<p>Aerosols are solid or liquid particles that span in size from a few nanometers to tens of micrometers. They are important in both outdoor and indoor environments due to their impact on climate and human health. Many aerosol processes of interest to the environment, health, and filtration are strongly size-dependent. Aerosol particle size distributions (PSDs) provide a basis to better understand outdoor and indoor air pollution sources, evaluate human exposure to air pollution, and aerosol deposition in the human respiratory tract and filters in building ventilation systems. Monitoring the temporal evolution of aerosol PSDs enable for characterization of dynamic aerosol transport and transformation processes, such as direct emissions, nucleation, condensation, coagulation, resuspension, deposition, and filtration. This thesis presents three studies on the PSDs of: (i.) urban aerosols in cities around the world in order to identify geographical trends in the shape and magnitude of PSDs and to frame future research needs for PSD observations at a global-scale, (ii.) synthesized salt particles from a novel thermal aerosol generator to evaluate its suitability for air filter testing, and (iii.) indoor biological particulate matter (<i>bio</i>PM) to characterize transient inhalation exposures of infants and adults to resuspended <i>bio</i>PM from carpet dust induced by crawling and walking.</p> <p>First, this thesis presents the current state-of-knowledge of urban aerosol PSDs by critically analyzing <i>n</i>=793 urban aerosol PSD observations made between 1998 to 2017 in <i>n</i>=125 cities in <i>n</i>=51 countries across the following regions of the world: Africa (AF), Central, South, and Southeast Asia (CSSA), East Asia (EA), Europe (EU), Latin America (LA), North America, Australia, and New Zealand (NAAN), and West Asia (WA). Prominent geographical variations in the shape and magnitude of urban aerosol number and mass PSDs were identified and significant variations in number PSDs were found between cities in EU and NAAN with those in CSSA and EA. Most PSD observations published in the literature are short-term, with only 14% providing data for longer than six months. There is a paucity of PSDs measured in AF, CSSA, LA, and WA, demonstrating the need for long-term aerosol measurements across wide size ranges in many cities around the globe. Inter-region variations in PSDs have important implications for population exposure, driving large differences in the urban aerosol inhaled deposited dose rate received in each region of the human respiratory system. Similarly, inter-region variations in the shape of PSDs impact the penetration of urban aerosols through filters in building ventilation systems, which serve as an important interface between the outdoor and indoor atmospheres. The results of this critical review demonstrate that global initiatives are urgently needed to develop infrastructure for routine and long-term monitoring of urban aerosol PSDs spanning the nucleation to coarse modes.</p> <p>Second, this thesis evaluates a newly designed commercially available thermal aerosol generator for ageing air filters in building heating, ventilation, and air-conditioning (HVAC) systems. The physical characteristics of the synthesized salt aerosol (NaCl and KCl) under different generator operational conditions were evaluated. The shape of the number and mass PSD output of the thermal aerosol generator are similar to those found in outdoor (urban) and indoor air and can be modulated by varying the rate at which the salt stick is fed into the flame. The morphology of the NaCl and KCl particles varied with size, with compact spherical or cubic structures observed below 100 nm and agglomerates observed above 100 nm. The thermal aerosol generator is a cost-effective technique for rapid ageing of HVAC filters with a PSD that more accurately represents, compared to conventional loading dusts, what filters encounter in real HVAC installations.</p> <p>Lastly, this thesis characterizes infant and adult inhalation exposures and respiratory tract deposited dose rates of resuspended <i>bio</i>PM from carpets. Chamber experiments were conducted with a robotic crawling infant and an adult performing a walking sequence. Breathing zone (BZ) size distributions of resuspended fluorescent biological aerosol particles (FBAPs), a <i>bio</i>PM proxy, were monitored in real-time. FBAP exposures were highly transient during periods of locomotion. Both crawling and walking delivered a significant number of resuspended FBAPs to the BZ, with concentrations ranging from 0.5-2 cm^-3. Infants and adults are primarily exposed to a unimodal FBAP size distribution between 2 and 6 μm, with infants receiving greater exposures to super-10 μm FBAPs. In just one minute of crawling or walking, 10²-10³ resuspended FBAPs can deposit in the respiratory tract, with an infant receiving much of their respiratory tract deposited dose in their lower airways. Per kg body mass, an infant will receive nearly four times greater respiratory tract deposited dose of resuspended FBAPs compared to an adult.</p>

