Background

Understanding and predicting global climate change may be one of the most complex scientific challenges we face today. MIT’s School of Science recently launched the Lorenz Center, a new climate think tank devoted to fundamental inquiry. By emphasizing curiosity-driven research, the Center fosters creative approaches to learning how climate works.

This year’s lecturer, Timothy Palmer, discussed the key sources of uncertainty in making such predictions, how we estimate their impact and how we might reduce forecast uncertainties.

Edward Lorenz's pioneering work on systems with unpredictable and chaotic evolutions was motivated by skepticism about the use of statistical models to predict next month's weather. And yet, on the web and elsewhere, one can find predictions not only of next month's weather, but also of the human effect on long-term climate. Can we have any confidence at all in long-range predictions of weather? And should we believe these estimates of human-induced climate change, or is the whole notion of predicting long-term changes in climate misguided and unscientific?