Response of inorganic fine particulate matter to emission changes of sulfur dioxide and ammonia: the eastern United States as a case study.

ABSTRACT

A three-dimensional chemical transport model (PMCAMx) was used to investigate changes in fine particle ([PM.sub.2.5]) concentrations in response to changes in sulfur dioxide (S[O.sub.2]) and ammonia (N[H.sub.3]) emissions during July 2001 and January 2002 in the eastern United States. A uniform 50% reduction in S[O.sub.2] emissions was predicted to produce an average decrease of [PM.sub.2.5] concentrations by 26% during July but only 6% during January. A 50% reduction of N[H.sub.3] emissions leads to an average 4 and 9% decrease in [PM.sub.2.5] in July and January, respectively. During the summer, the highest concentration of sulfate is in South Indiana (12.8 [micro]g x [m.sup.-3]), and the 50% reduction of S[O.sub.2] emissions results in a 5.7 [micro]g x [m.sup.-3](44%) sulfate decrease over this area. During winter, the S[O.sub.2] emissions reduction results in a 1.5 [micro]g x [m.sup.-3] (29%) decrease of the peak sulfate levels (5.2 [micro]g x [m.sup.-3]) over Southeast Georgia. The maximum nitrate and ammonium concentrations are predicted to be over the Midwest (1.9 [micro]g x [m.sup.-3] in Ohio and 5.3 [micro]g x [m.sup.-3] in South Indiana, respectively) in the summer whereas in the winter these concentrations are higher over the Northeast (3 [micro]g x [m.sup.-3] of nitrate in Connecticut and 2.7 [micro]g x [m.sup.-3] of ammonium in New York). The 50% N[H.sub.3] emissions reduction is more effective for controlling nitrate, compared with S[O.sub.2] reductions, producing a 1.1 [micro]g x [m.sup.-3] nitrate decrease over Ohio in July and a 1.2 [micro]g x [m.sup.-3] decrease over Connecticut in January. Ammonium decreases significantly when either S[O.sub.2] or N[H.sub.3] emissions are decreased. However, the S[O.sub.2] control strategy has better results in July when ammonium decreases, up to 2 [micro]g x [m.sup.-3] (37%), are predicted in South Indiana. The N[H.sub.3] control strategy has better results in January (ammonium decreases up to 0.4 [micro]g x [m.sup.-3] in New York). The spatial and temporal characteristics of the effectiveness of these emission control strategies during the summer and winter seasons are discussed.

INTRODUCTION

Atmospheric particles have adverse effects on human health and have been implicated in the formation of acid rain and acid fogs, visibility reduction, and changes of the energy balance of the planet. Particulate matter (PM) less than 2.5 [micro]m in size ([PM.sub.2.5]), ozone, and other pollutants are related through a complex web of common emissions and precursors, photochemical production pathways, and meteorological processes. Therefore reductions in emissions of one pollutant can lead to changes (either positive or negative) in the concentrations of other pollutants. In the eastern United States, [PM.sub.2.5] is primarily anthropogenic, and inorganic species account for approximately half of the total mass. (1) Oxidation of sulfur dioxide (S[O.sub.2]) is the primary source of sulfate (S[O.sub.4.sup.2-]), which is the predominant inorganic aerosol component in the eastern United States. S[O.sub.2] emissions reduction is expected to be an effective control strategy for reducing [PM.sub.2.5]. However, reductions in S[O.sub.4.sup.2-] aerosol can lead to higher levels of gas-phase ammonia (N[H.sub.3]), which can then increase ammonium nitrate. (2,3) Ammonium nitrate, a potential contributor to inorganic [PM.sub.2.5], can be controlled through either a reduction of ammonia or oxides of nitrogen (N[O.sub.x]) emissions, or both.

