This thesis aims to improve the understanding of El Niño Southern Oscillation (ENSO) diversity, in its future change, modeling and predictability. How might ENSO change in the warming climate? To reach a comprehensive understanding, a set of empirical probabilistic diagnoses (EPD) is introduced to measure the ENSO behaviors as to tropical Pacific sea surface temperature (SST) climatology, annual cycle, ENSO amplitude, seasonal phase locking, diversity in peak location and propagation direction, as well as the El Niño-La Niña asymmetry in amplitude, duration and transition. This diagnosis is applied to the observations, and consistency with previous studies indicates it is valid. Analysis of 37 CMIP5 model simulations for the 20th century and the 21st century shows that, other than the projected increase in SST climatology, changes in other aspects are largely model dependent and generally within the range of natural variation. The change favoring eastward propagating El Niños is the most robust seen in the SST anomaly field. To what extent can we trust the future projection? CMIP5 models show large spreads in terms of 20th century ENSO performance. So a data-driven approach called Empirical Model Reduction (EMR) is carried out, by fitting a low-dimensional nonlinear model from the observation with a representation of memory effect and seasonality. The stochastic simulation of EMR is able to reproduce a realistic ENSO diversity statistics and a reasonable range of natural variation, thus provides an additional benchmark to evaluate the CMIP5 model biases. What are the key model components leading to a good performance to simulate and predict ENSO? Using a suite of models under the aforementioned framework of EMR, control experiments are conducted to advance the understanding of ENSO diversity, nonlinearity, seasonality and the memory effects. Nonlinearity is found necessary to reproduce the ENSO diversity features by simulating the extreme El Niños. Nonlinear models reconstruct the skewed distribution of SST anomalies and improve the prediction of the El Niño-La Niña transition. Models with periodic terms reproduce the SST seasonal phase locking but do not improve the prediction appreciably. Models representing the ENSO memory effect, based on either the recharge oscillator (multivariate model with tropical Pacific subsurface information) or the time-delayed oscillator (multilevel model with SST history information), both improve the prediction skill dramatically. Models with multiple ingredients capture several ENSO characteristics simultaneously and exhibit overall better prediction skill. In particular, models with a memory effect show an alleviated skill drop during the spring barrier and a reduced prediction timing delay. One new ENSO prediction target is to predict not only the occurrence and amplitude of El Niño (EN) but also the peak location is at the central Pacific (CP) or the eastern Pacific (EP). Many prediction models have difficulty with it, which motivates the investigation on whether such ENSO diversity has intrinsically limited predictability. Here three aspects are addressed including the source/limit of predictability, time range and uncertainty. Approaches are combined including linear inverse modeling, singular vector analysis and probabilistic measure. The results show that two similar initial conditions with western Pacific SST warming anomalies may finally develop to either CPEN or EPEN. The equatorial Pacific subsurface evolution is important to tell the final outcome. Restricted by the chaotic property, the prediction horizon appears to be ~4 months before CPEN and ~7 months before EPEN. A flavor prediction model using data's transition probabilities is introduced as a new benchmark for probabilistic prediction.

The sustainable management of the environment has emerged as a major conceptual and policy challenge. In this thesis, I discuss both general theoretical considerations and aspects of a striking and important case study. In chapters 1-4, I study the dynamics of collective decisions on long-term investments in public goods, in situations in which different stake-holders disagree on the relative valuation of short-term costs and future benefits. The discussion is inspired by climate change economics but is general. Chapters 5-8 discuss aspects of the depletion of groundwater in India, one of the most dramatic resource sustainability crisis in the world, and analyze empirically the implications for irrigation, in terms of sustainability, efficiency and reliability.

