Applications of the Regional Atmospheric Modeling System (RAMS) to Provide Input to Photochemical Grid Models for the Lake Michigan Ozone Study (LMOS)

Abstract

In spite of ongoing control measures, episodes of elevated boundary layer ozone in violation of national ambient air quality standards continue to occur in the Lake Michigan region. The Lake Michigan Ozone Study is a multiyear effort that included a 1991 field program, among the purposes of which were the collection of sufficient data to initialize and evaluate a photochemical modeling system to be used for regional emission control decisions. The Regional Atmospheric Modeling System was configured to provide the meteorological input. Four multiday episodes were simulated, and model performance was evaluated against a dedicated surface network of meteorological sensors and supplemental aircraft measurements. A nested grid configuration allowed a 4-km mesh size over the lake, while the synoptic environment was simulated using a U.S.-scale, 80-km outer grid. Using only synoptic-scale surface and upper-air data for four-dimensional data assimilation, the model showed considerable skill in simulating the shallow lake-breeze circulation of 16 July 1991 as well as regional mixing depths (typically to within 10%). Over 11 days, average absolute temperature errors were generally between 1° and 2°C with rmse wind speed errors on the order of 1.5 m s?1. Sensitivity tests confirmed that the development of mesoscale circulations can be strongly influenced by the specification of regional soil moisture in the model.