Session Presentations

Impacts of Future Energy Transition on the U.S. Air Quality: Projections of Emissions and Air Quality in 2050

Kai Wang¹, Yang Zhang¹, Shen Wang², Benjamin Hobbs², Hugh Ellis², Emily Fisher², Ken Gillingham³, and Michelle Bell³

¹Department of Civil and Environmental Engineering, Northeastern University, Boston, MA

²Department of Environmental Health and Engineering, Johns Hopkins University, Baltimore, MD

³School of Environment, Yale University, New Haven, CT

Understanding impacts of emission changes--especially those caused by the ongoing energy transition--on future air quality and human health is challenging due to the complex interactions among population and economic growth, energy demand, air pollution, and climate change. In this study, emission change factors (ECFs) of major air pollutants for energy-related sectors such as power plants, industrial, commercial, and residential, etc. are used in conjunction with the Sparse Matrix Operator Kernel Emissions model to project future emissions. The ECFs are generated through the downscaling approach based on the year 2050 projections from the National Energy Modeling System under a set of energy transition scenarios representing different drivers and shifts in the energy production/consumption in the future. The scenarios examined include a reference scenario without the clean power plan (refnocpp), a scenario with “abundant natural gas” (highNG), a scenario with “high electric vehicles penetration” (highEV), a scenario with “port electrification” (port), and a scenario with “high building energy efficiency” (highEE). Multiple 5-year (2008-2012) simulations using the Weather Research and Forecasting model with Chemistry (WRF/Chem) are carried out over the U.S. at a horizontal grid resolution of 36-km using current-year emissions based on the National Emission Inventory 2011 and projected future emissions in 2050. Our objectives are to 1) examine the emission changes for key air pollutants between future and past years under various energy transition scenarios, and 2) quantify the impacts of those emission changes on future air quality and human health. The preliminary results show large wide-spread reductions of CO, NOx, SO₂, and primary PM_2.5 emissions (~ 40-60% for gases and ~20% for aerosols) over the U.S. in 2050 for all scenarios, mainly due to coal power plants retirement and electric vehicles penetration, but much smaller reduction (~10%) for VOCs because the increases of their emissions from natural gas/oil production over source regions such as TX, PA, OH, and KY offset the decreases caused by power plant, industry, and transportation over the rest of domain. WRF/Chem predicts wide-spread domain-mean and maximum reductions of max 8hr O₃ by < 2.0 ppb (~3.0%) and up to 8 ppb (~15%), respectively, and those for PM_2.5 are by < 1.0 ug m^-3 (~10%) and up to 6 ug m^-3 (~50%), respectively, for all scenarios. The highEE scenario shows the largest reduction over the U.S. These reductions indicate the important impacts of energy transition on future air quality. The benefits of reduced air pollution on human health and economy under different scenarios will be further assessed in our future work.

Decarbonization will lead to more equitable air quality in California

Shupeng Zhu¹, Andrea Carlos-Carlos², Michael Mac Kinnon¹, Scott Samuelsen¹ and Steven J. Davis^{2, 3}

¹ Advanced Power and Energy Program, University of California, Irvine, CA 92697, USA

² Department of Earth System Science, University of California, Irvine, Irvine, CA 92697

³ Department of Civil and Environmental Engineering, University of California, Irvine, Irvine, CA 92697

Climate policies enacted to mitigate emissions of greenhouse gases (GHG) further attain improvements in regional air quality that provide public health benefits to exposed populations. However, the current literature lacks an understanding of these co-benefits within the framework of socially vulnerable populations, including how portfolios of strategies within mitigation pathways can be designed to achieve maximum benefits within disadvantaged communities (DAC). Here, we introduce a Lorenz Curve based novel method (Suits Index) for quantifying and assessing the health co-benefits that accrue from regional air quality improvements associated with different long-term, low-carbon scenarios in California. The electrification of building appliances (HBE) yields 14.7% higher total co-benefits for DAC, as does the use of zero-emission renewable nature gas technologies (NBE). However, the considered distribution of co-benefits differs across DAC, e.g., the HBE is less progressive in promoting environmental justice compared to the NBE with a more flatted Lorenz Curve and a Suits Index 6.5% lower. Additionally, a cost/benefit analysis is performed to provide a comparison of the value of health benefits relative to the cost of implementing GHG mitigation strategies economy-wide. The results demonstrate the importance of including environmental justice analyses when evaluating climate mitigation policies.

