Tuesday, 14 January 2020: 9:30 AM
151A (Boston Convention and Exhibition Center)
James Johnstone, Climate Forecasting Applications Network, Seattle, WA
Large societal benefits might result from accurate forecasts of changes in the Atlantic Multidecadal Oscillation (AMO), which describes alternating ~30-40 year excursions of North Atlantic sea surface temperatures and correlated anomalies of tropical hurricane activity and continental temperatures and drought. However, debate continues over the AMO’s primary physical causes, particularly the role of the regional atmosphere. Some studies have proposed a primary and direct atmospheric driver of AMO SST changes via surface heat flux forcings. Others suggest a key role for variable ocean heat transports by the Atlantic Meridional Overturning Circulation (AMOC), possibly due to time-integrated atmospheric forcing of ocean dynamics with a decadal delay, behavior suggestive of long-range predictability.
Here we present observational evidence that historical AMO transitions were not incremental or gradual changes, as produced by slow ocean processes, but a series of abrupt, step-like SST shifts that reorganized the regional climate within just a few months to a few years, behavior more consistent with atmospheric processes. Based on spatio-temporal patterns of coupled atmosphere-ocean change, it is proposed that that AMO transitions begin as stochastic atmospherically-forced SST perturbations that are occasionally sustained by positive feedbacks as anomalous multidecadal climate states. Such behavior has important implications for AMO forecasting approaches, and suggests that patterns of short-term atmospheric variability may hold the key to prediction of low-frequency North Atlantic climate changes.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner