Grantee Research Project Results
1998 Progress Report: Development and Application of an Air Quality Modeling System with Integrated Meteorology, Chemistry, and Emissions
EPA Grant Number: R825388Title: Development and Application of an Air Quality Modeling System with Integrated Meteorology, Chemistry, and Emissions
Investigators: Xiu, Aijun , Coats, Carlie J. , Mathur, Rohit
Current Investigators: Xiu, Aijun , Mathur, Rohit , Hanna, Adel , Coats, Carlie J.
Institution: MCNC / North Carolina Supercomputing Center
EPA Project Officer: Hahn, Intaek
Project Period: October 1, 1996 through September 30, 1999
Project Period Covered by this Report: October 1, 1997 through September 30, 1998
Project Amount: $372,830
RFA: Exploratory Research - Air Engineering (1996) RFA Text | Recipients Lists
Research Category: Land and Waste Management , Air , Safer Chemicals
Objective:
To develop a fully-integrated, physically and numerically consistent, regional-scale coupled atmospheric dynamics and chemistry modeling system.Progress Summary:
In the second year of this project, an initial version of an air quality modeling system with integrated meteorology, chemistry, and emissions has been developed and tested. The modeling system attempts to incorporate recent advances in dynamic meteorological modeling, atmospheric transport/chemistry/deposition modeling, and emissions processing and modeling in a single consistent modeling framework.
Accomplishments and Research Results:
- Model Formulation and Components
The integrated air quality modeling system development has been largely based on the further development and refinement of three existing models: the Penn State/NCAR Mesoscale Model MM5 (Grell et al. 1994), the Multiscale Air Quality Simulation Platform (MAQSIP) air quality model (Odman and Ingram 1996), and the Sparse Matrix Operator Kernel Emissions (SMOKE) model (Coats 1996). The starting point in the development of the ntegrated model described in here is the MM5, which is a nonhydrostatic prognostic meteorology model with four dimensional data assimilation (FDDA) and multiple nesting capabilities. We have included directly into the MM5, modules to represent tracer transport through advection and turbulent mixing, dry-deposition, and gas-phase chemistry. The incorporation of cloud effects on tracer distributions through cloud transport, aqueous chemistry, and scavenging are underway. The current version of the model uses the Bott advection scheme (Bott, 1993), while turbulent transport of tracers is represented through a K-theory based formulation. Gas phase chemistry is represented by a modified and updated version of the CBM-IV mechanism of Gery et al. (1989); the modifications and updates are described in Kasibhatla et al. (1997). The solution of the ordinary differential equations representing the chemical transformations between model species is achieved through concepts presented in Mathur et al. (1998). While our current version of the integrated model uses the various process representations and numerical schemes described above, it should be pointed out that the current structure of the model is highly modular. This modularity provides the options to update individual components at a later stage, if desired. For example, the chemical mechanism can be changed to the RADM2 mechanism of Stockwell et al (1990), or alternatively the advection scheme can be replaced by an alternate scheme with minimal effort if desired.In order to facilitate the coupling of meteorological and chemical tracer transport and chemistry calculations in the integrated model, a Meteorology CouPLing module (MCPL) that fits directly into MM5 has been developed (Coats et al., 1998). MCPL is designed for extremely easy insertion into the MM5 source code and is callable at a variety of times scales from the MM5 advection-step frequency on up (Coats et al., 1998). MCPL uses the EPA Models-3/North Carolina Supercomputing Center (NCSC) Environmental Decision Support System (EDSS) Input-Output Applications Programming Interface (I/O API) (Coats et al., 1993). MCPL is principally controlled by environmental variables and by ASCII tables and can be configured to write data either to buffered files (for on-line integrated chemistry calculations) files or to files on disk (for offline chemistry calculations). At the same time the land-use data coming from one of the MM5 preprocessors need to be processed for dry deposition velocity calculation in MAQSIP.