Future in- and out-of-state air pollutant emissions under decarbonization pathways for Massachusetts

Kaixin Huang

Daniel H. Loughlin

Matthew J. Eckelman

Many U.S. states have adopted the goal of achieving carbon neutrality in the coming decades. Pathways for achieving these targets would likely yield a reduction of air pollutant emissions as well. We assess several state-level decarbonization pathways for Massachusetts, including electrification of buildings and transportation and decarbonization of the electric grid, estimating their impacts on in- and out-of-state emissions of carbon dioxide (CO2) and several air pollutants. The analysis applies the GCAM-USA integrated assessment model with state-level resolution, and considers interactions between state and existing regional policies such as the Regional Greenhouse Gas Initiative and the Transportation and Climate Initiative. The most aggressive state decarbonization pathway that was examined would achieve only a 65% of reduction in CO2 emissions by 2050, indicating the need for further decarbonization options within the industry and non-road transportation sectors. The air pollution co-benefits under these decarbonization pathways are significant, especially for PM2.5. While policies focusing only on MA are shown to produce in-state reductions of CO2 and air pollutant emissions, a portion of these reductions was countered by emission increases in other states, indicating the need for coordinated planning to prevent leakage.

Modeling the Air Quality Impacts of Future Energy Scenarios

Uma Shankar^1,2, Kristen Brown³, Ben Murphy², George Pouliot², Kristen Foley², Dan Loughlin², and Chris Nolte²

¹ORISE Post-doctoral Research Fellow 

²Office of Research and Development, U. S. Environmental Protection Agency, RTP, NC, USA

³Department of Civil Engineering, University of Texas at San Antonio, San Antonio, TX, USA

Many areas of the United States remain out of compliance with the National Ambient Air Quality Standards. Additionally, some states have announced goals or adopted legislation to reduce emissions of carbon dioxide (CO2) and other greenhouse gases (GHGs). Human-Earth system models such as the Global Change Analysis Model (GCAM) can simulate the evolution of the energy system and estimate the associated emissions of both GHGs and traditional air pollutants. A tool has been developed to bridge GCAM and the Community Multiscale Air Quality Model using the Detailed Emission Scaling, Isolation, and Diagnostics (DESID) module, facilitating the simulation of the air quality impacts of alternative future energy scenarios, and better coordination of long-term air, climate and energy planning. In this work we demonstrate the application of this tool to examine the ozone and PM2.5 impacts at 2050 of a suite of GCAM scenarios, such as implementation of a Clean Energy Standard, high levels of electrification in the transportation and building sectors, and a moderate GHG reduction target.

The impact of improving public transportation on decreasing air pollution and greenhouse gases emissions: the case of Sao Paulo, Brazil

Flávia Noronha Dutra Ribeiro

Pedro Gerber Machado

Júlio Barboza Chiquetto

Dominique Mouette

Natasha Murgu Valdambrini

Jacyra Soares

The transportation sector is one of the major sources of greenhouse gases (GHG) emissions in the energy sector. In the Metropolitan Area of São Paulo (MASP), home to more than 20 million people, private cars contribute to 70% of vehicular air pollutants emissions and, even using biofuels such as ethanol, to more than 40% of GHG emissions. Therefore, the use of public transportation must be encouraged. However, to achieve this goal, investment in infrastructure is needed. An expansion of subway and metropolitan train lines was planned to take place until 2025, particularly towards the periphery of the MASP, but the actual implementation is delayed. The present work addressed the impact on air pollutants and GHG emissions of fully implementing the planned expansion of public transport. Emission scenarios were produced with SMOKE, using the expected emissions for the business as usual case and reduced emissions for the improved public transportation system case. Air quality simulations were performed using WRF-CHEM model and meteorology from 2018 was considered. While the general reduction of air pollutants concentration would be small, citizens near the planned stations would benefit from improved air quality. Considering GHG, the emission reduction is small but necessary to mitigate climate change. The results show that investment in public transportation has the potential to decrease GHG and air pollution emissions while increasing equality in transportation access.

Analyzing air emission co-benefits of transportation decarbonization scenarios for New York City using an energy systems model

Ozge Kaplan and Andrew Zalesak

Many areas of the US require further emissions reductions to meet and maintain the National Ambient Air Quality Standards (NAAQS). Specifically, there has been a persistent NAAQS non-attainment issue surrounding the Northeast and Mid-Atlantic region. The majority of emissions that affect air quality originate from the energy system. New technologies for producing and using energy have the potential to dramatically alter future emissions. EPA recently developed an energy systems model for urban scale analysis for the State of New York focusing on New York City, City based Optimization Model for Energy Technologies (COMET), built on MARKAL/TIMES framework. In an earlier analysis, we highlighted the importance of increased energy efficiency and electrification of transportation at city scale to reach decarbonization targets where additional NOx reductions were observed due to reduced fuel consumption in heavy duty fleets and electrification of busses and LDVs. However, air pollutant emissions are estimated using regional average emission factors, emission factors from vehicles in NYC could be quite different from regional averages due to different driving conditions and vehicle type, age and condition. We will present a more refined analysis of light- and heavy-duty on-highway mobile emissions projections based on updated emission factors where we simulated county level data with the EPA’s MOVES emissions model and other relevant sources. This includes the effects of EPA’s Tier 3 standards set new vehicle emissions and fuel sulfur standards to reduce emissions beginning in 2017. The updated future year criteria air pollutant reductions due to decarbonization pathways would greatly aid in understanding future trends and help to refine policies so that greenhouse gas and air pollutant goals can be achieved in an equitable manner.