<p>Urban meteorology has developed in parallel to other sub-fields in the science, but in many ways remains poorly described. In particular, the study of urban rainfall modification remains behind compared to other comparable features. Urban rainfall modification refers to the change of a precipitation feature as it crosses an urban area. Typically, this manifests as rainfall initiation, local suppression, local invigoration, and/or storm morphology changes. Research in the prior decades have shown urban rainfall modification to arise from a combination of land-atmosphere and aerosol-cloud interaction. Urban areas create a greater surface roughness, which produces local convergence and divergence, modifying local thunderstorm inflow and morphology. The land surface also generates vertical velocity perturbations which can act to initiate or modify existing convection. Urban aerosols act as CCN to perturb existing cloud and precipitation characteristics. Higher CCN narrows the cloud droplet distribution, creating more smaller cloud droplets, and initially reducing precipitation efficiency by keeping more liquid water in the cloud than what would form into rain. The CCN-cloud interaction eventually increasing heavy rainfall production as graupel riming is enhanced by the narrower cloud droplet distribution, leading to more larger raindrops and higher rain in areas.</p><p>This dissertation addresses the observation and modeling of urban thunderstorm interaction from both the land surface and aerosol perspective. It reassesses the original urban rainfall anomaly: The La Porte Anomaly. First analyzed in the late 1960s, the La Porte Anomaly was ultimately dismissed by 1980 as either a temporary, biased, or otherwise unexplainable observation, as the process level understanding had yet to be explained. The contemporary analysis utilizes all existing data and objective optimal interpolation to show that a rainfall anomaly downwind of Chicago has indeed existed at least since the 1930s. The current rainfall anomaly exists as a broad region of warm season rainfall downwind of Chicago that is 20-30% greater than the regional average. Using synoptic parameters, the rainfall anomaly is shown to be independent of wind direction and most closely associated with local land surface forcing. Weekdays, where local aerosol loading has been measured at 40% or more greater than weekends, have up to 50% more warm season rainfall than weekends. The analysis is able to show that there is a land surface and aerosol contribution to the rainfall anomaly, but cannot unambiguously separate them.</p><p>In order to separate the land surface and aerosol effects on urban rainfall distribution, a numerical model was improved to better handle urban weather interaction. The Regional Atmospheric Modeling System (RAMS 6.0) was chosen for its base land surface and cloud physics parameterization. The Town Energy Budget (TEB) urban canopy model was coupled to RAMS to handle the urban land surface. The Simple Photochemical Module (SPM) was coupled with the cloud physics to handle conversion of surface emissions to CCN. The model utilized an external traffic simulation to create a realistic diurnal and weekly cycle of surface emissions, based on human behavior. The new Urban RAMS was used to study the land surface sensitivity of city size and of aerosol loading in two studies using the Real Atmosphere Idealized Land surface (RAIL) method, by which all non-urban features of the land surface are removed to isolate the urban effects. The city size study determined that the land surface of a given city eventually has a maximum effect on thunderstorm modifying potential, and that rainfall does not continue to increase or decrease locally for cities larger than a certain size based on that storm’s own motion. The aerosol-cloud analysis corroborated previous observations on the non-linear effects of aerosol loading on clouds. It also demonstrated that understanding the aerosol effect in an urban environment requires high resolution observations of precipitation change. In a single thunderstorm, regions can be both impacted by local rainfall rate increases and decreases from urban aerosols, leading to little total change in precipitation. But the rainfall rate changes can significantly affect soil moisture and drought potential in and around urban areas.Following the idealized studies, the historical and current La Porte Anomaly was simulated to separate the land surface from the aerosol factors near the Chicago area. The Urban RAMS model was deployed on a real land surface with full model physics. Simulations with 1932, 1962, 1992, and 2012 land covers were run over an exceptionally wet Aug. 2007 to approximate the rain variability for an entire summer season. Surface emissions were also varied in the 2012 land cover for variable aerosol loading. The simulations successfully reproduced the location of the downwind rainfall anomaly in each land cover scenario: farther east toward La Porte in 1932, moving southwestward to its current location by 2012. Doubling surface emissions eliminated the downwind anomaly, as was observed during the highest pollution decade of the 1970s. Eliminating surface emissions also decreased the downwind anomaly. As the land cover at the upwind edge of Chicago became more connected from the 1932 to 2012 land cover scenarios, a local upwind rainfall anomaly developed, moving westward with urban expansion. The results of these simulations enabled the conclusions that a) at the upwind edge, the land surface dominates urban rainfall modification, b) the aerosol loading sustains and increases the locally downwind rainfall increase, and c) that the total modification distance is static on given day and given urban footprint. A more expansive city does not produce a rainfall anomaly more distantly downwind, but rather the distance of rainfall modification moves to where the upwind edge of the city begins.