Ventilation and dissolved oxygen cycle in Lake Superior: Insights from a numerical model

Abstract

Ventilation and dissolved oxygen in Lake Superior are key factors that determine the fate of various natural and anthropogenic inputs to the lake. We employ an idealized age tracer and biogeochemical tracers in a realistically configured numerical model of Lake Superior to characterize its ventilation and dissolved O 2 cycle. Our results indicate that Lake Superior is preferentially ventilated over rough bathymetry and that spring overturning following a very cold winter does not completely ventilate the lake interior. While this is unexpected for a dimictic lake, no part of the lake remains isolated from the atmosphere for more than 300 days. Our results also show that Lake Superior's oxygen cycle is dominated by solubility changes; as a result, the expected relationship between biological consumption of dissolved O 2 and ventilation age does not manifest.

Light may have triggered a period of net heterotrophy in Lake Superior

“…We argue that an increase in water clarity, in association with the climate change and teleconnection effects of higher water temperatures and reduced ice and cloud cover, is likely to be the strongest driver of this heterotrophy, causing the drawdown of the deep‐water DOC pool by direct photomineralization and/or photodegradation of recalcitrant DOC. Both in situ (Weiler , Russ et al ) and modeling studies (Bennington et al ; Matsumoto et al ), as well as this study (Fig. ), describe the tendency for Lake Superior to be more autotrophic during the ice‐free season and indicate that its most heterotrophic period occurs during the winter months.…”

“…Lake Superior's DOC concentrations in 1998 averaged 1.3 mg L 21 (Biddanda et al 2001;Biddanda and Cotner 2003;Urban et al 2005), providing a total DOC pool of 17 Tg C (Urban et al 2005). While DOC drawdown rates in 1998 were sufficient to explain hypolimnetic O 2 depletion rates measured in 2000 and 2001, only 5-10% of the DOC pool was being mineralized by bacteria within the average stratified period (Biddanda et al 2001;McManus et al 2003). A rough annual carbon deficit during the heterotrophic period of 0.9 Tg C can be calculated as the difference between total estimated carbon losses (surface CO 2 emissions 5 1.3 Tg C yr 21 ; Atilla et al 2011; OC burial 5 0.5 Tg C yr 21 ; outflows 5 0.1 Tg C yr 21 ; Fig.…”

“…A suite of studies published in the early 2000s found that measured community respiration (CR) rates tended to be higher than GPP rates, indicating a state of net heterotrophy (Cotner et al ; Russ et al ; Urban et al a ). A state of net heterotrophy was further suggested by full‐lake carbon budgets (Cotner et al ; Urban et al ), and studies measuring and/or modeling surface dissolved oxygen (O 2 ) or CO 2 emissions from the lake to the atmosphere (Atilla et al ; Bennington et al ; Matsumoto et al ). However, prior studies describe Lake Superior as net autotrophic, being typically super‐saturated with O 2 and undersaturated with CO 2 relative to the atmosphere (Dobson et al ; Weiler ; Kelly et al ).…”

“…Given its strong link to cyclical climatic forces, its large size, and low anthropogenic impact, Lake Superior is an ideal ecosystem for studying the effects of climate and water physical characteristics on the metabolic balance of large oligotrophic aquatic environments. While disagreements concerning the net heterotrophy of the oligotrophic subtropical gyres largely revolve around the different methodologies being adopted (Williams ; Duarte et al ; Williams et al ), differing reports of Lake Superior's net heterotrophy appear to follow a timeline where studies using data from the late 1990s onward tend to report or assume a state of net heterotrophy (e.g., Urban et al ; Bennington et al ; Matsumoto et al ), while earlier reports imply a state of net autotrophy (Weiler ; Kelly et al ). Such disagreements might result from gradual improvements in our understanding of the biogeochemistry of Lake Superior, though we instead predict that the long‐term condition and potential short‐term shifts in the lake have likely been overlooked, as they remain unaddressed in the literature.…”

Toward a Comprehensive Water-Quality Modeling of Barnegat Bay: Development of ROMS to WASP Coupler

“…The Regional Ocean Modeling System (ROMS; Shchepetkin and McWilliams, 2005), an open-source ocean modeling system, has been successfully applied in many coastal, lacustrine, and estuarine applications (Banas, MacCready, and Hickey, 2009;Defne, Haas, and Fritz, 2011;Grifoll et al, 2013;Li et al, 2016;Matsumoto, Tokos, and Gregory, 2015;Ralston et al, 2013;White and Matsumoto, 2012). With the addition of the wetting and drying capability , ROMS is suitable for modeling tidal dynamics and water circulation, temperature, salinity, and sediment transport in a shallow, back-barrier estuary.…”