Development of risk-based air quality management strategies under impacts of climate change

Abstract

Climate change is forecast to adversely affect air quality through perturbations in meteorological conditions, photochemical reactions, and precursor emissions. To protect the environment and human health from air pollution, there is an increasing recognition of the necessity of developing effective air quality management strategies under the impacts of climate change. This paper presents a framework for developing risk-based air quality management strategies that can help policy makers improve their decision-making processes in response to current and future climate change about 30–50 years from now. Development of air quality management strategies under the impacts of climate change is fundamentally a risk assessment and risk management process involving four steps: (1) assessment of the impacts of climate change and associated uncertainties; (2) determination of air quality targets; (3) selections of potential air quality management options; and (4) identification of preferred air quality management strategies that minimize control costs, maximize benefits, or limit the adverse effects of climate change on air quality when considering the scarcity of resources. The main challenge relates to the level of uncertainties associated with climate change forecasts and advancements in future control measures, since they will significantly affect the risk assessment results and development of effective air quality management plans. The concept presented in this paper can help decision makers make appropriate responses to climate change, since it provides an integrated approach for climate risk assessment and management when developing air quality management strategies.

Implications:

Development of climate-responsive air quality management strategies is fundamentally a risk assessment and risk management process. The risk assessment process includes quantification of climate change impacts on air quality and associated uncertainties. Risk management for air quality under the impacts of climate change includes determination of air quality targets, selections of potential management options, and identification of effective air quality management strategies through decision-making models. The risk-based decision-making framework can also be applied to develop climate-responsive management strategies for the other environmental dimensions and assess costs and benefits of future environmental management policies.

Introduction

Climate change has been forecast to influence air quality on regional and global scales through perturbations in precursor emissions as well as atmospheric physical and chemical processes (Jacob and Winner, 2009; Weaver et al., 2009). Two air pollutants of most concern are ozone and PM_2.5 (particulate matter with an aerodynamic diameter less than 2.5 μm) due to their adverse human health effects. Health effects of ozone include increased rates of hospital admissions, exacerbation of respiratory illnesses, and others (Bell et al., 2004, 2005; U.S. Environmental Protection Agency [EPA], 2006a). Studies have demonstrated that PM_2.5 leads to premature death, increased risks of heart attacks, adverse effects on lung function, and other health effects (Dominici et al., 2006). Overall, the most significant health effect associated with both PM_2.5 and ozone air pollution is premature death. The American Lung Association (2011) has reported that roughly half the people (∼154.5 millions) in the United States live in the 366 counties where they are exposed to levels of air pollution that exceed the National Ambient Air Quality Standards (NAAQS) for ozone or particulate matter (American Lung Association, 2011).

Air quality studies show that projected higher temperatures and more stagnant meteorological conditions could increase ground-level ozone concentrations in the future (Hogrefe et al., 2004; Leung and Gustafson, 2005). Higher volatile organic compound (VOC) emissions (including anthropogenic and biogenic VOC emissions), due to increases in temperatures, can increase peak ozone concentrations in urban areas where peak ozone levels are often limited by the availability of VOC (Jacob and Winner, 2009). Ambient PM_2.5 concentrations also change in response to modifications in emissions, stagnation, precipitation, and shifts of thermodynamic equilibriums between gas- and condense-phase semivolatile compounds (Liao et al., 2006). Further, several studies have shown that warmer climate is expected to increase wildfire severities and, therefore, wildfire-related air pollutant emissions in the future (Fried et al., 2004; Spracklen et al., 2009). Overall, the elevated air pollutant levels attributed to climate change are expected to exacerbate the adverse health effects of air pollution (Bell et al., 2007). These issues have led the U.S. Environmental Protection Agency (EPA) to conclude, “America's communities face serious health and environmental challenges from air pollution and the growing effects of climate change” (EPA, 2010a).

To protect the environment and human health from air pollution, there is an increasing recognition of the necessity of incorporating the impacts of climate change into regional air quality planning processes. The goal of this paper is to provide an integrated approach for decision makers developing effective air quality management strategies that take in to account the impacts of climate change.