</p><p></p><p>The modeling work ends with a two-city simulation in the southeast United States, of a bow-echo forming near Memphis, TN and crossing Birmingham, AL before splitting. Simulations were performed on different surface emissions rates, land covers where Birmingham did not exist, and a novel approach with two inner emitting grids over both Birmingham and Memphis. A storm tracking algorithm enabled one-to-one comparisons of point simulated storm characteristics between scenarios. The results of most scenarios only corroborated previous research, showing how increased aerosol loading changes cloud and rainfall characteristics until the highest aerosol loading shuts down riming and rainfall enhancement. However, the two most accurate simulations, where the storm forms and splits over Birmingham, were a non-urban higher rural aerosol scenario and the scenario with Memphis also emitting pollution. In order to split the storm over Birmingham, the upwind cloud characteristics were primed by higher upwind aerosols, either from a realistic city upwind or unrealistically high rural aerosols. The conclusions produced by this study demonstrated the importance of aerosol cloud interaction, perhaps equal with land surface, but also the need for far upwind information for a storm in a given city. Memphis and Birmingham are separated by over 300km, far exceeding the threshold thought to connect two cities by mutual rainfall modification.</p><p>The overall conclusions of the research presented in this dissertation shows a more unified approach to the effects of urban rainfall modification. The upwind edge of a city is a fixed location, and a thunderstorm begins modifying at that point. The thunderstorm usually produces a local rainfall maximum at the upwind edge, due to the vertical velocity of the urban land surface. The urban aerosols proceed to narrow the cloud droplet distribution, locally reducing rainfall as the storm passes over the urban area. Eventually the enhanced rainfall from enhanced riming produces a maximum somewhere downwind. However, “downwind” is a location relative to the storm’s motion and could exist anywhere over the urban footprint or downwind in a rural region. The climatological location of increased rainfall is an average of every storm in a season and beyond. The results of each part of the study provide a way to continue the research presented here.</p><br>

Atmosphärische Aerosolpartikel können durch ein breites Spektrum an natürlichen oder anthropogenen Emissionen mit unterschiedlich hohen Konzentrationen in die Atmosphäre freigesetzt werden. Sie beeinflussen den Strahlungshaushalt und damit auch das Klima der Erde und können außerdem durch ihre Präsenz in der Atmosphäre Wechselwirkungen mit Mensch und Natur, also dem gesamten Ökosystem haben. Seit dem Jahr 2010 gelten in der EU Grenzwerte für die PM10 Tagesmittelkonzentration, die jedoch bereits wenige Monate nach Beginn der Gültigkeit an vielen Messstationen überschritten wurden. Das Ziel der vorliegenden Arbeit war eine objektive Bewertung der Herkunft und des Zustandes der an einem Messort ankommenden Luftmasse und der damit verbundenen Schadstoffniveaus. Im ersten Teil der Arbeit wurden PM10 Messdaten aus fünf Jahren in und um Leipzig sowie analog in fünf verschiedenen Regionen deutschlandweit in Bezug auf PM10 Grenzwertüberschreitungen untersucht. Es wurden Rückwärtstrajektorien für eine Clusteranalyse verwendet, mit der spezifische Wetterlagen unterschieden wurden und diesen dann die einzelnen Messtage mit den zugehörigen Schadstoffkonzentrationen zugeordnet wurden. Hierbei wurde deutlich, dass durch entsprechende meteorologische Bedingungen sowohl lokal als auch regional emittierte Schadstoffe in Bodennähe akkumulieren oder aber auch räumlich verteilt werden können. Außerdem wurde in dieser Arbeit eine Modellvalidierung vorgestellt. Es wurde das Modellsystem COSMO-MUSCAT/ext-M7 verwendet, dessen Ergebnisse mit Beobachtungsdaten verglichen wurden. Als erstes wurde die Beschreibung der meteorologischen Bedingungen, dann die räumliche Verteilung von PM10, die chemische Partikelzusammensetzung sowie physikalische Aerosolparameter wie Partikelanzahl, -volumen und -durchmesser verglichen. Die Ergebnisse deuten darauf hin, dass nach wie vor Probleme bei der Beschreibung der mikrophysikalischen Aerosoleigenschaften bestehen. Die Größenordnungen der verglichenen Parameter können vom Modell abgebildet werden, dennoch müssen Modellergebnisse nach wie vor mit Vorsicht interpretiert werden, insbesondere hinsichtlich von Prozessen, bei denen die Partikelanzahl eine Rolle spielen.

