Unsteady convective exchange flows in cavities

Sturman, J.; Ivey, Gregory

doi:10.1017/s002211209800175x

Cited by 21 publications

(29 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This buoyancy-driven circulation embraces the entire basin and gradually weakens when the water temperature approaches the value T m . This feature agrees very well with previous experiments on horizontal convective exchange-flows in similar geometries: the basin-wide circulation and the dependency of its intensity on the buoyancy flux through the surface were also observed in experiments in both long basins with a differentially heated horizontal bottom [13,19,25] and in basins with an inclined bottom and a uniform heat flux through the surface [6,7,29].…”

Section: Experimental Observationssupporting

confidence: 91%

On the fine structure of the thermal bar front

2011

View full text Add to dashboard Cite

The thermal bar-a hydrodynamic phenomenon, arising in natural basins due to successive changes of the water temperature across the temperature of maximum density (T m , which is close to 4 • C)-has been studied in laboratory experiments and by numerical simulations. The experiments were performed in a rectangular tank with an inclined bottom, filled with water with initial temperature T 0 < T m and then heated at the surface. During the heating a basin-wide circulation develops, consisting of down-slope cascades in regions where T < T m , a subsurface off-shore jet in the region where T > T m , and a compensating flow at intermediate depths towards the shallow part of the tank, supplying both off-shore flows with waters from deeper regions. Analysis of the water temperature and density fields as well as the currents has revealed that the location of the convergence zone of the surface current (when formed) does not coincide with that of the Tm-isotherm. The thermal bar front is typically understood as a convergence zone near the 4 • C-isotherm, formed due to the effect of cabbeling. Our experiments demonstrate, however, that the front is associated with the leading edge of the subsurface current. The increasing distance between the 4 • Cisotherm and the subsurface jet has been recorded in the laboratory experiments. Numerical simulation results corroborate the laboratory experiments. A scaling analysis predicts the speed of propagation of this frontal zone to be U ∼ [g × ρ/ρ ×H] 1/2 , where H is the depth (increasing with time) of the upper thermo-active layer, ρ 0 a reference density, and ρ is the characteristic horizontal density difference across the front. A combined analysis of laboratory, field and numerical data has corroborated this law. N. Demchenko (B) · I. Chubarenko P.P.Shirshov Institute of Oceanology RAS, Atlantic Branch, Prospect Mira, 1, Kaliningrad,

show abstract

Section: Experimental Observationssupporting

confidence: 91%

On the fine structure of the thermal bar front

2011

View full text Add to dashboard Cite

show abstract

“…The particle image velocimetry (PIV) method, developed by Stevens and Coates (1994), was used to measure velocities as described in Sturman and Ivey (1998). Temperatures within the nonslope region of the tank were measured by two fastresponse thermistors placed 10 and 410 mm from the end of the slope.…”

Section: Methodsmentioning

confidence: 99%

“…Convective flushing timescales in wetlands have previously been quantified by Sturman et al (1999), who adapted scaling arguments from convective dynamics on geophysical scales (Sturman and Ivey 1998). However, the work of Sturman et al (1999) did not take into account the effect of emergent vegetation on the convective flushing characteristics.…”

Section: Acknowledgmentsmentioning

confidence: 99%

The effect of emergent vegetation on convective flushing in shallow wetlands: Scaling and experiments

