Highlighting a novel method presented by Robb Wills
Key Points & Overview
- A new statistical method is presented that can distinguish natural fluctuations in climate from climate change.
- The method demonstrates that a mode of climate variability called the Pacific Decadal Oscillation (PDO) is confined to the midlatitude North Pacific, separate from El Niño variability of the tropical Pacific.
Being able to attribute the amount of warming that is the result of anthropogenic CO2 emissions helps remove uncertainty in future-temperature projections and allows policymakers to understand the effects of climate change at a regional level—which is often where it has the biggest impact. Yet, separating observed temperature change into a component caused by anthropogenic global warming and a component caused by natural climate variability (e.g., El Niño) has remained an arduous task.
The standard approach is to run a global climate model numerous times with minor perturbations to its initial conditions, to simulate many different possible realizations of natural climate variability. By comparing these simulations to simulations where the global climate model is forced with increasing CO2, the effect of global warming can be determined. However, this method is costly, as these global climate models take large amounts of computational power to run.
In a new paper, Robb Wills, a postdoctoral researcher in the Department of Atmospheric Sciences, along with a prodigious list of climate science co-authors, leaves behind the computationally expensive model pursuit for a simple statistical method. The method, deemed low-frequency component analysis (LFCA), sorts modes of variability based on their dominant time scale, distinguishing global warming from natural variability based on their differences in temperature pattern and time scale. This approach separates observed sea-surface temperature (SST) variability into components due to the Pacific Decadal Oscillation (PDO), the El Niño-Southern Oscillation (ENSO), and global warming, allowing a diagnosis of the trend in SSTs that can be attributed to anthropogenic influence. It also shows that the natural decadal SST fluctuations of the North Pacific are more persistent and more independent of El Niño than previously thought, which may aid in predicting decadal fluctuations in climate over North America.
This study has marked implications that will contribute to refining our view of the climate system and reducing uncertainty in future temperature projections.