I’ve made a healthy journalistic living off of the Climate Prediction Center’s long-lead forecasts, seasonal prognostications about whether we can expect things to be wetter or drier next winter. It got so bad for a while that my sister started calling me Niña-boy.

The thing is, people want to know what to expect. Regular people do, I mean. As I’ve learned over the years, the people with real skin in the game – farmers and water supply managers, for example – are more cautious, planning their lives in a way that is robust to the range of variability they know to expect.

But regular folks want to know what it’s going to be like this winter, and I’ve used a lot of ink trying to tell them.

A new paper in the Bulletin of the American Meteorological Society by Robert E. Livezey and Marina M. Timofeyeva looks back at the first decade of serious seasonal prediction, since the Climate Prediction Center started doing this back in the mid-1990s. Turns out that a lot of the skill in the forecast, they write, can be explained by simple trend analysis:

The inescapable conclusion is that this lead-independent skill comes from use of long-term trends to make the forecasts and we show that these trends are almost entirely associated with climate change.

See? Climate change is good for something.