In a feature article in this issue, veteran climate scientist Tom M.L. Wigley evaluates the performance of climate models—two of them his own—asking how the models’ predictions made thirty years ago stand up today. They have performed remarkably well, he concludes; they’ve stood the test of time.
Now a new study has been published evaluating the predictive performance of climate models all the way back fifty years, when the models were then still rudimentary. The conclusions are similar: Even fifty-year-old climate models, basic as they were, correctly predicted current global warming.
The researchers compared annual average surface temperatures across the earth today to the surface temperatures predicted in seventeen forecasts drawn from fourteen separate computer models released between 1970 and 2001. Most of the models accurately predicted recent global surface temperature observations. For ten forecasts, there was no statistically significant difference between their output and historic observations.
The findings are published in the December 4, 2019, Geophysical Research Letters (“Evaluating the Performance of Past Climate Model Projections”) by Zeke Hausfather of UC Berkeley, Henri Drake of MIT and Woods Hole Oceanographic Institution, Tristan Abbot of MIT, and Gavin Schmidt of NASA Goddard Institute for Space Studies.
“We find that climate models published over the past five decades were generally quite accurate in predicting global warming in the years after publication, particularly when accounting for difference between modeled and actual changes in atmospheric CO2 and other climate drivers,” the authors wrote in their paper. “This research should help resolve public confusion around the performance of past climate modeling efforts, and increases our confidence that models are accurate projecting global warming.”