Hidden Costs of Energy: Unpriced Consequences of Energy Production and Use (2010)

The Air Pollution Emission Experiments and Policy (APEEP) analysis model (Muller and Mendelsohn 2006, 2009) is a traditional integrated assessment model (Mendelsohn 1980; Nordhaus 1992; Burtraw et al. 1998; EPA 1999). Like other integrated assessment models, APEEP connects emissions of air pollution through air-quality modeling to exposures, physical effects, and monetary damages. Making these links requires the use of findings reported in the peer-reviewed literature across several scientific disciplines.

APEEP is designed to calculate the marginal damage of emissions for nearly 10,000 distinct (individual and aggregated sources of air pollution in the contiguous United States. APEEP computes marginal damages of six pollutants: sulfur dioxide (SO₂), volatile organic compounds (VOCs), nitrogen oxides (NO_x), fine particulate matter (PM_2.5), coarse particulate matter (PM₁₀),¹ and ammonia (NH₃).

The individual and aggregate sources are defined by the U.S. Environmental Protection Agency (EPA 2009). Sources of emissions include both county-aggregated ground-level sources as well as point sources. Ground-level sources include vehicles, residences, and small industrial or commercial facilities without a smokestack. Emissions from individual ground-level sources are aggregated at the county level by EPA. Point sources are dif-

ferentiated by effective stack height and by location because the height of emissions affects the dispersion patterns from these sources. Emissions from point sources with an effective height of less than 250 m are aggregated to the county level, as are emissions from point sources with an effective height of 250 to 500 m. In contrast, point sources with an effective height of greater than 500 m, such as certain power plants and other large industrial facilities, are modeled individually—that is, APEEP does not aggregate emissions from these sources; they are modeled separately for each facility.

The air-quality models in APEEP use the emission data provided by EPA to estimate corresponding ambient concentrations in each county in the coterminous states. The accuracy of the estimated pollution levels produced by the APEEP model has been statistically tested against the Community Multiscale Air Quality (CMAQ) model (Byun and Schere 2006), which is considered the state-of-the-art air-quality model. The results of these statistical comparisons are shown in the accompanying materials to Muller and Mendelsohn (2006).

APEEP can be used to compute the marginal damage of emissions on a source-specific basis. This approach isolates the source-specific damage per ton for each of the six pollutants covered by the model. To calculate marginal damages, APEEP uses the following algorithm: First, APEEP estimates total damages due to all sources in the model, producing its baseline (observed) emissions (EPA 2009); next, APEEP adds 1 ton of one pollutant from one source and recomputes total damages. The marginal damage is the damage that occurs after adding 1 ton of pollutant minus the damages due to the baseline emissions. The algorithm isolates the contribution of a single ton of emissions from each source to total national damages. This approach captures the formation of secondary pollutants, such as sulfates and nitrates (constituents in PM_2.5) as well as tropospheric ozone (O₃) that are formed by the emissions of other substances. APEEP attributes the damage due to such secondary pollutants back to the source of emissions. As shown in Equation 1, the marginal damage is computed by adding the changes in damages across the complete set of receptor counties. (Receptor counties are those counties that receive emissions from a source.)

(1)

where

MD_i,p = damage per ton of an emission of pollutant (p) from source (i).

D_r = total dollar damage that occurs at receptor county (r).

bp = 2002 baseline emissions of p.

ep = 2002 baseline emissions plus 1 ton of p from i.

After computing the marginal damage of emissions for a specific pollutant from source i, this experiment can be repeated for each of the six pollutants covered in APEEP and the approximately 10,000 distinct (individual and grouped) sources in the United States. The total 10,000 sources encompass all anthropogenic emissions of these six pollutants in the lower 48 states. It is important to note that the APEEP model, in its current form, does not test for interactions among emissions of multiple pollutants in terms of the damages that such emissions cause. The model is designed to simulate the emissions of 1 ton of one specific pollutant from a particular source and to estimate its impact rather than the emissions of multiple pollutants from a source and estimating their cumulative impact.

The following section briefly highlights the basic structure of the model and some of its most important assumptions. The model uses data on emissions (excluding carbon monoxide and lead and including ammonia) that contribute to the formation of criteria air pollutants. The data were provided by EPA’s 2002 National Emission Inventory (EPA 2009). Concentrations due to the baseline levels of emissions are estimated by the air-quality models in APEEP. The air-quality modeling module makes use of a source-receptor matrix framework. That is, the marginal contribution of emissions in a source county (s) to the ambient concentration in a receptor county (r) is represented as the s,r element in a matrix. Using a linear algebraic approach, APEEP multiplies the matrix times an emission vector to generate a vector of predicted ambient concentrations. When the emission vectors represent changes to existing emissions, the corresponding estimated concentrations reflect changes to the baseline levels, or existing concentrations. When the emission vectors represent the emission rates, then predicted concentrations reflect those rates, not changes to concentrations.

