Hostname: page-component-7b7567c8c-fvjsw Total loading time: 0 Render date: 2026-11-16T07:10:11.278Z Has data issue: false hasContentIssue false

The propagation of gravity currents in a circular cross-section channel: experiments and theory

Published online by Cambridge University Press:  09 January 2015

S. Longo*
Affiliation:
Dipartimento di Ingegneria Civile, Ambiente Territorio e Architettura (DICATeA), Università di Parma, Parco Area delle Scienze, 181/A, 43124 Parma, Italy
M. Ungarish
Affiliation:
Department of Computer Science, Technion, Israel Institute of Technology, Haifa 32000, Israel
V. Di Federico
Affiliation:
Dipartimento di Ingegneria Civile, Chimica, Ambientale e dei Materiali (DICAM), Università di Bologna, Viale Risorgimento, 2, 40136 Bologna, Italy
L. Chiapponi
Affiliation:
Dipartimento di Ingegneria Civile, Ambiente Territorio e Architettura (DICATeA), Università di Parma, Parco Area delle Scienze, 181/A, 43124 Parma, Italy
A. Maranzoni
Affiliation:
Dipartimento di Ingegneria Civile, Ambiente Territorio e Architettura (DICATeA), Università di Parma, Parco Area delle Scienze, 181/A, 43124 Parma, Italy
*
†Email address for correspondence: sandro.longo@unipr.it
Rights & Permissions [Opens in a new window]

Abstract

High-Reynolds number gravity currents (GC) in a horizontal channel with circular/semicircular side walls are investigated by comparing experimental data and shallow-water (SW) theoretical results. We focus attention on a Boussinesq system (salt water in fresh water): the denser fluid, occupying part of the depth or the full depth of the ambient fluid which fills the remaining part of the channel, is initially at rest in a lock separated by a gate from the downstream channel. Upon the rapid removal of the gate (‘dam break’), the denser ‘current’ begins propagating into the downstream channel, while a significant adjustment motion propagates upstream in the lock as a bore or rarefaction wave. Using an experimental channel provided by a tube of 19 cm diameter and up to 615 cm length, which could be filled to various levels, we investigated both full-depth and part-depth releases, considered the various stages of inertial-buoyancy propagation (in particular, the initial ‘slumping’ with constant speed, and the transition to the late self-similar propagation with time to the power $3/4$ ), and detected the transition to the viscous-buoyancy regime. A first series of tests is focused on the motion in the lock while a second series of tests is focused on the evolution of the downstream current. The speed of propagation of the current in the slumping stage is overpredicted by the theory, by about the same amount (typically 15 %) as observed in the classical flat bottom case. The length of transition to viscous regime turns out to be ${\sim}[\mathit{Re}_{0}(h_{0}/x_{0})]^{{\it\alpha}}$ ( $\mathit{Re}_{0}=(g^{\prime }h_{0})^{1/2}h_{0}/{\it\nu}_{c}$ is the initial Reynolds number, $g^{\prime }$ is the reduced gravity, ${\it\nu}_{c}$ is the kinematic viscosity of the denser fluid, $h_{0}$ and $x_{0}$ are the height of the denser current and the length of the lock, respectively), with the theoretical ${\it\alpha}=3/8$ and experimental ${\it\alpha}\approx 0.27$ .

Information

Type
Papers
Copyright
© 2015 Cambridge University Press 
Figure 0

Figure 1.A schematic description of the lock release problem and a sketch of circle-segment cross-section for three values of the parameter ${\it\beta}$. The point $C$ is the centre of the circle, the cross area of the channel is $A_{T}=A+A_{a}$.

Figure 1

Figure 2.Schematic description of the types of flow fields. (a) Type 1: rarefaction wave into the reservoir (position $L$) and subcritical GC front (position $N$). (b) Type 2: moving jump (bore) at position $L$, subcritical GC front. (c) Type 3: moving jump (bore) at position $L$, critical GC front. (d) Type 4: same as type 3, with additional stationary jump at position $x=0$. The type 4 flow is a rare occurrence, for ${\it\beta}$ very close to 2 and $H$ very close to 1 only. Here $z$ is scaled with $h_{0},x$ with $x_{0}$. Point $L$ denotes the transition from the reservoir to the activated (moving) domain, $N$ the GC front on the opposite boundary of the activated domain; $M$ denotes the left boundary of the core, and $Q$ the left boundary of the expansion domain following the GC front (when present).