Constructing a detailed understanding of the heterogeneous oxidation of atmospheric organic aerosols, both from a mechanistic and kinetic perspective, will enable researchers to predict the fate and lifetime of atmospheric gases and the particles with which they interact. In an effort to develop a more complete understanding of the interfacial reactions of ozone with vinyl-containing organic thin films, self-assembled monolayers that contain vinyl groups positioned precisely at the gas/surface interface were synthesized as model systems for atmospheric organic aerosols. To isolate the reactions of background gases with ozone or surface products, an ultrahigh vacuum surface analysis instrument was designed and constructed to explore the reactions of ozone with the atmospheric model systems. The surface reactions can be monitored in real-time with reflection absorption infrared spectroscopy (RAIRS) and mass spectrometry. The chemical identity of adsorbates on a surface can also be determined before or after a reaction with X-ray photoelectron spectroscopy (XPS). Disordering of the monolayers concurrent with the disappearance of the vinyl group was observed with RAIRS. New bands within the RAIR spectra were observed and assigned to carbonyl or carboxylic acids bound to the surface. Little oxidation of the sulfur head groups and no significant loss of carbon during the reaction was observed with XPS. A mechanism is proposed that includes the cross linking of the hydrocarbon chains within the monolayer, which impedes further oxidation of the sulfur head group and limits desorption of the chains. By RAIRS, the kinetics of the oxidation of the vinyl groups were tracked and an observed rate constant was determined by monitoring the changes in IR absorbance of the C=C bond. With the aid of the rate constant, an initial reaction probability for the collisions of ozone with vinyl groups positioned precisely at an interface was determined. The reaction probability is approximately three orders of magnitude greater than the reaction probability for an analogous gas-phase reaction, which demonstrates that the gas/surface interface plays an important role in this reaction. The results presented in this thesis should help develop a more detailed understanding of the interfacial reactions of pure ozone at surfaces. / Ph. D.

<div> <div> <div> <p>The mid-Miocene Climatic Optimum (MMCO), 17-14.50 million years ago, is studied using general circulation models (GCMs). This period of time is characterized by enhanced warming in the deep ocean and in the mid-to-high latitudes. Previous GCMs fail to accurately represent the warmer climate of the MMCO, providing evidence that other warming feedbacks are missing in the models. This study focuses on cloud feedbacks by modifying the Community Earth System Model (CESM 1.0) to explore the MMCO climate. We implement modifications in pre-industrial (284.7 ppm CO2) and modern slab ocean cases (367.0 ppm CO2, 400 ppm CO2, and 800 ppm CO2). One modified case showing the most potential implements an aerosol de- pendent ice nucleation mechanism and a theory based cloud phase separation. This modified case allows the model predicted aerosol concentrations to interact with the cloud microphysics and provide more realistic cloud water contents. The data shows an increase in surface temperature and increase in upper atmospheric cloud fraction when compared to the control case. Preliminary results suggest that this model is able to capture the mid-to-high latitude warming trends and weaker equator to pole temperature gradient. </p> </div> </div> </div>

Nowadays, megacities all over the world are facing air quality issues, especially regarding the control of secondary pollutants such as tropospheric ozone (O3) but mainly fine particles (PM2.5 ; 2.5 m in diameter), as they have important impacts on both human health and climate change. Understanding the evolution of these particles in the atmosphere requires the description of emission sources as well as the physicochemical processes involved in their formation, growth and removal. In this study, the Weather Research and Forecasting with Chemistry (WRF-Chem) community model, a state-of-the-art coupled meteorology-chemistry modelling system, along with experimental data collected during the Narrowing