Monte carlo simulation of regional aerosol: transport and kinetics

Summary and Discussion

A redesigned and re-calibrated Monte Carlo model for the simulation of regional scale transport, transformation, and removal of atmospheric pollutants was presented. The model was developed in a modular fashion, separating the calculation of the pollutant transport from the kinetics. The pre-computed transport data could then be used for the calculation of kinetic processes. This procedure was based upon the assumption that pollutant species do not influence meteorology to a significant degree. Transport calculations were conducted using 3-D gridded mean wind fields, and simulating horizontal and vertical diffusion. The kinetic processes were calculated through kinetic rate equations, where the rate coefficients were based upon space and time dependent environmental variables.

The model was used to simulate the - system over the Eastern US during 1992. Comparisons, of simulated seasonal concentrations to observations, showed that the model was able to reproduce the spatial patterns and magnitudes. Comparisons, performed with daily observations, showed that the model adequately simulated concentrations in New England with seasonal r² = 0.56 - 0.83. In the East Central US, the seasonal r² = 0.35 - 0.61. Comparisons of simulated total wet deposition rates to observed rates also showed strong agreement. Correlations of the weekly data in both New England and the East Central had seasonal r² = 0.5 - 0.94.

A possible explanation, for the better correlations of the concentrations to observations in New England than the East Central domain, is that New England may be more dominated by regional scale phenomena than the East Central domain. There are few large sulfur emitting sources in the New England region. High and low concentrations of will tend to correspond to the transport of and from high and low emitting regions. The East Central domain, on the other hand, has large sources within and around the domain. Also, the Great Smoky Mountains are at the eastern edge of the domain. This is a complex situation, and the proper simulation of the concentrations is more likely to be dependent upon mesoscale transport and kinetics processes than in New England.

The Monte Carlo model was designed in such a manner as to allow for the investigation and quantification of the physical/chemical processes in the source receptor relationship. This was demonstrated by examining the role of source emissions, pollutant transport, and kinetics processes responsible for the source contributions of and to a Massachusetts receptor site during the quarter 3, 1992. It was found that the concentrations were highly influenced by nearby sources, such as those in Connecticut and Vermont. This was a result of the inverse relationship of the transit and kinetic probabilities with distance in the source receptor relationship. The high emissions in the Ohio River Valley also lead to significant contributions to the Massachusetts receptor concentration. The concentrations were dominated by emissions from more distant sources, particularly from the Ohio River Valley. This was due to the fact, that the concentrations of secondary species in a plume first increase and then decrease with time. Consequently, emissions from distant sources generally have less of a likelihood of impacting the receptor, but when they do, a higher fraction of the emissions is in the form of secondary pollutants than is the case with emissions from nearby sources.

Acknowledgment. Although the research described in this article has been funded in part by the U.S. Environmental Protection Agency under cooperative agreement CR 823756 and CR 818969, it has not been subjected to the Agency's peer and administrative review and therefore may not necessarily reflect the views of the Agency, and no official endorsement should be inferred. The information in this document has also been funded in part by Colorado State University Subcontract G-2887-4.

REFERENCES

1. Ottar, B. Ambio. 1976, 6, 203-206.

2. Hidy, G.M.; E.Y. Tong; P.K. Mueller. Design of the Sulfate Regional Experiment (SURE). Electric Power Research Institute, 1976; EC-125.

3. Pack, D.H.; G.J. Ferber; J.L. Heffter; K. Telegadas; J.K. Angell; W.H. Hoecker; L. Machta. Atmos. Environ. 1978, 12, 425-444.

4. Eliassen A. Atmos. Environ.. 1978, 12, 479-487.

5. Fisher, B.E.A. Atmos. Environ.. 1978, 12, 479-487.

6. Venkatram, A.; B. Ley; S. Wong. Atmos. Environ. 1982, 16, 249-257.

7. Draxler R.R.; Elliott W.P. Atmos. Environ. 1977, 11, 35-40.

8. Husar, R.B., D.E. Patterson, J.D. Husar, N.V. Gillani. Atmos Environ. 1978, 12, 549-568.

9. Husar, R.B.; D.E. Patterson; W.E. Wilson. A semi-empirical approach for selecting rate parameters for A Monte Carlo regional model. CAPITA Publication: Washington University St. Louis MO, 1986, 86r2.

10. Rolph, G. D.; Roland R Draxler; Rosa G. de Pena. Atmos. Environ. 1992, 26, 73-93.

11. Chang, J.S.; R. A. Brost; I.S.A. Isaksen; S. Madronich; P. Middleton; W. R. Stockwell; C.J. Walcek.. J. Geophys. Res. 1987, 92, 14681-14700.