Variability is inherent to any living system, and adaptation, or changing one's behavior in response to variability, is an important way to reduce or eliminate possible adverse consequences of change. Adaptation is particularly important to consider in the face of contemporary climate change, as individuals and communities may be able to adapt their behavior in response to weather variability and reduce or possibly eliminate predicted adverse impacts. To gain a more mechanistic understanding of which factors may lead to enhanced adaptive capacity of individuals and communities to future change, this dissertation uses a multi-disciplinary and multi-scale approach to broadly examine which social, economic, biophysical, and perceptional factors are associated with agricultural adaptation to current weather variability. The results from this dissertation generally show how adapting agricultural practices, like changing cropping patterns or increasing irrigation, can reduce the vulnerability of farmers to weather variability. Importantly, however, we show that adaptation is not simply about adopting appropriate technical solutions like sowing weather-appropriate crops or irrigating optimally, it is also about the complex set of economic, social, and perceptional factors that influence farmer decision-making and adaptive capacity. A global literature review highlights important biases and gaps in our current knowledge about climate change adaptation research in the agricultural sector. Based on these findings, we offer recommendations for future research that may result in a more process-based understanding of adaptation, including conducting multi-disciplinary studies that simultaneously consider the social, economic, biophysical, and perceptional factors that are associated with adaptation, and understanding how weather variability and change influence well-being to more accurately identify which individuals, households, or communities are best able to adapt. Using these recommendations, we design a case study that examines how farmers alter their cropping strategies in response to monsoon variability in Gujarat, India. Much of our research is focused on India given that over 50% of the nation practices smallholder agriculture and is particularly sensitive to climate variability and change. Through this work, we find that farmers altered their cropping decisions in response to a delayed monsoon onset, by increasing irrigation, switching crop type, and/or delaying crop sowing, and these strategies, particularly increasing irrigation, were adaptive considering yield and profit in the year of our study. These results highlight the importance of considering farmer behavior and decision-making in models that estimate future weather and climate impacts on agricultural production. While household-level surveys allow one to assess individual-level decision-making, they are difficult to implement over large spatial and temporal scales. Thus we develop a remote sensing algorithm that quantifies cropped area of smallholder farms over large spatial and temporal scales using readily-available MODIS imagery. Given the importance of irrigation as an adaptation strategy, we link these cropped area maps with rainfall and irrigation data at the village scale across all of India to assess the relative impact of different types of irrigation (e.g. groundwater versus canal) on winter cropped area and its sensitivity to rainfall variability. Overall, we find that deep well irrigation is both associated with the greatest amount of winter cropped area, and is also the least sensitive to monsoon and winter rainfall variability. However, the relative benefit of deep well irrigation varies across India, with the largest benefits seen in the regions that are facing the greatest levels of groundwater depletion. This work highlights the critical importance of groundwater for agriculture in India, and suggests that future work should identify ways to use groundwater more efficiently, increase the recharge rate of groundwater, or improve the performance of canal irrigation in order to maintain similar levels of production in the face of climate variability and change over the upcoming decades. While this dissertation focuses on agricultural adaptation to weather variability, the methods and implications derived from this dissertation are applicable more broadly to the study of resilience and adaptive capacity of social-ecological systems to global environmental change. In a rapidly changing global system, using a multi-disciplinary, multi-scale, and coupled systems approach similar to the one employed in this dissertation will help better understand and identify possible ways to enhance the ability of societies to adapt to global environmental change.

