Patterns and rates of biological nitrogen fixation during secondary succession in a lowland tropical rain forest

Background/Question/Methods

Intact and regenerating tropical rain forests are thought to be an important sink for atmospheric carbon dioxide, but low nutrient availability may limit future carbon uptake in this biome. Nitrogen (N) inputs via biological N fixation (BNF) have been estimated at 15 to 35 kg N ha^-1 y^-1, which would be sufficient to fuel significant and sustained rates of new biomass production well into the future. Surprisingly, however, very few studies have empirically measured rates of BNF in tropical forests, and current biome-scale estimates are extrapolated from a few measurements in a handful of sites. Here, our objective was to quantify the rates and relative contribution of N fixation from symbiotic versus free-living sources in a set of lowland tropical forest sites in southwestern Costa Rica. We quantified BNF in both primary forests and in a set of replicated sites comprising a secondary successional forest chronosequence (5 – 60 y). We used a sampling method (systematic cluster sampling) that allowed us to quantify symbiotic N fixing nodules independent of putative N fixer locations, and we measured symbiotic and free-living N fixation rates over the course of a full year using the acetylene-reduction assay.

Results/Conclusions

Our results show that, although legumes were abundant in all sites, symbiotic N fixation rates were lower than previous tropical forest estimates would suggest. Free-living N fixation in soil and plant litter was a greater source of N than symbiotic N fixation when expressed on an areal basis. Along the chronosequence, primary forests had both the lowest total rates of BNF and the lowest ratio of symbiotic-to-free living BNF. Biological N fixation rates were higher in recently disturbed secondary forests than primary forests, but still were much lower than rates previously reported for the biome as a whole. Overall, our results add to a small but growing body of evidence suggesting that rates of symbiotic N fixation in tropical forests may not be as high previously reported. These results also suggest that purportedly high N availability in tropical forests is not simply a product of continuously high rates of BNF.