Procedures for Developing Air Quality Management Strategies Accounting for Climate Change

For air quality management, the impacts of climate change can be viewed as risks to air quality and air pollution–related human health, and the development of climate-responsive air quality management strategies should be considered in two stages: risk assessment and risk management (Figure 1). Since developing climate-responsive air quality management strategies depends on an understanding of forecasted impacts of climate change, the risk assessment includes identification and quantification of the climate change impacts and associated uncertainties. The risk management for air quality under the impacts of climate change includes determination of air quality targets, selections of potential air quality management options, and identification of preferred air quality management strategies through risk-based decision-making models. Each of the steps of the risk assessment and risk management process is discussed below.

Figure 1. Framework for developing air quality management strategies under the impacts of climate change.

Step 1: Identification and quantification of the impacts of climate change and associated uncertainties

The effects of climate change on air quality are typically estimated using global- or regional-scale atmospheric chemical transport models (CTMs) that are driven by general circulation model (GCM) simulations of future climate change (Jacob and Winner, 2009). To investigate regional air quality impacts of climate change, future meteorological fields from the GCM simulations need to be downscaled to regional scales using meso-scale meteorological models, such as the Fifth-Generation National Center for Atmospheric Research (NCAR)/Penn State Mesoscale Model (MM5) (Leung and Gustafson, 2005) and the Weather Research and Forecast (WRF) model (Lo et al., 2008). To investigate the impacts of climate change alone on air quality, two scenarios need to be studied in combined GCM-CTM analyses. First, a base-case scenario of air quality is simulated using present-day emission inventories (EIs) of air pollution precursors (e.g., oxides of nitrogen [NO_x], sulfur dioxides [SO₂], volatile organic compounds [VOCs], etc.) and present-day climatic conditions. The other scenario uses present-day emission inventories but future climatic conditions (Figure 2), which are projected using GCM simulations assuming greenhouse gas (GHG) emissions will follow the Intergovernmental Penal on Climate Change's (IPCC) Special Report on Emission Scenarios (SRES) (Nakicenovic and Swart, 2000).

Figure 2. Examination of impacts of climate change alone on air quality.

Due to a wide range of the main driving forces of future GHG emissions, the IPCC SRES developed six different emission scenario groups (A2, B1, B2, and three groups [i.e., A1FI, A1B, and A1T] within the A1 family) specifying GHG emission levels and increases in temperatures based on different assumptions in energy uses, economic growths, as well as the other social and technological changes. Among the six IPCC SRES emission scenarios, the A2 and A1B are the most widely used scenarios in assessing the impacts of climate change on air quality (Jacob and Winner, 2009). The A2 scenario projects the highest increase in GHG emissions and temperatures among the six scenarios and is considered as the “worst-case scenario” in terms of the impacts of climate change on air quality. On the other hand, the A1B scenario projects mid-level increases in GHG emissions and temperatures, and it is considered as a more neutral case if only one of the six emission scenarios is chosen for studying the climate change impacts on air quality. Comparisons between the base-case (i.e., present-day climate and emission inventories, Scenario PP in Table 1) and future scenarios with future climate and present-day emission inventories (Scenario FP in Table 1) show how climate change alone will affect air quality (Figure 2). Several studies, based on the IPCC SRES emission scenarios, have found that climate change alone is expected to increase ground-level ozone concentrations through changes in meteorological conditions, atmospheric chemical processes, and precursor emissions induced by warmer climate (Hogrefe et al., 2004). On the other hand, climate change may increase or decrease PM_2.5 levels due to changes in precipitation, which is a significant removal mechanism for PM_2.5 (Tagaris et al., 2007), and warming, leading to increased PM volatilization (Dawson et al., 2007; Mahmud et al., 2010). Generally speaking, air quality, especially ozone, in urban areas is more responsive to the changes in climatic conditions because of higher emissions of its precursors (Jacob and Winner, 2009; Liao et al., 2009b). It is important to note a new set of IPCC climate change scenarios, Representative Concentration Pathways (RCPs), is under development, and expected to provide a more comprehensive analysis of potential changes in future climate (Inman, 2011). Therefore, the RCPs should be used in climate change–air quality analyses when they become available.

