Interdependency in Multimodel Climate Projections: Component Replication and Result Similarity

Abstract Multimodel ensembles are the main way to deal with model uncertainties in climate projections. However, the interdependencies between models that often share entire components make it difficult to combine their results in a satisfactory way. In this study, how the replication of components...

Full description

Saved in:
Bibliographic Details
Main Author: Julien Boé
Format: Article
Language:English
Published: Wiley 2018-03-01
Series:Geophysical Research Letters
Subjects:
Online Access:https://doi.org/10.1002/2017GL076829
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Abstract Multimodel ensembles are the main way to deal with model uncertainties in climate projections. However, the interdependencies between models that often share entire components make it difficult to combine their results in a satisfactory way. In this study, how the replication of components (atmosphere, ocean, land, and sea ice) between climate models impacts the proximity of their results is quantified precisely, in terms of climatological means and future changes. A clear relationship exists between the number of components shared by climate models and the proximity of their results. Even the impact of a single shared component is generally visible. These conclusions are true at both the global and regional scales. Given available data, it cannot be robustly concluded that some components are more important than others. Those results provide ways to estimate model interdependencies a priori rather than a posteriori based on their results, in order to define independence weights.
ISSN:0094-8276
1944-8007