différentielle (DOAS). Les concentrations de formaldéhyde intégrées le long du chemin optique moyen dans l’atmosphère sont dérivées à partir des spectres mesurés, et ensuite transformées en colonnes verticales par le biais de facteurs de conversion appelés facteurs de masse d’air. Ces derniers sont calculés à l’aide d’un modèle de transfert radiatif, en tenant compte de la présence de nuages, des propriétés de la surface terrestre et la distribution verticale supposée du formaldéhyde, fournie par le modèle IMAGES. Un des objectifs principaux de la thèse a été d’optimiser les paramètres d’inversion pour H2CO, et ceci pour les trois senseurs, tout en tenant compte des spécificités de chaque instrument. Ces efforts ont conduit à la création d’un jeu de données homogène, couvrant la période de 1996 à 2010. Les colonnes sont fournies avec un bilan d’erreur complet, incluant les erreurs liées à l’inversion des concentrations dans les spectres, ainsi que celles provenant de l’évaluation des facteurs de masse d’air. La série temporelle des observations de GOME, SCIAMACHY et GOME-2 présente une bonne cohérence et stabilité sur toute la période. Nous montrons aussi que la meilleure couverture terrestre de GOME-2 entraîne une réduction significative du bruit sur les observations moyennées, permettant une meilleure identification des sources d’émission. Notre jeu de données est exploité pour étudier la distribution régionale du formaldéhyde, ainsi que ses variations saisonnières et interannuelles, principalement liées aux variations de température et aux feux de végétation, mais aussi aux activités anthropiques. De plus, en s’appuyant sur la qualité de la série temporelle de 15 ans, nous présentons la première analyse des variations à long terme des concentrations de H2CO. Des tendances positives, de l’ordre de 1.5 à 4% par an, sont observées en Asie, en particulier dans l’est de la Chine et en Inde, liées à l’augmentation des émissions anthropiques d’hydrocarbures dans ces régions. Finalement, nos données ont été largement exploitées par le groupe de modélisation de l’IASB pour faire des études de modélisation inverse des émissions de NMVOCs. Les résultats démontrent le haut potentiel des données satellitaires pour contraindre les inventaires d’émissions dues à la végétation et aux feux de biomasse, particulièrement dans les écosystèmes tropicaux, en Asie du sud-est, et dans le sud-est des Etats-Unis.

The issue of odour, greenhouse gas emissions and indoor air quality in swine buildings have become a great concern for the neighbouring communities as well as governments. Air dispersion models have been adopted widely as an approach to address these problems which determine science-based distance between livestock production site and neighbours. However, no existing model considers the diurnal and seasonal variations of odour, gas (ammonia, hydrogen sulphide, greenhouse gas), and dust concentrations and emissions, which may cause great uncertainty. The primary objective of this project is to monitor and model the diurnal and seasonal variations of odour, gases, and dust concentrations and emissions from nursery, farrowing, and gestation rooms. Additionally, this study tried to quantify the greenhouse gas contribution from swine buildings and evaluate the indoor air quality of swine barns. <p>Strip-block experimental design was used to measure the diurnal variation of odour and gas concentrations and emissions in PSC Elstow Research Farm. It was found that: 1) odour and gas concentrations in winter were significantly higher than those in mild and warm weather conditions for all three rooms (P<0.05); 2) the nursery room had higher level of odour and gas concentration and emission than the other two types of rooms, no significant difference existed between the farrowing and gestation rooms (P>0.05); 3) significant diurnal variations occurred in August and April (P<0.05) for odour and some gas concentrations and emissions, while no significant diurnally variations were found in February (P>0.05); 4) apparent diurnal variation patterns were observed in August and April for NH3, H2S and CO2 concentrations, being high in the early morning and low in the late afternoon; 5) positive correlation was found between odour concentrations and NH3, H2S, and CO2 concentrations, respectively. <p>A whole year ( August 2006 to July 2007) monitoring of odour, gas and dust concentrations and emissions revealed that: 1) significant seasonal effect on odour and gas concentrations and emissions, total dust concentrations and dust depositions were observed (P<0.05), but no specific variation pattern was discovered for odour and gas emissions; 2) the total greenhouse gas emission from all the rooms in the gestation, nursery and farrowing area was 2956 CO2 equivalent tons per year, where gestation area, nursery area, and farrowing area accounted for 39.3 %, 37.2% and 23.5%, respectively; the CO2 emission contributed 53.4% to the total greenhouse emission, and CH4 contributed to 43.9%, 2.7% for N2O; N2O could be considered negligible; 3) indoor air quality of the swine barn met the requirements set by the Occupational Health and Safety Regulations (1996) of Saskatchewan for NH3, H2S, and CO2. <p>Statistical models were developed for each type of room to predict the odour and gas concentrations and emissions based on four variables: ventilation rate, room temperature, ambient temperature, and animal unit. The predicted results showed agreeable with measured values for most models (R2 = 0.56-0.96). Generally, gas prediction models performed better (R2=0.61-0.96) than odour prediction models (R2=0.56-0.85).<p>This study was conducted in the province of Saskatchewan throughout one year and the results could be used as representative data for Canada Prairies. Due to the large diurnal and seasonal variabilities of odour emissions, it was recommended to take multiple measurements of odour emission rate under different weather conditions in order to improve the accuracy of air dispersion modeling.