Anthropogenic climate change risks significant changes in the global distribution of precipitation. Across the western United States, modelling studies show significant reductions in wetness that imply weighty societal and ecological impacts. But the validity of the model projections need to be ground-truthed. Paleo-hydroclimate data are useful reference points to assess a model’s ability to hindcast past hydroclimate. If the hindcast matches the paleodata, it brings confidence to a model’s ability to predict future hydroclimatic change. The foremost metric of hydroclimate in the geologic record is the surface area of lakes in hydrologically closed basins. In such basins, a lake’s surface area is determined by the balance between precipitation and evaporation. The lake will expand when the balance is positive, and it will contract when the balance is negative. In this dissertation, I develop a 25-9 ka record of lake fluctuation from the Mono Basin, a hydrologically closed basin in east-central California. I deduced the fluctuations using three pieces of evidence: stratigraphy; geomorphology; and geochronology. These pieces of evidence were determined from a study of the Mono Basin’s Late Pleistocene lithostratigraphic unit: the Wilson Creek Formation. There are 19 tephra intercalated in the Wilson Creek tephra. They are named by their reverse depositional order (Ash 19 is the oldest and Ash 1 is the youngest). Uncertainty on their ages cause confusion as to the paleo-hydroclimate record of the Mono Basin. The age of Ash 19, for example, is important because its deposition marks the onset of relatively high lake levels that occurred during the last glaciation. There are two principal interpretations of Ash 19’s age: 40 ka, which is based on lacustrine macrofossil 14C data; and 66 ka, which is underpinned by paleomagnetic intensity data. In chapter 2, I tested these end-member interpretations. I used the U/Th method to date carbonate deposits that underlie and cut across Ash 19. The U/Th data show that Ash 19 must have been deposited between these two dates: 66.8 ± 2.8 ka; and 65.4 ± 0.3 ka. These dates are, therefore, more consistent with the 66 ka interpretation of Ash 19’s age. Thus the onset of relatively high lake levels in the Mono Basin corresponds with the rapid drawdown of atmospheric CO2 during Marine Isotope Stage 4. The coincidence between the drop in atmospheric CO2 and lake level rise is suggestive of a causal link. In chapter 3, I determined Mono Lake's fluctuations 25-9 ka. This time encompasses three climatic intervals: the coolest time of the last glaciation, termed the Last Glacial Maximum (LGM); the period corresponding to the rapid termination of the last glaciation, termed the deglaciation; and the early Holocene, a period of inordinate warmth that immediately followed the last glaciation’s termination. In this study, I used stratigraphic and geomorphic evidence in conjunction with 14C and U/Th dates. I measured the 14C dates on bird bones and charcoal. And I measured the U/Th dates on carbonates. Together the data showed that the lake's rises and falls concurred with North Atlantic climate. Periods of aberrant warmth in the North Atlantic concurred with low stands of Mono Lake. On the other hand, extreme cooling in the North Atlantic correlated with Mono Lake high stands. The timing of these lake fluctuations also corresponds with variations in other tropical and mid-latitude hydroclimatic records. The global harmony in the hydroclimatic records suggests a unifying conductor. I hypothesize that the conductor is tropical atmospheric circulation. In chapter 4, I present evidence on the peculiar case of an extreme low stand of Mono Lake. The low stand is dubbed the “Big Low”. The principal evidence underpinning the Big Low derives from a sedimentary sequence exposed along the canyon walls of Mill Creek. The strata show that the lake fell below 1,982 m between the deposition of Ashes 5 and 4—making this low stand the lowest recognized level of Mono Lake during the Wilson Creek Formation. Observations from dispersed sequences corroborate this interpretation. And three data constrain the age of the Big Low to be between ~24.4-20.5 ka: a carbonate U/Th date on a littoral conglomerate associated with the Big Low; a carbonate U/Th date that underlies Ash 4; and a carbonate U/Th date that cuts across Ash 5. Thus the interval that the Big Low must occur within encompasses the LGM. The timing of this low stand, therefore, corresponds with summer temperature minima, suggesting that the fall was due not to an increase in evaporation but due to a decrease in precipitation. This finding is counter to conventional wisdom: that the LGM was a relatively wet interval. In addition, both the documentation of a low stand during glacial maximum conditions and the inference that precipitation must have been reduced are contrary to previous published interpretations from model and paleoclimatic data. These discrepancies raise significant questions about our understanding of the regional expression and forcing of hydroclimate across the western United States during the LGM. Because of this period’s importance to ground-truthing climatic models, additional evidence on the geographic extent of this unexpected result is essential.