- Initial Testing and Model Applications
Initial tests with the integrated model were performed for a geographical domain over the Eastern United States. The MM5 was run in a one-way nest mode wherein a coarse domain with horizontal resolution of 108 km was used for providing hourly boundary conditions for a nested fine domain utilizing a horizontal resolution of 36 km. The integrated model domain is located within the 36 km MM5 domain; this set-up is essential for specifying meteorological variables at the boundary of the integrated model. The vertical domain ranging from the surface to 100 mb was discretized using 22 levels of varying thickness. Since we only run the integrated model on the 36 km domain, we will only describe the model physics and FDDA options in MM5 for this domain. The analysis nudging is used for data assimilation and Kain-Fritsch cloud parameterization for cloud process. A modified version of Blackadar PBL scheme (Blackadar 1979) is implemented in MM5 to overcome the problem with sudden collapse of PBL height in the original scheme (Alapaty, personal communication). Since the linkages with the SMOKE emissions processing system are currently under development, in our current testing and simulations with the integrated model, we use input emissions computed in an offline mode.A starting point in the testing of the integrated model was motivated by the need to connect the dynamics with the tracer transport calculations in a consistent manner. MM5 uses a centered difference scheme for advecting scalar fields. The scheme is not positive definite, nor does it conserve mass exactly (Jakobs and Tilmes, 1995). Chemical tracers, however, exhibit significant spatial gradients and consequently it is desirable to advect their fields with a higher order accurate scheme that has minimal numerical diffusion. While the three-dimensional wind velocity and density fields satisfy the discretized form of continuity equation for air used in MM5, the use of a different numerical advection scheme for tracer transport can lead to mass- consistency errors in the model. Two test cases were examined to investigate the magnitude of this inconsistency. In the first case, an inert tracer mass was injected into the modeled domain at initialization and the total domain mass was tracked for a 24 hour period. In the second test, the MM5 generated wind fields were used to advect a tracer field with uniform mixing ratio throughout the modeled domain. While in the first case, the total mass was conserved exactly, in the second test case the tracer gradually became ?unmixed? relative to the background. This is attributed to mismatch between the divergences in the air and tracer continuity equations, locally. We are currently investigating alternate approaches for a correction scheme for this problem. One approach is to minimally adjust the wind fields so that the resulting advection is mass-consistent. Lu et al. (1997) have propose a correction to the density fields to maintain mass-consistency between air and tracer advection; the extension of this scheme to the integrated model is currently under development. Byun (1998) also studied mass conservation problems in air quality models and proposed a mass correction algorithm.
In parallel to the tracer transport tests, a series of model simulations were also performed for a five day period from July 9-13, 1995. The intent of these simulations were primarily to (i) check the internal consistency of the integrated model; and (ii) to provide a preliminary qualitative examination of the ability of the integrated model to simulate regional tropospheric ozone production and distributions over the eastern United States. The model predicted afternoon average ozone distribution for the model surface layer on July 10, 1995 showed elevated ozone distributions being downwind of metropolitan Atlanta and also downwind of source regions of the lower Ohio River Valley. The spatial distribution of predicted ozone is in reasonable qualitative agreement with that of ozone observations from the AIRS data base. A scatter plot comparison of afternoon average predicted ozone with observations from the AIRS data base showed that while majority of the points are within the 25 % comparison limits, overprediction of ozone is noticeable. This overprediction in part can be attributed to the absence of any cloud processes, especially photolysis attenuation and cloud transport, in the current version of the model.
We also compared the "online" and "offline" simulations for the same episode (July 9-13, 1995) and found the most differences occurred at times when the planetary boundary layer (PBL) evolves in the morning and collapses in the afternoon. This is because at those times the integrated model (with the "online" approach) can catch the PBL development better than the "offline" model which interpolates PBL height hourly.
References:
Blackadar, A.K., 1979: High resolution models of the planetary boundary layer. In Advances in Environmental Science and Engineering; Pfafflin and Ziegler, Eds.; Gordon and Briech Sci. Publ., New York, 1979, pp 50-85.
Bott, A., 1993: The monotone area-preserving flux-form advection algorithm: reducing the time-splitting error in two-dimensional flow fields. Mon. Wea. Rev., 121, 2637-2641.
Byun, D.W., 1998: Dynamically consistent formulations in meteorological and air quality models for multi-scale atmospheric studies: II. Mass conservation issues. J. Atmos. Sci., in press.
Coats, C.J., Jr., 1996: High performance algorithms in the Sparse Matrix Operator Kernel Emissions (SMOKE) modeling system. In Preprint for Ninth AMS Joint Conference on Applications of Air Pollution Meteorology with A&WMA, American Meteorological Society: Atlanta, GA, 1996; pp 584-588.