Figure 2

Figure 3.Schematic description of the experimental apparatus.

Figure 3

Figure 4.Pictures of the motion (left to right) in the lock (a) and downstream (b). Experiment 2, ${\it\beta}=1$, $H=1.1$, $({\it\rho}_{c}-{\it\rho}_{a})/{\it\rho}_{c}=4.1\,\%$, $\mathit{Re}_{0}=13.8\times 10^{3}$, type 3 profile (see figure 2).

Figure 4

Figure 5.The shape of the left-moving profiles for different values of $H$ and for $1-R=7.9\,\%$. The experiment with $H=1.0$ refers to ${\it\beta}=2$ (Experiment 34) whereas the other three experiments refer to ${\it\beta}=1.5$ (Experiments 16, 18 and 20). The type is in figure 2.

Figure 5

Table 1.Parameters of the experiments in the long lock configuration. The internal radius of the cross-section is $r^{\ast }=9.5~\text{cm}$, the length of the lock is $x_{0}=100~\text{cm}$. Here $1-R=({\it\rho}_{c}-{\it\rho}_{a})/{\it\rho}_{c},g^{\prime }=(1-R)g$ is the reduced gravity, $\mathit{Re}_{0}=Uh_{0}/{\it\nu}_{c}$ is the Reynolds number with ${\it\nu}_{c}$ the kinematic viscosity of the denser fluid, $U=\sqrt{g^{\prime }h_{0}}$ and $T=x_{0}/U$ are the velocity and the time scale, respectively. The type refers to the classification sketched in figure 2. The superscript $a$ in the first column indicates that a video is available as supplementary material.

Figure 6

Figure 6.The shape of the right-moving reflected bore/wave. Caption as in figure 5.

Figure 7

Figure 7.Detection of the interface profiles in the lock with a time step of 0.5 s. Experiment 3 with ${\it\beta}=1$, $H=1.1$, $1-R=10.6\,\%$, $\mathit{Re}_{0}=18.4\times 10^{3}$.

Figure 8

Figure 8.Velocity $V_{f}$ (a) and height $h_{\ast }$ (b) of the bore or rarefaction wave in the lock as a function of $H$ for ${\it\beta}=1,\,1.5,\,2.0$. For the rarefaction wave, $h_{\ast }=0$. Symbols and lines represent experimental results and theoretical predictions, respectively. The type of profile is indicated as t2–t4. The error bars indicate one standard deviation.

Figure 9

Figure 9.The relative error in predicting the speed of the left-moving bore or rarefaction wave as a function of the initial Reynolds number of the current $\mathit{Re}_{0}$ (a) and of the non-dimensional group $\mathit{Re}_{0}(h_{0}/x_{0})$ (b).

Figure 10

Figure 10.Speed $V_{b}$ (a) and height $h_{\ast \ast }$ (b) of the reflected bore/wave in the lock as a function of $H$ for ${\it\beta}=1,\,1.5,\,2.0$. For the reflected wave, $h_{\ast \ast }=0$. Symbols and lines represent experimental results and theoretical predictions, respectively. The error bars indicate one standard deviation.

Figure 11

Figure 11.The speed of the bore (empty circles) and of the reflected bore (filled squares) as functions of time for Experiment 4 (a) and Experiment 18 (b). The dashed lines are the average values.

