Intermittent oxygen flux from the interior into the bottom boundary of lakes as observed by eddy correlation

Turbulent oxygen transport from the overlying stratified water column into the bottom boundary layer (BBL) on the slope of a medium-sized lake was investigated using the eddy correlation (EC) technique. The seicheinduced oscillatory flow of the BBL, with a period of ,1 d, was identified as the mechanism driving turbulent oxygen transport. Sporadic short-term EC vertical oxygen fluxes exceeded the sedimentary oxygen uptake of 13 6 2 mmol m22 d21 calculated from sediment oxygen profiles by more than a factor of three. The average EC flux over half of a seiching period was 9.2 mmol m22 d21 similar in range to the flux into the sediment; however, these two fluxes do not have to coincide spatially and temporally. The EC oxygen flux was only significant when the deep basin-scale currents exceeded a velocity of 2 cm s21 and the corresponding bottom shear was sufficient to produce active turbulence. Below this threshold, decaying turbulence resulted in oxygen fluxes lower than 3.5 mmol m22 d21, with an even lower average flux of 0.8 mmol m22 d21 observed during reversals of the seiching. At low velocities, the weak turbulence is insufficient to transport dissolved oxygen through the stratified top of the BBL (stability N2 < 2.4 3 1024 s22), even though turbulence was found in the inertial subrange and periodical bottom convective mixing was still present. The EC technique provided valuable data on the temporal variability of oxygen transport related to the BBL hydrodynamics and flux pathways. The eddy correlation (EC) technique is a well-established method to determine atmospheric fluxes of water vapor, carbon dioxide, nitrous oxide, and others between water and air or soil and air (e.g., Famulari et al. 2004; Lee et al. 2004). This method is based on the simultaneous measurements of the fluctuations of turbulent velocity (uj9) and concentration (c9): The product of both variations results in the momentary turbulent fluxes of a substance horizontally and vertically (u9c9 and w9c9), respectively. The net turbulent flux is obtained by averaging this fluctuation product over a given time span. Berg et al. (2003) were the first to determine sedimentary oxygen uptake in aquatic systems with an EC approach in fluvial and coastal marine sites. The resulting oxygen flux estimates agreed well with those determined with traditional techniques, like flux chambers and the evaluation of oxygen profiles in the sediment. Kuwae et al. (2006) recorded time series between 12 and 40 min over a tidal mudflat during daytime and nighttime and showed that the EC method could also be used to investigate photosynthesis. McGinnis et al. (2008) recorded the diurnal cycle of photosynthesis and respiration in a run-of-the-river hydropower reservoir in Switzerland. These studies revealed the excellent potential of the EC technique for quantifying solute dynamics under turbulent conditions. Here we use the EC technique to examine oxygen flux in a turbulent system that demonstrates strong intermittency. Specifically, we examined the bottom boundary layer (BBL) of lakes where there is strong variability due to sporadic energy input. Lake Alpnach, a prealpine lake with regular, nonsteady basin-scale deep currents, was chosen as the study site because convective winds from the surrounding mountains force the lake-internal seiching that generates regular BBL dynamics (Gloor et al. 2000). Two different seiche modes were observed in Lake Alpnach: the first horizontal, first vertical seiche mode with a typical period of 8 to 12 h and the first horizontal second vertical seiche mode with a period of approximately 1 d (Münnich et al. 1992). Acknowledgments We thank Christian Dinkel, Michael Schurter, and Daniela Richter for support in the field; John Little for improving the English of the manuscript; and two anonymous reviewers and Josef D. Ackerman for valuable remarks. D.F.M. was supported by the Swiss National Science Foundation (grants 200020103827.1 and 200020-111763.1) and A.B. by Eawag. Limnol. Oceanogr., 53(5), 2008, 1997–2006 E 2008, by the American Society of Limnology and Oceanography, Inc.