Previous lectures have discussed the observations of ozone made to date and the physics and chemistry controlling the distribution of ozone. Manmade chemicals have clearly had an impact on stratospheric ozone, the importance of which is discussed in Chapter 1. With the advent of high-speed, computationally powerful computers, it has become possible to run mathematical simulations or models of Earth's atmosphere that allow us (1) to test existing observations against theory; and (2) to make future trend predictions of different aspects of the atmosphere based on different scenarios. The former are referred to as process studies, and as an example, we might try to model air parcel trajectories around the polar vortex and compare the results to actual observations. The latter are referred to as assessment studies, and as an example, we might try to model the global temperature by making different assumptions about carbon dioxide content over the next 50 years.
Models used in process studies attempt to recreate as closely as possible, within the computational limits imposed by the computers used, existing observations of the atmosphere. Such studies are done in order to gain understanding of the various physical, dynamical, radiative, and chemical processes at work. Different models will emphasize different processes. We can improve upon our understanding of these processes by knowing how well our current theories compare to actual observations. Process modeling studies of stratospheric ozone are usually done using one of three types of photochemical models that attempt to simulate ozone photochemistry. These include the photochemical box model, the trajectory model, and the zonal mean or 2-dimensional model. Their descriptions and uses are discussed in Section 3.0.
Models used in assessment studies are designed to simulate the same dynamical, physical, and chemical processes of the atmosphere as in processing studies, but with some variable of Earth's atmosphere changed. The studies are done to make predictions about the atmosphere based on different scenarios. The most frequent simulation has been the scenario of a doubling of the carbon dioxide content of the atmosphere over some time period (typically 25 to 100 years) in order to predict how the general circulation and climate might change. The models allow us to determine the atmospheric response to a certain course of action, such as curbing carbon dioxide output from industrial society. The predictions then allow society to reach consensus on whether the potential environmental benefits of a certain course of action outweigh the economic costs associated with that action.
The key feature of assessment studies are their predictive capability. One way to gain confidence in the forecasting accuracy of the model is to "predict the past" in a procedure known as hindcasting. If our models can accurately simulate the known climate in the past, we can be better assured that the simulations of the future climate are reasonable. Other types of predictions include extrapolation of past trends into the future, and still others make forecasts on the basis of different scenarios altogether.
Assessment studies of future stratospheric ozone amount are based on different assumptions about future atmospheric levels of ozone destroying chlorofluorocarbons (CFCs). (See Chapters 5 and 11 for discussion of CFCs.) Hindcasting of ozone trends from the past and extrapolation of past ozone trends into the future are done in such studies. Often with ozone trend studies, different assumptions about the rate of change in global CFC concentration are made. These are examples of the different scenarios mentioned previously.
Upon completing this chapter, the reader should be able to