Use the Fourier transform convention , with inverse factor . The initial profile occupies only the vertical mode . Its transform is
The transformed evolution is a harmonic oscillator for each , with
Therefore inversion gives
Using the nonnegative even frequency instead gives the same cosine. At , with , recovering the specified initial profile and its zero time derivative.
Choose the rightgoing signed dispersion relation . The horizontal phase velocity and group velocity are
The leftgoing branch has their negatives. For a convention using positive frequency at both signs of , multiply each displayed speed by . For every nonzero , on the rightgoing branch; the horizontal group-speed bound for a confined internal wave is , approached as .
Substitute into the transformed equation. It yields
The nonzero eigenfunctions of this Dirichlet boundary value problem are , . Hence, with and ,
The physical vertical velocity includes the factor . The zero horizontal wavenumber is the zero-frequency limiting case, not a propagating oscillation.
For , direct differentiation gives
Thus the transformed partial differential equation and its impermeable-wall boundary conditions are
The positive exponential factor does not change where the vertical velocity vanishes.
Write the density and pressure as and , respectively. The background satisfies hydrostatic balance, , and . Linearizing the Euler equations for an inviscid fluid, incompressibility and advection of density gives
Differentiate the horizontal equation in and use incompressibility to obtain . Differentiate the vertical equation in time and twice in :
Dividing by gives the equation for internal waves in an exponentially stratified incompressible fluid,
The first-derivative term retains the variation of inertial density with height; no Boussinesq approximation has been made.
For the th normal mode, the denominator just obtained is
Its forcing coefficient is nonzero for and . Therefore the periodic response grows like as approaches : this is resonance. At exact resonance there is no bounded periodic particular solution in the linear inviscid model; the resonant amplitude instead contains a secular term proportional to or . Physical damping and nonlinear effects eventually matter. Away from resonance, undamped free oscillations can persist alongside the periodic particular solution.
Put and
This polynomial is a harmonic function: . Also , and . Thus satisfies Laplace's equation with homogeneous Neumann boundary conditions on both side walls and the bottom. Its Fourier cosine series consequently has the form
The surface condition becomes , where
For a nonzero forcing frequency away from the natural frequencies, the coefficients of the oscillating-wall forcing of a rectangular tank are
Here in the definition of . In fact for , since . The zero-mode coefficient ensures that the mean surface elevation is in complex amplitude, exactly the first-order volume conservation adjustment to the moving wall.
The wall position is , so its velocity is . The kinematic boundary condition equates the normal fluid velocity to the normal wall velocity. Linearizing the evaluation point to the mean wall position gives
The other wall and bottom retain and , while the free surface gives .
Apply separation of variables to Laplace's equation. The side-wall Neumann boundary conditions give horizontal factors with . The bottom condition fixes the vertical factor to . Substitution into the surface condition in i gives the normal modes of surface gravity waves in a rectangular tank:
The corresponding surface elevation follows from . The spatially uniform mode has zero frequency; its constant elevation changes total fluid volume, so it is absent when that volume is fixed. An additive constant in changes no fluid velocity.
The velocity potential gives horizontal and vertical velocity components . The kinematic boundary condition at each fixed impermeable wall is zero normal fluid velocity, hence
Let be the surface elevation. To first order in its amplitude, the kinematic boundary condition for a free-surface graph gives at . The dynamic boundary condition for an inviscid interface fixes the surface pressure to atmospheric pressure. The unsteady Bernoulli equation, after choosing the time-dependent gauge of , then gives at : the quadratic velocity term is second order in amplitude. Differentiate this last equation in time and eliminate to obtain the linearized free-surface boundary conditions in combined form,
At the interface, the tangential velocity is and the normal velocity remains zero. The preceding error function expression gives
Hence, within the leading boundary layer model,
Allowing the formally discarded finite-Reynolds number terms gives an additional order- error, smaller than the order- correction under the assumed ordering. The coefficient is .
Thus positive surface shear increases the outward surface speed linearly: for it increases , and for it makes more negative. The stagnation point remains at . A negative coefficient would result from using the printed condition literally, and would describe the different half-line problem identified in part (ii).
Set . The boundary value problem becomes
For , direct substitution gives . This growing polynomial solution is useful for reduction of order.
Write . Substitution into the linear ordinary differential equation cancels the terms proportional to and gives
An integrating factor for this first-order equation for yields
The far-field boundary condition selects the convergent integral from :
Any additional multiple of would grow at . The chosen expression tends to zero: for ,
Here and . Thus the far-field boundary condition is verified without a cancellation of large terms.
At , differentiating gives , so . Using the antiderivative supplied in the paper gives the explicit answer
Here is the error function. Finally,
recovers the unique correction satisfying . The decay bound also shows that approaches a finite constant at , as allowed by the normal velocity matching. This is the weak surface-shear perturbation of stagnation flow.
If the printed condition is instead imposed literally on a positive half-line, reduction of order gives
It satisfies but grows as , so it cannot describe the fluid domain specified in this paper. This makes the consequence of the printed sign error explicit.
