A Long-Term WRF Meteorological Archive for Dispersion Simulations: Application to Controlled Tracer Experiments

Abstract

AbstractA long-term archive of meteorological data using the Weather Research and Forecasting (WRF) Model was created to provide data that are compatible with the Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) dispersion model and to serve as initial and boundary conditions for simulations at a finer resolution. On the basis of these WRF data, generated with a variety of planetary boundary layer (PBL) schemes and nudging options, the HYSPLIT model was run to simulate four controlled tracer experiments?the Cross Appalachian Tracer Experiment (CAPTEX), the Across North America Tracer Experiment (ANATEX), a 1980 release in Oklahoma City, Oklahoma (OKC80), and the Metropolitan Tracer Experiment (METREX)?covering different time periods with diverse durations, including a summer day, several days in autumn, three months during winter, and one full year, respectively. The evaluation of the WRF results utilizing conventional observations showed a similar statistical performance for the different PBL schemes. Given the limited information the meteorological evaluation alone can provide, the authors used the dispersion evaluation with measurements from multiple tracer experiments to gain further insight into the most appropriate WRF configuration to generate reasonable data for dispersion applications. The dispersion simulations that were based on WRF data generated equal or slightly better statistical performance than did those driven by the North American Regional Reanalysis (NARR) dataset. The statistical comparison showed a mixed impact for the dispersion results driven by the nonnudged and nudged WRF data. The main advantage of the WRF data is the availability of hourly meteorological data from 1980 to the present and the inclusion of additional variables that are relevant to atmospheric dispersion and are not available from NARR. This WRF dataset will be accessible online, providing additional capabilities for using different meteorological inputs and a variety of options to compute the HYSPLIT mixing parameters.