Table 1. Summary of climate change–air quality simulations

Scenario	Climatic Condition	Emission Inventory	Future Air Quality Impacting Factor
PP	Present-day	Present-day	—
FP	Future	Present-day	Climate change
FF	Future	Future	Climate change and planned emission controls

Another important task, and one that is directly policy relevant, for developing climate-responsive air quality management strategies is to assess how climate change will affect current air pollution improvement plans. The base-case scenario, which applies present-day climatic conditions and present-day emission inventories, is also required in this assessment. However, the other scenario required in this assessment applies future climatic conditions and projected emission inventories of air pollution precursors, which are estimated by taking into account existing control plans for future air pollutant emissions (Scenario FF in Table 1). Liao et al. (2007) apply the IPCC A1B climate scenario and projected emission inventories, including currently planned emission control strategies, in CTM simulations for the year 2050. The results show that currently planned air pollutant emission control strategies will still be effective in reducing ozone and PM_2.5 levels in the United States, but the “climate penalty” is expected to partially offset the effectiveness of currently planned air quality management plans.

The uncertainty in climate change forecasts is another important consideration in assessing the impacts of climate change on air quality. Recent studies show that modeling results of near-surface ozone air quality over Europe and the United States are highly sensitive to the IPCC climate scenarios chosen in climate change–air quality assessments (Lin et al., 2008; Meleux et al., 2007). Liao et al. (2009) present that uncertainties in climate change forecasts, derived using probabilistic distributions of projected climate fields, led to differences in modeled ozone concentrations of up to 10 ppb, which is about one-seventh of current ozone National Ambient Air Quality Standards (NAAQS) of 75 ppb (parts per billion) in 2050 in urban areas in the United States. Such differences related to the uncertainty in climate forecasts are expected to significantly affect the risk assessment results and, therefore, should be taken into account in the air quality decision-making processes. One common approach for quantifying a possible range of the impacts of climate change on air quality is to assess results of multiple GCM-CTM simulations using different models and climate change scenarios (Jacob and Winner, 2009). A possible approach to quantify the uncertainty in air quality modeling is to calculate the probability of air quality responses to emission controls through air quality sensitivity analyses and Monte Carlo simulations (Digar et al., 2011).

Step 2: Determination of air quality targets

The second step for developing the climate-responsive air quality management strategies is to set future air quality targets with the consideration of the impacts of climate change. However, the determination of the air quality target is not a straightforward task and should take into account the flexibility of effective policy-making, feasibility of potential air pollution control measures, and uncertainties in decision-making due to incomplete information and knowledge. Furthermore, the targets should be fully discussed across government agencies, the public, and stakeholders, since they dominate the development and implementation of the management strategies. Another important consideration for setting the air quality targets under the impacts of climate change is the time horizon for implementing management strategies, and it is usually chosen as a compromise between being far enough in the future to experience nontrivial climate modifications, yet is still within a reasonable time period for long-term air quality planning (i.e., 20–50 years). The time frame of 20–50 years is beyond that typically considered in current air quality planning to meet the NAAQS, which typically do not address the impacts of climate change. Looking further in to the future, however, is consistent with other types of planning conducted by states, including transportation planning, and could provide a direction for development of long-term climate-responsive air quality management strategies, since climate change is assessed based on emission projections for 20–100 years. In this paper, we consider two types of air quality targets with the consideration of the impacts of climate change. The first type of targets is targets that offset the adverse effects of climate change on air quality. Liao et al. (2010) choose offsetting the impacts of climate change on ground-level ozone and PM_2.5 air quality as targets and the results show that an annual cost of about $9 billion (1999$) will be required to offset the climate impacts on ground-level ozone and PM_2.5 air pollution in 2050 over six regions and five urban areas in the United States (Liao et al., 2009a).