A changing climate may portend increasing disaster risk across many countries and business enterprises. While many aspects of the hazards, exposure and vulnerability that constitute disaster risk have been well studied, several challenges remain. A critical aspect that needs to be addressed is the rapid response and recovery from a climate-induced disaster. Often, governments need to allocate funds or design financial instruments that can be activated rapidly to mobilize response and recovery. The proposed research addresses this general problem, focusing on a few selected issues. First, there is the question of how to rapidly detect and index a climate hazard, such as a flood, given proxy remote sensing data on attributes that may be closely related to the hazard. The second is the need to robustly estimate the return periods of extreme climate hazards, and the temporal changes in their projected frequency of occurrence using multi-century climate proxies. The third is the need to assess the potential losses from the event, including the disruption of services, and cascading failure of interlinked infrastructure elements. The fourth is the impact on global and regional supply chains that are induced by the event, and the associated financial impact. For each of these cases, it is useful to ground an analysis and the development of an approach around real world examples, which can then collectively inform a strategy for emergency response. Here, this will be pursued through an analysis of flooding in the Philippines, livestock mortality induced by drought and freezing winter in Mongolia, Hurricane Sandy impacts in New York, supply chain impacts in Thailand, and an end to end analysis of the potential process using data from Thailand and Bangladesh. Collectively, these analyses are expected to inform climate hazard planning and securitization processes with broad applicability at a regional to national level.

Weather and climate events are costly to society both financially and in terms of human health and well being. The costs associated with extreme climate events have motivated governments, NGOs, private investors, and insurance companies to use the data and tools at their disposal to estimate the past, present, and future hazards associated with a wide range of natural phenomena in an effort to develop mitigation and/or adaptation strategies. The nonstationary nature of climate risks requires the use of numerical climate models, often general circulation models (GCMs), to project future risk. The climate risk field, however, currently finds itself in a predicament because GCMs can be biased and do not provide a clear way to credibly estimate their uncertainty with respect to simulations of future surface climate conditions. In response to this predicament, I lay the groundwork for a set of GCM credibility assessments by identifying the large-scale drivers of surface climate events that evolve over a range of timescales ranging from daily to multi-decadal. I specifically focus on three types of climate events relevant to the water and energy sectors: 1) seasonal precipitation, which impacts drinking water supplies and agricultural productivity; 2) extreme precipitation and the costly associated riverine flooding; and 3) temperature, wind, and solar radiation fields that modulate both electricity demand and potential renewable electricity supply. In chapter I, I derive a set of atmospheric indices and investigate their efficacy to predict distributed seasonal precipitation throughout the conterminous United States. These indices can also be used to diagnose the impact of tropical sea surface temperature heating patterns on conterminous United States precipitation. This is particularly of interest in the aftermath of the unexpected precipitation patterns in the conterminous United States during the 2015-2016 El Niño event. I show that the set of atmospheric indices, which I derive from zonal winds over the conterminous United States and portions of the North Atlantic and Pacific oceans, can skillfully predict precipitation over most regions of the conterminous United States better than previously recognized mid-latitude atmospheric and tropical oceanic indices. This work contributes a set of intermediate atmospheric indices that can be used to assess the efficacy of forecasting and simulation climate models to capture signal that exists between tropical heating, mid-latitude circulation, and mid-latitude precipitation. In chapter II, I first show that the frequency of regional extreme precipitation events, which are predictive of riverine flooding, in the Ohio River Basin are poorly simulated by a GCM relative to historical precipitation observations. I then illustrate that the same GCM is much better able to simulate the statistical characteristics of a set of atmospheric field-derived indices that I show to be strongly related to the precipitation events of interest. Thus, I develop a statistical model that allows for the simulation of the precipitation events based on the GCM's atmospheric fields, which allows me to estimate future hazard based on credibly simulated GCM fields. Lastly, I validate the fully Bayesian statistical model against historical observations and use the statistical model to project the future frequency of the regional extreme precipitation events. I conclude that there is evidence of increasing regional riverine flood hazard in the Central US river basin out to the year 2100, but that there is high uncertainty regarding the magnitude of