A comparative study of prognostic meteorological and of air quality model predictions with NEOPS 1999 observations

This study presents a comparative evaluation of prognostic mesoscale meteorological and of photochemical gas/aerosol air quality model predictions with data from the North East Oxidant and Particle Study (NE-OPS) research program over Philadelphia, PA. Model simulations were performed for a two week period from 11th July 1999 00 UTC to 25th July 1999 11 UTC corresponding to the field study days for NEOPS. The MM5 model was applied with 14 layers in the vertical direction and the results were,compared with aircraft, RASS, wind profiler, lidar and tethersonde balloon data collected by the NE-OPS program. Comparisons with aircraft data indicate that while the MM5 model successfully reproduces the observed temperature values, this is not the case with relative humidity values. The virtual temperature profiles predicted by the model compare very well with RASS data while the wind components calculated by the model are only in partial agreement with the wind profiler data. However the mixing ratio and temperature profiles obtained from lidar compare well with the model results. The model predicted meteorological variables are only in partial agreement with the tethersonde balloon observations with both relative humidity and wind speed being underestimated by the model. US EPA's Community Multiscale Air Quality (CMAQ) model, a component of the Models-3 system, and MCNC's Multiscale Air Quality Simulation Platform (MAQSIP) were used to simulate gaseous and aerosol phase air quality dynamics for the same domain. The modal aerosol model included as part of the current release of CMAQ is used in the CMAQ simulations while the dynamic sectional aerosol developed at the University of Delaware (UDAERO) is adopted in the MAQSIP simulations. The emissions data were processed from the National Emissions Trends (NET) inventory using MCNC's Sparse Matrix Operator Kernel Emissions (SMOKE) modeling system. Fourteen layers in the vertical direction and three levels of nested domains are used, with grid resolution of 36 km for the outermost domain, 12 km for the intermediate domain and 4 km for the innermost domain encompassing the metropolitan Philadelphia area. The model predictions were compared with chemically and temporally resolved pollutant concentration measurements obtained through the NEOPS study to evaluate the performance of the models in capturing the 3-dimensional regional scale dynamics of ozone and particulate matter.