Nitrogen (N) fixation – the biochemical conversion of N₂ gas to ammonia – influences many different aspects of terrestrial ecosystem form and functioning – including net primary productivity, carbon (C) storage, and ecosystem responses to global environmental change. It has been recognized for some time that N fixing organisms hold an advantage under N-poor/phosphorus (P)-rich conditions; and this line of reasoning reliably explains patterns of fixation in some sites, especially those associated with major disturbances and early in ecosystem succession. At the same time, symbiotic N-fixing plants (i.e. Fabaceae) are widely distributed across sites that appear high in N and low in P, but almost entirely absent from those that are low in N and high in P. It thus appears that a globally applicable framework of N fixation must be able to capture myriads of conditions, with mechanistic explanations for presence and absence, persistence and disappearance.

Results/Conclusions

We argue that the challenge of N fixation rests on the integration of biophysical and biogeochemical controls, that, when combined, offer reasonable explanations for patterns of N fixation across Earth’s major biomes. In particular, evidence indicates that low temperatures limit N fixers from mature forests at high latitudes, and the ability of fixers to invest N into P acquisition appears vital to persistent N fixer populations in many lowland tropical environments. These biophysical and biogeochemical mechanisms provide a means for representing N fixation in predictions of global CO₂ rise and climate change. By analyzing the results from the coupled carbon-climate simulations by 8 different models, for example, we found that all models overestimated C uptake by the land biosphere and underestimated the atmospheric CO₂ concentrations and global warming due to the positive carbon-climate feedback by 2100. When available N and its response to increasing atmospheric CO₂ and temperature are considered, we found that the additional warming from the positive carbon-climate feedback by year 2100 will result in 1.2 ^oC warming, rather than 0.7 ^oC warming as predicted by the coupled carbon-climate models without nutrient limitation.  We suggest that continued work on N fixation will improve climate change predictions – with much of the challenge still lying in finding effective ways to quantify N fixation across Earth’s diverse ecosystems.