the trend. This work suggests that the identification of atmospheric circulation patterns that modulate the probability of extreme precipitation and riverine flood risk may improve flood hazard projections by allowing risk analysts to assess GCMs with respect to their ability to simulate relevant atmospheric patterns. In chapter III, I present the first comprehensive assessment of quasi-periodic decadal variations in wind and solar electricity potential and of covariability between heating and cooling electricity demand and potential wind and solar electricity production. I focus on six locations/regions in the conterminous United States that represent different climate zones and contain major load centers. The decadal variations are linked to quasi-oscillatory variations of the global climate system and lead to time-varying risks of meeting heating + cooling demand using wind/solar power. The quasi-cyclical patterns in renewable energy availability have significant ramifications for energy systems planning as we continue to increase our reliance on renewable, weather- and climate-dependent energy generation. This work suggests that certain modes of low frequency climate variability influence potential wind and solar energy supplies and are thus especially important for GCMs to credibly simulate. All of the investigations are designed to be broadly applicable throughout the mid-latitudes and are demonstrated with specific case studies in the conterminous United States. The dissertation sections represent three cases where statistical techniques can be used to understand surface climate and climate hazards. This understanding can ultimately help to mitigate and adapt to climate variabilities and secular changes, which impact society, by assisting in the development, improvement, and credibility assessment of GCMs capable of reliably projecting future climate hazards.

The Asian monsoon-affected area is one of the most vulnerable regions in the world facing hydroclimate changes. Anthropogenic climate change, particularly the emissions of greenhouse gases (GHGs) and aerosols, exerts significant impacts on monsoon rainfall and circulation. Understanding the effects of external forcing on monsoon rainfall is essential for improving the predictability, constraining the uncertainty, and assessing the climate risks. In this dissertation, I use a combination of observations, outputs from multiple Coupled Model Intercomparison Project - Phase 5 (CMIP5) models, and idealized atmospheric general circulation model (AGCM) experiments to examine the Asian summer monsoon variability and change. The main focus is understanding the responses to GHGs and anthropogenic aerosols and their differences for both the historical period and future projections. The Asian monsoon is an interactive system influenced by multiple natural and anthropogenic factors. GHGs and aerosols induce significantly different changes in monsoon rainfall through both thermodynamical and dynamical processes. These changes can be further separated into the fast adjustments related to radiation and cloud processes and the slow response due to changes in sea surface temperature (SST). This thesis provides a detailed analysis of the multiple physical processes entangled in the total response, advancing our mechanistic understanding of the effects of external forcing on the Asian monsoon system and the associated uncertainties. In Chapter 2, I first analyze the monsoon-ENSO (El Nino - Southern Oscillation) relationship in observations and CMIP5 models to determine the role of natural variability. Separating the natural and forced components shows that natural variability plays a dominant role in the 20th century, however enhanced monsoon rainfall associated with global warming may contribute to a weakened ENSO-monsoon relation in the 21st century. In Chapter 3, I examine the physical mechanisms causing the changes of the Asian summer monsoon during the 20th and 21st century using observations and CMIP5 models, attributing the rainfall changes to the relative roles of thermodynamic and dynamic processes. CMIP5 models show a distinct drying of the Asian summer monsoon rainfall during the historical period but strong wetting for future projections, which can be explained by the strong aerosol-induced dynamical weakening during the 20th century and the thermodynamic enhancement due to GHGs in the 21st century. In Chapters 4 and 5, I further use multiple AGCMs to separate the total monsoon response into a fast adjustment component independent of the sea surface temperature (SST) responses, and a slow response component associated with SST feedbacks. For GHGs (Chapter 4), the fast and slow monsoon circulation changes largely oppose each other, leading to an overall weak response and large inter-model spread. For aerosols (Chapter 5), the strongly weakened monsoon circulation over land due to aerosols is largely driven by the fast adjustments related to aerosol-radiation and aerosol-cloud interactions. Finally in Chapter 6, I design idealized AGCM experiments with prescribed SSTs using the Community Atmosphere Model (CAM5) and the Geophysical Fluid Dynamic Laboratory Model (GFDL-AM3) to investigate the relative roles of uniform SST warming/cooling as well as global and regional SST patterns in shaping the differing monsoon responses. While GHGs-induced SST changes affect the monsoon largely via the uniform warming effect, for aerosols the SST spatial pattern plays the dominant role through changes in atmospheric circulation.

