Some fluid modeling studies of flow and dispersion over two-dimensional low hills

A wind tunnel study of turbulent flow over a two-dimensional ridge

“…Although two-dimensional laminar separation of thin boundary layers is well understood inasmuch as triple deck theory (Meyer, 1983), an asymptotic theory of comparable rigour to boundarylayer theory, can be combined with boundary-layer theory to predict the position of separation and the subsequent development of the wake, three-dimensional laminar separation is much less predictable and no theory of rough-wall turbulent separation exists at all (Smith, 1986). We mentioned earlier that the intermittent separation bubble observed in this experiment is consistent with other measurements in both the field and tunnel (Finnigan, 1988), while the persistence of the mean flow deficit we observe, as well as the increased turbulence in the wake, accord well with data obtained behind both steeper and rougher ridges by Britter et al (1981), Castro and Snyder (1982) and Arya et al (1987). They are also qualitatively very similar to data obtained for flow normal to the long axis of a model of the roughly elliptical Asker-vein Hill by Bowen and Teunissen (1986), although their model was aerodynamically smooth.…”

“…They are also qualitatively very similar to data obtained for flow normal to the long axis of a model of the roughly elliptical Asker-vein Hill by Bowen and Teunissen (1986), although their model was aerodynamically smooth. Since these examples of wake flow range from wakes following steady separation (Arya et al, 1987;Castro and Snyder, 1982) to those developing without separation (Arya et al, 1987;Bowen and Teunissen, 1986), we can draw the encouraging conclusion that wake parameters not far behind the hill are, in principle, predictable. The wake behind a hill is treated at length by Taylor (1988).…”

“…For example, on a wind-tunnel model hill of similar profile to the present one but with a maximum slope angle of 16" and covered with a model plant canopy that produced a value h/z0 = 50, a steady separation bubble extended about 2h downwind of the crest (unpublished work by J. J. Finnigan, M. R. Raupach, and Y. Brunet). In contrast, the wind-tunnel model of Arya et al (1987) with maximum slope 16" and h/z0 = I20 displayed only intermittent separation, as did the model of Britter et al (1981) with a slope of 12.5" and h/z0 = 500…”

“…This pattern of near-surface response has been observed whenever measurements can be made in the region of increased shear on the hill top and decreased shear upwind, i.e., on Asker-vein and Cooper's Ridge, on Blasheval (Mason and King, 1985) and in the wind-tunnel simulations of Kettle's Hill (Teunissen et al, 1982)) of Asker-vein ) on the rough two-dimensional triangular ridge studied by Snyder and Britter (1987), and on the three rough curved ridges tested by Arya et al (1987). But we emphasize that it is a region inaccessible to wind-tunnel models which are aerodynamically fully rough.…”

A Wind Tunnel Study to Design Large, Open-top Chambers for Whole-tree Pollutant Exposure Experiments

“…24 Turbulence is enhanced in a cavity wake zone and entrains tracer material from the free-stream wind field, thereby increasing the concentration of a tracer in the cavity zone. 37 Wind tunnel studies by Davis and Rogers 20 show that a 45-degree frustum slightly reduces the entrainment of external air into a model chamber. Fluid dynamic studies on an aerofoil in a wind tunnel show that flow separation occurs when the angle of incidence is about 15 degrees.…”