Project Focus
Spatio-temporal variability and environmental implicaitons of diapycnal mixing in Lake Simcoe 

Background and Motivation
Quantifying the extent of diapycnal mixing (K_z) in the water column of a lake has significant environmental implications. For example, the turbulent transport rate of DO from the well-oxygenated epilimnion to the hypolimnion is controlled by the magnitude of diapycnal mixing rates (K_z).  During periods of strong stratification, the thermocline can act as a barrier to the vertical turbulent transport of DO from the surface waters and increase the likelihood of low DO in the hypolimnion.

Lake Simcoe is a large dimictic lake located in the province of Ontario, Canada and has a history of significant hypolimnetic DO depletion during the strongly stratified summer season. The contribution of the turbulent DO fluxes across the thermocline in the hypolimnetic DO levels in the lake has not been explored in detail. In addition, the spatio-temporal variability of K_z as a function of the external environmental forcing  in any stratified lake  has not been studied in great detail. A number of studies quantified K_z in other North American lakes. However, most of these studies rarely exceeded a few days in duration. The main driving force for the diapycnal mixing in Lake Simcoe during summer stratification is from vertical shear due to the wind forcing. Thus, K_z is expected to correlate with the dimensionless Lake number (L_N), which quantifies the relative strength of the wind forcing in relation to the stratification.

The main objectives of this study were as follows:
1) To examine the spatial and temporal variability of  K_z in the stratified water columns of Lake Simcoe,
2) To relate the K_z variability with the environmental forcing through the Lake number L_N, and
3) To determine the contribution of turbulence transport of DO across the thermocline on the hypolimnetic DO budget.    

Methods

• The project was mainly based on a 40-day long field study in late summer of 2011. In this field study, high-sampling frequency records of water currents and temperature profiles were acquired using a moored acoustic Doppler current profiler (ADCP) and a chain of fast-response temperature loggers respectively.
• The acquired data were processed to quantify the relative strengths of the stratification in relation to the velocity shear, expressed by the dimensionless gradient Richardson number (Ri_g). The calculated Ri_g was used as a proxy to estimate K_z from a Ri_g-based  K_z model, recently prooposed for the ocean. The model paramters were first tuned up for the lake, based on temperature microstructure-based K_z measurements. 
• Hourly wind data from a nearby buoy, DO profile data from the Ontario Minstry of Environment, and long term three-hourly NARR wind data were also used.
  

Figure 1. Profiles of mean (a) temperature, (b) buoyancy frequency, (c) current shear, and (c) diapycnal mixing rates. Values at the thermocline are shown by an arrow in each figure. Error bars represent the standard deviation. Cartoon in  bottom figure shows Kz values in different layers of stratified waters.	Figure 2. Variation of Kz in the epilimnion in (b) , the thermocline in (c), and the hypolimnion in (d), with the LN shown in (a). LN-1  vs wind speed is shown in (e)
Figure 3a. time series of the distribution of dissolved oxygen [DO (mg L-1)]       in the water column for April 27 – November 24, 2011.	Figure 3b. variation of depth-averaged epilimnetic and hypolimentic DO concentration with time

Key Findings

• K_z exhibited large spatial and temporal variability. Spatially, the nearly-isothermal epilimnion showed high values K_z ~10^-5-10^-4 m² s^-1. Time-averaged K_z was ~10^-6 m² s^-1 and ~10^-5 m² s^-1 in the metalimnion and the hypolimnion respectively. In both regions, there was temporal variability in K_z of as much as one order of magnitude depending on how energetic the movements of the thermocline were. Sporadic mixing events at the benthic layer were also observed. The variability suggests that mixing between the bottom and the surface layers that occurs in Lake Simcoe is neither frequent nor fully absent, but is mainly episodic in nature [Fig. 1].
• Hourly mean K_z averaged over the epilimnion, the hypolimnion and the metalimnion correlated with the L_N, which mainly dependent on the wind speed and duration, rather than changes in depth of the thermocline [Fig. 2]
• Analysis of the observed K_z variability with DO distributions for an extended period [Fig. 3] suggested that vertical turbulent transport of dissolved oxygen was 10% of the total oxygen depleted in the hypolimnion for the observation period. We concluded that at least some of the observed year-to-year variability of hypolimnetic oxygen depletion could be due to the variation in turbulent transport of dissolved oxygen across the thermocline.