Air quality impacts of electric vehicle adoption in California

T. Nash Skipper, Armistead G. Russell

School of Civil & Environmental Engineering, Georgia Institute of Technology, Atlanta, GA

Sales of electric vehicles (EVs) have increased in recent years and are expected to continue to rise. The governor of California has issued an executive order requiring, by 2035, all new cars and passenger trucks sold in California to be zero-emission vehicles. EVs are often touted for the climate benefits of reduced CO2 emissions, but EV adoption also has implications for air quality since emissions from onroad mobile sources contribute significantly to air pollution. We use the Community Multi-scale Air Quality Model (CMAQ) to estimate the impacts on PM2.5 and ozone of varying levels of EV adoption in the state of California. Simulations are conducted for one historic year (2016) and one future year (2028). The PM2.5 impacts of onroad emissions from tailpipe and non-tailpipe (i.e., brake and tire wear) emissions are also quantified. Results indicate that PM2.5 and ozone reductions from EV adoption scale approximately linearly with increasing EV adoption in much of California. Results also show that non-tailpipe emissions contribute up to 0.5 µg/m³ to annual mean PM2.5 concentrations.

Life-Cycle Environmental Impacts of Liquid Hydrogen Production Pathways in La Guajira Colombia

Amanda Ullman, Noah Kittner

As countries seek to transition their energy sectors towards zero-emissions energy systems, alternative energy carriers, like liquid hydrogen production, are becoming increasingly attractive for export and domestic use. Countries like Colombia, which export large levels of raw coal, must consider strategies to support a just transition for regions that depend economically on coal exports. Development of a hydrogen production industry is one such strategy discussed to support a just transition for coal dependent communities and to contribute to global environmental goals. This paper uses a life cycle assessment (LCA) methodology to examine the greenhouse gas emissions, water consumption, and trace metal emissions associated with six potential Colombian liquid hydrogen production strategies: (1) electrolysis powered by the country’s national electricity grid, (2) onsite electrolysis powered by electricity produced by a wind farm, (3) offsite electrolysis powered by electricity produced by a wind farm, (4) electrolysis powered by electricity produced from a coal-fired power plant, (5) coal gasification without carbon capture and storage (CCS), and (6) coal gasification with CCS. The paper finds that wind-powered electrolysis outperforms the coal- and grid-powered scenarios in almost every environmental category, while coal-powered electrolysis consistently performs the poorest, with life-cycle mercury, nickel, and arsenic emissions of particular concern. While coal gasification with CCS reduced gasification CO2 emissions by 35%, the scenario’s VOC emissions were 62% greater than the gasification without CCS scenario, due to the increased inputs required to meet the same energy output. Wind-based power-to-hydrogen is also found to use at least 100 times less water than coal- or grid-based hydrogen production pathways. For Colombia to benefit from a hydrogen-based decarbonization transition, community-focused planning and wind-based hydrogen systems should be prioritized over grid- or fossil-fuel-based hydrogen production pathways.

Examining the Representation of Extreme Events in 4-km Dynamically Downscaled Simulations within the Southeast U.S.

Megan Mallard, Tanya Spero, Jared Bowden, Anna Jalowska, Geneva Gray, and Gregory Tierney

As powerful computational resources become more available, finer resolution grid spacing can be used for long-term regional climate simulations. The use of increased spatial resolution to dynamically downscale global climate model output to finer regional domains has the potential to improve representation of extreme events. Presumably, the ability of the regional model to add value increases because mesoscale phenomena and finer-scale topographic and land use/land cover features can be better resolved. However, use of finer scale domains should be investigated to ensure that the chosen physics options are appropriate and that the computational expense is justified by improved results.

Here, use of a 4-km domain for simulating extreme events with the Weather Research and Forecasting (WRF) model over the southeastern U.S. is explored and contrasted with results from a 12-km WRF simulation. A global reanalysis dataset, the 0.75° × 0.75° ERA-Interim, is used to drive 3-year simulations over a historical period (2015-2017). A 12-km domain covering the entire contiguous U.S. is run in a configuration similar to previous downscaling work, and then one-way nesting is used to implement a 4-km domain over the Southeast. This small ensemble of 4-km runs compares the use of the Multi-scale Kain–Fritsch scheme to an explicit approach to represent convective processes. These simulations will be compared with observation-based 2-m temperature and precipitation from PRISM (Parameter-elevation Relationships on Independent Slopes Model) data. The ability of these simulations to capture extremes in daily minimum and maximum temperatures, as well as heavy rainfall events, will be assessed.

