Decadal Prediction: The New Kid on the Block

Climate scientists generally group future outlooks of the earth's climate into two, and now possibly three, time-scales. First, there's short term, or seasonal forecasting, which covers the next month to a year into the future. The International Research Institute for Climate and Society and other centers in the U.S. and around the world issue new seasonal forecasts every month. They tell us the likelihood a given place will see above- or below-average temperatures and rainfall. These forecasts are incredibly useful for projecting crop yields in an upcoming growing season, for example, or for managing water reservoirs. They can even help humanitarian organizations prepare better for the possibility of extended droughts or extreme weather events, which can wreak havoc on vulnerable communities.

At the other end of the time scale, we have long-term climate projections, such as those summarized by the Intergovernmental Panel on Climate Change. They model climate scenarios far into the future, 50-100 years, to give us an indication of average temperature and other climate characteristics on regional and planetary scales. But the associated uncertainties of long-term projections are greater than those of seasonal forecasts.

In between these two time scales is the "new kid on the block": decadal, or near-term climate change prediction. It covers 1-10 years, and in some cases up to 20 years, into the future.

"Climate change projections are important for planning and adaptation, but in order to make it through the next 50 years, you first have to get through the next ten," says IRI's Lisa Goddard, the lead author of a new essay on decadal prediction in this month's issue of the Bulletin of the American Meteorological Society. "The intersection of climate variability and change is crucial at this time scale."

Research on decadal prediction has created a buzz in the climate science community because it is still relatively new and experimental, and because it's in high demand for the planning and decision making needed on longer term development projects. For example, a country that wants to construct a dam to meet future water and energy demands, or that is planning how it will expand agricultural production, needs to understand what climate conditions might be like in 10, 20 or 30 years. Seasonal forecasts and long term climate projections are useful, but inadequate in these cases.

Goddard, who leads IRI's Climate Program, writes that while decadal prediction is the "fascinating baby that all wish to talk about" in the research community, it shouldn't be viewed as competing for research dollars and attention with its "older sibling", seasonal forecasting.

"The main point we want to get across is that research dollars invested in decadal prediction efforts benefit seasonal forecasting as well, and vice versa," she says.

For example, both fields depend on the same type of 'general circulation models' used for climate projections, and they make use of the same global observing systems, which gather massive amounts of data on land, ocean and atmosphere conditions around the world that can be used to better understand and simulate physical processes. Improvements in either of these efforts will improve short-term and near-term prediction ability.

The excitement over decadal prediction is good for seasonal forecasting too, the authors write, because it draws attention to the need for climate information and its use. This can bring in funding and new research to strengthen the whole climate prediction family.

"As we research decadal variability and the potential for prediction, we also gain a better understanding of the role that this kind of variability has on year-to-year impacts, such as droughts and floods," they write.

Read the new paper online for free here.