Three-Dimensional Representation of Large-Scale Structures Based on Observations in Atmospheric Surface Layers

Large amounts of synchronous multipoint measurement data were acquired from long-term observations in high-Reynolds-number atmospheric surface layers (Re-tau similar to O(10(6))) at the Qingtu Lake observation array site. Based on these experimental data, the general shapes of three-dimensional (3-D) large-scale coherent structures are constructed coupled with the symmetry of two-point streamwise velocity correlations in the spanwise direction, where the spanwise symmetry is confirmed to be applicable in the atmospheric surface layer flow by experimental verification. The full extent of the 3-D structure is much larger than a 2-D slice at the condition point and increases significantly with the reference height in the logarithmic region. Moreover, the streamwise length and the spanwise width of the horizontal section are found to vary systematically with the relative height to the condition point. These variations are parametrized and can subsequently be used to infer the horizontal section at different relative heights. Then, a plausible empirical model is proposed by taking into consideration the streamwise shift caused by the inclination angle, and the 3-D structure can be predicted by this model using the primary flow parameters and location information. This study contributes to a better understanding of dominant structures in the outer region of wall turbulence at high Reynolds numbers.