Coats, C.J., Jr., A.F. Hanna, D. Hwang, and D.W. Byun, 1993: Model engineering concepts for air quality models in an integrated environmental modeling system. In Transactions, AWMA Specialty Conference on Regional Photochemical Measurement and Modeling Studies, Air & Waste Management Association: San Diego, CA, 1993; pp 213-222.
Coats, C.J., Jr., J.N. McHenry, A. Lario-Gibbs, and C.D. Peters-Lidard, 1998: MCPL(): A drop-in MM5-V2 module suitable for coupling MM5 to parallel environmental models; With lessons learned for the design of the weather research and forecasting (WRF) model. In Preprints of the Eighth PSU/NCAR Mesoscale Model Users' Workshop, Mesoscale and Microscale Meteorology Division, National Center for Atmospheric Research: Boulder, CO, 1998; pp 117-120.
Gery, M.W., G.Z. Whitten, J.P. Killus, and M.C. Dodge, 1989: A photochemical kinetics mechanism for urban and regional scale computer modeling. J. Geophys. Res., 94(D10), 12,925-12,956.
Grell, G.A., J. Dudhia, and D.R. Stauffer, 1994: A Description of the Fifth-Generation Penn State/NCAR Mesoscale Model (MM5). NCAR Technical Note NCAR/TN-398+STR, 1994.
Jakobs, H.J., and S. Tilmes, 1995: New advection schemes in MM5. In Preprints, The fifth PSU/NCAR Mesoscale Model Users' Workshop, Mesoscale and Microscale Meteorology Division, National Center for Atmospheric Research: Boulder, CO, 1995, pp 2-5.
Kasibhatla, P., W.L.,Chameides, B. Duncan, M. Houyoux, C. Jang, R. Mathur, T. Odman, and A. Xiu, 1997: Impact of inert organic nitrate formation on ground-level ozone in a regional air quality model using the Carbon Bond Mechanism 4, Geophys. Res. Lett., 24, 3205-3208.
Lu, R., R.P. Turco, and M.Z.Jacobson, 1997: An integrated air pollution modeling system for urban and regional scales: 1. Structure and performance. J. Geophys. Res., 102, 6063-6079.
Mathur, R., J. Young, K. Schere, and G. Gipson, 1998: A comparison of numerical techniques for solution of atmospheric kinetic equations. Atmos. Environ., in press.
Odman, M.T., and C.L. Ingram, 1996: Multiscale Air Quality Simulation Platform (MAQSIP): Source Code Documentation and Validation. ENV-96TR002-v1.0. Available from MCNC, P.O. Box 3021 Cornwallis Road, Research Triangle Park, NC 27709.
Pleim, J.E., and A. Xiu, 1995: Development and testing of a surface flux and planetary boundary layer model with explicit soil moisture parameterization for application in mesoscale models. J. Appl. Meteor., 34, 16-32.
Stockwell, W.R., P. Middleton, and J.S. Chang, 1990: The second generation regional acid deposition model chemical mechanism for regional air quality modeling. J. Geophys. Res., 95(D10), 16,343-16,367.
Future Activities:
The existing version of the integrated air quality modeling system provides an essential base for further enhancement and refinement of the integrated modeling system. Preliminary testing and applications of the integrated model have been motivated by the need to test the data flow and internal consistency of the model. These tests have also revealed areas of further refinement such as the connection between the MM5 predicted density and wind fields with tracer advection in a mass-consistent manner; alternate approaches for this are currently under development and testing and will be incorporated in the model. The further development of the model will focus on four major aspects: (1) the inclusion of aerosol dynamics based on the method of Binkowski and Shankar (1995); (2) completion of the incorporation of cloud processes impacting chemical tracer concentrations and distributions and development of appropriate linkages; (3) the integration of the SMOKE emissions processing system within the model; and, (4) the linkage the land-surface model of Pleim and Xiu (1995) with the MM5 and the biogenic emissions modeling component.Journal Articles:
No journal articles submitted with this report: View all 8 publications for this projectSupplemental Keywords:
RFA, Scientific Discipline, Air, particulate matter, air toxics, Environmental Chemistry, Environmental Monitoring, tropospheric ozone, Engineering, meteorology, air quality models, ambient air, emission-based modeling, ozone, chemical composition, air pollution models, air quality data, atmospheric aerosols, atmospheric aerosol particles, atmospheric chemistry, engineering modelsProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.