{"id":"40ec17a5-bb22-4050-8669-43f3b2827c9d","arxiv_id":"2607.07443","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":6,"one_line_summary":"A fully one-sided diffuse-interface immersed boundary method coupled with an explicit wall model and a turbulent shear-stress tensor model achieves accurate WMLES in channel flows and over a NACA23012 airfoil.","lead":"The paper develops a wall-modeling approach for large-eddy simulation that confines immersed-boundary forcing to the solid side of an interface, eliminating cross-boundary diffusion that degrades wall-shear-stress predictions. Engineers simulating high-Reynolds-number turbulent flows over complex geometries on Cartesian grids may benefit from improved near-wall accuracy without body-fitted meshes.","discovery_kind":"unclear","skeptic_critique":{"model":"glm-5.2","headline":"The tau-model's equilibrium assumptions (constant total shear stress, linear modeled stress profile) are the weakest link, but they are standard limitations of equilibrium wall models validated within the paper's stated scope.","rationale":"The reader correctly identified the most load-bearing concern: the tau-model's linear profile and equilibrium assumptions are untested for non-equilibrium flows. I agree with this assessment. The concern is real but represents a scope limitation rather than a fundamental flaw. Within the validated regime (channel flow at multiple Re, inclination angles, grid resolutions; attached airfoil flow), the method performs well with Cf errors below 1% and Cl errors below 5%. The derivation is internally consistent: the WS coupling (Eq. 42) is parameter-free and cleanly derived from the integral condition (Eq. 37); the FODIBM one-sided spreading with scaling factor φl (Eq. 24) and improved Lagrangian weight Wl (Eq. 25) is well-established from [38]; the tau-model adaptation from sharp-interface [22] to diffuse-interface IBM is straightforward. The comparison between FODIBM and DIBM (Section 3.1.3, Tables 7-8) clearly demonstrates the advantage of one-sided spreading, particularly at low reference heights (Cf error 3.75% vs 12.7% at dref=2Δx). The main gaps are: (1) no separated-flow validation, (2) no sensitivity analysis for tripping parameters on the airfoil, (3) no shipped code or data for reproducibility, and (4) relatively short averaging time for the airfoil (10 flow-through times). These gaps justify the CONDITIONAL verdict but do not warrant a stronger downgrade. The paper makes a solid contribution to the IBM-WMLES program, and the concern about non-equilibrium flows is explicitly acknowledged as future work by the authors.","tokens_in":24233,"tokens_out":6837,"duration_ms":461769,"concrete_test":"Run the periodic hill case (Mellen et al.) or backward-facing step at Re where DNS/experimental data exists — a canonical separated flow with strong adverse pressure gradient. Compare the tau-model's predicted velocity profile, Reynolds shear stress, and reattachment length against DNS. If the reattachment length error exceeds ~15% or the velocity profile shows log-layer mismatch in the separated region, the equilibrium assumptions in Eqs. (49) and (53) are load-bearing and the method's scope claim needs narrowing. As a cheaper diagnostic: in the existing channel flow data, replace the linear fs profile (Eq. 49) with the exact DNS total shear stress profile shape and recompute; if Cf changes by less than 1%, the linear assumption is not load-bearing even in equilibrium.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The reader correctly identifies the linear tau-model profile assumption (Section 2.4.3, Eq. 49/50) as the least secure point. I would sharpen this: the tau-model (Eq. 54) relies on two equilibrium assumptions, both untested beyond attached flows. First, Eq. (53) estimates τw via a mixing-length approximation at the reference height, which assumes the total shear stress is approximately constant in the inner layer (Eq. 51) — an equilibrium assumption that fails in strong adverse pressure gradients and separated flows. Second, the modeled turbulent shear stress τmodel_ξη is assumed to vary linearly from τw at the wall to zero at the reference height (Eq. 49, fs = max((dref−η)/dref, 0)), following Tamaki and Kawai [22]. In equilibrium channel flow, the total shear stress is indeed approximately linear, making both assumptions reasonable — and the channel flow validation (Cf errors < 1%, good velocity and Reynolds stress profiles) confirms this. However, the airfoil case at α=6.2° is also an attached flow, so neither assumption is tested in non-equilibrium conditions. The paper explicitly acknowledges this by listing separated flows as future work. This is a genuine scope limitation rather than an internal inconsistency: the derivation is clean, the WS coupling (Eq. 42) is parameter-free, and the FODIBM one-sided spreading is well-motivated. The concern is about generalization, not correctness within the validated regime. Additionally, the airfoil tripping parameters (At=0.25, δ*=2.4Δx, Section 3.2) are empirically chosen without sensitivity analysis, but the Cl results are similar with and without tripping (0.754 vs 0.756), suggesting the global force prediction is not strongly sensitive to these