12. Venkatram, A.; P.K. Karamchandani; P.K. Misra. Atmos. Environ. 1988, 22, 737-747.

13. Middleton P.; J. S. Chang; J.C. Del Conal; H. Geiss; J. M. Rosinkski. Atmos. Environ. 1988, 22, 1195-1208.

14. Patterson, D.E.; R.B. Husar; W.E. Wilson; L.F. Smith. J. Appl. Meteor. 1981, 20, 404-420.

15. Hall, C.D. Quart. J. Roy. Meteor. Soc. 1975, 101, 235-244.

16. Alsmiller, F.S.; Alsmiller R.G.; H.W. Bertini; C.L. Begovich. J. Appl. Meteor. 1979, 18, 17-26.

17. Boughton, B.A.; J.M. Delaurentis; W.E. Dunn. Boundary-Layer Meteorology. 1987, 40, 147-163.

18. Wiel, J.C. Atmos. Environ. 1994, 28, 3433-3448.

19. Luhar, A.K.; Rao, K.S. Atmos. Environ. 1994, 28, 3417-3431.

20. National Research Council. Science and Judgment in Risk Assessment. National Academy Press: Washington DC, 1994.

21. Hoke J.E.; N.A. Phillips; G.J. DiMego; J.J. Tuccillo; J.G. Sela. Wea. and Forecasting 1989, 4, 323-334.

22. Rolph G.D. National Climatic Data Center Report, 1992; TD-6140.

23. Pielke, R.A.; W.R. Cotton; R.L. Walko; C.J. Tremback; W.A. Lyons; L.D. Grasso; M.E. Nicholls; M.D. Moran; D.A. Ewlsey; T.J. Lee; J.H. Copeland. Meteorol. Atmos. Phys. 1992, 49, 69-91.

24. Lenschow, D.H.; Stephens, P.L. Boundary layer Meteor. 1980, 19, 509 - 532.

25. Willis, G.E.; Deardorff, J.W. J. Atmos. Sci. 1974, 31, 1297-1307.

26. Gillani, N.V. Atmos. Environ. 1978, 12, 569-588.

27. Gryning S.E., A.A.M. Holtslag, J.S. Irwin, and B. Sivertsen. Atmos. Environ. 1987, 21, 79-89.

28. Smith, F.B.; Hunt, R.D. Atmos. Environ. 1978, 12, 461-477.

29. Carras J.N.; Williams D.J. Atmos. Environ. 1980, 15, 2205-2217.

30. Munn, R.E.; Bolin B. Atmos. Environ. 1971, 5, 363-402.

31. Stocker, R.A.; M. Uliasz; R.A. Pielke. “Aerosol and Atmospheric Optics: Radiative Balance and Visual Air Quality, Volume A.” In Proceedings of the International Specialty Conference, Snowbird, Utah, September 1994.

32. Schichtel, B.A. The retrieval of pollutant emission fields from ambient concentration precipitation chemistry data. Ph.D. diss., Washington University, 1996.

33. National Center for Atmospheric Research (NCAR), DS505.0 NCDC TD3240, US Controled Hourly precipitation, Daily 1948-Cont.

34. National Renewable Energy Laboratory (NREL). September 1992. User’s Manual: National Solar Radiation Database (1961-1990), Volume 1.0. Distributed by National Climatic Data Center, Federal Building, Asheville, NC 28801.

35. Irving, Patricia M. Acidic Deposition: State of Science and Technology, Vol. I, Report 1, (Emissions). Government Printing Office: Washington, DC, 1991; 20402-9325.

36. Sisler J.F.; D. Huffman; D.A. Latimer; W.C. Malm; and M. Pitchford. Spatial and temporal patterns and the chemical composition of the haze in the Unites States: An analysis of data from the IMPROVE network, 1988-1991. CIRA, CSU, Fort Collins, CO, 1993; #ISSN No. 0737-5352-26.

37. Poirot, R. L.; P. J. Galvin; N. Gordon; S. Quan; A.V. Arsdale; R.G. Flocchini. “Annual and seasonal fine particle composition in the Northeast: second year results from the NESCAUM monitoring network.” In the A&WMA conference, Vancouver, Canada, 1991; 91-49.1.

38. Husar R.B.; Elkins J.B.; Wilson W.E. “US visibility trends, 1960-1992, regional and national.” In the 87th Annual Air & Waste Management Meeting, Cincinnati, OH, 1994.

39. US Environmental Protection Agency. AIRS User’s Guide, Volume AQ1: AQS Data Dictionary, February, 1994, EPA-454/B-94-005.

40. National Atmospheric Deposition Program/National Trends Network. NADP/NTN Coordination Office, Natural Resource Ecology Laboratory, Colorado State University, Fort Collins, CO, 1993.

41. Husar, R.B.; Husar, J.D. Fine particle maps derived from regional PM2.5 and visibility data. Final Report to Radian Corporation, NC, 1995; EPA Contract Number 68-D2-0160, Work Assignment Number 2032.

42. Lamb, R.G.; Seinfeld, J.H. Environmental Science & Technology, 1973, 7, 253-261.

43. Husar, R.B. Possible remedies to “acid rain.” CAPITA Publication: Washington University St. Louis MO, 1983; 83p4.

Figure CaPTIONS

Figure 1. A depiction of the Monte Carlo approach simulating the physical and chemical processes applied to a quantum.

