Peatlands are the planet’s biggest terrestrial carbon sink, and understanding how they store and cycle carbon is critical for understanding climate change. But Jonathan Nichols, a biogeochemist at Columbia University’s Lamont-Doherty Earth Observatory, noticed a pattern when reading about peatlands: everyone cites the same statistics over and over again. He began to wonder whether those numbers were “the best we can do,” he says. Nichols and coauthor Dorothy Peteet of NASA Goddard Institute for Space Studies created a probabilistic distribution that included data from peatlands that had been left out of previous studies because their data were too sparse. The new approach nearly doubled the estimated amount of carbon held in northern peatlands (Nat. Geosci. 2019, DOI: 10.1038/s41561-019-0454-z). The study also has implications for paleoclimatology; it suggests that the expansion of peatlands and their carbon accumulation rate increased much more quickly after the end of the last ice age than once thought. The inclusion of sparsely sampled sites in the updated estimate sacrifices some precision in favor of improved accuracy, Nichols admits. However, this is a trade-off that offers a “more complete look at carbon in peatlands” that will help refine future climate models, he says.