Why a mid-depth stress-free boundary condition is incorrect for Ekman flows

Christian Puntini; Luigi Roberti

REVIEW 1 major objections 3 minor 24 references

Why a mid-depth stress-free boundary condition is incorrect for Ekman flows

T0 review · 1 major / 3 minor · reviewed 2026-08-03 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper proves that imposing a stress-free boundary condition at the base of the Ekman layer forces the modeled current to intensify beneath that depth, a 'reverse Ekman spiral' that contradicts the assumption that wind-driven flow is ne

desk verdict A clean, correct theorem about the Ekman ODE, but the 'unphysical' label depends on an unstated assumption that the same ODE holds below the chosen depth. read the letter →

arxiv 2607.29261 v1 pith:2XSCPXPG submitted 2026-07-31 physics.flu-dyn math-phmath.APmath.MPphysics.ao-ph

Christian Puntini , Luigi Roberti This is my paper · ORCID

classification physics.flu-dynmath-phmath.APmath.MPphysics.ao-ph

keywords Ekmanlayerstress-freeboundaryconditiontransportwind-drivencurrentseddyviscosityWKBapproximationreversespiraloceanmodelling

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper targets a boundary condition commonly used in ocean models: setting the vertical shear of the horizontal velocity to zero at an intermediate depth -D, taken as the base of the wind-driven Ekman layer. The authors prove that any solution of the f-plane Ekman equations with this condition must have its speed minimum exactly at -D, with the current strengthening again below -D (a 'reverse Ekman spiral') and the wind-current angle decreasing again. This contradicts the very premise of the condition, namely that wind-driven flow is negligible below -D. They then show, via a WKB estimate for slowly varying deep density, that the physically motivated no-slip condition at the true seafloor gives Ekman transport nearly perpendicular to the wind when the ocean is deep, with an exponentially small error.

What carries the argument

The central mechanism is an integral identity obtained by multiplying the complex Ekman equation by U and integrating from -D to z: this yields m(z)|U|'(z)|U|(z) = ∫ m|U'|² and m(z)φ'(z)|U|² = f∫ρ|U|², both evaluated from -D. Because the integrands are positive (m,ρ>0), these identities force |U| to have a minimum and φ (relative to the wind) to have a maximum at z=-D. The second piece of machinery is a WKB ansatz, U(z)≈κ(z)^{-1/2} exp(±∫κ), with κ = sqrt(ifρ/m), valid when |ρ'/ρ| ≪ |κ|; this is used to estimate the bottom shear U'(-H) under a no-slip condition and show it is exponentially small for deep oceans.

What would settle it

An observed current profile in the open ocean that exhibits a shear-free point at mid-depth while the speed continues to decay beneath it, and where the local dynamics are independently verified to be governed by the linear Ekman balance (1), would contradict the paper's central claim.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes Theorem 1: for the Ekman system (mU')' = ifρU with a stress-free condition U'(-D)=0 at an intermediate depth -D (with m>0, ρ>0), the modulus |U| is strictly decreasing from the surface to -D and strictly increasing from -D toward the seafloor; in the Northern Hemisphere the angle between the wind stress and the current varies in the opposite sense, with a maximum at -D. Consequently the flow below -D is not negligible but grows, which is the 'reverse Ekman spiral' the authors deem unphysical. The argument is a two-line integration by parts: multiplying the ODE by U and integrating yields sign-definite integral expressions for r'(z) and φ'(z). The paper

Load-bearing premise

The argument that the deepening current below -D is unphysical assumes the same Ekman equation—linear, with positive eddy viscosity and density—holds all the way from -D down to the seafloor; in the real ocean the deep flow is often geostrophic or otherwise outside that balance, in which case the reverse spiral may not be realized and the boundary condition might merely be an imperfect but not hopeless truncation.

