Abstract
We describe in this paper a set of general circulation model experiments on possible climate changes caused by man's generation of thermal energy or pollution. Three experiments were carried out: one in which we introduced only a small initial error, one in which we added the expected ultimate levels of thermal energy generation, and one in which we added a negative amount of thermal energy. In an three experiments, we obtained the same results, indicating that the thermal pollution effect is probably small compared to the natural fluctuations of the model. We also discuss some limitations of the present model for inferring the proper climatic-change response.