Some recent studies have suggested that climate change may accelerate in the 21st century due to positive feedback processes if increases of greenhouse gas concentrations in the atmosphere are not greatly curbed. Through use of the NASA GISS Model II, several simulations were performed increasing carbon dioxide concentrations in the atmosphere exponentially at several rates from 1991-2050, followed by no change in concentrations from 2051-2200. The purpose of this work is to investigate the ability of the atmosphere to regain an equilibrium state once increases in carbon dioxide concentrations are ceased following external forcings of varying strengths. An emphasis was placed on climate change within the tropical north Atlantic basin.

An equilibrium state can be defined, in this sense, as a state in which variability in atmospheric properties due to interannual and interdecadal oscillations greatly exceeds variability due to external forcing. In order to determine the variability in atmospheric properties from external forcing, an ensemble with varying random initial conditions was used. The ensemble mean was used to evaluate the climate change due to external forcing, which is the same among ensemble members. The amount of variability due to interannual and interdecadal oscillations was determined through the range and standard deviation among ensemble members. The ensemble contained six members, each varying only in initial condition. The effects of different external forcing were assessed by simulating several ensembles, varying in the degree of carbon dioxide increases at the beginning of the simulation.