Ecosystem metabolism, or the balance between gross primary production and respiration, has rarely (if at all) been compared within environments exhibiting alternative stable states. Shallow lakes typically exist in two alternative states that differ by their dominating primary producers: a clear-water, submerged macrophyte-dominated state and a turbid-water, phytoplankton-dominated state. Because of their highly productive nature, prairie pothole shallow lakes likely play an important role in global carbon cycling; however, their metabolism has not been thoroughly investigated. We estimated the metabolism of 9 lakes (4 clear and 5 turbid) across all seasons (spring, summer, fall, winter) periodically over 5 years, using in situdiel changes in oxygen and temperature. We used an information-theoretic approach to assess relationships between gross primary production (GPP), respiration (R), net aquatic production (NAP), and a suite of predictive models, with models comprised of ecosystem state, season, solar radiation, lake to watershed ratio, % agriculture in the watershed, chlorophyll-a, dissolved organic carbon (DOC),and turbidity.
Results/Conclusions
NAP of both alternative states was similar, and therefore was not influenced by the dominating primary producer type. The lakes were autotrophic in the summer, but heterotrophic in the spring, fall, and winter. In contrast to most lakes, metabolism was relatively neutral on an annual basis, as GPP and R closely tracked each other. The best predictor of NAP was an interaction between season and solar radiation, with solar radiation positively related to NAP across all seasons, with the highest rate of increase in the fall, followed by the summer, spring, and winter. Similar to NAP, GPP was best predicted by season and solar radiation, with the highest production rates in the summer, then spring, fall, and winter. R was best predicted by season and DOC, with the highest rates in the spring, then summer, fall, and winter. Contrary to most lakes, R decreased with increasing DOC concentration across all seasons, suggesting that R drove DOC concentration via consumption. Our study is one of the first to evaluate the metabolic rate of ecosystems under alternative states. Our analysis indicated there was no effect of state on lake metabolism, and abiotic factors ultimately constrained the metabolism of these biologically saturated ecosystems.