the Uncertainties on Aerosol and Climate Change in São Paulo State (NUANCE-SPS, FAPESP thematic project) campaigns performed in 2012 and 2014, were used in order to examine the main properties of atmospheric aerosol particles over the Metropolitan Area of São Paulo (MASP), in southeastern Brazil, where changes in fuel blend and consumption in recent years have affected the evolution of pollutant concentrations. The combined application of aerosol data and WRF-Chem simulations made it possible to represent some of the most important aerosol properties such as particle number concentration (PNC) and cloud condensation nuclei (CCN) activation, besides allowing us to evaluate the contributions of anthropogenic and biomass burning sources to the PM2.5 loadings in the MASP. For instance, on-road vehicles have a potential to form new particles between 20 and 30 % in relation to the total PM2.5 mass, whereas biomass burning, on average, accounted for 824 % (515 g m3 ) of it. In addition, biomass burning accounted for up to 20 % of the baseline PNC- and CCN-weighted relative differences over the MASP (2300 cm3 and 1400 cm3 , respectively). The results indicate the potential importance of biomass burning sources for air quality in the MASP, and underscore the need for more accurate representations of aerosol emissions to reduce uncertainties in the model predictions. / Atualmente, megacidades ao redor do mundo enfrentam problemas de qualidade do ar, especialmente aqueles relacionados ao controle de poluentes secundários como o ozônio troposférico (O3), mas principalmente partculas finas (PM2.5 ; 2.5m em diâmetro). As partculas finas têm impactos significativos na saúde humana bem como no clima, através das mudanças climáticas. O entendimento da evolução destas partculas na atmosfera requer a descrição de suas fontes de emissão e também dos processos fsico-qumicos envolvidos na sua formação, crescimento e remoção. Neste estudo, o modelo Weather Research and Forecasting with Chemistry (WRF-Chem), um sistema de modelagem atmosférica estado-da-arte, juntamente com dados coletados durante duas campanhas experimentais realizadas em 2012 e 2014 no âmbito do projeto Narrowing the Uncertainties on Aerosol and Climate Change in São Paulo State (NUANCE-SPS, projeto temático FAPESP), foram usados para analisar as principais propriedades do aerossol atmosférico na Região Metropolitana de São Paulo (RMSP), no sudeste do Brasil, onde mudanças na composição e no consumo dos combustveis têm afetado a evolução da concentração de poluentes. A aplicação combinada de dados experimentais e simulações numéricas com o WRF-Chem permitiu a representação de algumas das propriedades mais importantes do aerossol atmosférico como concentração do número de partculas e ativação de núcleos de condensação de nuvens, assim como a avaliação da contribuição de fontes antropogênicas e de queimadas na concentração do PM2.5 . Por exemplo, fontes veiculares têm potencial para formar novas partculas entre 20 e 30 % em relação à massa total do PM2.5, enquanto fontes de queimadas contriburam, na média, entre 824 % (515 g m3 ) da massa total deste poluente. Outrossim, fontes de queimadas contriburam em até 20 % das concen- trações base do número de partculas e de núcleos de condensação de nuvens sobre a RMSP (2300 cm3 e 1400 cm3, respectivamente). Os resultados indicam o potencial impacto das queimadas na qualidade do ar na RMSP, e enfatizam a necessidade de aprimoramentos nos modelos de emissão de aerossol, visando reduzir incertezas nas previsões do modelo.

A significativa emissão veicular na Região Metropolitana de São Paulo (RMSP), com mais de 7 milhões de veículos e uma população da ordem de 18 milhões de habitantes, fazem desta uma área crítica do ponto de vista de níveis de poluentes atmosféricos. Neste trabalho foi obtida a determinação quantitativa de fontes de poluentes atmosféricos na RMSP, em particular do material carbonáceo na fração fina (PM2.5) do aerossol atmosférico, focando na componente veicular. Como parte do projeto FONTES, coordenado pela Petrobrás, PUC-Rio e IFUSP, foram operadas por 1 ano quatro estações de amostragem localizadas em Congonhas (CGH), Ibirapuera (IBP), Cerqueira César (FSP) e Instituto de Física da USP (IFP), no período entre agosto de 2011 e janeiro de 2014. A concentração