Editorial extensions

If this is right

The stress-free condition at an intermediate depth is internally inconsistent with the assumption that the flow is negligible below that depth; any model that imposes it will generate spurious sub-Ekman currents.
Ocean models should not impose an artificial Ekman depth as a boundary; instead the Ekman depth should emerge from the vertical structure of the eddy viscosity (e.g., a KPP-style profile).
With no-slip at the true seafloor, the Ekman transport is nearly orthogonal to the wind in deep water, matching field observations, with an error that decays exponentially in (H-D)/δ.
The theorem is robust to the details of m(z) and ρ(z): it requires only positivity, so the inconsistency is structural rather than an artifact of a particular profile.
The orthogonality of Ekman transport and wind is recovered under no-slip even with depth-dependent density, contrary to what one might expect from the variable-density transport formula.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

A natural extension is to replace the stress-free condition with a matching condition to a geostrophic interior; the theorem suggests any derivative-zero condition at finite depth will misrepresent the deep flow unless the interior dynamics differ from the Ekman balance.
The result implies that coarse-resolution models that parameterize the Ekman layer with a fixed depth and shear-free base will systematically misplace the level of no motion, with consequences for the vertical structure of the simulated currents.
The WKB estimate suggests a testable quantitative prediction: departures of the Ekman transport angle from 90° should be negligible in the deep open ocean but become significant where H-D is small (shelves and coastal regions), consistent with the coastal circulation literature.
One could test the theorem directly in a laboratory or numerical setting: solve (1)–(2)–(6) and measure |U| below -D; the predicted increase is independent of the viscosity profile, so any observed decay below -D would indicate that the Ekman balance itself breaks down there.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

The paper's real contribution is a simple but useful theorem: in the f-plane Ekman model with positive depth-dependent eddy viscosity and density, imposing U'(-D)=0 forces |U| to decrease down to -D, hit a minimum, and then increase again below it. The proof is a one-line integration by parts after multiplying by the conjugate, and it's correct and self-contained. As far as the citations go, the specific statement is new, though it's an immediate corollary of the technique in the authors' companion paper [6]. That's fine.

The paper also contains a WKB estimate showing that if you instead impose no-slip at the bottom and the ocean is deep, U'(bottom) is exponentially small, so the Ekman transport is nearly orthogonal to the wind. The estimate is standard but competently carried out.

Where I part company with the authors is the headline conclusion that the stress-free condition is "necessarily unphysical." The theorem says: within the single ODE (1), if U'(-D)=0, then |U| grows below -D. That is a statement about that ODE on an interval that extends below -D. If you are modeling the whole water column with the same turbulent-viscosity balance, then yes, the flow does not vanish below -D. But if the deep ocean is in geostrophic balance, or if the model domain ends at -D and the interior is handled differently, then the stress-free condition does not force any reverse spiral. The paper never acknowledges that this transfer requires an assumption about the deep-ocean momentum balance; it simply asserts that below -D the eddy viscosity drops to its molecular value and (11) applies. That is not enough. So the physically strong claim is not supported; the mathematically precise claim is sound.

There's also a small slip in the WKB section: to bound |U_D|, the authors cite Theorem 1, but that theorem assumes stress-free at -D, which does not hold in the no-slip case. The exponential estimate is so strong that the slip doesn't change the conclusion, but it should be fixed.

Overall, the paper is a good candidate for publication after revision. The core theorem is a genuine contribution. The physical narrative needs to be more careful: either restrict the "unphysical" claim to models that solve (1) below -D, or justify that balance. I'd send it to a fluid-dynamics refereed journal, expecting the authors to tone down the interpretation and fix the WKB citation.

Referee Report

1 major / 3 minor

Summary. The paper studies the f-plane Ekman equations with depth-dependent density and eddy viscosity, in complex notation (Eq. (7)). Its main theorem proves that if a stress-free boundary condition is imposed at an intermediate depth z = -D, then the speed |U| decreases with depth down to -D, attains a strict minimum there, and increases again below -D; the direction angle behaves analogously. The authors interpret this 'reverse Ekman spiral' as making the mid-depth stress-free condition unphysical. They then use a WKB approximation to show that, for a deep ocean with no-slip bottom boundary and with eddy viscosity falling to the molecular value below the Ekman depth, the bottom velocity derivative is exponentially small, so the Ekman transport is nearly orthogonal to the wind stress. The paper concludes that the no-slip bottom condition is the physically correct choice and that choosing an a priori Ekman depth with a stress-free condition is incorrect.