Estimating heat related exposures: A case study of the 2012 Chicago Heatwave

Kaiyu Chen, Mengjiao Huang, Colton Coon, Andrew J. Newman, Lyndsey A. Darrow, Matthew J. Strickland, Heather A. Holmes

Heat-related impacts on human health have received increasing attention in recent years. Abnormal increases in temperature cause additional morbidity and mortality in highly urbanized regions and induce severe economic loss. Human activities, especially urban development, increase both the frequency and intensity of heatwaves (HW) and urban heat islands (UHIs). An extreme HW event occurred in 2012 summer that caused temperatures of more than 40℃ in Chicago, Illinois, USA, which is a highly urbanized city impacted by UHIs. Various numerical models, including the High Resolution Land Data Assimilation System (HRLDAS) and Weather Research and Forecasting (WRF) Model, were used to evaluate the model performance in simulating the HW and UHI. Sensitivity testing of three different WRF boundary layer physics options was done to determine the impact of increasing model complexity in simulating urban meteorology. The application of a multi-layer urban canopy model (MLUCM) in WRF performed best. MLUCM results showed that the HW caused rural temperatures to increase by ~4℃, whereas urban Chicago had lower magnitude increases from the HW (~2-3℃ increases). Nighttime UHI intensity ranged from 1.44 to 2.83℃ during the study period. Spatiotemporal investigations of the temperature changes are used to quantify the excessive heat factor (EHF) that estimate heat-related exposure. The EHF during the HW episode provides a risk map indicating that while urban Chicago had higher heat-related stress during this event, the rural area also had high risk, especially during nighttime in central Illinois. This study provides a reliable method to estimate spatiotemporal exposures for future studies of heat-related health impacts.

Valuing the Impacts of Agricultural Soil Nitrogen Emissions on Air Quality and Climate

Lina Luo, Rice University

Daniel S. Cohan, Rice University

Agriculture has long been neglected in air quality management and climate mitigation. However, fertilizer-intensive agriculture harms air quality and climate as the largest source of reactive nitrogen emissions in the United States (U.S.). Almost half of the nitrogen added to croplands is lost to the environment via runoff, sedimentation, or air emissions of nitric oxide (NO), ammonia (NH₃), and nitrous oxide (N₂O). Emission rates respond nonlinearly to meteorological conditions, soil properties, and farming practices. Once emitted to the air, NO and NH₃ can contribute to the formation of ozone (O₃) and fine particulate matter (PM_2.5), with impacts that vary with atmospheric conditions. By contrast, N₂O is a long-lived greenhouse gas whose radiative impact is independent of the timing and location of its origin. Thus, the relative emission rates and impacts of agricultural NO, NH₃, and N₂O may vary strongly by time and region. Understanding those variations can help inform the development of effective mitigation strategies. However, most prior studies have considered only individual pollutants or neglected the influences of farming practices and spatial and temporal variability.

In this study, we develop an integrated assessment framework to compute emission rates and associated impacts of NO, NH₃, and N₂O from croplands, accounting for the influence of farming practices, atmospheric conditions, and interannual variability. This framework consists of a process-based biogeochemical model, Fertilizer Emission Scenario Tool for CMAQ (FEST-C), to simulate nitrogen emissions; a reduced-form air quality health effects model, Air Pollution Emission Experiments and Policy Analysis (APEEP), to estimate monetized impacts of NO and NH₃ on air quality and health; and the social cost of N₂O for comparing the relative importance of climate impacts with the air quality impacts. We have updated FEST-C by adjusting its crop demand-driven fertilizer database to consider fertilizer sales data and modifying its nitrogen schemes to include NO and N₂O from nitrification NO from denitrification, adapting mechanisms from the DayCent model. Using this integrated framework, we compute adverse impacts of agricultural reactive nitrogen emissions across the contiguous U.S. for three years. We show that the social costs of greenhouse gases (N₂O) are close to the monetized air pollution impacts associated with NO emissions but are lower than air pollution health impacts from NH₃. Overall, climate impacts are relatively comparable to climate impacts for croplands in the Mountain and Northern Plains regions. However, in other regions, air quality impacts significantly exceed climate impacts. We also characterize how interannual variability of emissions rates and associated damages depend on both fertilizer input and meteorological conditions. Our results demonstrate that the selection of mitigation strategies should be regionally specific, due to the strong spatial variability in emission rates and the sensitivity of air quality and health to NO and NH₃.

