17 April 2017

A Cheaper Way to Understand Teleconnections in Climate Models

Scientists find a way to simultaneously characterize input-output relationships across different timescales in a single simulation.


More understanding is needed about how climate models respond to perturbations, particularly in regions far from where the perturbations happened.


By being smart about how to ask questions of climate models, researchers can get more information out of them with fewer simulations. This is particularly important given the intense computational expense of climate models.


A study led by researchers at the U.S. Department of Energy’s (DOE) Pacific Northwest National Laboratory (PNNL) introduced system identification techniques to climate science in which input-output relationships across different timescales can be simultaneously characterized in a single simulation. This method, involving small perturbations (in space and time) of an input field while monitoring output fields to quantify responses, allows modelers to identify different climate response timescales to force the climate without substantially pushing it far away from a steady state. Researchers used this technique to determine the steady-state responses of low cloud fraction and latent heat flux to heating perturbations over 22 regions spanning the Earth’s oceans. Findings showed similarities between the response characteristics and those of step-change simulations, but the new method simultaneously characterized the responses for 22 regions. Using this technique, researchers can estimate the timescale over which the steady-state response emerges. This methodology could be useful for a wide variety of purposes in climate science, including characterization of teleconnections (far-field effects) and uncertainty quantification to identify the effects of climate model tuning parameters.

Ben Kravitz
Pacific Northwest National Laboratory
Kravitz, B. "Technical note: Simultaneous fully dynamic characterization of multiple input–output relationships in climate models." Atmospheric Chemistry and Physics 17, 2525-2541 (2017). [10.5194/acp-17-2525-2017].