Dispersion simulations using HYSPLIT for the Sagebrush Tracer Experiment

Abstract

The Sagebrush experiment, led by NOAA's Field Research Division of the Air Resources Laboratory, consisted of five releases (intensive observation periods, or IOPs) of a chemically inert trace gas on five days in October 2013. All releases occurred in the afternoon under either near neutral stability conditions with high wind speeds or unstable conditions with low wind speeds. The sampling network for the tracer concentrations covered distances 200 m–3200 m from the release location and samples were obtained in 10-min averages. HYSPLIT, NOAA's transport and dispersion model, was used to simulate the spatial and temporal distribution of the tracer. The dispersion simulations were driven by WRF meteorological data with 27-km to 333-m grid spacing and using the inline and offline approaches as well as different planetary boundary layer schemes and a large-eddy simulation parameterization. Comparisons with measured wind speeds showed that none of the WRF PBL schemes or the large-eddy simulation parameterization was able to reproduce the rapid increase in high wind speeds observed during IOP3. The dispersion results were compared with the tracer measurements obtained during the experiment. The HYSPLIT dispersion simulations for IOP3, driven by the WRF data generated with various PBL schemes, showed greater concentration variability than the simulations performed for IOP5. The comparison between the inline and offline HYSPLIT simulations showed that the inline approach statistically outperformed the offline approach in three out of four IOPs because the tight coupling between the advection and dispersion processes implemented in the inline approach produced higher simulated concentrations close to the release location.