Jan 6 2020
According to a group of meteorologists, the Pacific Decadal Oscillation (PDO) and the Atlantic Multidecadal Oscillation (AMO) do not seem to exist.
Ocean with white caps. Image Credit: National Science Foundation.
Meteorologists believe that this finding holds implications for the possibilities of decade-scale predictability of climate and also for the validity of earlier studies that attribute past trends to such hypothetical natural oscillations.The scientists used a combination of climate model simulations and observational data to demonstrate that no reliable evidence exists for longer-term or decadal internal oscillatory signals that can be distinguished from climatic noise—arbitrary year-to-year variation. The El Niño/Southern Oscillation (ENSO) appears to be the only verifiable oscillation.
According to scientists, “A distinct—40-to 50-year timescale—spectral peak that appears in global surface temperature observations appears to reflect the response of the climate system to a combination of anthropogenic and natural forcing rather than any intrinsic internal oscillation.”
The study has recently been reported in the Nature Communications journal on January 3rd, 2020.
If there were Pacific Decadal or Atlantic Multidecadal oscillations, the proof of their existence would have been available across the range of present sophisticated climate model simulations, the researchers further stated.
Given the current sophistication of climate models as seen in their ability to capture the El Niño/Southern Oscillation, we would expect to see consistent evidence for oscillations across a suite of climate models. We found no such evidence.
Michael E. Mann, Distinguished Professor, Department of Meteorology and Atmospheric Science, Penn State
The MTM-SVD technique is a tool jointly developed by Mann in the mid-1990s, and to date, it has been utilized in over 50 peer-reviewed articles across various fields. With the help of the MTM-SVD technique, the scientists looked at the long-term “control” simulation-produced global surface temperature data as well as observational data dating back over 150 years.
The control simulations—which lack the exterior drivers used in the models—come from the latest global climate model intercomparison projects (CMIPS).
We found a tendency in the control models for oscillations in the three to seven-year ENSO band. However, we found no other signals, no Pacific or Atlantic climate variability on decadal or longer timescales that could be characterized as a true oscillation. Such variability was essentially indistinguishable from random noise.
Michael E. Mann, Distinguished Professor, Department of Meteorology and Atmospheric Science, Penn State
In the “forced” suite of CMIPS simulations, the climate models are powered with external factors, including volcanoes and increased pollution caused by humans. From help by simulations, scientists effectively demonstrated that the four-decade or five-decade spectral peak, which is occasionally linked with the AMO, is, in fact, an object of the slowdown in warming between the 1950s and 1970s.
This warming was caused by the accumulation of sulfur “aerosol” pollutants that cool the surface of the Earth. The 1970s saw the passage of the Clean Air Act that eliminated the cooling effect, which resulted in increased greenhouse gas warming. The slowdown and further acceleration of warming masquerade as an evident “oscillation.”
Our study provides another line of evidence that purported decadal and longer timescale internal oscillation in climate that have been identified through analysis of observational data are in fact mostly a result of external influences like greenhouse gas and aerosol emissions by humans.
Byron A. Steinman, Study Co-Author and Associate Professor, Earth and Environmental Science, University of Minnesota Duluth
Sonya K. Miller from BASF also worked on the new project. The study was funded by the National Science Foundation.