Blanket claim that climate models are flawed dismisses their key role

Climate scientists analyzed Jeff Jacoby’s column on climate modeling (“Climate science far from settled”) for climatefeedback.org, a site where scientists collaboratively comment on the accuracy of climate change media coverage, and rated it “very low” in scientific credibility.

Jacoby’s central claim is that “climate projections using models based on [the role of CO2] have consistently failed, nearly always predicting far more warming than has occurred,” which he supports by linking to a 2013 study led by John Fyfe. However, in Climate Feedback’s review, Fyfe described this statement as “false.” Jacoby makes a blanket claim that the models are flawed, but his column does not offer supporting evidence. If one looked for evidence, one would find a more nuanced reality that, despite their limitations, the models remain useful tools to understand climate changes. In reality, “climate model projections from the 1980s have essentially come true,” as professor Andreas Schmittner noted in the same review.

The column argues that because the climate is “unfathomably complex,” “the odds that a climate model . . . would be reliable” are extremely low. However, Jacoby’s analogy of the mathematical probability of correctly guessing a sequence of random independent variables is irrelevant, since climate models are based not on the flips of a coin, but on the laws of physics.


As The Boston Globe’s Twitter account claims, #FactsMatter. To gain correct scientific understanding of how the world works, sound logic is also required.

Emmanuel Vincent


Scott Johnson

Science editor

Climate Feedback

Merced, Calif.