The manuscript presents a detailed, multi-diagnostic analysis of a single 3D radiation-hydrodynamics Athena++ snapshot of a red-supergiant donor undergoing Roche-lobe overflow. The authors (i) reconstruct an effective Roche potential and mask material outside the L1 equipotential (Sec. 2.1–2.2), (ii) quantify the morphology and intermittency of the mass-flux field and show a strongly filamentary, bursty outflow rather than a smooth stream (Sec. 3.1–3.2, Table 2, Fig. 2), (iii) apply the Q-criterion to assess flow topology and argue the L1 stream is shear-dominated without large coherent vortices (Sec. 2.3, Sec. 3.3, Fig. 4), and (iv) introduce an upstream “source fingerprinting” procedure tracing stream cells back to a photospheric surface and comparing the resulting footprint’s properties to the global surface (Sec. 2.5, Sec. 3.4, Table 3, Fig. 5). The headline interpretation is that localized convective upwellings (lower density/pressure, higher outward radiation flux, more positive radial velocity) preferentially supply material to the L1 stream. The overall narrative is compelling and the analysis toolkit is potentially valuable, but several conclusions are currently overextended given the reliance on a single snapshot and the under-specification/validation of key threshold-based steps (Roche masking, Q-structure identification, and source tracing). The paper would be substantially strengthened by (a) clearer documentation of the underlying simulation and gravity/potential consistency, (b) robustness checks across nearby snapshots and parameter/threshold choices, and (c) quantitative presentation of the force/work decomposition that underpins the “driving engine” claims (Sec. 2.4, Sec. 3.4, Sec. 4).

This section audits symbolic/analytic mathematical consistency (algebra, derivations, dimensional/unit checks, definition consistency).

Maths relevance: light

The paper contains a small set of central analytic definitions (coordinate transform, Roche potential, mass flux, Q-criterion, force densities, and power density) used to construct masks/diagnostics on simulation data. There are few step-by-step derivations; most mathematics is definitional and relies on correct frame/notation choices.

### Checked items

• ✔ Spherical-to-Cartesian transform (Eqs. (1)–(3), Sec. 2.1, p.2)

• Claim: Defines $X = r \sin\theta \cos\phi$, $Y = r \sin\theta \sin\phi$, $Z = r \cos\theta$ for mapping the spherical grid to Cartesian coordinates.
• Checks: algebra, notation consistency
• Verdict: PASS; confidence: high; impact: minor
• Assumptions/inputs: $\theta$ is the polar (colatitude) angle and $\phi$ is the azimuthal angle in the simulation’s spherical grid.
• Notes: Standard coordinate relations; no internal conflicts elsewhere in the paper.

• ✖ Roche potential: companion gravitational term appears dimensionally wrong (Unnumbered equation for $\Phi_{\rm R}$ in Sec. 2.1, p.2)

• Claim: Defines the effective Roche potential as the sum of two point-mass gravitational potentials and a centrifugal potential term.
• Checks: dimensional/units consistency, notation/parentheses consistency
• Verdict: FAIL; confidence: medium; impact: critical
• Assumptions/inputs: $\Phi_{\rm R}$ is intended to have units of specific potential energy (e.g., $\mathrm{cm}^2\,\mathrm{s}^{-2}$ in cgs)., Distances are measured from the donor-centered Cartesian coordinates defined by Eqs. (1)–(3).
• Notes: As written in the provided PDF text, the companion term appears as “$-\sqrt{GM_2} / ((X-D)^2+Y^2+Z^2)$” (or equivalently with the square-root only over $GM_2$ rather than over the distance). This is not dimensionally consistent with a gravitational potential term, which should scale like $-GM_2 / \sqrt{(X-D)^2+Y^2+Z^2}$. The missing/incorrect square-root placement (or missing parentheses) makes $\Phi_{\rm R}$ ambiguous/incorrect in the document as provided.