Modeling the Source Sectors Contribution to Nitrogen Deposition in U.S. Hydrological Regions

Sharmin Akter and Kristina Wagstrom

The excessive deposition of atmospheric nitrogen containing species to aquatic systems and watersheds can lead to harmful algae growth and loss of biodiversity, particularly in coastal waterways. It increases the risk of acidification and hypoxia by reducing oxygen levels for living organisms in marine environments. It is important to determine the major species, source sectors, and source regions responsible for atmospheric nitrogen deposition to develop effective watershed management systems. We use the Comprehensive Air Quality Model with extensions (CAMx) version 6.0, along with the Particulate Matter Source Apportionment Technology (PSAT), to identify and separate source sector contributions to atmospheric nitrogen deposition. We model the amount of atmospheric nitrogen deposition from electricity generating units, biogenic emissions, area fugitive dust, on-road, residential wood combustion, agricultural emissions, non-point source oil and gas, point source oil and gas, fires, non-road mobile equipment sources, marine vessels and locomotives, and other non-point sources in the contiguous United States, southern Canada, and northern Mexico. We use emissions, meteorology, boundary conditions, and ozone column inputs from the United States Environmental Protection Agency’s 2011 Modeling Platform. We evaluate the wet and dry deposited mass of atmospheric nitrogen containing species against measurements. This information will aid environmental regulators in developing watershed management plans to protect the health of aquatic and terrestrial ecosystems. The Upper Mississippi hydrologic region experienced the highest flux of gaseous ammonia deposition, and the dominant source sector was agricultural ammonia emissions. The Ohio region received the highest flux of reactive gaseous nitrogen, mostly from on-road emissions. Reactive gaseous nitrogen depositions from biogenic and electricity generating source sectors were significant throughout the United States.

RESPONSE SURFACE METHODOLOGY TO EVALUATE THE EFFECTS OF GEOMETRIC PARAMETERS ON THE ELECTROSTATIC PRECIPITATION

Raíssa G. S. A. Andrade¹, Vádila G. Guerra¹

¹Department of Chemical Engineering, Federal University of São Carlos, São Carlos, SP, Brazil

ABSTRACT –  Electrostatic precipitator (ESP) is an equipment used on air pollution control systems to reduce the particulate material concentration. The main components of the ESP are the discharge and collecting electrodes, which are negatively charged and grounded, respectively. Due to the electrode different polarities, an electric field is generated inside the ESP duct and the particles are electrically charged and collected. The collection efficiency is influenced by operating conditions and geometric parameters, which have not been widely discussed for nanoparticles. In this study, the influence of the number and diameter of the discharge electrodes on the performance of the ESP, with different air velocities and electric field intensities, were evaluated through an ANOVA analysis and the response surface methodology. The inlet air that enters the experimental apparatus passes through purification filters, to remove impurities, while the NaCl aerosol is generated and mixed with the purified air. After that, the excess moisture is retained by a diffusion dryer and the particles are neutralized by a Krypton-85 source and enter the ESP, which is connected to a high voltage power supply. The phenomenon of electrostatic precipitation occurs inside the ESP duct and the particles are collected. The outlet airflow is neutralized, with a source of Americium-241, to prevent imprecise results. A Scanning Mobility Particle Sizer (SMPS) and an Ultrafine Condensation Particle Counter, operating for diameters between 5.83 and 228.8 nm, were used to determine the collection efficiency of ESP. The experiments were performed in a wire-plate single-stage ESP with two copper collecting plates spaced in 6.5 cm. The number of wires analyzed was 1 and 2, with the wire-spacing of 6.5 cm, for the wire diameters of 0.3 and 0.4 mm, the air velocities of 1.03 and 4.08 cm/s, and the electric fields of 3.08 and 3.38 kV/cm. The overall efficiencies results were applied on a 24 statistical analysis to determine the parameters with the most significant effects. It was evaluated the influence of each variable on the collection efficiency through the analysis of the effects, the standard error, and the statistical significance (p-value), with a confidence interval of 95%. The ANOVA analysis and the Pareto chart results showed that all factors were significant for the particle collection, especially the electric field and wire diameter. The wire diameter and the air velocity presented a negative effect, which means that the particle collection efficiency reduced (14-30%) with the increase of these parameters. On the other hand, the number of wires and electric field showed a positive effect, with an increase of over 36% on the particle collection. Through the analysis of the response surfaces, it was concluded that the highest collection efficiencies were achieved for the wire diameter of 0.3 mm and the electric field of 3.38 kV/cm. Therefore, it is possible to obtain collection efficiencies higher than 95% with a small number of wires and operating conditions that contribute to the electrostatic precipitation of particles.

Keywords: Electrostatic precipitator, geometric parameters, response surface.

