Laboratory for Atmospheric and Space Physics, Department of Astrophysical and Planetary Sciences, University of Colorado, Boulder, 80309, USA.
Phys Rev Lett. 2011 Nov 18;107(21):214501. doi: 10.1103/PhysRevLett.107.214501. Epub 2011 Nov 14.
The mixing properties of turbulent flows are, at first order, related to the dynamics of separation of particle pairs. Scaling laws for the evolution in time of the mean distance between particle pairs <r(2)>(t) have been proposed since the pioneering work of Richardson. We analyze a model which shares some features with 3D experimental and numerical turbulence, and suggest that pure scaling laws are only subdominant. The dynamics is dominated by a very wide distribution of "delay times" t(d), the duration for which particle pairs remain together before their separation increases significantly. The delay time distribution is exponential for small separations and evolves towards a flat distribution at large separations. The observed <r(2)>(t) behavior is best understood as an average over separations that individually follow the Richardson-Obukhov scaling, r(2) ∝ t(3), but each only after a fluctuating time delay t(d), where t(d) is distributed uniformly.
湍流的混合性质在第一阶上与颗粒对的分离动力学有关。自 Richardson 的开创性工作以来,人们已经提出了用于预测颗粒对之间的平均距离 <r(2)>(t) 随时间演化的标度律。我们分析了一个与 3D 实验和数值湍流具有某些共同特征的模型,并表明纯标度律只是次要的。动力学主要由非常宽的“延迟时间” t(d) 分布主导,即颗粒对在显著分离之前保持在一起的持续时间。对于小的分离,延迟时间分布呈指数分布,而对于大的分离,它则演化成平坦分布。观察到的 <r(2)>(t) 行为最好被理解为对个别遵循 Richardson-Obukhov 标度律的分离的平均,即 r(2)∝t(3),但每个分离都要在波动的延迟时间 t(d)之后,其中 t(d) 是均匀分布的。