• ⚠ Roche potential: frame/origin consistency of centrifugal term (Same Roche potential definition, Sec. 2.1, p.2)

• Claim: Uses $-(1/2)\Omega^2(X^2+Y^2)$ as the centrifugal potential while stating the domain is centered on the donor.
• Checks: definition consistency, missing assumptions check
• Verdict: UNCERTAIN; confidence: low; impact: critical
• Assumptions/inputs: The stated Cartesian coordinates $(X,Y,Z)$ are centered on the donor star (Sec. 2.1)., The binary rotates with angular velocity $\Omega$ in a co-rotating frame.
• Notes: If $(X,Y,Z)$ are donor-centered, the rotation axis for the centrifugal potential is not necessarily through the donor (unless the donor is at the barycenter). The paper does not specify whether the centrifugal term is computed about the barycenter or about the donor-centered origin, nor does it include any offset/linear inertial terms that may be needed in a non-barycentric rotating frame. Without an explicit frame definition, it is not possible to verify that the $\Phi_{\rm R}$ used to compute $\Phi_{\rm L1}$ and the Roche-lobe mask is self-consistent.

• ✔ Roche-lobe escape criterion (End of Sec. 2.1, p.2)

• Claim: Defines a boolean mask is_outside_RL true where $\Phi_{\rm R} > \Phi_{\rm L1}$ to identify material outside the donor Roche lobe.
• Checks: logical consistency, sign convention sanity check
• Verdict: PASS; confidence: medium; impact: moderate
• Assumptions/inputs: $\Phi_{\rm L1}$ is the value of $\Phi_{\rm R}$ at the L1 saddle point., The potential convention is such that ‘higher’ equipotentials correspond to larger $\Phi_{\rm R}$ (less negative).
• Notes: Given the paper’s own convention of negative gravitational terms, selecting $\Phi_{\rm R} > \Phi_{\rm L1}$ to indicate crossing the L1 equipotential is consistent. This pass is conditional on $\Phi_{\rm R}$ being correctly defined and computed.

• ✔ Mass flux definition (Secs. 2.2 and 3.2, pp.3–5)

• Claim: Defines mass flux vector $\vec{j} = \rho \vec{v}$ and magnitude $j_{\rm mag} = |\vec{j}|$ for morphology and PDF statistics.
• Checks: dimensional/units consistency, notation consistency
• Verdict: PASS; confidence: high; impact: minor
• Assumptions/inputs: $\rho$ is mass density; $\vec{v}$ is velocity.
• Notes: Standard definition; no internal contradictions.

• ✔ Velocity gradient decomposition (Sec. 2.3, p.3)

• Claim: Defines $J = \nabla \vec{v}$, $S = \frac{1}{2}(J + J^T)$, $W = \frac{1}{2}(J - J^T)$.
• Checks: algebra, definition consistency
• Verdict: PASS; confidence: medium; impact: moderate
• Assumptions/inputs: $J^T$ denotes transpose in the chosen basis., $\nabla \vec{v}$ is computed with the appropriate metric terms in spherical coordinates (as stated).
• Notes: Algebraic definitions are correct. The correctness of ‘including all metric terms’ is implementation-dependent and not verifiable from the paper, but the symbolic decomposition is consistent.

• ✔ Q-criterion formula (Secs. 2.3 and 3.3, pp.3–6)

• Claim: Defines $Q = \frac{1}{2}(||W||_F^2 - ||S||_F^2)$ and interprets $Q > 0$ as vorticity-dominated regions.
• Checks: algebra, sign/limiting sanity check
• Verdict: PASS; confidence: high; impact: moderate
• Assumptions/inputs: $||\cdot||_F$ is the Frobenius norm.
• Notes: If pure rotation: $S = 0$, $Q > 0$; if pure strain: $W = 0$, $Q < 0$. Interpretation matches the definition.