The second type of targets includes those achieve prescribed air quality standards in the future. GCM-CTM studies find that climate change alone will increase summertime surface ozone in polluted areas by 1–10 ppb by 2050 (Jacob and Winner, 2009). Such increases in surface ozone concentrations are expected to create further challenges to protecting the environment and human health from air pollution. Our study shows that, under the impacts of climate change, reductions in anthropogenic precursor emissions will have similar benefits to air quality improvement compared with current conditions (Liao et al., 2007). When the effects of the climate penalty are considered, the same magnitude of emission controls may not achieve prescribed air quality targets, since climate change could increase air pollutant levels in some areas. As such, additional air pollution controls for offsetting the air quality penalty induced by climate change should be considered in control strategy development as well as cost-and-benefit analyses for future attainment of air quality standards.

Step 3: Selections of potential air quality management options

Ambient ozone is formed through photochemical reactions mainly involving reactions of VOCs and NO_x in the air. SO₂, in addition to NO_x and VOCs, is a major precursor for ambient secondary PM_2.5 formation (Seinfeld and Pandis, 2006). Ammonia (NH₃) is another important precursor for nitrate and sulfate PM_2.5 formation. However, NH₃ is not included in the discussion. Ambient PM_2.5 could also be directly emitted from sources (e.g., road dust, motor vehicles, coal combustion, wood burning, etc.), and it is recognized as primary PM_2.5 emissions. NO_x is mainly emitted from mobile vehicles and power plants. The largest anthropogenic emission source of SO₂ is electricity generation from coal. VOCs in the air come from both biogenic and anthropogenic sources. Major anthropogenic sources of VOCs are solvent uses and motor vehicles (Table 2). Since biogenic VOC and the other precursors emitted from natural sources typically are not controlled as part of air quality management plans, this paper only discusses how to develop climate-responsive air quality management strategies by controlling those precursors emitted from anthropogenic sources and those that can be managed by human actions, i.e., prescribed forest burns.

Table 2. Major precursors of ambient ozone and PM_2.5, source categories, and potential management options

Precursors	Major sources ^a	Category of management options
Sulfur dioxides (SO2)	EGUs ^b	- Reduction-by-regulations- Cap-and-trade- Individual actions
Nitrogen oxides (NOx)	On-road vehicles, nonroad equipment and EGUs	- Reduction-by-regulations- Cap-and-trade for EGUs- Individual actions
Volatile organic compounds (VOC)	Solvent uses, on-road vehicles, nonroad equipment and fires	- Reduction-by-regulations- Individual actions- Wildfire management
Primary fine particulate matters (PM2.5)	Road dust, industrial processes, EGUs and fires	- Reduction-by-regulations- Cap-and-trade for EGUs- Wildfire management- Individual actions
Notes: ^a http://www.epa.gov/air/emissions/index.htm; ^bEGUs = electric generating units.

A variety of options are available for combating air pollution under the impacts of climate change, for example, individual actions, cap-and-trade approaches, and reduction-by-regulations. The cap-and-trade and reduction-by-regulation approaches are policy-driven strategies and widely used by air quality management agencies. For example, the EPA's Acid Rain Program uses the cap-and-trade approach to achieve significant environmental and public health benefits through reductions in SO₂ and NO_x emissions (EPA, 2010b). Since air quality in different regions may have different responses to common air pollution control measures, all possible management options should be considered in the selective processes to simultaneously improve air quality in multiple regions. Major categories of climate-responsive air quality management options are discussed below.

Reduction-by-regulations

A traditional approach for decreasing air pollutant levels is reduction-by-regulations, which decreases emissions through the imposition of emission standards or limitations. The effectiveness of this approach is more robust than other types of emission control strategies, since emission control measures and targets are prescribed by air quality management agencies. However, this type of control measures is less flexible and the least-cost regulations may not be achieved.