The high northern latitudes of North America are undergoing rapid climatic change with acute impacts to the ecosystems in which millions of songbirds breed each year. The goal of this dissertation is to improve understanding of how concurrent and interacting changes in environmental and land surface conditions influence annual movements and habitat selections of long distance migratory birds who must navigate the mosaic of changing North American ecosystems. Chapter 1 presents novel automated bioacoustic methods for estimating arrival dates of the songbird community to their arctic breeding grounds. Automated acoustic networks could vastly expand the spatiotemporal coverage of wildlife observations. However, the enormous datasets that autonomous recorders typically generate demand automated analyses that remain largely undeveloped. Chapter 1 demonstrates novel machine learning and signal processing techniques for estimating songbird community arrival dates near Toolik Field Station which agreed well with traditional survey estimates and were strongly related to the landscape’s snow free dates. Daily variations in vocal activity were more strongly influenced by environmental conditions prior to egg-laying dates. The success of the approaches presented in Chapter 1 indicate that variation in songbird migratory arrival can be detected autonomously. Widespread deployment of this advance could provide avian monitoring on a scale large enough to enable global-scale understanding of how climate change influences migratory timing of avian species. Chapter 2 examines potential future changes in habitat suitability for for two songbirds breeding throughout North America’s high northern latitudes – a tundra-nesting species (Lapland Longspurs (Calcarius lapponicus)) and a shrub-nesting species (White-crowned Sparrows (Zonotrichia leucophyrs)). By the late 21st century, models based on both climate and vegetation projected habitat suitability for Lapland Longspurs decreased across nearly all of the study domain (54-96%), while that for White-crowned Sparrows decreased in 69% of North America’s high northern latitudes. For both species, currently unsuitable habitats in northern Canada and Alaska are projected to provide suitable breeding habitat in the future. In contrast, models based solely on climate showed more drastic declines in habitat suitability for both species (Lapland Longspur, ~100% and White-crowned Sparrow ~80%). This discrepancy between model projections demonstrates that the future availability of suitable songbird breeding habitat for both species will be strongly dependent on how both the vegetation and climate– as opposed to climate alone - of northern ecosystems respond to ongoing climate change. Chapter 3 investigates the environmental and ecological drivers of migratory movements of songbirds breeding at high northern latitudes. For North America alone, there is overwhelming evidence of major shifts in seasonality of meteorological conditions, snow cover, and vegetation phenology. Few studies have focused on how this suite of changes impacts long distance migratory species that annually navigate throughout the spatially and temporally dynamic mosaic of ecosystems because of technological constraints in animal tracking. However, recent advances in GPS technology have generated units small enough to be placed on songbird species. In 2016-2018 a total of 55 American robins (Turdus migratorius) were tracked during their spring migration through the Canadian boreal forest en route to their breeding grounds. We found a significant trend towards earlier arrival of robins to the Canadian boreal forest over the past quarter-century, consistent with advances in spring environmental conditions. Robin stopover timing at our tagging site was delayed in response to later seasonal snowmelt, but triggered by adverse environmental conditions. Individuals breeding in regions with shorter snow-free seasons moved faster than individuals breeding in areas with longer snow-free seasons and selected locations with less favorable environmental conditions. Overall, arrival timing to breeding grounds was negatively related to snow depth and positively related to snowmelt timing. Migratory movements and timing of American robins are highly tied to seasonal environmental dynamics en route to their breeding grounds. Our findings present a unique, mechanistic understanding of how migratory birds navigate highly dynamic ecosystems. In light of rapid global change, the use of multi-disciplinary, spatially explicit approaches similar to the ones used in this dissertation will be critical for understanding how avian taxa breeding at high northern latitudes may respond to ongoing and future change. This is important for investigating both regional and global impacts because species breeding in arctic-boreal zones perform key ecosystem services around the globe.