• ✔ Force density definitions (Sec. 2.4, p.3)

• Claim: Defines ${\bf F}_{\rm grav} = -\rho \nabla \Phi_{\rm R}$, ${\bf F}_{\rm gas} = -\nabla p_{\rm g}$, ${\bf F}_{\rm rad} = -\nabla\cdot {\bf P}_{\rm r}$ as the dominant force densities.
• Checks: dimensional/units consistency, notation consistency
• Verdict: PASS; confidence: medium; impact: moderate
• Assumptions/inputs: $\Phi_{\rm R}$ is specific potential; $p_{\rm g}$ is gas pressure; ${\bf P}_{\rm r}$ is radiation pressure tensor.
• Notes: Each term has force-per-volume units ($\rho \times$ acceleration; pressure gradient; divergence of a pressure tensor).

• ✔ Power density (work rate per unit volume) (Sec. 2.4, p.3)

• Claim: Defines $W_i = {\bf F}_i \cdot \vec{v}$ and interprets sign as kinetic-energy source/sink locally.
• Checks: dimensional/units consistency, sign convention sanity check
• Verdict: PASS; confidence: high; impact: minor
• Assumptions/inputs: ${\bf F}_i$ is force density; $\vec{v}$ is velocity.
• Notes: ${\bf F}_i \cdot \vec{v}$ yields power per unit volume. The sign interpretation (positive when force aligns with velocity) is correct.

• ⚠ Kurtosis reference value for Gaussian (Sec. 3.2, p.5)

• Claim: States that a Gaussian distribution has kurtosis $3$ and uses this to interpret large kurtosis as intermittency.
• Checks: definition consistency, missing assumptions check
• Verdict: UNCERTAIN; confidence: medium; impact: minor
• Assumptions/inputs: Kurtosis is the 4th standardized moment (Pearson), not excess kurtosis.
• Notes: The comparison to $3$ is correct under the Pearson (non-excess) definition, but the paper does not specify the convention used by the moment calculator; without that, the textual interpretation could be offset by $3$.

• ⚠ Undefined radial radiation flux symbol (Sec. 3.4 and Table 3, pp.6–7)

• Claim: Compares ‘radial radiation flux ($F_{r,{\rm rad}}$)’ between source regions and the full surface.
• Checks: symbol/definition consistency
• Verdict: UNCERTAIN; confidence: high; impact: moderate
• Assumptions/inputs: $F_{r,{\rm rad}}$ is a component of a radiation flux vector available from the simulation or derived from radiation moments.
• Notes: Methods define radiation pressure tensor ${\bf P}_{\rm r}$ and radiation force density $-\nabla\cdot {\bf P}_{\rm r}$ but never define a radiation flux vector or how $F_{r,{\rm rad}}$ is computed/sign-conventioned. This blocks a purely analytic check of consistency between the ‘radiation force’ and ‘radiation flux’ quantities discussed.

### Limitations

• The paper provides few explicit derivation steps; most mathematical content is definitional, so deeper algebraic verification is limited.
• Key correctness of the Roche-lobe mask hinges on the exact Roche potential formula and reference frame; the provided text appears ambiguous/incorrect for the companion term and under-specified for the rotating-frame origin/axis.
• Implementation details (finite differencing, metric terms in spherical coordinates, connected-component segmentation) are intentionally out of scope and cannot be analytically validated here.

This section audits numerical/empirical consistency: reported metrics, experimental design, baseline comparisons, statistical evidence, leakage risks, and reproducibility.

16 candidate numeric checks were evaluated: 14 PASS, 1 FAIL, and 1 UNCERTAIN. The only failing check is a text claim of an “almost $20\times$” flux increase (computed $17.747\times$ from the reported means). Several internal consistency/inequality checks for Table 1 and ratio/threshold checks for Table 2/3 pass as stated.

