Over the past several decades, a large body of research has examined how biodiversity loss influences ecosystem functioning, as well as the cascading impacts on the goods and services ecosystems provide to humanity. The form of biodiversity-ecosystem function relationships commonly described in studies suggests that initial losses of biodiversity have relatively small impacts on ecosystem functions, but beyond some threshold, increasing losses lead to accelerating rates of change. Parameter estimates for this 'saturating' curve suggest that few species are required to maximize the rate of any given process, such as primary production. However, some have questioned whether a saturating relationship between diversity and productivity is an artifact of overly simplified experiments, imposed by the spatial homogeneity or short time-scales of experiments that minimize opportunities for expression of niche differences.
We used a biogeographic dataset to test two alternative hypotheses about how the diversity of primary producers relates to the production of biomass in unmanipulated ecosystems: (H1) Compared to experiments that show a saturating curve, biomass production will be a stronger, monotonically increasing function of algal diversity in natural lake ecosystems. (H2) Alternatively, any potential relationship of biomass production to species diversity in natural lakes will be insignificant and undetectable as other environmental drivers of production (e.g., nutrients, temperature, light) overwhelm any influence of diversity. To distinguish among these hypotheses, we used Structural Equations Modeling (SEM) to quantify statistical relationships between species richness of algae, biomass production, and environmental parameters measured in the U.S. EPA's Lakes Survey, which sampled 1028 freshwater lakes across the U.S.
Results/Conclusions
Initial SEMs have revealed statistically significant relationships between algal species diversity and biomass production in lakes (p<0.05). Statistically significant relationships between biomass production and total concentration of nutrients (p<0.05) and between biomass production and temperature (p<0.05) were also found. A statistically significant relationship was not found between biomass production and light. One caveat is that the models constructed thus far have had low overall explanatory power, which suggests caution is warranted when interpreting these results (Χ2=452, p=0.00). We are continuing to refine the SEMs, and construct alternative models, that increase the amount of variation in biomass production that can be explained. Further modifying of SEMs will highlight the magnitude of the impacts of biodiversity and environmental variables on the production of biomass in 'real-world' ecosystems. This will have important implications in understanding the applicability of prior experiments in 'real-world' ecosystems.