Abstract
Previous field and laboratory studies have indicated that flow and turbulence inside urban areas and, in particular, in street canyons, is very complex and is associated with wakes and vortices developing near buildings. However, a number of open questions still exist, primarily with regard to which parameters determine the structure of street-canyon flow. The paper presents results from high-resolution wind measurements in a downtown urban street canyon in Oklahoma City, Oklahoma, that were conducted during the Joint Urban 2003 tracer experiment. Data collected with sonic anemometers on two towers installed on opposite sites of the street canyon, each with five different measurement levels, have been analyzed, and the variation of in-canyon flow and turbulence parameters with wind direction and atmospheric stability is discussed. It was found that the street-canyon flow is strongly channeled and its direction is determined by the along-canyon component of the above-roof-level winds. As a consequence, the direction of the street-level winds changes abruptly, and small variations of the upwind direction can significantly alter the in-canyon flow properties. Contrary to results from studies with idealized street canyons, the along-canyon flow components remained significant even for conditions with winds approaching the street at almost perpendicular angles. For such wind directions, a tendency toward development of street-canyon vortices with pronounced vertical motions have been found. However, the complex building geometries at the chosen measurement site enhance the complexity of the flow patterns, and situations with a classic street-canyon vortex rotating in the street could not be identified. In addition, the comprehensive dataset from the Joint Urban 2003 field campaign allowed detailed study of the influence of boundary layer stability on flow in the urban canopy layer. It has become clear that very different conclusions can be drawn with regard to these effects depending on the choice of reference values used in the analysis of the street-canyon data. Using winds from higher elevations in the atmospheric boundary layer (250 m) as reference data, one would conclude that atmospheric stability strongly influences in-canyon flow and, in particular, turbulence. However, only minor stability effects are still seen after normalization with wind speed values at average roof-level height (80 m). This choice allows one to conclude that the in-canyon flow is primarily driven by the boundary layer wind field at average roof level.
The Spanwise Variation of Roof-Level Turbulence in a Street-Canyon Flow