### Checked items

• ⚠ C1 (Page 4, §3.1 (gravitational environment))

• Claim: “We calculate the potential at L1 to be $\Phi_{\rm L1} = -3.828$ (in code units)”
• Checks: repeated-constant consistency
• Verdict: UNCERTAIN
• Notes: Cannot verify repeated-constant consistency: no document text/table corpus to scan for other $\Phi_{\rm L1}$ occurrences.

• ✔ C2 (Page 4, Table 1 (Density $\rho$ row))

• Claim: Table 1 reports Density ($\rho$): Mean $8.342\times 10^{-2}$, Std Dev $3.557\times 10^{-1}$, Min $1.000\times 10^{-8}$, Max $6.495\times 10^{0}$.
• Checks: range and positivity checks
• Verdict: PASS
• Notes: All inequality checks satisfied.

• ✔ C3 (Page 4, Table 1 (Gas Pressure $p_g$ row))

• Claim: Table 1 reports Gas Pressure ($p_g$): Mean $4.737\times 10^{-2}$, Std Dev $1.444\times 10^{-1}$, Min $1.000\times 10^{-12}$, Max $1.213\times 10^{0}$.
• Checks: range and positivity checks
• Verdict: PASS
• Notes: All inequality checks satisfied.

• ✔ C4 (Page 4, Table 1 (Radial Velocity $v_r$ row))

• Claim: Table 1 reports Radial Velocity ($v_r$): Mean $-5.345\times 10^{-2}$, Std Dev $4.056\times 10^{-1}$, Min $-2.396\times 10^{0}$, Max $2.537\times 10^{0}$.
• Checks: range checks (signed variable)
• Verdict: PASS
• Notes: All inequality checks satisfied. Also $|\text{mean}|/$std $= 0.13178$ (informational).

• ✔ C5 (Page 4, Table 1 (Theta Velocity $v_\theta$ row))

• Claim: Table 1 reports Theta Velocity ($v_\theta$): Mean $1.140\times 10^{-4}$, Std Dev $3.999\times 10^{-1}$, Min $-2.606\times 10^{0}$, Max $2.629\times 10^{0}$.
• Checks: range checks (signed variable)
• Verdict: PASS
• Notes: All inequality checks satisfied.

• ✔ C6 (Page 4, Table 1 (Phi Velocity $v_\phi$ row))

• Claim: Table 1 reports Phi Velocity ($v_\phi$): Mean $-1.481\times 10^{-1}$, Std Dev $7.258\times 10^{-1}$, Min $-3.850\times 10^{0}$, Max $1.960\times 10^{0}$.
• Checks: range checks (signed variable)
• Verdict: PASS
• Notes: All inequality checks satisfied.

• ✔ C7 (Page 4, Table 1 (Radiation Energy $E_r$ row))

• Claim: Table 1 reports Radiation Energy ($E_r$): Mean $2.268\times 10^{-2}$, Std Dev $6.587\times 10^{-2}$, Min $1.726\times 10^{-15}$, Max $5.830\times 10^{-1}$.
• Checks: range and positivity checks
• Verdict: PASS
• Notes: All inequality checks satisfied.

• ✖ C8 (Page 5, §3.2 + Table 2 (mean flux ratio claim))

• Claim: “When we isolate a region immediately surrounding the L1 point… The mean flux is almost 20 times higher” (Global mean $0.028256$ vs L1 Region mean $0.501464$).
• Checks: ratio recomputation
• Verdict: FAIL
• Notes: Computed ratio (L1/global) $= 17.747168742921858$; exceeds the $10\%$ relative tolerance for “almost $20\times$”.

• ✔ C9 (Page 5, §3.2 + Table 2 (kurtosis 'over 1000'))

• Claim: “The PDF of $j_{\rm mag}$… is characterized by an extremely high kurtosis of over 1000.” Table 2 lists global kurtosis $1033.6$.
• Checks: threshold check
• Verdict: PASS
• Notes: Strict greater-than threshold check.