Significance. The mathematical core of the paper is sound and elegant. Theorem 1 is a parameter-free identity: multiplying Eq. (7) by the complex conjugate of U and integrating gives a signed relation that forces the qualitative behaviour of r(z) and phi(z). This is a useful and nontrivial caution for modelers who solve the full-column f-plane Ekman equation with a stress-free condition at an intermediate depth. The WKB estimate in the second half is also constructive and gives an explicit, quantitative bound on the bottom stress in the deep-ocean no-slip case. However, the paper's headline conclusion—that the stress-free condition is 'necessarily unphysical'—goes beyond the theorem: it requires the additional, unproved premise that the same linear frictional ODE (1) remains valid below the Ekman layer all the way to the seafloor. If that premise is not supplied or the claim is not restricted, the central message overreaches.

major comments (1)

[Abstract and Section 'Main Result' (Eq. (7))] The inference from Theorem 1 to 'necessarily unphysical' is load-bearing and not justified. The theorem is a conditional statement about solutions of the ODE (7) on the full interval (-H,0), and the proof integrates from -D to z in (-H,0). The conclusion r'(z)<0 for z<-D is a mathematical consequence only when Eq. (1) is assumed to hold below -D. In the ocean, the abyssal flow is typically geostrophic, with horizontal pressure gradients that are absent from (1). Moreover, in many practical implementations of the stress-free condition at the base of the Ekman layer, the model domain ends at z=-D and no solution is defined below that depth; Theorem 1 does not apply to that truncated problem. The paper should either restrict its conclusion to full-column f-plane frictional models that solve (1) down to -H, or provide a physical justification that (1) describes the deep ocean. Without this,

minor comments (3)

[Proof of Theorem 1, Eq. (8)] The proof says 'multiply the first equation in (7) by U(z)', but Eq. (8) contains |U'(s)|^2 and |U(s)|^2, which is only consistent with multiplication by the complex conjugate \bar U(z). The subsequent extraction of the real and imaginary parts in (9)-(10) is correct with \bar U, but the notation should be fixed.
[Introduction, paragraph after Eq. (4)] The sentence 'a stress-free condition at -D analogous to (4)' should refer to Eq. (5), not Eq. (4), since (4) is the no-slip condition.
[WKB section, before Eq. (11)] The statement 'due to Theorem 1, we have |U_D|<|U(0)|' is formally invalid: Theorem 1 assumes the stress-free condition U'(-D)=0, whereas the WKB scenario imposes no-slip at -H. The same monotonicity can be obtained directly from the integral identity applied to the no-slip problem, or the remark can be omitted, as the WKB estimate does not depend on it.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the main theorem is proved in-paper from the stated ODE and boundary conditions; self-citations are not load-bearing.

full rationale

The central claim (Theorem 1) is derived directly from the model (7): multiplying (m U')' = i f rho U by the conjugate of U, integrating from -D to z, and using U'(-D)=0 gives equations (9)-(10), from which the sign pattern of r'(z) and phi'(z) follows. The boundary condition U'(-D)=0 is an input, and the monotonicity/minimum/reverse-spiral behaviour is a consequence, not an assumed conclusion. No parameter is fitted to data and then renamed a prediction. The observed 90-degree Ekman transport appears only as motivation for considering stress-free boundary conditions (Eq. (3) and the sentence 'measurements show that...'), not as an input to the proof. The citation to [6] is used only to say 'We argue along the lines of the proof of Theorem 1 in [6]' for a proof technique; the actual proof is reproduced in the paper, so this self-citation is not load-bearing. The WKB estimate in the final section is a self-contained asymptotic calculation under explicit assumptions (constant m below -D, |rho'/rho| << |kappa|); it invokes no fitted parameters. The physical interpretation that a mid-depth stress-free condition is 'unphysical' does rest on an extra-model assumption that Eq. (1) continues to hold below -D (or that the deep flow is described by the same frictional balance); this is a scope-of-validity caveat about the oceanographic setting, not circularity. The note 'due to Theorem 1, we have |U_D| < |U(0)|' in the WKB section technically misapplies Theorem 1 (which assumes U'(-D)=0) to the no-slip case, but this lapse is not load-bearing: the bound (12) is exponentially small regardless of the size of |U_D|, and it is a correctness slip, not a circular reduction. Overall, the derivation chain is internally self-contained; the only concerns are physical-domain assumptions, not circular reasoning.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The theorem itself is parameter-free; it only needs the f-plane Ekman ODE, positive m and rho, and the stated boundary conditions. The WKB section adds asymptotic assumptions about constant eddy viscosity and slow density variation below the Ekman depth, plus an unproved bound on |U_D|. No new physical entities are introduced.