The regulated control of pollutants is essential to maintaining good air quality in urban areas. A major concern is the formation of tropospheric ozone, which can be especially harmful to those with lung conditions and has been linked to the occurrence of asthma. Ozone is formed through reactions of oxidized volatile organic compounds with nitrogen oxides, and the accurate modeling of the process is necessary for smart and effective regulations. Ambient measurements are important to understanding the mechanisms involved in tropospheric chemistry. This dissertation describes the characterization of a novel instrument for the ambient measurement of dinitrogen pentoxide, N2O5, and the results of several field studies. This is an important intermediate in the major nighttime loss pathway of nitrogen oxides. The understanding of this process requires correct modeling formation, as any nitrogen oxides not removed at night will result in increased ozone formation at sunrise. Calibration studies have been performed in order to quantify the loss of reactive species within the instrument, and the sampling flow and N2O5 detection have been well characterized. The results of the laboratory measurements are presented. Results are presented from the SHARP Field Study in Houston, TX in the spring of 2009. N2O5 measurements are compared to measurements of other species, including nitric acid and nitryl chloride, which were performed by other research groups. Mixing ratios exceeding 300 ppt were observed following ozone exceedance days, and a dependence of the concentration on both wind speed and direction was noticed. There was a strong correlation determined between N2O5 with HNO3 and ClNO2 indicating both a fast heterogeneous hydrolysis and N2O5 as the primary source of the species. Observed atmospheric lifetimes for N2O5 were short, ranging from several seconds to several minutes. We have also investigated the presence of N2O5 in College Station, TX. Low mixing ratios peaking at approximately 20 ppt were observed, with longer atmospheric lifetimes of up to several hours. The role of biogenic emissions in the NO3-N2O5 equilibrium is discussed.

When modelling pollutants in the atmosphere, it is nearly impossible to get perfect results as the chemical and mechanical processes that govern pollutant concentrations are complex. Results are dependent on the quality of the meteorological input as well as the emissions inventory used to run the model. Also, models cannot currently take every process into consideration. Therefore, the model may get results that are close to, or show the general trend of the observed values, but are not perfect. However, due to the lack of observation stations, the resolution of the observational data is poor. Furthermore, the chemistry over large bodies of water is different from land chemistry, and in North America, there are no stations located over the great lakes or the ocean. Consequently, the observed values cannot accurately cover these regions. Therefore, we have combined model output and observational data when studying ozone concentrations in north eastern North America. We did this by correcting model output at observational sites with local data. We then interpolated those corrections across the model grid, using a Kriging procedure, to produce results that have the resolution of model results with the local accuracy of the observed values. Results showed that the corrected model output is much improved over either model results or observed values alone. This improvement was observed both for sites that were used in the correction process as well as sites that were omitted from the correction process.