choices.","agreement_with_reader":"agree"},"referee_report":{"model":"glm-5.2","summary":"This manuscript presents a wall-modeled large-eddy simulation (WMLES) approach built on a fully one-sided diffuse-interface immersed boundary method (FODIBM). The key contributions are: (1) a wall-shear-stress enforcement strategy (WS) that couples the wall-parallel IB forcing directly to the wall shear stress predicted by an explicit wall model, without introducing artificial parameters, and (2) a tau-model adapted from Tamaki and Kawai [22] that introduces a modeled turbulent shear-stress tensor to preserve the total shear-stress balance below the reference height. The method is validated in turbulent channel flow (Re_tau = 1000–10000) with both wall-aligned and wall-unaligned grids, and in flow over a NACA23012 airfoil at Re = 1.88×10^6. The FODIBM is shown to substantially outperform the conventional DIBM, particularly at low reference heights.","tokens_in":24524,"tokens_out":991,"duration_ms":361357,"significance":"The paper addresses a genuine limitation of diffuse-interface IBMs for WMLES — namely, cross-boundary diffusion that contaminates near-wall flow and degrades wall-shear-stress prediction. The WS coupling (Eq. 42) is parameter-free and well-motivated. The validation is thorough: grid convergence (Table 3), reference height sensitivity (Table 4), inclination angle robustness (Table 5), Reynolds number dependence (Table 6), and a direct DIBM–FODIBM comparison (Tables 7–8). The channel flow results are strong, with skin-friction coefficient errors below 1% for the WS+tau-model at Re_tau = 5200. The airfoil case provides a relevant industrial test case with lift coefficient errors below 5%. The tau-model is adapted from an independent reference [22], grounding the central claim externally rather than circularly.","major_comments":[{"comment":"§2.4.3, Eqs. (51)–(53): The tau-model relies on two equilibrium assumptions — constant total shear stress in the inner layer (Eq. 51) and a linear profile of modeled turbulent shear stress from tau_w at the wall to zero at the reference height (Eq. 49/50). Both are standard limitations of equilibrium wall models, and the paper explicitly lists separated flows as future work. However, the airfoil validation (§3.2) at alpha=6.2° is also an attached flow, so neither assumption is tested in non-equilibrium conditions. This is a genuine scope limitation rather than an internal inconsistency, but it should be stated more explicitly in the conclusions or the airfoil discussion. As written, the reader may infer broader applicability than is demonstrated. A brief sentence acknowledging that the airfoil case does not exercise the tau-model under adverse pressure gradients or incipient separation, ","section":null}],"minor_comments":[{"comment":"§2.4.2, Eq. (42): The statement that the reciprocity condition is 'no longer required' for the wall-parallel momentum forcing is stated without detailed justification. A brief remark on why the velocity boundary-condition error framework does not apply when the forcing is stress-linked would improve clarity.","section":null},{"comment":"§3.2: The tripping parameters (A_t=0.25, delta*_0=2.4*delta_x) are calibrated for this case. The sensitivity of the results to these parameters is not reported. A brief comment on robustness to tripping parameters would strengthen the airfoil validation.","section":null},{"comment":"Table 7: The theoretical bulk velocity of 24.85 is described as obtained from 'an empirical formula' but the specific formula or reference is not cited. Please clarify.","section":null},{"comment":"Figures 4–9: The axis labels and legends are small and difficult to read. Consider enlarging for the final version.","section":null},{"comment":"§2.4.3, Eq. (49): The blending function f_s is described as a 'simple linear form' adopted 'for convenience.' It would help to note that alternative forms (e.g., tanh-based) were not tested, so the sensitivity to this choice is unknown.","section":null},{"comment":"Reference [38] is cited as 'J. Comput. Phys. (2026) 114721' — please verify this is published or in press at the time of submission.","section":null}],"recommendation":"minor_revision","confidential_remarks":"The reader's report and stress-test note correctly identify the linear tau-model assumption as the weakest point. I agree this is a scope limitation, not an internal inconsistency. The derivation is clean, the WS coupling is genuinely parameter-free, and the FODIBM one-sided spreading is well-motivated. The paper is a solid contribution to the IBM-WMLES literature. The minor revisions requested are primarily about clarifying scope limitations and presentation; no load-bearing element needs rework."