Statistical Considerations in the Evaluation of Climatic Experiments with General Circulation Models,
RAND CORP SANTA MONICA CALIF
Pagination or Media Count:
Numerical calculations made with general circulation models contain short-term fluctuations that need to be taken into account when considering the significance of long-term climatic means. In particular, when such models are used in the study of climatic change resulting from altered forcing, quantitative evaluation of the effects of the fluctuations, as measured by the variance of model generated climatic estimates, is required. This report develops a rationale for planning the extent of climate simulation trials needed to establish the significance of any proposed climatic change experiment. The topic is treated from an elementary statistical viewpoint and the report is in part written to serve as an introduction to statistical questions that are important for climate modeling. Conclusions concerning the planning of climatic change experiments include a quantitative prescription for determining the extent of model calculations required to establish the presence of a climatic change between two numerical experiments at any chosen confidence level.