As one of the main contributors to greenhouse gas emissions, the electricity sector is anticipated to go through the following transitions in order to meet deep decarbonization targets for a sustainable future: 1) on the supply side, the electric grid is increasing its reliance on renewable generation, such as wind and solar; 2) on the demand side, heating is shifting from direct burning of fuel on site to electric, namely heat pumps. This dissertation evaluates the benefits of selected methods to alleviate pressing challenges associated with the electricity sector transitions on both the supply side and the demand side. First, on the supply side, the benefits of renewable generation forecasting coupled with storage are evaluated for an electric grid with high wind energy penetration and load following generation served by fossil fuels. A time series based forecasting method is found to have high forecasting accuracy and low computational costs. This methodology is applied to a real world situation in Sao Vicente, an island with 30% current wind energy penetration. The simulation results show that coupling forecasting and energy storage would further increase the wind penetration up to 38% without additional installation of wind turbines. Second, on the demand side, the benefits of demand side management using heat pumps enabled by the inherent thermal storage of the building envelope are evaluated during extreme cold events when the electric demand peaks and the wind power is often highly fluctuating. A second order thermal model is developed to thoroughly characterize the thermal inertia and leakage of the building envelope and quantify the amount of flexibility the building envelope is able to provide. This methodology is applied to five historical extreme cold events in New York City and the simulation results show that the requirements for short term ramping due to high wind variability are greatly reduced through the sequential controls of the heat pumps. This dissertation also studies the implications of the electricity sector transitions on the residential sector with regard to costs, energy, missions, and policy. Four representative residential city blocks located in three different climate regions of the United States are analyzed using fine spatial and temporal real historical consumption and weather data. Residential blocks in different climate regions have different weather patterns, demand profiles, and local renewable resources. Future energy scenarios with electric heating at high renewable penetration levels are modeled and compared for the representative residential city blocks. Detailed costs comparisons are evaluated for various technological interventions including 1) air source and ground source heat pumps; 2) battery and thermal storage; and 3) wind and solar generation. This dissertation finds that 1) the optimal wind and solar generation mix varies with location and amount of storage and 2) battery storage is more cost effective than thermal storage, ground source heat pumps, and overbuilt renewable generation. In addition, optimal pathways to deep decarbonization for these representative residential city blocks are proposed and compared. Strategic actions are identified for the homes and suggestions are discussed for policy makers and local utilities. This dissertation through its methodologies and analysis enables home owners and policy makers to make cost assessments in achieving the goals of deep decarbonization.

Atmospheric aerosols are tiny particles suspended in the atmosphere. They affect global air quality, public health and climate (Boucher et al., 2013; Myhre et al., 2013; Seinfeld and Pandis, 2016), thus playing a key role in the Earth system. However, due to the complexity of aerosol processes and climate change feedbacks, our understanding of aerosols in a changing world is still limited (Boucher et al., 2013). To understand the impact of organic aerosol volatility on particle microphysics and global climate, I developed a new aerosol microphysics scheme, MATRIX-VBS, and its evaluation and application are presented in this dissertation. MATRIX-VBS couples the volatility-basis set (VBS, Donahue et al., 2006) framework with the aerosol microphysical scheme MATRIX (Multiconfiguration Aerosol TRacker of mIXing state, Bauer et al., 2008) that resolves aerosol mass and number concentrations, size, and mixing state. With the inclusion of organic partitioning and photochemical aging of semi-volatile organic aerosols, aerosols are able to grow via organic condensation, a process previously not available in the original model MATRIX, where organic aerosols were treated as nonvolatile. Both MATRIX and MATRIX-VBS can be used as stand-alone box models or within a global model. After the development of MATRIX-VBS in the box model framework, both model’s simulations were performed and assessed on the box and global scales. On the box model scale, idealized experiments were designed to simulate different environments, clean, polluted, urban, and rural. I