Cap-and-trade systems

Air quality management can also be implemented though cap-and-trade systems, which are market-based approaches and flexible to achieve air quality targets at the least cost to society. In regional air quality management, the cap-and-trade systems have been successfully used to decrease SO₂ and NO_x emissions in the United States. The EPA's Cross-State Air Pollution Rule (CSAPR) requires a 73% reduction in SO₂ and a 54% reduction in NO_x emissions, using the cap-and-trade approach, from power plants in 2014 from 2005 levels in the eastern half of the United States (EPA, 2011). A distinct advantage of the cap-and-trade systems is that they place a definite limit on the aggregate emission from a particular type of sources regardless of future economic and population growth. The cap-and-trade systems are usually more cost-effective than the traditional reduction-by-regulation approaches, but they are only applied for reducing emissions from large stationary sources, such as power plants. A potential issue of cap-and-trade systems is that they do not guarantee that emission reductions will be achieved in the locations that have the most need for achieving specific air quality targets and are most useful in achieving broad regional reductions in air pollution.

Fire management

Wildfires and prescribed fires have significant impacts on air quality due to increased emissions of air pollutant precursors (Phuleria et al., 2005; Tian et al., 2008). One of the main objectives for wildfire management is to reduce the risk of negative impacts from unplanned fires on people, property, and ecosystems. A long-term (i.e., 50 years) study shows that appropriate wildfire management (e.g., prescribed burning) could reduce frequencies of wildfire incidences and areas burned by wildfires (Boer et al., 2009). On the other hand, fire emissions in some parts of the United States are dominated by prescribed fires. Prescribed burning management via a smoke management (e.g., by choosing specific days to burn with maximum dilution of the plumes and minimal downwind adverse impacts) can lead to significantly lower impacts from prescribed fires. As such, appropriate fire management is expected to improve air quality by reducing air pollutant emissions from both planned and unplanned fires, and it could be one of the options for air quality management under the impacts of climate change. However, developing air pollution control strategies through planned (i.e. prescribed fires) and unplanned (i.e., wildfires) fire management is less robust than the other approaches due to uncertainties in estimates of fire emissions and associated effects on air quality.

Individual actions

Air quality improvement can also be affected through actions of individuals. Since power plants, industry, and motor vehicles are major sources of many air pollution precursors, individual actions, induced by restrictions and economic disincentives, for improving air quality under the impacts of climate change could include energy saving, carpool and public transportation, and others. However, it is important to note that the benefits due to individual actions are very difficult to quantify and are highly uncertain.

To select potential air quality management options, responses of air quality to management options need to be quantified; this is usually done through sensitivity analyses (i.e., assessing responses of a pollutant concentration to its precursor emissions) and source apportionment techniques using air quality models. There are five main approaches to quantify how air pollutant concentrations respond to emission controls through air quality modeling: (1) Brute Force; (2) Response Surface Modeling; (3) Decoupled Direct Method (DDM); (4) Adjoint methods; and (5) Tagged Species Source Apportionment (TSSA). The Brute Force method is the most widely used technique to calculate sensitivities and involves one-at-a-time perturbation of model inputs or parameters (Russell et al., 1995). Response Surface Modeling has been developed to aggregate numerous individual simulations into a multidimensional air quality responses surface and allows for rapid assessment of effects of precursor emission reductions on air pollutant formation (EPA, 2006b). The DDM (Dunker et al., 2002a) and Adjoint (Hakami et al., 2007) methods calculate sensitivities by solving differential equations that are similar to the atmospheric chemical transport equation and are found to be more efficient than the Brute Force method (Hakami et al., 2003, 2007). The TSSA approach estimates the contributions of different sources to air pollutant concentrations by adding tracers to air quality models (Dunker et al., 2002b). Results of sensitivity analyses and source apportionment provide the responses of air pollutant concentrations to emission control measures and can help select potential climate-responsive air quality management options. However, air pollutants could have different responses to controls of common sources. For example, reductions in NO_x emissions decrease PM_2.5 levels but may increase or decrease peak ozone levels in Los Angeles when ozone formation is in the VOC-limited regime (Liao et al., 2008). Therefore, the effectiveness of management strategies for multipollutant air quality should be taken into account as a whole in the selections of potential management strategies.

