Adaptive nitrogen fixation strategies with realistic tradeoffs allow persistent N limitation and N richness

Nitrogen (N) fixation lies at the heart of two unresolved and counterintuitive ecosystem patterns: 1) N-fixing trees are absent from mature temperate forests, even though most of these forests are N-limited; and 2) Tropical forests, which contain many potentially N-fixing trees, are N-rich, suggesting there is more fixation than might be needed. To elucidate physiological and/or ecological mechanisms that could drive these patterns we developed a simple, resource-based evolutionary model of N fixation. We assumed that a plant can be limited by N or some other resource (or both) and that the plant can acquire N from the soil or from fixation, then we allowed the trait of N fixation to evolve. In this basic model there are two possible evolutionary outcomes (continuously stable strategies), neither of which can explain the counterintuitive ecosystem patterns: 1) a population of non-fixers that is limited by the non-N resource and 2) a population of fixers that fix just enough N to be co-limited by N and the other resource. However, introducing tradeoffs between N fixation and the rates of mortality or resource uptake allows new evolutionary outcomes (either evolutionarily stable or continuously stable strategies) that match the observed empirical patterns: populations of chronically N-limited non-fixers or chronically non-N-limited fixers. We show analytically that the magnitudes of the tradeoffs needed to produce these patterns are well within the plausible ranges for real plants and ecosystems.