Simulating wildfire smoke transport in the western United States: Comparison of two fire emissions inventories

Kaiyu Chen, Jingting Huang, Samantha Faulstich, Heather Holmes

The western U.S. has had increasing impacts from wildfire smoke in recent years due to global climate change. Emissions from wildfires contain large amounts of criteria air pollutants such as particulate matter (PM), sulfur dioxide (SO2) and ozone (O3). As one of the largest wildfires in 2016, the Soberanes Fire occurred in central California and caused profound impacts on air quality through its smoke plume transport. This research is an applied modeling study to estimate the impacts from Soberanes Fire using the Community Multiscale Air Quality (CMAQ) model with the meteorological inputs from the Weather Research and Forecasting (WRF) model. Wildfire emissions from the Fire INventory from NCAR (FINN) and the National Emissions Inventory (NEI) using the BlueSky framework are used in CMAQ and the outputs are evaluated using ground-based observations and satellite products. FINN uses thermal anomalies and land cover products from satellite remote sensing to estimate burned area and fuel loading, respectively. FINN uses emissions factors, for many chemical species, and vegetation estimates to provide fire emissions inputs to chemical transport models in near real-time. The NEI has point source day-specific fire emissions obtained from SMARTFIRE (Satellite Mapping Automated Reanalysis Toll for Fire Incident Reconciliation) with the estimated fire activities (fire locations, duration and size) serving as the initial data for the BlueSky framework. This framework is implemented in the Sparse Matrix Operator Kerner Emissions (SMOKE) modeling system to generate CMAQ ready fire emissions. In addition to the emissions inventory comparison, different speciation mechanisms (CB6 and SAPRC07) are also evaluated in this study to quantify the impacts of applying different speciation configurations on modeling wildfire smoke. Consequently, future work will use these results to develop a wildfire smoke forecasting framework and to simulate smoke transport to estimate the impacts on air quality and human health. Broadly, these findings are valuable to researchers interested in assessing fire and smoke impacts and provide evidence for further model studies of fire activities.

PM2.5 variability with meteorology from reanalysis and ground-based monitors across locations

Shreya Guha^1*, Lucas R.F. Henneman^1
1Department of Civil, Environmental and Infrastructural Engineering, George Mason University, Fairfax, VA
^*Corresponding author: sguha2@gmu.edu

Air pollution is one of the most important eco-environmental problems throughout the world and its concentrations are sensitive to both source emission levels and meteorological fluctuations. The purpose of this study is to identify differences in covarying meteorological factors between reanalysis and observed meteorological data sets across locations such that more effective and integrated air quality policies can be drafted in the future.

We utilize two approaches to quantify the effects of ground-based observed (from National Climatic Data Center [NCDC]) and reanalysis model values (taken from Climate Forecast System version 2 [CFSv2]) of meteorological variables on PM_2.5emission levels (downloaded from AirNow Beijing and the US Environmental Protection Agency [EPA]website) for the years 2011 to 2020 for the cities Beijing, China and Washington DC, USA. We consider temperature, precipitation, relative humidity, wind speed & direction at 2m above ground level and planetary boundary layer height. The first approach applies Kolmogorov-Zubrenko filters and linear regression models to study long-term, seasonal, and short-term components of the pollutant signals and its meteorological components. The second approach applies general additive models to quantify relationships between meteorology and PM_2.5fluctuations at the given locations with the two different sets of meteorological data. We perform a holdout analysis to evaluate the predictive performances of all models.

In Beijing, reanalysis meteorological data show very weak relationships with PM_2.5 concentrations, whereas the observed data are very significant for the short-term scale, with the average daily PM_2.5 increasing by 1µg m^-3 per ~3.5℃ average temperature increase. In continuing analysis, we are comparing the results between Beijing and Washington, DC. 

Quantifying impacts of climate change and variability on future US PM2.5 by dynamically downscaling a global chemistry-climate model in WRF and CMAQ

Surendra Kunwar¹, Jared Bowden², George Milly³, Michael Previdi³, Arlene Fiore³, Jason West¹

¹University of North Carolina at Chapel Hill, Chapel Hill, NC, USA

²North Carolina State University, Raleigh, NC, USA

³Columbia University, Palisades, NY, USA

Anthropogenic climate change and meteorology-sensitive natural feedback emissions like biogenic VOCs, lightning NO_x and sea salt will likely affect future PM_2.5 and O₃ air quality. But the noise from variability in meteorological variables like temperature, precipitation and wind speed can confound the signal of climate change on air quality. Previous studies of the impacts of climate change on air quality in the US have often downscaled only a small number of years from global models, without understanding the broader distribution of possible effects, thereby not clearly separating the influence of variability from that of climate change.

Here we use a combination of multidecadal global model ensemble simulations with downscaling by finer regional atmospheric models for selected years, to quantify the effects of climate change and variability on US PM_2.5 in the 2050s under climate change from the RCP8.5 scenario. We use simulations of the global chemistry climate model GFDL-CM3 at 2^o resolution for the period 2006-2100, where the three ensemble members vary only in initial conditions, providing an unprecedented dataset for defining PM_2.5 probability distributions. These simulations held anthropogenic emissions of PM_2.5 and O₃ precursors constant at present day levels, to isolate the influence of RCP8.5 climate change. Empirical Orthogonal Function analysis of the global model dataset has helped identify the Eastern US regions where PM_2.5 covaries spatiotemporally. We then select four present (2006-2020) and four mid-century (2050-2060) years representing median and high PM_2.5 levels in different US regions for downscaling by regional models.

