Abiotic and biotic constraints on nitrogen fixation and producer species assemblages in an N-limited river ecosystem

Background/Question/Methods

Nitrogen fixation has great potential to influence productivity, species assemblages, and food web interactions, particularly in N-limited stream ecosystems.  Preliminary data from the South Fork Eel River basin indicate that N-fixation rates are often substantial, but highly variable in space and time.  The interaction between physical network structure and temporal dynamics that shape producer species assemblages and N-fixation throughout the basin are likely linked to observed changes in both abiotic (light and temperature) and biotic (grazing) factors, as well as seasonal changes in the availability of stream water nitrogen and phosphorus.   Spatial patterns are also shaped by the magnitude and timing of spring floods that influence the establishment of both producer and consumer species. The main objective of this study was to investigate abiotic and biotic constraints on spatial and temporal patterns in N-fixation and species assemblage structure, and the consequences for nitrogen loading.  In summer 2010, we characterized N-fixation rates and areal cover of cyanobacteria in 5 streams ranging in drainage area from 0.58 – 148 km², measured diel variation in N-fixation, and studied the effect of nutrient enrichment and grazing on epilithic N-fixation.  N-fixation was measured in-situ using acetylene-reduction assays in gas-tight chambers.

Results/Conclusions

During summer 2010, N-fixing species assemblages were dominated by free-living cyanobacteria, in contrast with previous years where Epithemia spp. (diatoms with cyanobacteria endosymbionts) were prevalent.  N-fixation rates in 2010 were high across cyanobacterial taxa, but were significantly lower than areal rates observed in Epithemia during 2008 and 2009. Taking both N-fixation rates and areal cover into account, we estimated that N-fixation associated with Epithemia spp. in 2008 and 2009 contributed significantly more fixed nitrogen to the river ecosystem than either Rivularia or Nostoc spp. in 2010.  Furthermore, N-fixation rates exhibited a strong relationship with light over diel periods, declining rapidly under low light conditions. Nitrogen addition suppressed N-fixation rates and the abundance of N-fixing species, while P-addition had no effect on N-fixation.  Grazing significantly increased N-fixation rates in epilithic biofilms in summer 2010.  In contrast, in 2008, a high Epithemia year, grazing reduced N-fixation rates, suggesting that inter-annual shifts in dominant N-fixing taxa alter the potential for biotic constraints on N-fixation.   These results suggest that both nutrients and grazing may exert a strong effect on N-fixation rates through impacts on abundance, species composition, and activity of N-fixing species, forging a link between biofilm assemblage dynamics and biogeochemical cycling within stream ecosystems.