1.0 INTRODUCTION

Frequent high concentrations of ground-level ozone in many parts of eastern North America during the summer of 1988 led to intensified efforts to identify the sources of the precursors of this pollutant and address issues of emissions abatement. The Canadian Council of Ministers of the Environment (CCME) identified the Windsor-Quebec City corridor as the area in Canada where higher than acceptable concentrations of ground-level ozone occur most often and for the longest periods (CCME, 1990). Research was undertaken by the Canadian Institute for Research in Atmospheric Chemistry (CIRAC) to aid in the development of a management plan to control ozone precursors. Participants in the development of the Canadian Federal / Provincial NOx/VOC Management Plan under the CIRAC umbrella included York University and representatives from government and industry.

Field studies including intensive atmospheric chemistry and meteorological measurements were carried out during the summers of 1992 and 1993 based around a station at a rural site near Hastings, Ontario about 137 km northeast of Toronto (Figure 1.1). In 1993, the observation program at Hastings, named the Southern Ontario Oxidant Study (SONTOS) (Reid et al., 1996), was conducted simultaneously with a similar program called the Southeast Michigan Ozone Study (SEMOS) (Wolff et al., 1993) in adjacent southeastern Michigan. These projects were further augmented by a surface mesoscale meteorological observation program in southwestern Ontario called the Southern Ontario Oxidant Study - Meteorological Measurements (SOMOS) (Sills et al., 1994). Aircraft measurements of atmospheric trace gases were also made during SONTOS and SEMOS. Hourly averages of ground-level ozone in Ontario were often observed to exceed acceptable levels (82 ppb national standard), especially during the summer of 1993. The effects of the Great Lakes on the transport of ozone and its precursors, and in particular the roles of lake breezes and land breezes, were found to be important to the resolution of source areas of pollutants and dispersion patterns (Reid et al., 1996; Sills and Moroz, 1996).

To model lake-induced circulations, a numerical model developed by AES-RPN and UQAM called the Mesoscale Compressible Community (MC2) model is being used. The MC2 model is based on a fully-elastic non-hydrostatic model developed by Tanguay, Robert and Laprise (1990). The model solves a full set of Euler equations in three dimensions on a limited-area Cartesian domain of the polar projection. Time-dependent lateral boundary conditions are supplied by a hemispheric or global model. Model efficiency is increased by using a semi-Langrangian, semi-implicit numerical scheme. MC2 is coupled with the RPN/CMC physics library of physical parameterizations that are currently used for AES operational forecasting. Self-nesting capabilities are included so that high-resolution simulations can be made. A detailed description of the MC2 model formulation is given by Bergeron et al. (1994) while the physical parameterizations are described in detail by Mailhot (1994). Recently, a paper was published by Benoit et al. (1997) that describes and validates an early version of the MC2 model.

The object of this study is to evaluate the performance of the MC2 model during a well-documented lake breeze and elevated ozone event, to assess the impact of changes made to the model boundary-layer physics from version 3.2 to version 4.0, and to conduct tests to gauge the model's sensitivity to changes in geophysical fields including albedo, roughness, lake temperature, deep soil temperature and soil moisture availability.

Back to Main Page