A variety of approaches has been used to explore such source-ambient [PM.sub.2.5] concentration relationships. Several atmospheric models have been developed based on thermodynamic equilibrium principles to predict inorganic atmospheric aerosol behavior. These include: EQUIL, (4) MARS, (5) SEQUILIB, (6) AIM, (7) SCAPE, (8,9) SCAPE2, (10,11) EQUISOLV, (12) AIM2, (13) ISORROPIA, (14) GFEMN, (15) and EQUISOLV II. (16) Some of these models have been used directly to estimate the effectiveness of emissions control strategies. For instance SEQUILIB was used to evaluate the effects of emissions reductions of precursor species on ambient PM concentrations during the winter in Phoenix, AZ. (17) Ansari and Pandis (2) used an aerosol thermodynamic model (GFEMN) to estimate the conditions for a nonlinear response of [PM.sub.2.5] to changes in S[O.sub.4.sup.2-] concentration in the sulfate-nitrate-ammonium-water system. These responses are functions of temperature and relative humidity (RH), as well as the concentrations of S[O.sub.4.sup.2-], total nitrate (N[O.sub.3.sup.-]), and total N[H.sub.3]. As a measure of this nonlinear response West et al. (3) introduced the term "marginal [PM.sub.2.5]," defined as the local change in [PM.sub.2.5] resulting from a small change in the concentration of a simple chemical species. The conditions for nonlinear [PM.sub.2.5] response to S[O.sub.4.sup.2-] reductions were found to be common in the eastern United States during winter and uncommon during summer. A useful concept for the design of the control strategies for [PM.sub.2.5] is that of a limiting reactant. Available thermodynamic models can be used to determine if N[H.sub.3], sulfuric acid ([H.sub.2]S[O.sub.4]), or nitric acid (HN[O.sub.3]) are limiting the formation of particulate N[O.sub.3.sup.-], using measurements of the gas and PM concentrations of these precursors. (17,18) A thermodynamic equilibrium model (SCAPE2), was used to investigate the response of fine particulate N[O.sub.3.sup.-] to changes in concentrations of HN[O.sub.3], N[H.sub.3], and S[O.sub.4.sup.2-] in the southeastern United States. (19) San Martini et al. (20) used the inorganic aerosol model ISORROPIA to calculate the response of inorganic aerosols to changes in precursor concentrations in Mexico City. Vayenas et al. (21) introduced a Eulerian box model, TMR, to investigate the behavior of the S[O.sub.4.sup.2-]-N[H.sub.3]-HN[O.sub.3] system. This model relies on measured [PM.sub.2.5] precursor concentrations and accounts for the variable deposition rates between aerosol N[O.sub.3.sup.-] and gas-phase HN[O.sub.3].

There have been rather limited published efforts to investigate the changes in fine particle concentrations in response to changes in precursor emissions by using three-dimensional chemical transport models (CTMs). Meng et al. (22) applied a three-dimensional, size and chemically resolved CTM to examine how the chemical coupling between ozone and PM influences joint control efforts of the two pollutants over the South Coast Air Basin of California during a summer smog episode in 1987. The University of California-Davis/California Institute of Technology photochemical transport model was used to examine the effect of N[O.sub.x], volatile organic compounds (VOCs), and N[H.sub.3] emission control programs on the formation of particulate ammonium nitrate in San Joaquin Valley during a winter episode. (23) Only a few studies have used CTMs to evaluate control strategies over the eastern United States. (24) The significant change in S[O.sub.2], N[H.sub.3], and other emissions and the corresponding emission inventories has moved this nonlinear system to a different state and has properly changed its response to emission control strategies compared with those analyzed by older studies. (3) Moreover, the comparison between the predictions of a CTM and the often-used box models has not been thoroughly investigated.

The objective of this study was to estimate the response of fine particle mass to changes in S[O.sub.2] and N[H.sub.3] emissions in the eastern United States. The responses of [PM.sub.2.5] to VOC and N[O.sub.x] emission changes will be examined in detail in future work. A three-dimensional transport model (PMCAMx) is well suited for this purpose because it directly links emissions to [PM.sub.2.5] concentrations through detailed descriptions of the physics and chemistry of the atmosphere. Two months (July 2001 and January 2002) are simulated to investigate the seasonal dependence of the [PM.sub.2.5] responses to emission changes.

The remainder of this paper is organized as follows. First, there is a brief description of PMCAMx and the details of its application in this domain. In the next section, the spatial and temporal characteristics of the effectiveness of the different emission control strategies (50% reductions of S[O.sub.2] and N[H.sub.3]) to S[O.sub.4.sup.2-], N[O.sub.3.sup.-], ammonium, and total [PM.sub.2.5] mass are discussed separately. The effects of a coupled 50% reduction of both S[O.sub.2] and N[H.sub.3] emissions are analyzed in the following section. The linearity of the system response to these control strategies is investigated by examining the results of a 25% reduction of S[O.sub.2] and N[H.sub.3] emissions during both seasons. Finally, the strengths and limitations of each control strategy for the different seasons are discussed.

THE PMCAMX CHEMICAL TRANSPORT MODEL

Model Description