CONCEPTUAL MODELING PROTOCOL FOR THE NEIGHBORHOOD ASSESSMENT PROGRAM

1. CONCEPTUAL MODELING PROTOCOL FOR THE NEIGHBORHOOD ASSESSMENT PROGRAM

2. Objectives of Modeling Protocol • develop and evaluate methodologies to estimate annual average concentrations of various pollutants released from multiple sources at the neighborhood scale (Barrio Logan) • recommend a method to perform neighborhood assessments

3. Size of Modeling Domain • Micro-scale (Neighborhood scale) modeling - receptors located near emission sources, i.e., meters to few kilometers from the sources • Regional scale modeling - source and receptor distances of several kilometers, to hundreds of kilometers, to the size of air basins

4. Air Toxic Pollutants To Be Modeled • Micro-scale modeling - more than 100 pollutants, including: Acetaldehyde, Acrolein, Acrylonitrile, Arsenic, Benzene, Beryllium, 1,3-Butadiene, Cadmium, Chloroform, Hexavalent Chromium, Ethylene Dibromide, Dioxin, Diesel PM, Formaldehyde, Hydrazine, Lead, Manganese, Methylene Chloride, Mercury, Nickel, (PCBs), Perchloroethylene, Trichloroethylene, Vinyl Chloride • Regional modeling - 30 pollutants: Acetaldehyde, Acrolein, Arsenic, Benzene, Beryllium, 1,3-Butadiene, Cadmium, Carbon Tetrachloride, Chloroform, Diesel PM, Ethylene Dichloride, Ethylene Oxide, Formaldehyde, Hexavalent Chromium, Iron, Lead, Manganese, Mercury, Methylene Chloride, MTBE, Nickel, o-Dichlorobenzene, p-Dichlorobenzene, Perchloroethylene, Styrene, Toluene, Trichloroethylene, Vinyl Chloride, Xylenes, Zinc

5. Micro-scale Models to Estimate Annual Concentrations • Traditional Models • ISCST3 - U.S. EPA regulatory model • CALINE - U.S. EPA approved model for line sources • Advanced Models • AERMOD - recommended as replacement for ISCST3 • CALPUFF- recommended model for complex terrain and for long-range transport • New and Emerging Models • Lagrangian particle dispersion model - state-of-the-science short-range model that estimates concentrations at scales of meters to tens of meters from a source

6. Input Data For Micro-scale Modeling • meteorology (hourly surface observations from local airports and on-site measurements) • point source emissions (including some traditional “area wide” sources, e.g. dry cleaners) • mobile source emissions (hourly traffic volumes and emission factors) • source configurations (e.g., stack heights) • receptor locations

7. Micro-scale Modeling Domain - Mobile Sources

8. Micro-scale Modeling Domain - Point Sources

9. Regional Models To Estimate Annual Concentrations • Traditional Models UAM (Urban Airshed Model) - used for air quality planning • Advanced Models Models-3 - state-of-science model developed by U.S. EPA

10. Regional Scale Modeling Objective: Estimate annual concentrations due to transport of pollutants from distant sources or air basin(s) that contribute to neighborhood areas (background concentrations) Improvements to Regional Scale Modeling: • Incorporate SAPRC 99 and air toxic chemistry • Apply over air basin(s) using an entire year of data for meteorology and emissions

11. Inputs For Regional Scale Modeling • Development of Regional Emissions Inventory • Update the 1998 database to reflect seasonal variations in emissions and update toxic emissions as well • Utilize latest version of EMFAC2000 to estimate motor vehicle emissions • Development of Meteorological Inputs for Regional Air Quality Models • CALMET - a diagnostic model that requires observational data for winds and temperature • MM5 - an advanced state-of-science prognostic meteorological model to simulate winds and temperatures

12. Location of the coarse and fine grids used in MM5 model

13. CALMET - Wind Fields (Morning)

14. CALMET - WindFields (Afternoon)

15. MM5 - Wind Fields (Afternoon)

16. Regional scale air quality modeling domain

17. Model Performance Evaluation • A scientific process to ensure models are working properly and are estimating reliable concentrations • Compare model estimates of concentrations with observational data and conduct operational and diagnostic evaluation • Follow the procedures identified in ARB's Technical Guidance Document on modeling. • Standard statistical techniques such as bias and gross error will be calculated for annual model estimates of concentrations as well as for monthly and seasonal averaging times

18. Databases For Evaluating Model Performance Micro-scale models: • Conduct a field monitoring study to collect a database for short-range dispersion in urban neighborhood areas • Field study includes release of a tracer gas, bag sampling and continuous monitoring of the tracer gas, and meteorological measurements which include six sonic anemometers and a mini-sodar • Sampling will be conducted over a one week period for four seasons of year at several sites • Funding available for total of 20 weeks of sampling

19. Databases For Evaluating Model Performance Regional Models: • Use routine meteorological and air quality (criteria and toxic) data collected by air districts and ARB • Utilize data collected from recent studies (SCOS97 and CRPAQS/CCOS) and other data gathered from special monitoring studies

20. Recommendations From Study • Identify which micro-scale and regional models are best suited for assessing neighborhood impacts • Develop recommendations for guidelines to determine the cumulative impacts posed by air pollutants at the neighborhood-scale • Share modeling results and recommendations with U.S. EPA and other interested groups