Figure 12

Table 2.Parameters for all experiments in the short lock configuration. The internal radius of the circular cross-section is $r^{\ast }=9.5~\text{cm},x_{0}$ is the length of the lock, $1-R=({\it\rho}_{c}-{\it\rho}_{a})/{\it\rho}_{c}$, $g^{\prime }=(1-R)g$ is the reduced gravity, $\mathit{Re}_{0}=Uh_{0}/{\it\nu}_{c}$ is the Reynolds number with ${\it\nu}_{c}$ the kinematic viscosity of the denser fluid, $U=\sqrt{g^{\prime }h_{0}}$ and $T=x_{0}/U$ are the velocity and the time scale, respectively.

Figure 13

Figure 12.Photographs showing profiles of the head of the GC in Experiment 39, $\mathit{Re}_{0}=21.3\times 10^{3},x_{0}=100$ cm, $x_{0}/h_{0}=10.5$. The vertical dashed lines are 20 cm apart and the times since release are 1, 2, 3 and 4 s.

Figure 14

Figure 13.Photographs showing profiles of the head of the GC in Experiment 29, $\mathit{Re}_{0}=12.9\times 10^{3}$, $x_{0}=6~\text{cm}$, $x_{0}/h_{0}=0.8$. The vertical dashed lines are 20 cm apart and the times since release are 1, 2, 3 and 4 s.

Figure 15

Figure 14.Photographs showing the bottom view of the head of the GC in Experiment 15 (see the online supplementary movie), $\mathit{Re}_{0}=4.1\times 10^{3}$, $x_{0}=100$  cm. The dashed curves are 20 cm apart and the times since release are 1.6, 3.2, 4.8 and 6.4 s.

Figure 16

Figure 15.The distance of propagation of the currents measured from the gate. The solid line represents the slumping phase with $x_{N}\propto t$, the dotted line is the inertial-buoyancy phase with $x_{N}\propto t^{3/4}$ and the dashed line is the viscous-buoyancy phase with $x_{N}\propto t^{1/4}$.

Figure 17

Figure 16.The experimental speed of the GC front versus distance of propagation.

Figure 18

Figure 17.The experimental speed of the GC front during the slumping phase. The filled symbols refer to experiments with long lock ($x_{0}=100~\text{cm}$), the open symbols refer to experiments with short lock ($x_{0}=6{-}15~\text{cm}$). The three curves are the theoretical solution. The error bars indicate one standard deviation.

Figure 19

Figure 18.The experimental exponent ${\it\gamma}$ (a) and coefficient $K$ (b) for an advancement of the GC front of the form $x_{N}=Kt^{{\it\gamma}}$. In (a) the dash-dot line corresponds to ${\it\gamma}=1$, typical of the slumping stage, the dotted line corresponds to ${\it\gamma}=3/4$, typical of the propagation in the self-similar inertial-buoyancy regime, while the dashed line corresponds to ${\it\gamma}=1/4$, typical of the propagation in the buoyancy-viscous regime. Experiments with short lock, $x_{0}=6{-}15~\text{cm}$.

Figure 20

Figure 19.Transition to viscous regime as a function of $\mathit{Re}_{0}(h_{0}/x_{0})$. The bold line represents the theoretical estimate $x_{V}=0.5[\mathit{Re}_{0}(h_{0}/x_{0})]^{3/8}$, the dashed lines are the best-fit curves of equation $x_{V}=0.43[\mathit{Re}_{0}(h_{0}/x_{0})]^{0.47}$, $x_{V}=1.20[\mathit{Re}_{0}(h_{0}/x_{0})]^{0.28}$ and $x_{V}=1.20[\mathit{Re}_{0}(h_{0}/x_{0})]^{0.27}$ calculated assuming an exponent ${\it\gamma}=0.5$ (circles), ${\it\gamma}=0.7$ (squares) and ${\it\gamma}=0.75$ (crosses), respectively.

Supplementary material: PDF

Longo et al. supplementary material

Supplementary material and figs

Download Longo et al. supplementary material(PDF)
PDF 361.8 KB

Longo et al. supplementary movie file

Flow in the lock and downstream, Re=4100

Download Longo et al. supplementary movie file(Video)
Video 19.2 MB

Longo et al. supplementary movie file

Flow in the lock and downstream, test Re=20500

Download Longo et al. supplementary movie file(Video)
Video 27.1 MB