In addition to the effectiveness of the management strategies, uncertainties and costs of implementation of the strategies should be considered, since they can significantly affect the selection process of potential air quality management strategies. Generally speaking, projections in effectiveness of fire management and individual actions to climate change are more uncertain than reduction-by-regulation and cap-and-trade approaches. Reducing precursor emissions by a cap-and-trade system is generally more cost-effective than a reduction-by-regulation design, since it has more flexibility to achieve specific emission targets using the market-based approach. Final climate-responsive air quality management strategies would be combinations of different categories of management options, since the effectiveness, costs, and uncertainties vary from region to region. In addition, emissions of black carbon (BC) and other short-lived species that contribute air pollution and climate change are fostered by similar anthropogenic activities, such as transportation and biofuel burning (Bond et al., 2004). Reductions in BC emissions from fossil fuel combustion and biofuel sources are expected to improve air quality and mitigate climate change and may prove to be more desirable for decision makers.

Step 4: Identification of optimal air quality management strategies through risk-based decision-making models

The final step for developing climate-responsive air quality management strategies is to identifying “optimal” solutions. In regional air quality management, the optimal management strategies may represent combinations of various control measures that minimize costs of emission reductions or minimize the impacts of climate change on air quality when limited resources are available. Implementation of air quality management strategies under the impacts of climate change requires substantial investments by governments, private sectors, and stakeholders, all of whom face many other demands on their resources. As such, it is important to identify optimal climate-responsive strategies among a wide variety of air quality management alternatives. Several approaches can be used to identify optimal air quality management strategies through decision-making models. One common type of decision-making models is “least-cost,” which minimize costs of air pollution controls when achieving prescribed air quality targets. To develop least-cost management strategies for ozone and PM_2.5 air quality, we define the following decision variables and parameters.

	ϵ ij  Ratios of emissions reduced (ton) to total emissions (ton) for species i from region j (unitless)
	Cost ij  Cost functions of reductions in species i emissions from region j ($/ton)
	S O3, ij  Sensitivities of ozone to species i emissions from region j (ppb/100% reduction in species i emissions)
	S PM2.5, ij  Sensitivities of PM2.5 to species i emissions from region j ([μg/m3]/100% reduction in species i emissions)
	C O3 ,prior Prior O3 concentrations (ppb)
	C O3 ,target  Target O3 concentrations (ppb)
	C PM2.5 ,prior Prior PM2.5 concentrations (μg/m3)
	C PM2.5 ,target Target PM2.5 concentrations (μg/m3)
	Rij      Maximum available reduction ratios (i.e., ratios of maximum possible emission reductions [tons] to total emissions [ton]) for species i emissions from region j
	I Number of emission species
	J Number of regions

The complete mathematical form of the least-cost model for achieving prescribed ozone and PM_2.5 air quality targets in multiple areas is as follows.

Model 1: Least-cost model

(1)

Subject to:

(2)

(3)

(4)

where i = anthropogenic SO₂, NO_x, VOC, and primary PM_2.5 emissions and j = regions of emissions (e.g., Northeast United States, Southeast United States, etc.)

The cost estimate of reductions in species i emissions from region j is used as the objective function in the least-cost model (eq 1). Equation 2 represents that decreases in ozone concentrations, attributed to controls of anthropogenic NO_x and VOC emissions, should be larger than or equal to desired reductions in ozone levels, whereas eq 3 represents that decreases in PM_2.5levels, attributed to controls of anthropogenic SO₂, NO_x, VOC, and primary PM_2.5 emissions, should be larger than or equal to prescribed reductions in PM_2.5 levels. Furthermore, the calculation needs to include maximum available reduction ratios (i.e., ratios of emissions reduced to total emissions), accounting for technological, economic, and other limits (eq 4). Liao et al. (2010) use the least-cost model estimating that an annual cost of $9.3 billion (1999$) will be required for offsetting the impacts of climate change on ambient ozone and PM_2.5 in 2050 over six regions and five urban areas in the United States (Liao et al., 2009a). However, it is important to note that the air quality targets may not be achieved using known emission control measures, since the required magnitude of emission reductions may be beyond that where feasible controls have been identified.

