Drew Shindell argues that while one can never be certain that a climate model designed to recreate the effects of an ancient supereruption like Toba's turns in accurate results, one cannot ignore actual climate records stored in ice cores—which can paint quite a disquieting picture.