We use the Weather Research and Forecasting (WRF) model version 3.9.1.1 to dynamically downscale the coarse GFDL-CM3 meteorology of the eight selected years to a finer 12 km resolution for the CONUS. The GFDL chemistry and the WRF meteorology are then provided as inputs to the regional air quality model CMAQ to obtain US PM_2.5 levels at 12 km resolution for the selected years. CMAQv5.3.2 uses the BEIS model to generate biogenic emissions inline, and the Carbon Bond 6 (with Aerosol Module 7 and Aqueous Chemistry) chemical mechanism, which incorporates aqueous-phase and cloud aerosol chemistry. NEI emissions for year 2016 processed in SMOKE are used in all CMAQ simulations to represent present day emissions.

We then combine the broad probability distribution of mean annual PM_2.5 from the global model with the CMAQ-downscaled PM_2.5 values in selected years, to construct PM_2.5 distributions for the present and mid-century in individual 12 km grid cells. We do so by fitting distribution parameters that minimize percentile differences with the percentiles of the selected years from the global model. A Monte Carlo simulation of differences in fine resolution mean annual PM_2.5 distributions between the present and the future will be used to quantify the impacts of climate change and variability on US PM_2.5. Expressing the impact of climate change on PM_2.5 as probability distributions will then allow us to calculate the impact of climate change and variability on human mortality and visibility.

Analyzing the Transportation and Climate Initiative Program Using an Integrated Assessment Model
Zhuoran "Gray" Li* (presenter) and Dan Loughlin+

* Conducted while a graduate student at Duke University
Currently with the Environmental Defense Fund, Beijing, China
+ U.S. EPA Office of Research and Development

The Transportation & Climate Initiative (TCI) is a cap-and-invest program that is designed to reduce transportation-related CO2 emissions from the thirteen participating states. Electrification of the transportation sector has the potential to interact with the Regional Greenhouse Gas Initiative (RGGI) market, a cap-and-trade system for CO2 emissions from electricity generation. Potential impacts of this interaction could include leakage of CO2 emissions to other sectors and states. This study utilized a human-earth systems model to investigate TCI's regional emissions reduction potential and to quantify the impacts of the program under alternative assumptions. The results show that TCI would lower regional net CO2 emissions and increase market penetration of both heavy- and light-duty electric vehicles through 2050. Furthermore, emissions reduction and changes in transportation fuel mix would be accompanied by an increase in the RGGI CO2 allowance price, a moderate increase in the price of electricity, and an increase in the cost of on-road travel. In the freight sector, a portion of heavy-duty demand shifts from trucks to trains and marine vessels. Also, some of the CO2 reduction from on-road transportation is offset by the increases in emissions from non-road transportation and electricity generation. Pennsylvania, the only TCI state that does not currently belong to RGGI, experiences the highest increases in CO2 from the electric sector as well as the lowest percentage reduction of NOx emissions, reflecting both its own additional electricity demands from TCI and greater electricity exports to RGGI states. Extending the declining RGGI and TCI cap trends through 2050 leads to greater reductions in CO2 and air pollutant emissions. In addition, it further facilitates the energy transition from carbon-based fuels to electricity in the on-road transportation sector.

Changes in Mortality in Response to Decreases in Ozone and PM2.5 Concentrations Across the United States from 1990 to 2019

Revathi Muralidharan, Yuqiang Zhang, Omar Nawaz, Daniel Tong, Aaron van Donkelaar, Randall Martin, Marc L. Serre, and J. Jason West

O3 and PM2.5 concentrations in the United States have decreased significantly since 1990 due to the success of air quality regulations such as the 1990 Clean Air Act Amendments. It is estimated that between 1990 and 2019 maximum daily 8 hour average (MDA8) O3 decreased by 25%, and from 2000 to 2019 annual average PM2.5 concentrations decreased by 43%. A decrease in mortality is expected to correspond to this reduction in O3 and PM2.5 concentrations. In order to assess the health impacts of reduced air pollution across the United States, mortality attributable to O3 and PM2.5 in the continental United States is assessed each year from 1990 to 2019 using multiple concentration datasets for comparison. We create a 28-year dataset by performing Bayesian Maximum Entropy kriging on observations (1990-2017). We also use a long-term simulation (1990-2010) from a regional air quality model (CMAQ), as well as CMAQ simulations from the North American Chemical Reanalysis project (2009-2019), and a satellite-derived PM2.5 dataset (1999-2018). We also use annual county-level mortality statistics from the US Centers for Disease Control and Prevention, and separate the trends caused by the change in concentrations vs. the changes in population and baseline mortality rates. In using multiple concentration datasets, we aim to investigate whether trends in mortality are consistent across different concentration datasets and to account better for uncertainty. We hypothesize that the different datasets will show similar decreases in premature mortality.

