Metabolic ecology meets ecological stoichiometry: Predicting nutrient excretion rates of aquatic animals in the field

Background/Question/Methods

Metabolic ecology and ecological stoichiometry provide two important frameworks for understanding ecological interactions across many levels of organization. Metabolic ecology theory predicts that metabolic rates scale with body mass with an exponent of 0.75 (i.e., the slope of the log body mass vs log metabolic rate is 0.75) and that rates increase exponentially with temperature. Ecological stoichiometry theory posits that nutrient cycling rates of animals are a function of the imbalance in nutrient composition between consumer tissue and food resources. The integration of these two areas of ecology may yield important new insights into nutrient cycling, but an empirical integration is lacking. We compiled data from the literature to test the hypothesis that nutrient cycling rates of aquatic animals (freshwater and marine), in field settings, can be predicted with animal body size, temperature and animal body nutrient content. Our data set includes over 7000 nitrogen (N) and 6000 phosphorus (P) excretion measurements on aquatic animals ranging in body size from <1 ug to over 500 g dry mass, and inhabiting streams, lakes and oceans.

Results/Conclusions

Using the entire data set, log body mass and temperature together explained >87% of the variation in both log N and log P excretion rates of animals in nature. The slope of the log body mass vs log excretion rate relationships were significantly greater than 0.75 for both N (slope=0.84) and P (0.83) excretion. Invertebrates had lower N and P excretion rates than vertebrates, at a given body mass, and the slopes of the log mass vs log excretion rate relationships were also lower for invertebrates as a group (slope=0.67 for N and 0.68 for P) than for vertebrates as a group (0.81 and 0.76 for N and P). Relationships at finer taxonomic resolution (e.g., phylum, class, order) yielded a variety of slopes, but they were usually significantly different from 0.75. Using the entire data set, temperature effects were greater for N excretion (Q10=2.72) than for P excretion (Q10=1.71). Animal body nutrient content explained relatively little variation in excretion rates, using the entire data set, but had a significant effect at finer taxonomic resolutions. For example, for vertebrates, P excretion rate was negatively correlated with body P content, as predicted by ecological stoichiometry theory. Our results show that body size, temperature and body nutrient composition can successfully predict nutrient cycling rates of aquatic animals, in natural ecosystems.