em massa de material particulado fino (PM2.5), grosso (PM2.5-10), e inalável (PM10) foi determinada através de análise gravimétrica. Íons solúveis foram determinados por cromatografia iônica (IC), elementos traços por fluorescência de raios-X (XRF) e as concentrações de black carbon equivalente por refletância ótica. As componentes de carbono orgânico (OC) e carbono elementar (EC), bem como as diversas frações carbonáceas foram determinadas por análises termo-ópticas em equipamento da Sunset Inc., seguindo vários protocolos analíticos. As concentrações de gases tais como CO, NOx, e O3 foram fornecidas por estações de monitoramento da CETESB. Modelos receptores tais como APFA (Absolute Principal Factor Analysis) foram usados na determinação quantitativa de fontes de poluentes. Observou-se uma grande similaridade nas concentrações medidas nas estações, indicando uma homogeneidade nas concentrações e composição de aerossóis da moda fina (PM2.5) na RMSP. Nas estações amostradoras IFP, FSP e IBP foram observadas concentrações entre 10 e 12g m-3 na fração fina e na faixa de 16 a 18 g m-3 na fração grossa. Em CGH, observou-se uma concentração média de 34 g m-3, para PM10. O balanço químico de massa mostrou, na fração fina, impacto predominante de aerossóis orgânicos (~50%), EC (~20%) e sulfato (~20%). Na fração grossa verificaram-se concentrações dominantes de aerossóis de poeira do solo (~40%). A APFA identificou e quantificou o impacto das componentes veicular (~60%), ressuspensão de solo (~10%), emissões industriais e de sulfato (~20%), aerossol marinho (~5%) e aerossol secundário (~5%). O impacto da componente veicular é dominante na RMSP. A aplicação dos modelos receptores forneceu a caracterização do material carbonáceo de acordo com as fontes de emissões e um perfil de volatilidade do material carbonáceo. / The large vehicle fleet in the Metropolitan Region of São Paulo (RMSP), with more than 7 million vehicles and a population of about 18 million people, make this a critical area from the point of view of atmospheric pollutants levels. This work focused on the quantitative determination of air pollutant sources, focusing at the vehicular component and the carbonaceous material in the fine fraction (PM2.5) of the atmospheric aerosol of RMSP. As part of the FONTES research project, coordinated by Petrobrás, PUC-Rio and IFUSP, it was operated for 1 year four sampling stations located in Congonhas (CGH), Ibirapuera (IBP), Cerqueira César (FSP) and the Institute of Physics at USP (IFP) during the period from August 2011 to January 2014. The mass concentrations of fine (PM2.5), coarse (PM2.5-10) and inhalable (PM10) particulate matter was determined by gravimetric analysis. Soluble ions were determined by ion chromatography (IC), trace elements by X-ray fluorescence (XRF) and equivalent black carbon (EBC) concentration by optical reflectance. The organic carbon (OC) and elemental carbon (EC) components, as well as several carbonaceous fractions were determined by thermo-optical analysis using a Sunset Inc. equipment, following various analytical protocols. The concentration of gases such as CO, NOx, and O3 were obtained from CETESB air pollution monitoring stations. Receptors models such APFA (Absolute Principal Factor Analysis) were used for the quantification of the impacts of polluting sources. It was observed similar concentrations in the several sampling stations, showing uniformity in the concentrations and aerosol composition of PM2.5 in the RMSP. For the sites IFP, FSP and IBP were observed concentrations between 10 and 12 g m-3 in the fine fraction and in the range from 16 to 18 g m-3 in the coarse fraction. In the CGH site, there was an average concentration of 34 g m-3 of PM10. The chemical mass balance showed large presence of organic aerosols (~50%), EC (~20%) and sulfate (~20%) in the fine mode fraction. In the coarse fraction soil dust aerosols (~40%) dominates. The APFA identified and quantified the impact of vehicular components (~60%), soil dust (~10%), industrial emissions and sulfate (~20%), marine aerosol (~5%) and secondary aerosol (~5%). Vehicular emissions is the major air pollution component at the RMSP. The application of receptor models has provided the source characterization of carbonaceous material according to their volatility profile.

