What is it about?

Predicting the climate over a timescale of several years is enabled by slowly varying ocean processes. The problem is, ocean observations, that are essential to start predictions for such timescales, are limited in quality and quantity. Here, we use an approach in which we generate realistic ocean states in the model by prescribing only the wind field at the ocean surface. The resulting retrospective forecast can reproduce the observed variations in mid-latitude North Atlantic surface temperature for even more than 7 years.

Featured Image

Why is it important?

Being able to make these multiyear predictions may not only be of scientific but also of political, public, and economic interest, because North Atlantic surface temperatures are thought to influence mean climate and extreme weather events over the adjacent continents.


Predictions on the multiyear time range differ from weather forecasts and climate change projections. Weather forecasts (for the next days to weeks) depend primarily on the atmospheric initial state and climate projections (for the next decades to centuries) depend primarily on the greenhouse gas concentration. Multiyear predictions as analysed here, depend primarily on the initial state of the ocean. Studying them helps us to understand complex processes in the ocean and in the communication between ocean and atmosphere, which is the most exciting part of climate science to me.

Dr. Annika Reintges
GEOMAR Helmholtz Centre for Ocean Research Kiel

Read the Original

This page is a summary of: Wind Stress‐Induced Multiyear Predictability of Annual Extratropical North Atlantic Sea Surface Temperature Anomalies, Geophysical Research Letters, July 2020, American Geophysical Union (AGU),
DOI: 10.1029/2020gl087031.
You can read the full text:




The following have contributed to this page