Department of Climate and Space Sciences and Engineering in the College of Engineering at the University of Michigan


Assoc. Prof. Flanner in Proceedings of National Academy of Sciences

Posted: September 7, 2018

Assoc. Prof. Flanner in Proceedings of National Academy of Sciences

A recent paper co-authored by Associate Professor Mark Flanner explores the hypothesis that the typical magnitude of biases inherent in the coupled climate model is similar to the magnitude of the climate change that is expected on a centennial time scale.

From Prof. Flanner: 

"All climate models have biases and the typical magnitudes of these biases are similar to those of the climate changes that are expected over the next century. Using climate models for assessing future climate change therefore relies on the hypothesis that these biases remain stable (stationary) or vary predictably. Although this hypothesis cannot be tested directly, we compare the biases of individual models with respect to multimodel mean states under two very different climate states, i.e., current conditions and those simulated under 4 times pre-industrial CO2 concentrations. Our comparison shows that under very large climate change the bias patterns of key climate variables remain strikingly stationary. This provides new justification for using state-of-the-art climate models to simulate climate change, for selecting well-performing models for regional downscaling, and for extending bias corrections of the present state to perturbed states such as those simulated under increased greenhouse gas concentrations."

The paper is titled, “Striking stationarity of large-scale climate model bias patterns under strong climate change" and is on early release in the Proceedings of National Academy of Sciences this week.

Read the paper here:

Latest Headlines