Quantitative Analysis of Feedbacks in Climate Model Simulations

  1. A. Berger,
  2. R. E. Dickinson and
  3. John W. Kidson
  1. Michael E. Schlesinger

Published Online: 18 MAR 2013

DOI: 10.1029/GM052p0177

Understanding Climate Change

Understanding Climate Change

How to Cite

Schlesinger, M. E. (1989) Quantitative Analysis of Feedbacks in Climate Model Simulations, in Understanding Climate Change (eds A. Berger, R. E. Dickinson and J. W. Kidson), American Geophysical Union, Washington, D. C.. doi: 10.1029/GM052p0177

Author Information

  1. Department of Atmospheric Sciences and Climatic Research Institute, Oregon State University, Corvallis, Oregon 97331

Publication History

  1. Published Online: 18 MAR 2013
  2. Published Print: 1 JAN 1989

ISBN Information

Print ISBN: 9780875904573

Online ISBN: 9781118666517

SEARCH

Keywords:

  • Climate changes—Congresses

Summary

Why do climate models, even within the same category of the climate model hierarchy, simulate different climatic changes for the same forcing? The first step toward answering this question is to calculate quantitatively the feedbacks of the physical processes in the models. This paper therefore presents the concept and terminology of classical feedback analysis; three different feedback analysis methods, one for radiative-convective models and two others for general circulation models (GCMs); and the application of these feedback analysis methods to the simulations of CO2-induced climatic change. A rigorous intercomparison of the quantitative values of the feedbacks for the GCMs is made difficult by the fact that these feedbacks have been determined by two different methods. This notwithstanding, the intercomparison indicates that the contribution of the individual feedback processes to the simulated climatic changes is not the same for different models, even when they are within the same category of the climate model hierarchy. Consequently, based on these and future feedback analyses, the second step toward answering the above question is to intercompare the parameterizations of the most important feedback processes among themselves, with more-detailed models, and with observations.