Equifinal ecosystem productivity despite biodiversity variance in replicate, open ecosystems

Background/Question/Methods

A link between biodiversity and ecosystem function is a long-standing subject that attracts debate. Interestingly, most discussion has ignored theory of open systems (Bertalanffy 1950, Science 111:23-29), though it provides a thermodynamic null hypothesis that ecosystem structure and function are not linked. Specifically, open systems theory states that thermodynamic state is not affected by initial conditions, but depends only on available energy and throughput; thus, communities that vary in composition should achieve the same ecosystem productivity given the same thermodynamic conditions (“equifinality”). The alternative hypothesis expected by many investigators is that potential phylogenetic effects underlying community structure affect function. We tested the equifinality hypothesis for ecosystem structure and function by inoculating 40 cattle tanks in different ways (4 treatments X 10 replicates) in a long-term project. Treatments are: Control - small inocula may lead to chance structural variation through time; Extra Biomass - large inocula should make ecosystems structurally similar; Homogenization – start as Controls, to be mixed later; and Drought – start as Controls but to be disturbed later. We censused plant community compositions at the end of the first growing season, and estimated net primary productivity (NPP) every 2 weeks using diurnal dissolved oxygen readings.

Results/Conclusions

Community composition significantly varied among treatments (p=0.03) as a result of inoculation and colonizations, and structural differences continue to develop (e.g., Typha, Isoetes, Hydrilla dominating certain tanks). The Extra Biomass treatment had a head start in NPP due to biomass, but no significant difference existed among treatments for NPP within 2 months and thereafter. Instead, NPP was very consistent among all systems. We conclude that Bertalanffy's equifinality hypothesis is supported; initial and continuing differences in biodiversity do not affect ecosystem energetics. Subsequent treatments (Homogenization, Drought) and successional changes will further test this hypothesis. We conclude that structural measures of biodiversity need not translate to functional measures of ecosystems (e.g., NPP) for two theoretical and one philosophical reason: (1) thermodynamic theory does not require it, (2) hierarchy theory does not expect simple translation from one hierarchical level (community) to another (ecosystem), and (3) philosophically, biodiversity has its own values that may not be measured well in the currency of ecosystem energetics. Instead, biodiversity should be measured and valued in its own terms. We also note that we measured NPP; which is more specific than general “functioning” (i.e., “activities, processes, or properties of ecosystems”; Naeem et al. 2002).