Sea level rise, prolonged drought, shifts in fisheries regimes, and changes in drinking water availability are all threatswe would like to anticipate years to decades in advance. Called decadal predictions, very long-range forecasts can provide guidance for decision-makers and natural resource managers on planning for the types of impacts and issues listed above. However, despite the scientific community’s increased attention to decadal prediction, skillful decadal prediction remains a major challenge.
A new study discusses major advances in understanding decadal predictability and prediction of temperature over the past few years such as the global warming “hiatus”, cause of abnormal temperature fluctuations in the northern latitudes from 1900 to the present, and new methods for decadal predictions. The author of the study, Tim DelSole (George Mason University), finds that sources of decadal predictability are still highly debated but continued research in this area could lead to breakthroughs in the near future, that research stimulated by the “hiatus” has led to increased understanding of decadal variability, and that there are a number of open questions that are not currently a focus of coordinated research.
This study was supported by the CPO Modeling, Analysis, Predictions, and Projections (MAPP) Program.
Decadal prediction refers to predictions on annual, multi-year, and decadal time scales. This paper reviews major developments in decadal prediction that have occurred in the past few years, including attribution of temperature anomalies in northern latitudes, the recent slowdown in the rate of global warming (the “hiatus”), and mechanisms of decadal predictability that do not involve interactive ocean circulations. In addition, this paper discusses certain advances that, in the opinion of the author, have not been given the attention they deserve in previous reviews, including a unified framework for quantifying decadal predictability, empirical models for decadal prediction, defining improved indices of decadal predictability, and clarification of the relation between power spectra and predictability.