The model contains source-receptor matrices for the following pollutants in both summer and winter: NO_x → NO_x, SO₂ → SO₂. The matrix governing the relationship between NO_x emissions, VOC emissions, and O₃ concentrations is calibrated to the summer season. The matrices representing formation and transport of particles (PM_2.5 → PM_2.5, PM₁₀ → PM₁₀, NO_x → PM, SO₂ → PM, NH₃ → NH₄, VOC → PM) produce annual means.² There is a specific matrix in APEEP for each of the emission-concentration relationships shown above.

The particulate matter source-receptor matrices compute the ammonium-sulfate-nitrate equilibrium, which determines the amount of ambient ammonium sulfate (NH₄)₂, SO₄, and ammonium nitrate (NH₄NO₃) at each receptor county. The equilibrium computations reflect several fundamental

aspects of this system. First, ambient ammonium (NH₄) reacts preferentially with sulfate (H₂SO₄). Second, ammonium nitrate is only able to form if there is excess NH₄ after reacting with sulfate. Finally, particulate nitrate formation is a decreasing function of temperature, so the ambient temperature at each receptor location is incorporated into the equilibrium calculations. To translate VOCs emissions into secondary organic particulates, APEEP uses the fractional aerosol yield coefficients estimated by Grosjean and Seinfeld (1989). These coefficients represent the yield of secondary organic aerosols corresponding to emissions of gaseous VOCs.

APEEP simulates O₃ concentrations using an empirical model that translates ambient concentrations of VOC, CO, and NO_x into ambient O₃ concentrations. The model captures many of the factors contributing to ambient concentrations of O₃, VOC, CO, and NO_x. These factors include forests and agricultural land uses, which produce biogenic hydrocarbons, as well as the ambient air temperature and several geographic variables. For a complete depiction of the O₃ modeling in APEEP, see Muller and Mendelsohn (2006). The inclusion of both linear and quadratic forms for NO_x, CO, and VOC concentrations in the O₃ models allows for the nonlinearity known to exist in O₃ production chemistry (Seinfeld and Pandis 1998). Specifically, the quadratic forms capture titration. This approach is critical to accurately predict O₃ levels in certain urban areas, where research has shown that additional emissions of NO_x can result in reduced O₃ concentrations (Tong et al. 2006).

The source-receptor matrices in APEEP are derived from the Climatological Regional Dispersion Model (CRDM) (Latimer (1996). The original CRDM matrices have been calibrated to produce estimates of pollution levels that are in good agreement with the predictions produced by CMAQ. The correlations between APEEP’s predicted surfaces and CMAQ’s are especially strong for annual mean PM_2.5 levels and summer mean O₃ levels; the correlation coefficients are 0.82 and 0.77, respectively.³ The matrices have been expanded in scope to encompass nearly 10,000 sources and source areas.

Following the estimation of ambient concentrations, exposures are computed by multiplying county-level populations times county-level pollution concentrations. In APEEP, populations include number of people (differentiated by age),⁴ crops produced, timber harvested, an inventory of anthropogenic materials, visibility resources, and recreation usage (for each

county in the contiguous United States). Each type of exposure is computed separately for each pollutant. The sources for each of these inventories are documented in Muller and Mendelsohn (2006).

In the next stage of the APEEP model, peer-reviewed concentration-response functions are used to translate exposures into the number of physical effects, including premature mortalities, cases of illness, reduced timber and crops yields, enhanced depreciation of anthropogenic materials, reduced visibility, and recreation usage. The studies that provide the concentration-response functions related to human health impacts are listed in Table C-1.

The final stage of the APEEP model attributes a dollar value to each of these physical effects. For effects on goods and services traded in markets (decreased crop yields, for example), APEEP multiplies the change in output due to exposures to air pollution times the market price. For nonmarket goods and services, APEEP uses valuation estimates from the nonmarket valuation literature in economics. APEEP values premature mortality risks using the value of a statistical life (VSL) approach (Viscusi and Aldy 2003). APEEP uses EPA’s preferred VSL, which is equivalent to approximately $6 million (year 2000 real U.S. dollars). APEEP provides the option of using a VSL estimate of approximately $2 million from Mrozek and Taylor (2002) as an alternative to the EPA’s VSL. The values attributed to chronic illnesses, such as bronchitis and asthma, are also derived from the nonmarket valuation literature. Acute illnesses are valued with cost of illness estimates. Each of the values applied to human health effects in APEEP are shown in Table C-3.