Figure 2. The Monte Carlo Model data flow chart.

Figure 3. The dynamics of the planetary boundary layer and of pollutants being emitting into this layer as simulated by the CAPITA Monte Carlo Model. The mixing heights were taken from the NGM Eulerian database (21, 22), averaged over the Southwest, during the summer of 1992.

Figure 4. Illustration of transport in the Monte Carlo simulation. The left figure displays the dispersion of a puff over three days of travel without horizontal eddy diffusion and the height history of each particle in the puff. The right figure is a similar puff, but an eddy diffusion coefficient of 10,000 m²/sec was used.

Figure 5. A snap shot of the simulation of the atmospheric flow over the western US, during January 16 at noon GMT. The Lagrangian data were created from RAMS Meteorological data.

Figure 6. The variation of six airmass history variables for a quantum that was released from St. Louis on July 15, 1992 at 10:00 GMT.

Figure 7. The variation of the kinetic rate coefficients along a quantum’s trajectory and the corresponding quantum’s sulfur budget. The quantum was released from St. Louis on July 15, 1992 at 10:00 GMT.

Figure 8. The simulated concentration fields during quarters 1 and 3 1992, and the corresponding “observed” isopleth maps generated from the fusion of IMPROVE and NESCAUM fine sulfur measurements and airport visibility data.

Figure 9. The spatial domains used to average the simulated and observed concentrations and wet deposition rates. The symbol  represents the IMPROVE and NESCAUM aerosol monitoring sites, the NADP wet deposition sites, and ^ the center of receptor grid cells.

Figure 10. Comparison of the daily averaged simulated concentrations, during 1992, to the NESCAUM/IMPROVE measurements over a New England and East Central domain, as defined in Figure 9.

Figure 11. The simulated total wet deposition rates during quarters 1 and 3 1992, and the corresponding observed total wet deposition rates generated from the NADP observations.

Figure 12. Comparison of the simulated weekly total wet deposition rates, during 1992, to NADP observations over a New England and East Central domain, as defined in Figure 9.

Figure 13. The source attribution of the and column concentrations at a receptor located in Massachusetts. a) 1985 NAPAP Emission rates. b) Transit probabilities, P_t. c) kinetic probabilities, P_k. d) source attribution map. e) kinetic probabilities, P_k. f) source attribution map. The X on each map identifies the location of the Massachusetts receptor.

Figure 5. A snap shot of the simulation of the atmospheric flow over the western US, during January 16 at noon GMT. The Lagrangian data were created from RAMS Meteorological data.

Figure 8. The simulated concentration fields during quarters 1 and 3 1992, and the corresponding “observed” isopleth maps generated from the fusion of IMPROVE and NESCAUM fine sulfur measurements and airport visibility data.

Figure 11. The simulated total wet deposition rates during quarters 1 and 3 1992, and the corresponding observed total wet deposition rates generated from the NADP observations.

Figure 13. The source attribution of the and column concentrations at a receptor located in Massachusetts. a) 1985 NAPAP Emission rates. b) Transit probabilities, P_t. c) kinetic probabilities, P_k. d) source attribution map. e) kinetic probabilities, P_k. f) source attribution map. The X on each map identifies the location of the Massachusetts receptor.

Figure 13. Continued

Figure 11. Continued

Figure 8. Continued



Figure 1. A depiction of the Monte Carlo approach simulating the physical and chemical processes applied to a quantum.



Figure 2. The CAPITA Monte Carlo Model data flow chart.



Figure 3. The dynamics of the planetary boundary layer and of pollutants being emitting into this layer as simulated by the CAPITA Monte Carlo Model. The mixing heights were taken from the NGM Eulerian database (21, 22), averaged over the Southwest, during the summer of 1992.



Figure 4. Illustration of transport in the Monte Carlo simulation. The left figure displays the dispersion of a puff over three days of travel without horizontal eddy diffusion and the height history of each particle in the puff. The right figure is a similar puff, but an eddy diffusion coefficient of 10,000 m²/sec was used.





Figure 6. The variation of six airmass history variables for a quantum that was released from St. Louis on July 15, 1992 at 10:00 GMT.



Figure 7. The variation of the kinetic rate coefficients along a quantum’s trajectory and the corresponding quantum’s sulfur budget. The quantum was released from St. Louis on July 15, 1992 at 10:00 GMT.



Figure 9. The spatial domains used to average the simulated and observed concentrations and wet deposition rates. The symbol  represents the IMPROVE and NESCAUM aerosol monitoring sites, the NADP wet deposition sites, and ^ the center of receptor grid cells.




Figure 10. Comparison of the daily averaged simulated concentrations, during 1992, to the NESCAUM/IMPROVE measurements over a New England and East Central domain, as defined in Figure 9.




Figure 12. Comparison of the simulated weekly total wet deposition rates, during 1992, to NADP observations over a New England and East Central domain, as defined in Figure 9.