EVALUATION OF PM2.5 SOURCE REGIONS OVER THE MISSISSIPPI GULF COAST USING WRF/HYSPLIT MODELNG APPROACH

摘要

Fine particulate matter (PM2.5) is formed by precursors such as sulfur dioxide, nitrogen oxides, volatile organic compounds, and trace metals, which are emitted largely from intense industrial operations and transportation activities. PM2.5 is known to cause serious respiratory problems. Evaluation of source regions and assessment of the contribution of various sources in the Mississippi Gulf Coast region will be useful for implementation of regulatory and mitigation measures. In the present study, output from the HYbrid Single- Particle Lagrangian Integrated Trajectory (HYSPLIT) Model driven by the Weather Research and Forecasting (WRF) Model is used as a tool to assess source location, transportation trends, and the extent of contribution to PM2.5. Meteorological and air pollution (PM2.5 sulfate) observations were collected during a joint atmospheric dispersion field experiment in summer 2009 at two locations in southern Mississippi. The WRF model was used to simulate meteorological parameters at a resolution of 4 km and was designed to have three two-way interactive nested domains with horizontal resolutions of 36, 12 and 4 km, and the inner-most domain covered the region of interest. The vertical resolution included 43 levels, with 33 levels confined to below 500 hPa to simulate boundary layer flow characteristics. The model was integrated for 72 hours starting from 00UTC on 16, 17 and 18 June 2009. The initial and boundary conditions were adopted from National Centers for Environmental Prediction Final Analyses (NCEP FNL) data available at one degree horizontal resolution. The model derived meteorological fields were validated with in-situ observations collected at the two locations. The HYSPLIT model was integrated with WRF model-derived meteorological fields to identify the source location using backward trajectory analysis. The backward trajectories were plotted for every one hour with different heights in the mixed layer. Trajectories were plotted for a 24-hr period starting from two observations, and a cluster analysis method was used to estimate the probable contribution from each source. As a second step, forward trajectories were plotted with different identified source locations using data from observations at elevated point sources. Concentration levels were calculated and compared with in-situ observations to examine and assess possible relative contributions from different sources.