Influence of land use and occupationon aerosol chemical composition in the eastern part of Sao Paulo city

Maria Carla Queiroz Diniz de Oliveira, Regina Maura de Miranda - University of São Paulo, São Paulo, SP, Brazil

The city of São Paulo is one of the main financial and mercantile centers in South America, impacted mainly by heavy trafficand industrial emissions. Thus, to better understand the sources of pollution emissions, fine particulate matter (PM2.5) samples werecollected in the eastern region of the city of São Paulo during 2015. Particle concentrations, chemical composition, Wind speed and direction were related to land use and occupation in order identify the main sources that influence air quality at the site. Episodes with high concentrations of PM2.5 (concentrations above 25ug/m3) indicated higher concentrations of the species S, Si, K and SO4-2. During 2015, World Health Organization (WHO) air quality standard for the pollutant (25ug/m3) has been exceeded for 17 days. The results indicated that vehicle traffic has the greatest influence on the site, but residential waste burning in the surrounding residences and the airport also have a strong contribution.

Effectiveness of Decarbonization Pathways on Eleven Northeastern States Greenhouse Gas Reduction Goals

Sarah Simm¹, Carol Lenox², & Dan Loughlin²

¹ORAU Student Service Contract, U.S. EPA, ORD, CEMM, AESMD, ESAB

²U.S. EPA, ORD, CEMM, AESMD, ESAB

Increased awareness of the impacts of global climate change has resulted in many US states committing to various greenhouse gas reduction (GHG) and decarbonization goals. In this study, GLIMPSE is used to analyze various decarbonization pathways in eleven Northeastern states. GLIMPSE is a decision support modeling tool being developed by the EPA to support state-level environmental and energy planning. GLIMPSE is a graphical interface for the Global Change Analysis Model (GCAM), an integrated assessment model (IAM) developed by the Pacific Northwest National Laboratory. GCAM links representations of the world's energy, economic, agriculture, and land use systems with a climate model, simulating their interactions and co-evolution. The aim of this study is to provide a greater understanding of the potential impacts these state GHG mitigation policies have on carbon dioxide (CO2) and air pollutant reductions and what additional efforts might be needed for states to reach their ultimate goals. The states studied in this analysis include Connecticut, Delaware, Massachusetts, Maine, Maryland, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, and Vermont. The decarbonization pathways explored include: increased electrification and efficiency in the building sector, increased electrification in the passenger vehicle sector, renewable portfolio standards, clean energy standards including carbon capture and storage technologies and biomass use. These pathway choices are applied to each individual state, both alone and in combination, and the reduction in CO2 is analyzed at the state level, providing insight into what type of pathways could be most effective for each state.

Envi Met Model Studies of Urbanization Impact on Climate Change

G S N V K S N SWAMY UNDI

The present reflects the dynamics of urbanization on climate change. Sustainable urbanization includes environmental factors that need to be considered for urban development. The complexity of urban structure brings considerable changes to the urban atmosphere, micrometeorology, and dispersion conditions in a significantly considerable behavior known as urban hotspots. An increase in temperature and air pollution levels influence regional and global trends, so it is important to correlate urbanization's impact on climate change indicators. This present paper aims to present the concept of sustainable urban planning, the effects of climate change, and importance of urban heat islands (UHI) in territorial planning. This chapter draws attention to the impact of land use and land cover change on UHI characterization, measurements, and impact assessment have been discussed in the present chapter. The paper presents a system that can be used for the assessment of the land use pattern effect on the UHI. The system integrates analysis of land use modeling and an indicator-based assessment. The applicability of the model is illustrated through developing scenarios for the future land use allocation of the urban zone. The results show scenarios of the UHI effect and outdoor thermal comfort. Moreover, for each scenario of urban design patterns accounting for adaptation to climate change could be implemented. This study provides a guideline for local authorities to create more sustainable urban structures and adapt to climate change.

US Clean Energy Futures - an analysis of different energy policies towards cleaner air

Petros Vasilakos, Huizhong Shen, Nash Skipper,  Armistead Russell

In this presentation, we will be comparing different policies aimed at controlling EGU emissions between the present day and 2050. Most of these policies are aimed at reducing CO₂ levels, but at the same time produce significant co-benefits for other criteria pollutants that we will be mostly focusing on. We find that even modest policies can provide both climate and air quality benefits at a small cost, indicating that an 80% clean energy sector by 2030 is both viable and financially beneficial (though the reduction in health associated mortality costs).