Water-level fluctuations due to reservoir management could substantially affect the timing and magnitude of reservoir methane (CH) fluxes to the atmosphere. However, effects of such fluctuations on CH emissions have received limited attention. Here we examine CH emission dynamics in six Pacific Northwest U.S. reservoirs of varying trophic status, morphometry, and management regimes. In these systems, we show that water-level drawdowns can, at least temporarily, greatly increase per-area reservoir CH fluxes to the atmosphere, and can account for more than 90% of annual reservoir CH flux in a period of just a few weeks. Reservoirs with higher epilimnetic [chlorophyll a] experienced larger increases in CH emission in response to drawdown (R = 0.84, p < 0.01), suggesting that eutrophication magnifies the effect of drawdown on CH emission. We show that drawdowns as small as 0.5 m can stimulate ebullition events. Given that drawdown events of this magnitude are quite common in reservoirs, our results suggest that this process must be considered in sampling strategies designed to characterize total CH fluxes from reservoirs. The extent to which (and the mechanisms by which) drawdowns short-circuit connections between methanogenesis and methanotrophy, thereby increasing net CH fluxes to the atmosphere, should be a focus of future work.

No abstract

Reservoirs are a globally significant source of methane (CH 4 ) to the atmosphere. However, emission rate estimates may be biased low due to inadequate monitoring during brief periods of elevated emission rates (that is, hot moments). Here we investigate CH 4 bubbling (that is, ebullition) during periods of falling water levels in a eutrophic reservoir in the Midwestern USA. We hypothesized that periods of water-level decline trigger the release of CH 4 -rich bubbles from the sediments and that these emissions constitute a substantial fraction of the annual CH 4 flux. We explored this hypothesis by monitoring CH 4 ebullition in a eutrophic reservoir over a 7-month period, which included an experimental water-level drawdown. We found that the ebullitive CH 4 flux rate was among the highest ever reported for a reservoir (mean = 32.3 mg CH 4 m -2 h -1 ). The already high ebullitive flux rates increased by factors of 1.4-77 across the nine monitoring sites during the 24-h experimental waterlevel drawdown, but these emissions constituted only 3% of the CH 4 flux during the 7-month monitoring period due to the naturally high ebullitive CH 4 flux rates that persist throughout the warm weather season. Although drawdown emissions were found to be a minor component of annual CH 4 emissions in this reservoir, our findings demonstrate a link between water-level change and CH 4 ebullition, suggesting that CH 4 emissions may be mitigated through water-level management in some reservoirs.

No abstract