The studies that provide the concentration-response functions for the remaining welfare effects are listed in Table C-2. Because PM_2.5 is a subset of PM₁₀, APEEP avoids double counting of damages due to PM_2.5 and PM₁₀. Specifically, APEEP estimates mortality impacts associated with emissions of PM_2.5, and the model measures chronic morbidity impacts of PM₁₀. In reporting the morbidity damages due to emissions of PM₁₀, APEEP nets out the mortality damages due to PM_2.5. In effect, the damages for PM₁₀ are expressed as PM₁₀-PM_2.5.

Each of the other nonmarket impacts of air pollution modeled by APEEP (impaired visibility and reduced recreation services) are also expressed in dollar terms. The values used in the APEEP model corresponding to these welfare effects are displayed in Table C-4.

Abbey, D.E., F. Peterson, P.K. Mills, and W.L. Beeson. 1993. Long-term ambient concentrations of total suspended particulates, ozone, and sulfur dioxide and respiratory symptoms in a nonsmoking population. Arch. Environ. Health 48(1):33-46.

Atteras, L., and S. Haagenrud. 1982. Atmospheric corrosion testing in Norway. Pp. 873-892 in Atmospheric Corrosion, W.H. Ailor, ed. New York: Wiley.

Bell, M.L., A. McDermott, S.L. Zeger, J.M. Samet, and F. Domenici. 2004. Ozone and short-term mortality in 95 U.S. urban communities, 1987-2000. JAMA 292(17):2372-2378.

Burnett, R.D., M. Smith-Doiron, D. Steib, S. Cakmak, and J. Brook. 1999. Effects of particulate and gaseous air pollution on cardiorespiratory hospitalizations. Arch. Environ. Health 54(2):130-139.

Burtraw, D., A. Krupnick, E. Mansur, D. Austin, and D. Farrell. 1998. Costs and benefits of reducing air pollutants related to acid rain. Contemp. Econ. Policy 16 (4):379-400.

Byun, D.W., and L.K. Schere. 2006. Review of the governing equations, computational algorithms, and other components of the models-3 Community Multiscale Air Quality (CMAQ) modeling system. Appl. Mech. Rev. 59(2):51-77.

Chestnut, L.G., and R.D. Rowe. 1990. Economic valuation of changes in visibility: A state of the science assessment for NAPAP. Pp. 27-153 to 27-175 in Report 27. Methods for Valuing Acidic Deposition and Air Pollution Effects. Acidic Deposition: State of Science and Technology, Vol. 4. Control Technologies, Future Emission, and Effects Valuation, P.M. Irving, ed. Washington, DC: U.S. National Acid Precipitation Assessment Program.

EPA (U.S. Environmental Protection Agency). 1999. The Benefits and Costs of the Clean Air Act 1990 to 2010, EPA Report to Congress. EPA-410-R-99-001. Office of Air and Radiation, Office of Policy, U. S. Environmental Protection Agency, Washington, DC. November 1999 [online]. Available: http://www.epa.gov/oar/sect812/1990-2010/chap1130.pdf [accessed Sept. 16, 2009].

EPA (U.S. Environmental Protection Agency). 2009. National Emissions Inventory (NEI) Data & Documentation. Office of Air Quality Planning and Standards, U. S. Environmental Protection Agency, Washington, DC [online]. Available: http://www.epa.gov/ttnchie1/net/2002inventory.html [accessed Sept 16, 2009].

Grosjean, D., and J. Seinfeld. 1989. Parameterization of the formation potential of secondary organic aerosols. Atmos. Environ. 23:1733-1747.

ICP (International Co-operative Programme). 2000. International Co-operative Programme on Effects of Air Pollution on Materials, including Historic and Cultural Monuments: Results [online]. Available: http://www.corr-institute.se/ICP-Materials/web/page.aspx?pageid=59263 [accessed Apr. 15, 2010].

Kengen, S. 1997. Forest Valuation for Decision-Making: Lessons of Experience and Proposals for Improvement. Food and Agriculture Organization of the United Nations, Rome, Italy. February 1997 [online]. Available: ftp://ftp.fao.org/docrep/fao/003/W3641E/W3641E00.pdf [accessed Sept. 15, 2009].

Latimer, D.A. 1996. Particulate Matter Source-Receptor Relationships Between all Point and Area Sources in the United States and PSD Class I Area Receptors. Prepared for Office of Air Quality Planning and Standards, U.S. Environmental Protection Agency, Research Triangle Park, NC. September 1996.

