![[logo] US EPA](https://webarchive.library.unt.edu/eot2008/20081107121516im_/http://www.epa.gov/epafiles/images/logo_epaseal.gif){kind=logo}
Final Report: Development and Testing of a State-of-the-Art PMx Particulate Matter Module for Regional and Urban Air Pollution Models
EPA Grant Number: R824793Title: Development and Testing of a State-of-the-Art PMx Particulate Matter Module for Regional and Urban Air Pollution Models
Investigators: Pandis, Spyros N.
Institution: Carnegie Mellon University
EPA Project Officer: Katz, Stacey , Robarge, Gail
Project Period: October 1, 1995 through September 30, 1998
Project Amount: $412,041
RFA: Air Pollutants (1995)
Research Category: Air Quality and Air Toxics
Description:
Summary/Accomplishments (Outputs/Outcomes):Airborne particulate matter (PM) continues to pose serious health risks for susceptible members of the US population and for sensitive ecosystems. Recent EPA regulations have created a strong interest in better understanding atmospheric aerosol processes and the relationship of ambient particulate concentrations to both natural and anthropogenic emission fluxes. The current regulations are expressed as PMx standards, where x is a diameter less than 10 micrometers. Modeling PMx concentrations requires reevaluation of the aerosol modeling techniques used in urban and regional air quality models, since fine aerosol mass is more sensitive to processes like growth, coagulation, and nucleation. In addition particles larger than x cannot be neglected in the simulations since they compete with the smaller particles in processes like condensation and coagulation. Despite significant advances in recent years, our ability to simulate atmospheric aerosol processes both accurately and efficiently has remained limited. The goal of this project was the development of a state-of-the-art "smart" PMx atmospheric aerosol module, that is able to internally use suitable assumptions and approximations reducing computational cost, but at the same time provides the required accuracy for the atmospheric conditions where these approximations are inappropriate. The key element of the modeling approach will be the accurate estimation of the errors associated with the various assumptions and approximations used in the competing modeling approaches.
Our first task was the development of benchmark aerosol thermodynamics model for the evaluation of the approaches currently used in Eulerian atmospheric chemistry models [Ansari and Pandis, 1998a]. These existing models calculate the composition of the atmospheric aerosol system by solving a set of algebraic equations based on reversible reactions derived from thermodynamic equilibrium. Some approaches rely on an a priori knowledge of the presence of components in certain relative humidity regimes, and often fail to accurately predict deliquescence point depression and multistage aerosol growth. The present approach (GFEMN), relying on adjusted thermodynamic parameters of solid salts and a state-of-the-art activity coefficient model, directly minimizes the Gibbs free energy (according to thermodynamic equilibrium principles) given temperature, relative humidity, and the total (gas plus aerosol) ammonia, nitric acid, sulfate, sodium, and hydrochloric acid concentrations. A direct minimization while requiring no additional assumptions allows the elimination of many of the assumptions used in previous models. The current model (GFEMN) agrees with experimental results for single and multicomponent salt systems reproducing observed multistage growth behavior patterns, efflorescence behavior, and deliquescence point depression over a broad temperature range.
The above accurate but computationally intensive algorithm was used as the basis for the development of a computationally efficient but rigorous thermodynamic model [Nenes et al., 1998]. One of the main features of the model is the implementation of mutual deliquescence of multicomponent salt particles, which lowers the deliquescence point of the aerosol phase. The model was used to examine the behavior of four types of tropospheric aerosol (marine, urban, remote continental, and non-urban continental. The major advantage of the new model, ISORROPIA, is its speed coupled with its stability and robust convergence to the solution. These advantages render it suitable for incorporation into urban and regional air quality models. ISORROPIA was embodied into the UAM-AERO air quality model [Nenes et al., 1999] and its performance was compared with two other thermodynamic modules currently in use, SEQUILIB 1.5 and SEQUILIB 2.1. The new model yields predictions that agree with the field measurements and the results of the other models, but at the same time proves to be much faster and computationally efficient. Using ISORROPIA accelerates the thermodynamic calculations by more than a factor of six, which the overall speedup of UAM-AERO is at least twofold. This speed up is possible by the optimal solution of the thermodynamic equations, and the usage of pre-calculated tables, whenever possible.
A detailed comparison was conducted between four atmospheric equilibrium models: GFEMN, ISORROPIA, SCAPE2, and SEQUILIB [Ansari and Pandis, 1999a]. We examined model performance for representative atmospheric environments over an extended composition, temperature, and RH domain, and against observations in Southern California. The predictions of all four models for semivolatile aerosol components are in general agreement, but ISORROPIA, SCAPE2, and SEQUILIB do not adequately reproduce multistage deliquescence behavior for multicomponent systems. This difference will be expressed as errors in the predicted liquid water content concentrations for intermediate relative humidities. The most notable differences in model predictions occur for H+ and aerosol water concentrations; discrepancies in predictions of aerosol nitrate, ammonium, and total dry PM concentrations are not as significant. Against measurements taken during the Southern California Air Quality Study (SCAQS), all models qualitatively reproduce, but generally underpredict aerosol nitrate concentrations. Finally, based on its overall agreement with GFEMN and its computational efficiency, ISORROPIA appears to be the model of choice for use in large-scale aerosol transport models. In places where crustal material comprises a significant portion of total PM, use of SCAPE2 is recommended.