assumptions (5)

domain assumption Ekman f-plane equations (1): (m(z)u')' = -f rho(z) v, (m(z)v')' = f rho(z) u on -H < z < 0.
The stated problem is based on this reduced model; it assumes a flat surface, constant Coriolis parameter, and depth-dependent m and rho.
domain assumption m(z) > 0 and rho(z) > 0 throughout the water column.
Used when dividing by r and m in (9)-(10); without positivity the sign conclusions in Theorem 1 fail.
domain assumption The same governing ODE (1) applies below the chosen Ekman depth z = -D.
The 'unphysical' conclusion of Theorem 1 assumes the reverse spiral below -D is part of the same model; real ocean interiors may obey different dynamics.
domain assumption Below -D the eddy viscosity is constant and the density variation is slow enough for WKB validity, |rho'/rho| << |kappa|.
Needed for the no-slip orthogonality estimate; the paper states this is satisfied at great depths and uses a linear density example.
ad hoc to paper |U(-D)| < |U(0)| in the no-slip setting.
Used to bound the WKB estimate; the paper justifies it by Theorem 1, but Theorem 1 assumes U'(-D)=0, which the no-slip problem does not satisfy.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Why a mid-depth stress-free boundary condition is incorrect for Ekman flows." pith.science (2026). https://pith.science/paper/2XSCPXPG

@misc{pith2026260729261,
  author       = {Pith},
  title        = {Pith review of: Why a mid-depth stress-free boundary condition is incorrect for Ekman flows},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2XSCPXPG}},
  note         = {Machine review of arXiv:2607.29261}
}

abstract

We show that the assumption of a stress-free boundary condition at a finite intermediate depth, namely, at the bottom of the Ekman layer, in the analysis of wind-driven ocean flows necessarily leads to an unphysical current profile. Indeed, if the $z$-derivative of the fluid velocity vanishes at a given depth, then this depth necessarily corresponds to a minimum of the velocity profile, with the velocity increasing beneath it. Using a WKB ansatz based on the small variations of the ocean's water density at great depths, we also argue that a no-slip condition at the bottom of the ocean, if sufficiently deep, still effectively implies (up to a very small error) the orthogonality of the Ekman transport and the wind-stress.

Figures

Figures reproduced from arXiv: 2607.29261 by the authors.

**Figure 2.** FIG. 2. The Cartesian coordinate system associated with the [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗

**Figure 3.** FIG. 3. An illustration of the behaviour of the “reverse Ek [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

24 extracted references · 2 linked inside Pith

[6]

A. D. Jenkins and J. A. T. Bye, Some aspects of the work of V. W. Ekman, Polar Rec.42, 15 (2006)

2006
[1]

J. P. Pedlosky,Ocean Circulation Theory(Springer Berlin, Heidelberg, 1998). 6

1998
[2]

L. D. Talley, G. L. Pickard, W. J. Emery, and J. H. Swift,Descriptive Physical Oceanography: An Introduc- tion(Academic Press, San Diego, 2011)

2011
[3]

Siedler, S

G. Siedler, S. M. Griffies, J. Gould, and J. A. Church, eds.,Ocean Circulation and Climate: A 21st Century Perspective(Academic Press, Oxford, UK, 2013)

2013
[4]

V. W. Ekman, On the influence of the Earth’s rotation on ocean currents, Ark. Mat. Astr. Fys.2, 1 (1905)

1905
[5]

well behaved

for a concise overview of Ekman’s work) applies to the idealised setting of a homogeneous ocean forced by a steady, spatially uniform wind and characterised by a constant vertical eddy viscosity. The interplay between the Coriolis force and the frictional force induced by the wind stress gives rise to the characteristic dynamics of the Ekman currents. The...

arXiv 2026
[7]

Puntini, L

C. Puntini, L. Roberti, and E. Stefanescu, On large- scale wind-drift ocean currents: An asymptotic approach in spherical coordinates, (2026), arXiv:2602.06473 [physics.flu-dyn]

arXiv 2026
[8]

Roberti, The Ekman spiral for piecewise-constant eddy viscosity, Appl

L. Roberti, The Ekman spiral for piecewise-constant eddy viscosity, Appl. Anal.101, 5528 (2022)

2022

Show all 24 references

[9]

Constantin, Frictional effects in wind-driven ocean currents, Geophys

A. Constantin, Frictional effects in wind-driven ocean currents, Geophys. Astrophys. Fluid Dyn.115, 1 (2020)

2020
[10]

T. K. Chereskin, Direct evidence for an Ekman balance in the California Current, J. Geophys. Res.100, 18261 (1995)

1995
[11]

J. F. Price, R. A. Weller, and R. R. Schudlich, Wind- driven Ocean Currents and Ekman Transport, Science 238, 1534 (1987)

