Underestimating Internal Variability Leads to Narrow Estimates of Climate System Properties

Journal Article
Underestimating Internal Variability Leads to Narrow Estimates of Climate System Properties
Libardoni, A.G., C.E. Forest, A.P. Sokolov and E. Monier (2019)
Geophysical Research Letters , 46(16), 10000-10007 (doi: 10.1029/2019GL082442)

Abstract/Summary:

Abstract: Probabilistic estimates of climate system properties often rely on the comparison of model simulations to observed temperature records and an estimate of the internal climate variability. In this study, we investigate the sensitivity of probability distributions for climate system properties in the Massachusetts Institute of Technology Earth System Model to the internal variability estimate. In particular, we derive probability distributions using the internal variability extracted from 25 different Coupled Model Intercomparison Project Phase 5 models. We further test the sensitivity by pooling variability estimates from models with similar characteristics. We find the distributions to be highly sensitive when estimating the internal variability from a single model. When merging the variability estimates across multiple models, the distributions tend to converge to a wider distribution for all properties. This suggests that using a single model to approximate the internal climate variability produces distributions that are too narrow and do not fully represent the uncertainty in the climate system property estimates.

Citation:

Libardoni, A.G., C.E. Forest, A.P. Sokolov and E. Monier (2019): Underestimating Internal Variability Leads to Narrow Estimates of Climate System Properties. Geophysical Research Letters , 46(16), 10000-10007 (doi: 10.1029/2019GL082442) (https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2019GL082442)
  • Journal Article
Underestimating Internal Variability Leads to Narrow Estimates of Climate System Properties

Libardoni, A.G., C.E. Forest, A.P. Sokolov and E. Monier

46(16), 10000-10007 (doi: 10.1029/2019GL082442)
2019

Abstract/Summary: 

Abstract: Probabilistic estimates of climate system properties often rely on the comparison of model simulations to observed temperature records and an estimate of the internal climate variability. In this study, we investigate the sensitivity of probability distributions for climate system properties in the Massachusetts Institute of Technology Earth System Model to the internal variability estimate. In particular, we derive probability distributions using the internal variability extracted from 25 different Coupled Model Intercomparison Project Phase 5 models. We further test the sensitivity by pooling variability estimates from models with similar characteristics. We find the distributions to be highly sensitive when estimating the internal variability from a single model. When merging the variability estimates across multiple models, the distributions tend to converge to a wider distribution for all properties. This suggests that using a single model to approximate the internal climate variability produces distributions that are too narrow and do not fully represent the uncertainty in the climate system property estimates.

Posted to public: 

Thursday, November 14, 2019 - 10:46