Another useful approach for developing climate-responsive air quality management strategies is to minimize the adverse effects of climate change when considering scarce resources. In December 2009, The United Nations Climate Change Conference (COP15) in Copenhagen ended with an agreement, The Copenhagen Accord, which recognizes the importance of a two-degree increase in global temperatures for staving off the worst effects of climate change. According to The Copenhagen Accord, developed countries should commit to a goal of mobilizing jointly US$100 billion per year by 2020 for developing countries to mitigate climate change and adapt to the environmental consequences of climate change (United Nations Framework Convention on Climate Change [UNFCCC], 2009). One important issue regarding the success of The Copenhagen Accord is to identify the most efficient way to allocate the available resource. For developing climate-responsive regional air quality management strategies, the “most efficient way” means emission control alternatives that minimize the impacts of climate change when considering scarce resources. Here, we propose the second type of decision-making models, a resource allocation model, which identifies optimal air quality management strategies that minimize the climate change effects when considering the use of scarce resources. If the objective of the decision-making is to minimize the health effects of air pollution attributed to climate change, the complete form of the resource allocation model is as follows.

Model 2: Resource allocation model

(5)

Subject to:

(6)

(7)

where y _o is the baseline incidence rate of the health effect (i.e., the incidence rate (as a percent) before the change in pollutant levels), ΔX (unit: ppb for ozone and μg/m³ for PM_2.5) is the changes in pollutant levels attributing to climate change, the parameter β is the mortality toxicity coefficient of pollutants (unit: 1/[ppb] for ozone and 1/[μg/m³] for PM_2.5) from epidemiological studies (Bell et al., 2004; Pope et al., 2002), and population is the number of people affected by air pollution. Equation 5 is the objective function of the resource allocation model and represents the health incidences attributed to air pollution changes due to climate change. Equation 6 presents that the total cost of the implementation of air quality management strategies should be smaller than or equal to the budgets available for combating air pollution with the consideration of climate change. Similar to the least-cost model, the capacities of each of the air pollution reduction measures need to be included in the resource allocation model (eq 7). The purpose of the resource allocation model is to minimize the health incidence (eq 5) attributed to climate-related air pollution when constraints of resources (eq 6) and emission control efficiencies (eq 7) are satisfied. Overall, the success of the air quality management in response to climate change depends on the availability of necessary resources, not only financial resources, but also knowledge and technical capability. Through the risk assessment and risk management process, it is expected that climate-responsive air quality management strategies can be more economically effective, and an increasing number of governments and stakeholders are expected to be more willing to provide financial support to combat air pollution with the consideration of climate change. Finally, because the knowledge about future climate change impacts and the effectiveness of pollution control measures is expected to improve in the future, the development of air quality management strategies should evolve with time when more information of climate change and potential control measures are available (Figure 1).

Conclusions

Development of climate-responsive air quality management strategies is fundamentally a risk assessment and risk management process. Effective approaches for developing air quality management strategies under the impacts climate change first depend on an understanding of projected impacts of climate change on air quality at different spatial and temporal scales. Managing the risks of air quality related to climate change involves selecting potential air quality management options, determining air quality targets and, finally, identifying optimal air quality management strategies. The potential air pollution control options should be carefully selected and evaluated according to different air quality management targets and available resources for air quality improvement. The optimal management strategies can be identified based on least-cost or resource allocation models when considering scarce resources.

The main challenges for developing climate-responsive air quality management strategies relate to (1) the level of the uncertainty associated with climate change forecasts, which make it challenging to predict the impacts of climate change on air quality; (2) advances in future air pollution control technologies. Technological advances, and identifying approaches that can lead to decreased climate forcing and air pollution, could significantly affect selections of air quality management options, and their uncertainties should be taken into account the development of air quality management strategies. The concept presented in this paper can help decision makers make appropriate responses to climate change when developing air quality management strategies.

Acknowledgments

The authors thank EPA for providing funding under STAR grants RD83096001, RD82897602, and RD83107601. The views expressed in this paper are those of the authors and do not necessarily reflect the views or policies of EPA. Kuo-Jen Liao is also supported by the University Research Award from Texas A&M University–Kingsville. Efthimios Tagaris is also supported by the FP7-REGPOT-2008-1 grant 229773.