Lesser, V.M., J.O. Rawlings, S.E. Spruill, and M.C. Somerville. 1990. Ozone effects on agricultural crops: Statistical methodologies and estimated dose-response relationships. Crop Sci. 30(1):148-155.

McClelland, G.H., W.D. Schulze, D. Waldman, D. Schenk, J.R. Irwin, T. Stewart, L. Deck, and M.A. Thayer. 1993. Valuing Eastern Visibility: A Field Test of the Contingent Valuation Method. Prepared for Office of Policy Planning and Evaluation, U.S. Environmental Protection Agency, Washington, DC, by the University of Colorado. September 1993 [online]. Available: http://yosemite1.epa.gov/ee/epa/eerm.nsf/vwAN/EE-0008-1.pdf/$file/EE-0008-1.pdf [accessed Sept. 15, 2009].

McDonnell, W.F., D.E. Abbey, N. Nishino, and M.D. Lebowitz. 1999. Long-term ambient ozone concentration and the incidence of asthma in non-smoking adults: The AHSMOG study. Environ. Res. 80(1):110-121.

Mendelsohn, R. 1980. An economic analysis of air pollution from coal-fired power plants. J. Environ. Econ. Manage. 7:30-43.

Moolgavkar, S.H. 2000. Air pollution and hospital admissions for chronic obstructive pulmonary disease in three metropolitan areas in the United States. Inhal. Toxicol. 12(Suppl. 4):75-90.

Mrozek, J.R., and L.O. Taylor. 2002. What determines the value of life? A meta-analysis. J. Policy Anal. Manage. 21(2):253-270.

Muller, N.Z., and R.O. Mendelsohn. 2006. The Air Pollution Emission and Policy Analysis Model (APEEP): Technical Appendix. Yale University, New Haven, CT. December 2006 [online]. Available: https://segueuserfiles.middlebury.edu/nmuller/ APEEP_Tech_Appendix.pdf [accessed Oct. 7, 2009].

Muller, N.Z., and R.O. Mendelsohn. 2007. Measuring the damages from air pollution in the U.S. J. Environ. Econ. Manage. 54(1):1-14.

Muller, N., and R. Mendelsohn. 2009. Efficient pollution regulation: Getting the prices right. Am. Econ. Rev. 99(5):1714-1739.

Nordhaus, W.D. 1992. An optimal transition path for controlling greenhouse gases. Science 258 (5086):1315-1319.

Pope, C.A., R.T. Burnett, M.J. Thun, E.E. Calle, D. Krewski, K. Ito, and G.D. Thurston. 2002. Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. JAMA 287(9):1132-1141.

Pye, J.M. 1988. Impact of ozone on the growth and yield of trees: A review. J. Environ. Qual. 17:347-360.

Reich, P.B. 1987. Quantifying plant response to ozone: A unifying theory. Tree Physiol. 3(1):63-91.

Schwartz J. 1995. Short term fluctuations in air pollution and hospital admissions of the elderly for respiratory disease. Thorax 50(5):531-538.

Schwartz, J., D. Slater, T.V. Larson, W.E. Pierson, and J.Q. Koenig. 1993. Particulate air pollution and hospital emergency room visits for asthma in Seattle. Am. Rev. Respir. Dis. 147(4):826-831.

Seinfeld, J.H., and S.N. Pandis. 1998. Atmospheric Chemistry and Physics. New York: John Wiley & Sons.

Sheppard, L., D. Levy, G. Norris, T.V. Larson, and J.Q. Koenig. 1999. Effects of ambient air pollution on nonelderly asthma hospital admissions in Seattle, Washington, 1987-1994. Epidemiology 10(1):23-30.

Steib, D.M., R.T. Burnett, R.C. Beveridge, and J.R. Brook. 1996. Association between ozone and asthma emergency department visits in St. Jon, New Brunswick, Canada. Environ. Health Perspect. 104(12):1354-1360.

Tong, D.Q., N.Z. Muller, D.L. Mauzerall, and R.O. Mendelsohn. 2006. Integrated assessment of the spatial variability of ozone impacts from emissions of nitrogen oxides. Environ. Sci. Technol. 40(5):1395-1400.

Viscusi, W.K., and J.E. Aldy. 2003. The value of a statistical life: A critical review of market estimates throughout the world. J. Risk Uncertain. 27(1):5-76.

Woodruff, T.J., J.D. Parker, and K.C. Schoendorf. 2006. Fine particulate matter (PM2.5) air pollution and selected causes of postneonatal infant mortality in California. Environ. Health Perspect. 114(5):786-790.