Nitrous oxide (N2O) is an important greenhouse gas produced by natural and human sources and is also the main contributor to the depletion of the stratospheric ozone layer1. With a long atmospheric lifetime of at least 100 years and a global warming potential about 265–298 times as high as that of CO2, N2O has the potential to cause both short-term and long-lasting environmental problems. Agriculture is the main source of anthropogenic N2O emissions, primarily through the application of synthetic nitrogen (N) fertilizers and manure on agricultural land1. As the global amount of fertilizer used is expected to increase considerably over the next few decades, it is essential to establish reliable inventories of N2O in order to define effective mitigation strategies. Writing in Nature Climate Change, Rona Thompson and colleagues2 present new global estimates of N2O emissions and show that this greenhouse gas has increased substantially since 2009, at a faster rate than expected. Their result questions one of the main methods currently used for the inventory of N2O emissions at the global scale.