Coherent Structures over Complex Terrain and their Contribution to Vertical Fluxes of Momentum and Heat

Near-surface turbulent flow, typical for the atmospheric boundary layer (ABL), is usually characterized by a high degree of stochasticness, resulting in the necessity of a statistically oriented approach to describe the mean properties of such flow. Even so, this turbulent flow oftentimes achieves the form of organized, deterministically more describable formations. The most common example of these formations are coherent structures. They are defined as an organized three-dimensional region of turbulent flow in which some property (e.g. temperature) is highly correlated with itself at a time scale larger than the smallest scale of the flow. As such, their contribution to the vertical transport of momentum and scalars may be significant. Historically, daytime coherent structures (convective plumes) have received considerable attention over flat terrain and across the forest canopy interface, where most studies agree that their contribution to the vertical fluxes of momentum and sensible heat oftentimes exceeds 50 %. Surprisingly, convective plumes over complex terrain (e.g. a valley) have received very little attention. It is unknown to what degree convective plumes are influenced by slope and valley flows, as well as mountain wave activity - phenomena typical of complex terrain. These interactions may cause substantial deviations in the flux contributions from those observed over flat terrain. Here we present preliminary results, with the aim to elucidate the structure of complex terrain convective plumes (frequency of occurence, mean duration, typical dimensions) and their spatial distribution over the Owens Valley floor and sidewall.

The main dataset we analyze is comprised of high-frequency turbulence measurements obtained during the Terrain-Induced Rotor Experiment, conducted in the spring 2006 in Owens Valley, CA. Specifically, we analyze sonic temperature time series obtained on three 34-m towers, two of which were located along the valley floor while the third one was located on the valley's western sidewall. Each tower was equipped with 6 levels of CSAT-3 sonic anemometers, located on heights of 5, 10, 15, 20, 25 and 30 metres above ground level. Convective plumes are most easily discernible in the sonic temperature time series, with well defined ramp-like signatures. To determine their contribution to vertical transport of momentum and heat, we employ three methods: quadrant analysis, the Variable-Interval Time Averaging (VITA) and the wavelet covariance technique. Finally, we compare the flux contributions determined by these three methods.