Open burning of biomass and anthropogenic waste is a major source of aerosols at the biosphere-atmosphere interface, yet its impact on Earth’s climate and air quality is not fully understood due to the intricate feedbacks between the natural environment and human activities. Earth system models (ESMs) are a vital tool in the study of these aerosol-biosphere-atmosphere interactions. ESMs allow the estimation of radiative forcing and climate impacts in terms of changes to temperature and precipitation as well as the attribution to natural or anthropogenic drivers. To provide coherent results, however, ESMs require rigorous development and evaluation against observations. In my work I use the NASA-GISS ESM: ModelE. One of its strengths lie in its detailed aerosol schemes that include microphysics and thermodynamic partitioning, both necessary for the simulation of secondary inorganic aerosols. To overcome one of ModelE’s weaknesses, namely its lack of interactive biomass burning (BB) emissions, I developed pyre: ModelE’s interactive fire emissions module. pyrE is driven by flammability and cloud-to-ground lightning, both of which are calculated in ModelE, and anthropogenic ignition and regional suppression parameterizations, based on population density data. Notably, the interactive fire emissions are generated from the flaming phase in pyrE (fire count), rather than the scar left behind (burned area), which is commonly used in other interactive fire modules. The performance of pyrE was evaluated against MODIS satellite retrievals and GFED4s inventory, as well as simulations with prescribed emissions. Although the simulated fire count is bias-high compared to MODIS, simulated fire emissions are bias-low compared to GFED4s. However, the bias in total emissions does not propagate to atmospheric composition, as pyrE simulates aerosol optical depth just as well as a simulation with GFED4s prescribed emissions. Upon the development and evaluation of the fire-aerosol capabilities of ModelE, I have utilized it, with the EVA health model, to study the health impacts of outdoor smoke in 1950, 2015, and 2050. I find that chronic exposure to aerosols (PM2.5) is the main driver of premature deaths from smoke exposure, yet by 2050, acute exposure to ozone, formed downwind of BB smoke plumes, is projected to cause more premature deaths than exposure to PM2.5. I estimate the annual premature deaths from BB and waste burning (WB) smoke in 1950 to be ~41,000 and ~19,000, respectively, and in 2015 to be ~310,000 and ~840,000, respectively. By 2050 I project 390,000 and 1.5 million premature deaths from BB and WB respectively. In light of the growing impact of WB smoke exposure I identify the need to scale up viable waste management practices in regions of rapid population growth.

The impacts of biomass burning smoke on local rainfall and the structure of the atmospheric boundary layer have been actively studied in recent years. However, whether the large-scale biomass burning in the later dry season over Amazonia Region could influence the dry-to-wet transition season have not been examined. Previous observations have shown that the substantial increase of rainfall from dry to wet season over Amazonia are actually caused by small changes of the atmospheric thermodynamic structure relative to those over other monsoon regions. Consequently, the onset date of wet season can vary greatly as influenced by external or internal anomalous forcings. Thus, it is possible that the transition of the atmospheric thermodynamic structure and circulation from dry to wet season is also sensitive to the impacts of biomass burning smoke. To test this hypothesis, we have forced RegCM3 model with direct radiative forcing of smoke inferred from MODIS for the transition season (August to November). The comparison with control run helps us to examine the direct and semi-direct influences of smoke on the transition from dry to wet season. Our preliminary results show that the direct and semi-direct forcings of smoke could significantly influence the rainfall and related atmospheric and land surface conditions during the transition. However, these changes are sensitive to the prescribed vertical distribution of the aerosols.

Anthropogenic atmospheric aerosols (suspended particulate matter) can modify the radiative balance (and climate) of the Earth by altering the properties and global distribution of clouds. Current climate models however cannot adequately account for many important aspects of these aerosol-cloud interactions, ultimately leading to a large uncertainty in the estimation of the magnitude of the effect of aerosols on climate. This thesis focuses on the development of physically-based descriptions of aerosol-cloud processes in climate models that help to address some of such predictive uncertainty. It includes the formulation of a new analytical parameterization for the formation of ice clouds, and the inclusion of the effects of mixing and kinetic limitations in existing liquid cloud parameterizations. The parameterizations are analytical solutions to the cloud ice and water particle nucleation problem, developed within a framework that considers the mass and energy balances associated with the freezing and droplet activation of aerosol particles. The new frameworks explicitly account for the impact of cloud formation dynamics, the aerosol size and composition, and the dominant freezing mechanism (homogeneous vs. heterogeneous) on the ice crystal and droplet concentration and size distribution. Application of the new parameterizations is demonstrated in the NASA Global Modeling Initiative atmospheric and chemical and transport model to study the effect of aerosol emissions on the global distribution of ice crystal concentration, and, the effect of entrainment during cloud droplet activation on the global cloud radiative properties. The ice cloud formation framework is also used within a parcel ensemble model to understand the microphysical structure of cirrus clouds at very low temperature. The frameworks developed in this work provide an efficient, yet rigorous, representation of cloud formation processes from precursor aerosol. They are suitable for the study of the effect of anthropogenic aerosol emissions on cloud formation, and can contribute to the improvement of the predictive ability of atmospheric models and to the understanding of the impact of human activities on climate.