At zero surface shear stress, the solution connected to part (a) is
Indeed, it has , and . Insert the regular perturbation into the nonlinear ordinary differential equation. The terms are
The coefficient of therefore gives
Expanding the surface boundary conditions and the far-field matching condition gives
The printed limit is a sign error. It is present in the native PDF on paper page 3, not introduced by transcription. The fluid occupies , the inner variable is , and part (b)(iii) correctly matches at . We retain that physical domain throughout. A change to the positive coordinate would also change the differential equation and surface derivative signs; it cannot be made by changing only the far-field sign.
The inequality ensures that effects discarded at order in the boundary-layer scaling are smaller than the retained order- perturbation. The inequality justifies this regular perturbation. The two conditions control different approximations.
With and , incompressibility is automatic because and . In the streamwise boundary-layer equation,
Equating coefficients of gives
This coefficient comparison determines the equation even though the streamwise equation vanishes identically at the stagnation point .
The no-penetration boundary condition gives , the surface shear stress gives , and matching the tangential velocity to the outer solution gives . Thus
If the correction is integrable at the far end, then there, which is consistent with . The small-shear stress solution below has precisely this property.
In a thin boundary layer, take streamwise velocity and distance to be order one, set , and use incompressibility to set . The streamwise convective acceleration is order one, whereas its dominant viscous contribution after division by the Reynolds number is . Their dominant balance selects
With this choice, the exact rescaled Navier-Stokes equations are
Thus the leading pressure is independent of . Matching it to the outer solution from part (a), evaluated near , gives , hence . The leading boundary-layer equation is therefore
The exact surface shear stress condition rescales to . Since for all , its tangential derivative also vanishes there. The leading surface boundary conditions are
Matching to the outer solution requires
The second matching condition specifies the leading growth of , allowing a constant displacement correction. Requiring would impose an extra condition not supplied by the leading outer solution. This is the stress-driven stagnation-point boundary layer.
The held-flat interface imposes the no-penetration boundary condition, while its specified tangential traction gives a shear stress condition:
A differentiable boundary value implies , so the second boundary condition is also . There is no imposed no-slip boundary condition at this interface; tangential velocity is determined by the shear stress and the fluid equations.
The solution in part (a) has zero tangential traction, and so remains valid for the whole interface if and only if . At the single point the imposed shear stress vanishes for any , but this does not satisfy a boundary condition that must hold for every .
The volumetric flow rate is the integral of axial velocity over the cross-section:
The plane Poiseuille flow contribution is
For each correction eigenfunction,
Uniform convergence permits termwise integration. Multiplying these two integrals by gives
Consequently
Every summand in the subtracted quantity is positive. Thus the side walls reduce the flux below the infinite-width plane Poiseuille flow prediction. Its dimensions are volume per unit time, as required for a volumetric flow rate.
Subtract the particular plane Poiseuille flow profile from the full rectangular duct flow. The remainder is a harmonic function:
The data are even in and . The boundary value problem is unique: if two solutions differed by , Green's first identity and the zero Dirichlet boundary condition would give
Thus is constant, and its zero boundary values force . Reflection of a solution in either coordinate axis therefore gives the same solution. We may seek even separated eigenfunctions.
Use separation of variables with
Each summand solves the Laplace equation, vanishes at , and has the required evenness. The hyperbolic cosine ratio equals one at either side wall. The functions have orthogonality relation
They are the even Dirichlet eigenfunctions on , so their Fourier series represents the even boundary data. Projecting that data onto the th eigenfunction yields the Fourier coefficient
For completeness, two integrations by parts give the quadratic arch expansion in half-integer cosine modes: the displayed integral is , agreeing with the identity supplied in the paper.
Adding the particular solution gives
The coefficients are absolutely summable and the hyperbolic cosine ratio is at most one throughout the closed rectangle, giving uniform convergence and the prescribed boundary values. On compact subsets of the interior, the hyperbolic cosine ratio supplies exponential decay in , justifying differentiation of the series and the Laplace equation. The side-wall correction slows the plane Poiseuille flow enough to satisfy all four no-slip boundary conditions.
For a profile depending only on , the Poisson equation becomes . Integrating twice gives
The no-slip boundary conditions imply and . Thus
This is plane Poiseuille flow between the two horizontal walls. It is positive in the interior for , so the velocity points down the pressure gradient. It does not satisfy the side-wall no-slip boundary conditions, which is precisely what the correction in part (iii) repairs.
For the unidirectional flow , incompressibility gives . Hence , and the convective acceleration vanishes. The transverse components of the Navier-Stokes equation give , while the axial component gives
A steady-flow assumption is implicit in the question. Neither unidirectional flow nor a time-independent pressure gradient alone eliminates : a decaying transient can be superposed on the final pipe profile. The native PDF does not explicitly say steady flow in this part. For the intended steady flow, , and therefore
Taking the pipe walls to be stationary, the no-slip boundary condition is
The no-penetration boundary condition is already satisfied by the assumed velocity field. This is the Poisson equation defining rectangular duct flow.

There are unlisted articles, also show them or only show them.