Turbulent mixing is a key process influencing the dynamics of subaqueous gravity currents. In this study, the evolution of the local turbulent mixing dynamics along a lock-exchange gravity current propagating over a mild slope is statistically investigated by ensemble-averaging and spanwise-averaging 200 large eddy simulation results at two time steps characteristic of the slumping and inertial phases of the current's propagation. Premultiplied velocity and density transport spectra are computed and confirm that vertical turbulent mixing is mainly due to the large-scale structures – i.e. Kelvin-Helmholtz billows – with smaller scale turbulence acting to homogenise the density field. We show that, unlike large-scale oceanic underflows, the head holds a substantial portion of the total turbulent transport flux (~ 25%), and the decrease of the local vertical density transport flux in the body scales exponentially with distance from the head, when normalised by the current's length. Finally, the vertical density transport flux is shown to be well predicted by an eddy-viscosity model after appropriate tuning of the model constants.

湍流混合是影响水下重力流动力学的关键过程。在这项研究中，通过集合平均和展向平均 200 个大涡模拟结果在两个时间步长的坍塌和惯性特征，统计研究了局部湍流混合动力学沿着在缓坡上传播的锁交换重力流的演变。电流传播的阶段。计算预乘速度和密度输运谱，并确认垂直湍流混合主要是由于大尺度结构——即开尔文-亥姆霍兹波涛——较小尺度的湍流起到了均匀密度场的作用。我们表明，与大规模海洋底流不同，水头占总湍流输运通量的很大一部分（~25%），当用电流长度归一化时，身体局部垂直密度输运通量的减少随着与头部的距离呈指数增长。最后，在适当调整模型常数后，涡粘性模型可以很好地预测垂直密度输运通量。