Development of a Model for Predicting the Vertical Profile of Ozone Based on Ground-Level Ozone Observations and Cloud Cover

Ozone formed in a chemical reaction is vertically dispersed and mixed with the ozone mass aloft by turbulent convection due to solar radiation. The mixing height, which depends on the stability of the atmosphere, varies by day/night time and by season of year and is characteristic of season and day and night time from year to year. Ground level ozone is generally the combined result of dynamics of the atmosphere resulted from the turbulent convection in the atmospheric boundary layer (ABL), production and destruction processes such as transport from upper air mass, subsequent transport to the ground, photochemical production and chemical destruction. Such processes are a function of the photochemical activity of chemical species (NO_{x and VOC) present in the ABL and meteorological conditions that are recurrent phenomena in a year. The purpose of this study is to develop a model to predict high elevation ozone concentration for a particular time as a function of elevation based on ground level ozone observation with and without incorporating cloud cover that influences ozone concentration and the atmospheric stability, using a number of techniques. This study demonstrated that the overall regression model, which did not consider the effect of cloud cover, provided an approach that estimates the high elevation ozone concentrations based on the statistical results with 13.48 ppb prediction error and 0.66 index of agreement between the observed and predicted datasets. The "clear to partly cloudy" and "cloudy" day component regression models associated with cloud cover provided improvement over the overall regression model.}