Modeling Global Climate Change

Article excerpt

The industrial and scientific revolutions of the past two centuries have brought much of the physical world under our control. Yet our lives can still be upended by what insurance companies call "acts of God" - floods, fires, hurricanes, droughts, and so on. Does this mean that Earth's climate and weather remain forces of nature, entirely beyond our influence? Over the past few decades, a consensus has gradually emerged among scientists that human activities can indeed affect climate and that they are already doing so. One way humans might affect climate is by causing "global warming" (an increase in global average surface temperature). With observed surface temperatures rising, and governments considering large-scale, costly efforts to counteract this trend, it is increasingly important to understand how and why our climate is changing.

Earth's climate system is an extremely complex web of interconnected physical, chemical, and biological processes. Understanding the cause of a change in a single parameter, such as surface temperature, is quite involved. It requires first a knowledge of which processes govern the value, second an understanding of how each of those processes functions, and third an understanding of how they interact. The interactions are frequently non-linear, meaning that the net change from variations in several processes is not equal to the sum of all independent changes.

What sorts of processes are involved? There are many potential causes of global climate change. They include changes in emissions of certain gases into the atmosphere, the amount of ozone, atmospheric particles (aerosols, most often produced from pollution), the sun's brightness, and changes in natural patterns of climate variability such as the El Nino oscillation. We have to be concerned with anything that might respond to these changes, as well as with whatever processes control the "natural" climate state. Important physical processes include absorption and reemission of solar energy by Earth and the atmosphere, evaporation, rainfall and cloud formation, atmospheric winds and storms, and ocean currents, among many others. Chemical processes include the formation and destruction of ozone and aerosol particles. Biological processes include carbon dioxide intake and oxygen emission by plants, and methane emissions from wetlands and livestock. Though some individual processes, such as generation of wind by atmospheric heating, can be expressed through relatively simple mathematics, the entire system of equations is far too complex to be solved by hand. Rather, the only feasible solution is to create a computer model of Earth's climate system.


These models, called general circulation models (GCMs), attempt to include the relevant physical principles governing the behavior of the atmosphere and the oceans. Those principles are formulated into mathematical equations, which determine how a given set of initial conditions will change over time. Some processes obey fairly simple laws, such as the passage of Earth around the sun, which gives us seasons. Others, such as cloud formation, are extremely complex processes that are very difficult to understand and incorporate into models. Yet the only way to approach a real understanding of the climate system is to create a realistic model.

GCMs divide the atmosphere into a three-dimensional grid of boxes around Earth. By using observations, scientists can specify initial values for each box for quantities such as temperature, wind speed, pressure, and composition. Emissions at Earth's surface are also specified over time. For example, urban regions emit large amounts of pollutants. Based on physical laws, the computer program then calculates changes over an interval of time called the "time step." In general, a smaller time step allows interactions to take place more nearly simultaneously, as they do in the real atmosphere, and so produces more realistic results. …