Nonstationarity in the surface layer over complex terrain during T-REX

Numerical atmospheric models use surface layer parameterizations to relate turbulence fluxes to gradients of mean variables. These parameterizations rely on surface layer similarity theory which is developed for horizontally homogeneous terrain and stationary conditions. These conditions are often disturbed in hilly and mountainous terrain characterized by the presence of multi-scale wind systems. A lack of stationarity introduces uncertainty in the applicability of statistical turbulence theory in surface layer parameterizations, resulting potentially in a poor simulation of turbulent surface fluxes in numerical atmospheric models. Assessing the degree of nonstationarity therefore seems imperative to developing new and improved surface layer parameterizations for complex terrain.

In this this presentation we address the degree of nonstationarity and its spatial variability in the surface layer over complex terrain. Data from three 30 m towers with 6 levels of ultrasonic anemometers are used, collected during the Terrain-induced Rotor Experiment (T-REX) conducted in Owens Valley, California in March and April 2006. The analysis is performed on the entire two month period. The degree of nonstationarity is determined using different methods, such as introduced by e.g. Foken and Wichura (1996), Vickers and Mahrt (1997) and Nappo et al. (2010). We then evaluate existing surface layer similarity functions with data obtained in a variety of conditions with different degrees of nonstationarity. We will discuss how these results can be used to improve surface layer parameterization schemes in numerical atmospheric models.