investigated the evolution of organic aerosol mass concentration and volatility distribution among gas and aerosol phases, and results show that semi-volatile primary organic aerosols evaporate almost completely in the intermediate-volatility range and stay in the particle phase in the low volatility range. I also concluded that the volatility distribution of organics relies on emission, oxidation, and temperature, and the inclusion of organic aerosol volatility changes aerosol mixing state. Comparing against parallel simulations with the original model MATRIX, which treats organic aerosols as nonvolatile, I assessed the effect of gas-particle partitioning and photochemical aging of semi-volatile organics on particle growth, composition, size distribution and mixing state. Results also show that the new model produces different mixing states, increased number concentrations and decreased aerosol sizes for organic-containing aerosol populations. Monte-Carlo type experiments were performed and they offered a more in-depth look at the impact of organic aerosol volatility on activated number concentration, which is the number concentration of aerosols that are activated but has not yet formed into a cloud droplet. By testing multiple parameters such as aerosol composition, mass concentration and number concentration, as well as particle size, I examined the impact of partitioning organic aerosols on activated aerosol number concentration. I found that the new model MATRIX-VBS produces fewer activated particles compared to the original model MATRIX, except in environments with low cloud updrafts, in clean regions at above freezing temperatures, and in polluted areas at high temperature (310K) and extremely low humidity conditions. I concluded that such change is caused by the differences in aerosol number concentration and size between the two models, which would determine how many particles could activate. On the global scale, MATRIX-VBS was implemented in the NASA GISS ModelE Earth systems model. I assessed and evaluated the new model by comparing aerosol mass and number concentrations, activated cloud number concentration, and AOD against output from the original MATRIX model. Further, I evaluate the two models against observations of organic aerosol mass concentration from the aircraft campaign ATom (Atmospheric Tomography Mission), and aerosol optical depth from ground measurement stations from AERONET (Aerosol Robotic Network) as well as satellite retrievals from MODIS (MODerate resolution Imaging Spectroradiometer) and CALIPSO (Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations). Results show that organics in MATRIX-VBS experience more distant long-range transport, and their mass concentration increase aloft and decrease at the surface as compared to those in MATRIX. There are still underestimations in the vertical profiles of mass concentration in both models, especially in the high latitudes in the Northern Hemisphere and South Pacific Ocean basin, possibly due to the application of universal distribution of mass-based emission factors among different volatilities that perhaps is not realistic in all climate zones, thus affecting organic aerosol lifetime and transport. Just as the box model results, there are more particles and generally more activated ones (except for rare cases such as the highly polluted Eastern China) in MATRIX-VBS than in MATRIX. As for AOD comparisons, MATRIX-VBS have generally lower AOD than MATRIX, which can be due to smaller aerosols and different aerosol composition in the new model, which is also underestimating biomass burning in the Amazon and Congo basins. Compared to satellite retrievals from MODIS and ground measurements from AERONET, both models overestimate aerosol optical depth over anthropogenic polluted regions and biomass regions such as central Africa. Overall, both models also underestimate AOD as compared to AERONET in the winter (DJF), whereas they generally overestimate or estimate it well in other seasons. Even though during its initial evaluation, MATRIX-VBS does not seem to have improved from MATRIX on the global scale in representing the real world, it made the first key step in improving our understanding of organic aerosols on the process level. Changes in mass, number concentration, size distribution, and mixing state (composition) have great implications and impact on climate. Further studies are needed in examining and improving factors linked to the new representation of semi-volatiles in an aerosol microphysics model, including but not limited to the treatment of mass-based emission factor distribution among different organic volatilities and the size distribution of tiny organic particles that have evaporated but not completely. Challenges in evaluations of organic aerosol against measurements remain in that remote regions of significant interest lack available measurements, and additional field campaigns will be important for us to better understand real world conditions and shed light on model performance.