Due to increases in atmospheric carbon (C) concentrations, global and regional carbon budgets are needed to assess C sources and sinks. These budgets often over generalize or ignore freshwater systems, but current research suggests that these systems may play an important role in organic carbon sequestration and gaseous CO₂ release, and therefore should be incorporated into carbon budgets. Critical to forming accurate freshwater carbon budgets is inorganic carbon, specifically dissolved inorganic carbon (DIC) and gaseous carbon dioxide fluxes (pCO₂). Watershed land use likely plays an important role in shaping freshwater C budgets, but little is known about this linkage. Thus, our research seeks to determine whether two Ohio reservoirs of varying watershed land use are sources or sinks of DIC. The study reservoirs are Acton Lake, which has a watershed that is dominated by agricultural land (89%), and Burr Oak, which is mainly forested (81%). A high resolution sampling program (at least daily inflow and outflow samples) and continuous discharge data allowed us to calculate DIC loading into, and export from, the reservoirs. Due to high primary productivity of Acton and Burr Oak, we expected that both would have a higher DIC load than export because once in the lake the DIC is subject to uptake by algae for photosynthesis, precipitation as CaCO₃, or degassing as CO₂ into the atmosphere.

Results/Conclusions

Our results show higher DIC inputs than outputs for both reservoirs. In addition, the two reservoirs showed significant differences in the load of DIC: the DIC load for Burr Oak was consistently lower than that of Acton except during extended dry periods when Acton flow was negligible. The maximum DIC load for Burr Oak was 89.3 mgCm^-2day^-1, while Acton had a maximum DIC load of 838 mgCm^-2day^-1. Also, differences in watershed land use and productivity between the reservoirs lead us to predict that the difference between DIC load and export (i.e. relative retention and loss of DIC) would be greater for Acton than Burr Oak. Our data supporting this trend, paired with data showing no significant between-lake difference in pCO₂ venting, illustrates that Acton retains a higher percentage of DIC than Burr Oak (~15% vs. ~6%). Acton’s greater ability to take up DIC may be due to a higher level of primary production. Our data show that reservoirs can be significant in watershed-scale DIC cycling and that effects of reservoirs depend on watershed land use and lake productivity.