The effect of metastable equilibrium states (efflorescence) in comparison to stable (deliquescence) on the partitioning of nitrate was quantified based on the above aerosol thermodynamics models [Ansari and Pandis, 199b]. On average, efflorescence concentrations of aerosol nitrate are around 10% higher than those of deliquescence at low nitrate concentrations (less than 8 mg m-3), whereas for higher aerosol nitrate levels the deliquescence concentrations are 3% greater. In the low nitrate range, approximately 40% of the time deliquescence and efflorescence predictions have discrepancies over 20%. The largest discrepancies between the two equilibrium states occur at two sets of conditions: at temperatures above 295 K and intermediate RH (around 60%), and at intermediate temperatures (20-25oC) and low RH (less than 40%). In these two regions, average discrepancies of 1-2 mg m-3 between efflorescence and deliquescence aerosol nitrate concentrations are estimated. The potential existence of metastable solutions in Southern California, where pollutant levels are high, appears to have a small effect on nitrate partitioning. However, for areas characterized by moderate to low pollutant levels such as the Northeastern US, a significantly larger effect is predicted.
Our accurate aerosol thermodynamic model (GFEMN) was used to investigate the response of inorganic particulate matter concentrations with respect to the precursor concentrations of sulfuric acid, ammonia, and nitric acid over a range of temperatures and relative humidities [Ansari and Pandis, 1998b]. A graphical method was developed for the direct estimation of the response of PM to sulfate, and overall sensitivity to ammonia and nitric acid availability. The PM concentration level responds nonlinearly to sulfate and shows overall sensitivity to ammonia and nitric acid availability for specific atmospheric conditions and precursor concentrations. The generated diagrams were applied as a means of approximating the PM response to two urban polluted areas. In both cases, reductions in ammonia emissions have the most significant impact on the total PM level. However, such a reduction will result in significant increases in atmospheric acidity. The results of this analysis were applied to the NE US to determine how prevalent the conditions for nonlinearity are during the year [West et al., 1999]. Using the ambient measurements collected during the Eulerian Model Evaluation Field Study (EMEFS) from June 1998 to May 1990 we found that the conditions for a nonlinear response of fine aerosols to sulfate are widespread in the winter while they are rare during the summer. We identified locations where sulfate reductions will be up to 50% less effective than expected in reducing annual average PM2.5. In such conditions, a combination of controls on SO2 with NOx or ammonia or control in organics may be required to reduce PM2.5 levels.
The last task of the project was the development of the Multicomponent Aerosol Dynamics Model (MADM) [Pilinis et al., 1999]. The model simulates the complete atmospheric aerosol size/composition distribution by solving the condensation-evaporation equation. Condensable species may be organic and/or inorganic. For the inorganic constituents the new equilibrium model ISORROPIA is used to predict the physical state of the particle, i.e., whether the aerosol is an aqueous solution, solid, or partially aqueous and partially solid. The phase transitions between states are simulated for the first time in a dynamic framework. Due to its speed and modularity MADM can be used in 3D air quality models to predict the PMx concentration.
Journal Articles on this Report: 8 Displayed | Download in RIS Format
| Other project views: | All 8 publications | 8 publications in selected types | All 8 journal articles |
| Type | Citation | ||
|---|---|---|---|
|
|
Ansari AS, Pandis SN. Response of inorganic PM to precursor concentrations. Environmental Science and Technology 1998;32(18):2706-2714. |
R824793 (Final) |
|
|
|
Ansari AS, Pandis SN. Prediction of multicomponent inorganic atmospheric aerosol behavior. Atmospheric Environment Volume 33, Issue 5, February 1999, Pages 745-757. |
R824793 (Final) |
|
|
|
Ansari AS, Pandis SN. An analysis of four models predicting the partitioning of semi volatile inorganic aerosol components. Aerosol Science and Technology 1999;31(2-3):129-153. |
R824793 (Final) R826371 (Final) R826371C005 (Final) |
|
|
|
Ansari AS, Pandis SN. The effect of metastable equilibrium states on the partitioning of nitrate between the gas and aerosol phases. Atmospheric Environment 2000;34(1):157-168. |
R824793 (Final) R826371C005 (Final) |
|
|
|
Nenes A, Pandis SN, Pilinis C. ISORROPIA: A new thermodynamic equilibrium model for multiphase multicomponent inorganic aerosols. Aquatic Geochemistry 1998;4(1):123-152. |
R824793 (Final) |
|
|
|
Nenes A, Pandis SN, Pilinis C. Continued development and testing of a new thermodynamic aerosol module for urban and regional air quality models. Atmospheric Environment Volume 33, Issue 10, 1 May 1999, Pages 1553-1560. |
R824793 (Final) |
|
|
|
Pilinis C, Capaldo KP, Nenes A, Pandis SN. MADM-A new multicomponent aerosol dynamics model. Aerosol Science and Technology 2000, Volume: 32 , Number: 5 (MAY) , Page: 482-502. |
R824793 (Final) |
|
|
|
West JJ, Ansari AS, Pandis SN. Marginal PM2.5: Nonlinear aerosol mass response to sulfate reductions in the Eastern United States. Journal of the Air & Waste Management Association 1999;49(12):1415-1424. |
R824793 (Final) |
|
Air, Geographic Area, Scientific Discipline, RFA, Atmospheric Sciences, EPA Region, particulate matter, Environmental Chemistry, State, ambient measurement methods, susceptible populations, sensitive ecosytems, ambient air quality, Volatile Organic Compounds (VOCs), Pennsylvania, regional air quality models, Region 3, PMx mass concentration standard, mass monitoring, atmospheric chemistry, modeling studies, PA, anthropogenic emission fluxes, photochemical assessment monitoring, ambient aerosol
Progress and Final Reports:
Original Abstract