Shallow Water Equations

Scaling Navier-Stokes Equations

The scaling of the variables is as follows:

There is a condition where \(\frac{W}{D}\) can not be larger than \(\mathcal{O}(\frac{U}{L})\) which implies that:

\begin{equation} W \leq \mathcal{O}(\delta U) \label{eq: W and delta and U} \end{equation}

where \(\delta = \frac{D}{L}\), which is a small parameter.

The momentum equations that are proved in Navier-Stokes(21) are altered to match the scenario of our problem, which is in the presence of Coriolis force:

\begin{equation} \boxed{\begin{aligned} &\frac{\partial u}{\partial t} + u \frac{\partial u}{\partial x} + v \frac{\partial u}{\partial y} + w\frac{\partial u}{\partial z} - fv = -\frac{1}{\rho}\frac{\partial \tilde{p}}{\partial x} \\ & \frac{\partial v}{\partial t} + u \frac{\partial v}{\partial x} + v \frac{\partial v}{\partial y} + w\frac{\partial v}{\partial z} +fu = -\frac{1}{\rho}\frac{\partial \tilde{p}}{\partial y} \\ &\frac{\partial w}{\partial t} + u \frac{\partial w}{\partial x} + v \frac{\partial w}{\partial y} + w\frac{\partial w}{\partial z} = -\frac{1}{\rho}\frac{\partial \tilde{p}}{\partial z} \end{aligned}} \rightarrow \boxed{\begin{aligned} & \frac{U^2}{L} \quad \frac{U^2}{L} \quad \frac{U^2}{L} \quad \frac{UW}{D} \quad fU \quad \frac{P}{\rho L} \\ & \frac{U^2}{L} \quad \frac{U^2}{L} \quad \frac{U^2}{L} \quad \frac{UW}{D} \quad fU \quad \frac{P}{\rho L} \\ & \frac{UW}{L} \quad \frac{UW}{L} \quad \frac{UW}{L} \quad \frac{W^2}{D} \quad \frac{P}{\rho D} \end{aligned} } \end{equation}

The total pressure is defined by:

\begin{equation} p(x,y,z,t) = -pgz + \tilde{p}(x,y,z,t) \end{equation}

where the first term is the hydrostatic pressure and cancels out with gravitational force, and the second term is the dynamic pressure.

Using (1) to prove the orders of the partial derivatives of the velocities:

\begin{equation} \begin{aligned} &W \leq \mathcal{O}(\delta U) \\ &W \leq \frac{D}{L}U\\ &WU \leq \frac{DU^2}{L}\\ &\frac{WU}{D} \leq \frac{U^2}{L} \end{aligned} \end{equation}

Resulting in the following:

\begin{equation} \frac{UW}{D} = \mathcal{O}(\frac{U^2}{L}) \end{equation}

So the Pressure scale: \(P = \rho U (U,fL)_{max}\). This is so for the horizontal pressure gradient to enter as a forcing term in the horizontal momentum balance; otherwise the flow would be unaccelerated.

Rossby Number Condition

Scaling p

As mentioned the Rossby number is the ratio of the relative to the coriolis accelerration. By the assumption that \(R_o << 1\). Meaning that the characteristic pressure field:\(P = \rho \frac{U^2}{R_o}\).

Proof:

\[ P = \rho U (U,fL)_{max} = \rho U f L = \rho \frac{U^2}{R_o} \]

The Shallow-Water model condition: hydrostatic approximation

The z-component of the momentum equation:

\[\frac{\partial w}{\partial t}+u \frac{\partial w}{\partial x} + v \frac{\partial w}{\partial y} + w\frac{\partial w}{\partial z} = -\frac{1}{\rho} \frac{\partial \tilde{p}}{\partial z}\]

Now let us examine the following term (interms of the pressure gradient) from (5), (6), and (8):

\begin{equation} \begin{aligned} \rho \frac{\frac{\partial w}{\partial t}}{\frac{\partial \tilde{p}}{\partial z}} &= \mathcal{O}(\rho \frac{\frac{UW}{L}}{\frac{P}{D}})\\ &= \mathcal{O}(\rho \frac{\frac{UW}{L}}{\frac{\rho U^2}{R_o D}})\\ &= \mathcal{O}(\frac{R_o DW}{UL})\\ &= \mathcal{O}(\delta^2 R_o) \quad \text{$\delta = \frac{D}{L}= \frac{W}{U}$ which is the ratio of the vertical to horizontal} \end{aligned} \end{equation}

Hence we can conclude that that \(\frac{\partial \tilde{p}}{\partial z}\) is negligible to \(\mathcal{O}(\delta^2 R_o)\) or in terms of the total pressure:

\begin{equation} \frac{\partial p}{\partial z} = -pg + \mathcal{O}(\delta^2 R_o) \end{equation}

which is the hydrostatic approximation or the shallow-water model.

Horizontal Momentum Equations

Integrating (10) with respect to z:

\begin{equation} p = \rho g(h-z) +p_o \end{equation}

The horizontal pressure gradient is independent of z:

\begin{equation} \begin{aligned} &\frac{\partial p}{\partial x} = \rho g \frac{\partial h}{\partial x}\\ & \frac{\partial p }{\partial y} = \rho g \frac{\partial h}{\partial y} \end{aligned} \end{equation}

So the horizontal acceleration must be independent of z \(\implies\) it is consistent to consider that the horizontal velocities themselves remain z-independent if they are so initially.