• ✔ C10 (Page 5, §3.2 (Gaussian kurtosis comparison) + Table 2)

• Claim: Text compares global kurtosis to “the value of 3 for a Gaussian distribution”; Table 2 gives global kurtosis $1033.6$.
• Checks: order-of-magnitude / relative comparison
• Verdict: PASS
• Notes: Checked global_kurtosis $>$ gaussian_reference; factor over Gaussian $= 344.5333333333333$.

• ✔ C11 (Page 6, §3.4 bullet (1) + Table 3 (density ratio))

• Claim: “The mean density… within the source fingerprint are approximately 5 times lower than the respective surface averages” using Table 3 densities: Source $1.727\times 10^{-2}$ vs Full $8.428\times 10^{-2}$.
• Checks: ratio recomputation
• Verdict: PASS
• Notes: Computed full/source ratio $= 4.880138969310943$, consistent with “approximately $5\times$ lower” under the stated tolerance.

• ✔ C12 (Page 6, §3.4 bullet (1) + Table 3 (pressure ratio))

• Claim: “The mean … pressure within the source fingerprint are approximately 5 times lower than the respective surface averages” using Table 3 pressures: Source $9.657\times 10^{-3}$ vs Full $4.665\times 10^{-2}$.
• Checks: ratio recomputation
• Verdict: PASS
• Notes: Computed full/source ratio $= 4.830692761727244$, consistent with “approximately $5\times$ lower” under the stated tolerance.

• ✔ C13 (Page 6, §3.4 bullet (3) + Table 3 (radiation flux ratio))

• Claim: “The mean outward flux from the source regions is over 6.4 times greater than the surface average.” Table 3: $F_{r,{\rm rad}}$ source $2.837\times 10^{-3}$ vs full $4.428\times 10^{-4}$.
• Checks: ratio + threshold check
• Verdict: PASS
• Notes: Computed ratio (source/full) $= 6.406955736224029$, which is strictly greater than $6.4$.

• ✔ C14 (Page 7, Table 3 (Density stats))

• Claim: Table 3 Density ($\rho$) mean and std dev for source vs full: Source mean $1.727\times 10^{-2}$, std $5.516\times 10^{-2}$; Full mean $8.428\times 10^{-2}$, std $3.637\times 10^{-1}$.
• Checks: std-dev non-negativity and plausibility bounds
• Verdict: PASS
• Notes: All std dev values non-negative.

• ✔ C15 (Page 7, Table 3 (Pressure stats))

• Claim: Table 3 Pressure ($p_g$) mean and std dev for source vs full: Source mean $9.657\times 10^{-3}$, std $3.772\times 10^{-2}$; Full mean $4.665\times 10^{-2}$, std $1.416\times 10^{-1}$.
• Checks: std-dev non-negativity
• Verdict: PASS
• Notes: All std dev values non-negative.

• ✔ C16 (Page 7, Table 3 (Velocity mean within implied min/max not provided))

• Claim: Table 3 lists mean radial velocity: source $-1.481\times 10^{-2}$ vs full $-5.987\times 10^{-2}$; claim: “mean radial velocity in the source regions is significantly less negative.”
• Checks: signed comparison
• Verdict: PASS
• Notes: Checked $v_{r,{\rm mean,\,source}} > v_{r,{\rm mean,\,full}}$ (less negative); difference $= 0.04506$.

### Limitations

• Only parsed text was available; no access to underlying simulation data arrays, so any claim requiring recomputation from fields (Q-criterion segmentation, force balance maps, PDFs) cannot be verified.
• No plot digitization was performed; numeric checks are limited to explicit table values and explicit numeric statements in the text.
• Some qualitative phrases (e.g., 'significantly', 'conservative threshold', 'objects of significant volume') lack quantitative criteria, limiting what can be asserted with automated fast checks.
• Repeated-constant consistency checks (e.g., verifying whether $\Phi_{\rm L1} = -3.828$ is used consistently throughout) could not be completed because the checking routine could not scan a full document corpus for other occurrences.