Abstract Non‐stationarity challenges the applicability of turbulent similarity theory and flux estimation, especially in urban areas with highly heterogeneous underlying surfaces. Based on the observational data in Beijing, we investigated the characteristics and potential physical mechanisms of turbulent non‐stationarity in a megacity area. The results show that the proportion of non‐stationary turbulence in Beijing reached 52.41% in 2017. Strongly non‐stationary turbulence mainly occurs under strongly‐stable/strongly‐unstable stratification conditions. Strong non‐stationarity reduces the turbulent transport efficiency. This is characterized by the disintegration of turbulent coherent structures and the dominance of submeso motions. Furthermore, it was found that submeso motions (e.g., internal gravity waves and convective circulations) are the main cause of non‐stationary turbulence. After removing the influence of submeso motions based on the Hilbert‐Huang transform, the non‐stationary level is effectively reduced. The results are significant for accurately estimating turbulent fluxes and improving the parameterization of turbulent exchange processes in megacity areas.

Read original article