Association for Tropical Biology and Conservation

Biological nitrogen fixation (BNF) is the largest natural source of new nitrogen (N) that supports terrestrial productivity1,2, yet estimates of global terrestrial BNF remain highly uncertain3,4. Here we show that this uncertainty is partly because of sampling bias, asfield BNF measurements in natural terrestrial ecosystems occur where N fixers are 17 times more prevalent than their mean abundances worldwide. To correct this bias, we develop new estimates of global terrestrial BNF by upscaling field BNF measurements using spatially explicit abundances of all major biogeochemical N-fixing niches. We find that natural biomes sustain lower BNF, 65 (52-77) Tg N yr-1, than previous empirical bottom-up estimates3,4, with most BNF occurring in tropical forests and drylands. We also find high agricultural BNF in croplands and cultivated pastures, 56 (54-58) Tg N yr-1. Agricultural BNF has increased terrestrial BNF by 64% and total terrestrial N inputs from all sources by 60% over pre-industrial levels. Our results indicate that BNF may impose stronger constraints on the carbon sink in natural terrestrial biomes and represent a larger source of agricultural N than is generally considered in analyses of the global N cycle5,6, with implications for proposed safe operating limits for N use7,8.

Shifting the current research paradigm in ecology and evolutionary biology (EVE) to one that values replication of previously published studies requires both educating the EVE community about their importance and providing the incentives to carry them out. Academic societies are uniquely poised to do both, which is why we propose that the Association for Tropical Biology and Conservation (ATBC) spearhead a Reproducibility Project in Tropical Biology – the first such effort in environmental biology – to encourage replicating some of the most important studies in this discipline in sites across the globe.