1987
[12]

Puntini, Nonlinear dynamics of wind-drift currents at mid-latitudes, Nonlinear Anal

C. Puntini, Nonlinear dynamics of wind-drift currents at mid-latitudes, Nonlinear Anal. Real World Appl.90, 104557 (2026)

2026
[13]

Wang and R

W. Wang and R. X. Huang, Wind energy input to the Ekman layer, J. Phys. Oceanogr.34, 1267 (2004)

2004
[14]

Soloviev and R

A. Soloviev and R. Lucas,The Near-Surface Layer of the Ocean(Springer, Dordrecht, 2014)

2014
[15]

G. K. Vallis,Atmospheric and Oceanic Fluid Dynamics (Cambridge University Press, Cambridge, 2017)

2017
[16]

V. I. Shrira and R. B. Almelah, Upper-ocean Ekman cur- rent dynamics: a new perspective, J. Fluid Mech.887, A24 (2020)

2020
[17]

G´ erard-Varet and E

D. G´ erard-Varet and E. Dormy, Ekman layers near wavy boundaries, J. of Fluid Mech.565, 115 (2006)

2006
[18]

Deusebio, G

E. Deusebio, G. Brethouwer, P. Schlatter, and E. Lind- borg, A numerical study of the unstratified and stratified ekman layer, J. Fluid Mech.755, 672 (2014)

2014
[19]

J. C. McWilliams, E. Huckle, and A. F. Shchepetkin, Buoyancy Effects in a Stratified Ekman Layer, J. Phys. Oceanogr.39, 2581 (2009)

2009
[20]

J. F. Price and M. A. Sundermeyer, Stratified Ekman layers, J. Geophys. Res.104, 20467 (1999)

1999
[21]

D. M. Lewis and S. E. Belcher, Time-dependent, coupled, Ekman boundary layer solutions incorporating stokes drift, Dyn. Atmos. Oceans37, 313 (2004)

2004
[22]

Sentchev, M

A. Sentchev, M. Yaremchuk, D. Bourras, I. Pairaud, and P. Frauni´ e, Estimation of the Eddy Viscosity Pro- file in the Sea Surface Boundary Layer from Underway ADCP Observations, J. Atmos. Ocean. Technol.40, 1291 (2023)

2023
[23]

W. G. Large, J. C. McWilliams, and S. C. Doney, Oceanic vertical mixing: A review and a model with a nonlocal boundary layer parameterization, Rev. Geophys.32, 363 (1994)

1994
[24]

S. J. Lentz and M. R. Fewings, The wind- and wave- driven inner-shelf circulation., Annu. Rev. Mar. Sci.4, 317 (2012)

2012

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.

[1] [6]

A. D. Jenkins and J. A. T. Bye, Some aspects of the work of V. W. Ekman, Polar Rec.42, 15 (2006)

2006

[2] [1]

J. P. Pedlosky,Ocean Circulation Theory(Springer Berlin, Heidelberg, 1998). 6

1998

[3] [2]

L. D. Talley, G. L. Pickard, W. J. Emery, and J. H. Swift,Descriptive Physical Oceanography: An Introduc- tion(Academic Press, San Diego, 2011)

2011

[4] [3]

Siedler, S

G. Siedler, S. M. Griffies, J. Gould, and J. A. Church, eds.,Ocean Circulation and Climate: A 21st Century Perspective(Academic Press, Oxford, UK, 2013)

2013

[5] [4]

V. W. Ekman, On the influence of the Earth’s rotation on ocean currents, Ark. Mat. Astr. Fys.2, 1 (1905)

1905

[6] [5]

well behaved

for a concise overview of Ekman’s work) applies to the idealised setting of a homogeneous ocean forced by a steady, spatially uniform wind and characterised by a constant vertical eddy viscosity. The interplay between the Coriolis force and the frictional force induced by the wind stress gives rise to the characteristic dynamics of the Ekman currents. The...

arXiv 2026

[7] [7]

Puntini, L

C. Puntini, L. Roberti, and E. Stefanescu, On large- scale wind-drift ocean currents: An asymptotic approach in spherical coordinates, (2026), arXiv:2602.06473 [physics.flu-dyn]

arXiv 2026

[8] [8]

Roberti, The Ekman spiral for piecewise-constant eddy viscosity, Appl

L. Roberti, The Ekman spiral for piecewise-constant eddy viscosity, Appl. Anal.101, 5528 (2022)

2022