Decadal Climate Prediction and Variability

Decadal climate prediction aims to forecast climate conditions on timescales of one to ten years, bridging the gap between seasonal forecasts and long-term climate projections. These predictions harness initialised coupled atmosphere–ocean models to capture both the influence of external forcings and the intrinsic variability of the climate system. Key drivers include ocean heat content anomalies, large-scale modes such as the El Niño–Southern Oscillation and the Atlantic Multidecadal Variability, and associated teleconnections that spread signals globally. Skillful decadal forecasts can inform adaptation and mitigation planning by anticipating near-term shifts in temperature, precipitation and extreme events. Ongoing research focuses on improving model resolution, enhancing initialization methods and quantifying predictability limits imposed by internal chaos and model uncertainties.

Research from Nature Portfolio

Recent studies have demonstrated that forecast skill can vary markedly with the phase of dominant climate modes. One investigation showed that predictions initiated during El Niño or La Niña events yield substantially improved accuracy in projecting multi-year temperature and precipitation anomalies, revealing windows of enhanced predictability. Another project extended these principles to marine ecosystems, demonstrating that phytoplankton biomass in a high-latitude shelf sea can be predicted up to five years ahead by capturing nutrient advection from the Subpolar North Atlantic and the influence of summer sea-ice concentration on light availability. Foundational work on Pacific variability has further highlighted how upper ocean heat content anomalies can precondition transitions in decadal sea surface temperature patterns, underscoring the ocean’s memory as a cornerstone of near-term climate predictability.

Research from all publishers

Advances outside the portfolio have enriched the field by exploring varied regions and methodologies. Forecasts indicate accelerated warming in a major mountain region over the next decade, attributing added skill to rigorous initialization and the influence of Pacific decadal variability via atmospheric wave pathways. High-resolution ocean forecasts have been shown to reduce biases in Southern Ocean warming, thereby enhancing global signal-to-noise characteristics and improving simulations of tropical Pacific multidecadal variability. Evaluations of extreme event predictions reveal moderate skill in forecasting multi-year changes in temperature extremes, though precipitation extremes remain a greater challenge, emphasising the need for region-specific calibration and user-oriented climate services.

Decadal Climate Prediction and Variability publication trend

The graph below shows the total number of articles in decadal climate prediction and variability across all publications each year (not limited to Nature Index journals).

Technical terms

Decadal prediction: Forecasts of climate variability over lead times of one to ten years that combine observed initial conditions with projected external forcings.

Initialization: The assimilation of observed climate data into a model to establish an accurate starting state for forecasts.

Teleconnection: A large-scale climate linkage whereby anomalies in one region influence conditions in another via atmospheric or oceanic pathways.

Signal-to-noise ratio: A metric comparing the strength of the predictable climate signal to the background of internal variability within an ensemble.

Hindcast: A retrospective forecast produced by initializing a model with historical observations to evaluate its predictive skill.

References

1. Enhanced multi-year predictability after El Niño and La Niña events. Nature Communications (2023).
2. Phytoplankton abundance in the Barents Sea is predictable up to five years in advance. Communications Earth & Environment (2023).
3. Initialized decadal prediction for transition to positive phase of the Interdecadal Pacific Oscillation. Nature Communications (2016).
4. Accelerated warming of High Mountain Asia predicted at multiple years ahead. Science Bulletin (2024).
5. Reduced Southern Ocean warming enhances global skill and signal-to-noise in an eddy-resolving decadal prediction system. npj Climate and Atmospheric Science (2023).
6. Multi-annual predictions of the frequency and intensity of daily temperature and precipitation extremes. Environmental Research Letters (2023).