Oldham

Sturman

2001

Limnology & Oceanography

Self Cite

View full text Add to dashboard Cite

Many wetlands around the world are characterized by shallow water, dense vegetation in the littoral zones, no significant riverine inflow and minimal circulation. Recent research on the hydrodynamics of such wetlands has identified convective circulation as being important for flushing of the littoral zones. To quantify this process, a parameterization of the convective discharge per unit width, which had been previously developed for nonvegetated systems, was extended to include a drag coefficient dependent on Reynolds number and vegetation density. The drag coefficient also included the effect of anisotropic permeability of the vegetation. The effects of relatively dense emergent vegetation (ϳ17% by volume) on convective flushing of shallow wetlands with low-Reynolds number (ϳ100) flow was then investigated using experiments in a laboratory convection tank (0.5 by 2 by 0.1 m) and in a wetland mesocosm (5 by 15 by 1 m). Bottom convective currents of ϳ1-10 mm s Ϫ1 were measured in both the laboratory and the mesocosm. These currents resulted in the shallow, vegetated regions of the mesocosm being flushed in 4 h. The discharge per unit width (m 2 s Ϫ1 ) predicted by the developed parameterization compared favorably (R 2 ϭ 0.7) with the discharge per unit width measured in both the laboratory and the mesocosm. The short timescales of convective flushing, even in the presence of reasonably dense vegetation, indicate the likely significance of this mechanism in sheltered wetlands.Convective circulation occurs in aquatic systems when shallow waters heat or cool more rapidly than deeper waters, causing horizontal gradients in water temperature and density. The horizontal density differences drive convective currents, which allow increased flushing of the littoral regions. In wetlands, these littoral zones are typically characterized by dense emergent vegetation, yet the effect of vegetation on convective circulation has rarely been addressed. This paper develops parameterizations to describe that effect, then presents some preliminary laboratory and field data to test the parameterization.Convective currents have been frequently observed in deep lakes (e.g., Monismith et al. 1990;Nepf and Oldham 1997), and they have also been measured in shallow wetlands. Arnold and Oldham (1997) measured the distinctive signals of cool convective currents inserting along the bottom boundary of a shallow (2 m deep) wetland (Fig. 1) for Ͼ30% of the year. These bottom currents are typically 0 (1-10) mm s Ϫ1 (Sturman et al. 1999). Although this is low compared to currents measured in lakes dominated by river in-1

show abstract

“…Such flows are driven by spatial and temporal variation of buoyancy fluxes through the water surface and are affected by sloping bottom topography. We deal mainly with the steady state condition, although unsteady forcing can be an important feature of the field situation (e.g., Farrow and Patterson, 1993;Horsch et al, 1994;Sturman and Ivey, 1998).…”

Section: Introductionmentioning

confidence: 99%

Steady convective exchange flows down slopes

Sturman¹,

Oldham²,

Ivey³

1999

Aquat. sci.

View full text Add to dashboard Cite

Horizontal exchange flows driven by destabilising buoyancy fluxes through the surface waters of lakes and coastal regions of oceans are important in understanding the transport of nutrients, micro-organisms and pollutants from littoral to pelagic zones. Our interest here is in the discharge flow driven by cooling or destabilising forcing at the water surface in a water body with variable depth due to sloping bottom topography. Flow visualisation studies and measurements in a laboratory model enabled us to develop scaling arguments to predict the dependency of discharge upon surface forcing and the angle of bottom slope. The results were used to interpret both the laboratory measurements and field data from a small shallow lake with sloping sides and an essentially flat bottomed interior, as well as published results from the literature. The steady state horizontal exchange can be described by Q = 0.24 B 1/3 (l tan q/(1 + tan q)) 4/3 , where Q is the discharge rate per unit length of shoreline, q is the angle of the bottom slope, B is the surface buoyancy flux and l is the horizontal length of the forcing region over the slope. The flushing timescale of the wedge shaped littoral region was given by t f~l 2/3 (1 + tan q) 4/3 /(B tan q) 1/3 . While the buoyancy flux in the field is almost never constant in space or time and the slope from the shore is seldom uniform, we found that the exchange rate was relatively insensitive to buoyancy flux changes and only moderately sensitive to slope.

show abstract

Unsteady convective exchange flows in cavities

Cited by 21 publications

References 15 publications

On the fine structure of the thermal bar front

On the fine structure of the thermal bar front

The effect of emergent vegetation on convective flushing in shallow wetlands: Scaling and experiments

Steady convective exchange flows down slopes

Contact Info

Product

Resources

About