Plug in the horizontal pressure gradient in (2) , to get the Horizontal momentum equations:

\begin{equation} \begin{aligned} &\frac{\partial u}{\partial t} + u \frac{\partial u}{\partial x}+v \frac{\partial u}{\partial y}- fv = -g \frac{\partial h}{\partial x} \\ & \frac{\partial v}{\partial t} + u \frac{\partial v}{\partial x}+v \frac{\partial v}{\partial y}+fu= -g \frac{\partial h}{\partial y} \end{aligned} \end{equation}

Continuity Equation

We can integrate the incompressibility equation since the horizontal velocities are independent of z:

\begin{equation} \begin{aligned} & \int \frac{\partial w}{\partial z} dz = \int (- \frac{\partial u}{\partial x} - \frac{\partial v}{\partial y }) dz \\ & \boxed{w(x,y,z,t) = - z (\frac{\partial u}{\partial x} + \frac{\partial v}{\partial y}) +\tilde{w}(x,y,t)} \end{aligned} \end{equation}

Applying boundary conditions at \(z = h_B\):

\begin{equation} w(z=h_B) = \frac{D h_B}{Dt} = u \frac{\partial h_B}{\partial x} + v \frac{\partial h_B}{\partial y} \label{eq: h_B} \end{equation}

Then we can write \(\tilde{w}(x,y,t)\):

\begin{equation} \tilde{w}(x,y,t) = u \frac{\partial h_B}{\partial x} + v \frac{\partial h_B}{\partial y} + h_B(\frac{\partial u}{\partial x} + \frac{\partial v}{\partial y}) \end{equation}

We can write \(w(x,y,z,t)\):

\begin{equation} w(x,y,z,t) = (h_B -z) (\frac{\partial u}{\partial x}+\frac{\partial v}{\partial y}) + u \frac{\partial h_B}{\partial x} + v \frac{\partial h_B}{\partial y} \end{equation}

Note: The corresponding kinematic condition at \(z = h\):

\begin{equation} w|_{z=h} = \frac{Dh}{Dt} = \frac{\partial h}{\partial t}+ u \frac{\partial h}{\partial x}+v \frac{\partial h}{\partial y} \end{equation}

Subtracting (17) from (18), knowing that \(H = h-h_B\):

\begin{equation} \begin{aligned} \frac{\partial h}{\partial t} +u \frac{\partial h}{\partial x} + v \frac{\partial h}{\partial y} - [(h_B -z) (\frac{\partial u}{\partial x}+\frac{\partial v}{\partial y}) + u \frac{\partial h_B}{\partial x} + v \frac{\partial h_B}{\partial y}]&= \frac{\partial h}{\partial t} + u( \frac{\partial h}{\partial x} - \frac{\partial h_B}{\partial x}) + v( \frac{\partial h}{\partial y} - \frac{\partial h_B}{\partial y})\\& - (h_B -z) (\frac{\partial u}{\partial x} + \frac{\partial v}{\partial y}) \\ &= \frac{\partial H}{\partial t} + \frac{\partial (uH)}{\partial x} + \frac{\partial(vH)}{\partial y} =0 \\ &=\frac{D H}{Dt }+H(\frac{\partial u}{\partial x}+ \frac{\partial v}{\partial y}) = 0 \end{aligned} \end{equation}

Analysis of the continuity equation (19): Notice that if the local horizontal divergence of volume is positive, it must be balanced by decrease of the layer thickness, due to the drop in the free surface.

Hence as the cross section of A of a fluid column increases at a rate:

\begin{equation} \frac{1}{A}\frac{DA}{Dt} = \frac{\partial u}{\partial x} + \frac{\partial v}{\partial y} \end{equation}

The total thickness must decrease so that the volume \(HA\) remains constant, as following:

\begin{equation} \frac{1}{H} \frac{D H}{D t} + \frac{1}{A} \frac{DA}{Dt} =0 \end{equation}

We are done deriving the Shallow-Water equations. They are the following:

\begin{equation} \boxed{\begin{aligned} &\frac{\partial u}{\partial t} + u \frac{\partial u}{\partial x}+v \frac{\partial u}{\partial y}- fv = -g \frac{\partial h}{\partial x} \\ & \frac{\partial v}{\partial t} + u \frac{\partial v}{\partial x}+v \frac{\partial v}{\partial y}+fu= -g \frac{\partial h}{\partial y} \\ & \frac{\partial H}{ \partial t} + \frac{ \partial (uH)}{\partial x} + \frac{\partial (vH)}{\partial y} =0 \text{ or } \frac{D H}{Dt} + H(\frac{\partial u}{\partial x} + \frac{\partial v}{\partial y}) = 0 \end{aligned}} \end{equation}

Reflection

\begin{equation} \begin{aligned} w(x,y,z,t) &= \frac{Dz}{Dt} \\ &= \frac{(h_B - z)}{H}[\frac{-DH}{Dt}] + u \frac{\partial h_B}{\partial x} +v \frac{\partial h_B}{\partial y} \\ &= \frac{(z - h_B )}{H}[\frac{DH}{Dt}] + u \frac{\partial h_B}{\partial x} +v \frac{\partial h_B}{\partial y} \end{aligned} \end{equation}

which can be written as:

\begin{equation} \frac{D}{Dt} [\frac{z-h_B}{H}] =0 \end{equation}

This tells us that during stretching or contraction of each column the relative position of each column, the relative position of the fluid is unchanged.