},"author_rebuttal":{"model":"glm-5.2","summary":"We thank the referee for the careful reading and the positive assessment. The referee raises one major comment regarding the scope limitation of the tau-model validation, specifically that the airfoil case at α=6.2° is an attached flow and does not exercise the equilibrium assumptions under non-equilibrium conditions. We agree this should be stated more explicitly.","responses":[{"response":"We fully agree with this observation. The airfoil case at α=6.2° is indeed an attached-flow configuration, and the tau-model's equilibrium assumptions — constant total shear stress in the inner layer and the linear profile of the modeled turbulent shear stress — are not tested under adverse pressure gradients or incipient separation. We will add an explicit statement in both the airfoil discussion (§3.2) and the conclusions (§4) acknowledging this scope limitation. Specifically, we will note that the NACA23012 case at α=6.2° does not exercise the tau-model under non-equilibrium conditions, and that validation under adverse pressure gradients and separated flows remains necessary future work. This is consistent with the existing mention of flow separation as future work in the conclusions, but we agree it should be stated more precisely in the airfoil section as well.","revision_made":"yes","referee_comment":"§2.4.3, Eqs. (51)–(53): The tau-model relies on two equilibrium assumptions — constant total shear stress in the inner layer (Eq. 51) and a linear profile of modeled turbulent shear stress from τ_w at the wall to zero at the reference height (Eq. 49/50). Both are standard limitations of equilibrium wall models, and the paper explicitly lists separated flows as future work. However, the airfoil validation (§3.2) at α=6.2° is also an attached flow, so neither assumption is tested in non-equilibrium conditions. This is a genuine scope limitation rather than an internal inconsistency, but it should be stated more explicitly in the conclusions or the airfoil discussion. As written, the reader may infer broader applicability than is demonstrated. A brief sentence acknowledging that the airfoil case does not exercise the tau-model under adverse pressure gradients or incipient separation."}],"tokens_in":23766,"tokens_out":543,"duration_ms":53197,"standing_objections":[]},"desk_editor":{"model":"glm-5.2","letter":"Short version: this paper couples a fully one-sided diffuse-interface IBM (FODIBM) with an explicit wall model for WMLES, and the coupling is genuinely parameter-free. The channel flow validation is thorough and the improvement over conventional DIBM is real and well-demonstrated. The main limitation is that all validations are attached flows, so the key equilibrium assumptions in the tau-model are untested where they matter most. It deserves a serious referee. Here is the detail. What is new: (1) The wall-parallel IB forcing is linked directly to the wall shear stress from the wall model via Eq. (42), with no artificial parameters — this is the central contribution and it is clean. (2) The tau-model from Tamaki and Kawai (sharp-interface IBM) is adapted to the diffuse-interface setting, adding an anisotropic modeled shear stress tensor rather than an isotropic eddy viscosity. This avoids the log-layer mismatch that the mu-model produces, and the comparison in Fig. 4 and Table 2 makes the case clearly: Cf error drops from 8.4% (mu-model) to 0.6% (tau-model) at Re_tau=5200. What the paper does well: The sensitivity studies are the strongest part. Grid convergence (Table 3), reference height (Table 4), inclination angle (Table 5), and Reynolds number sweep (Table 6) all show the method is robust within its regime. The DIBM vs. FODIBM comparison (Tables 7-8) directly demonstrates that one-sided spreading reduces the reference height requirement — DIBM needs d_ref=3.5 dx to reach 1.7% Cf error while FODIBM gets there at 2.5 dx. That is a practical win. The airfoil case at alpha=6.2 degrees gives Cl errors below 5% on both grids, which is reasonable for WMLES at this Reynolds number. Soft spots, in proportion: The tau-model relies on two equilibrium assumptions — constant total shear stress in the inner layer (Eq. 51) and a linear profile of modeled stress from wall to reference height (Eq. 49). Both hold in equilibrium channel flow, and the channel results confirm this. But the airfoil case is also attached flow at moderate angle of attack, so neither assumption is tested in non-equilibrium conditions. The authors acknowledge this and list separated flows as future work, which is the right call. This is a scope limitation, not an internal inconsistency. The tripping parameters (A_t=0.25, delta*=2.4 dx) are empirically chosen without sensitivity analysis, but the Cl results with and without tripping are nearly identical (0.754 vs 0.756), so the global force prediction is not strongly sensitive to these choices. This is minor. No code or data is shipped, which is standard for this subfield but limits reproducibility. Who this is for: Researchers working on IBM-WMLES coupling, particularly those using diffuse-interface approaches on Cartesian grids. The method is implemented in an LBM solver but the formulation transfers to Navier-Stokes solvers. Recommendation: Accept for peer review. The derivation is correct, the validation is appropriate for the stated scope, and the FODIBM-WS coupling is a real methodological advance. A referee should push the authors to be more explicit about the equilibrium assumptions in the tau-model and whether they expect degradation in separated flows, but the paper as it stands is honest about its limitations.","headline":"Solid methodological contribution to diffuse-interface IBM-WMLES coupling. The wall-shear-stress enforcement (Eq. 42) is clean and parameter-free, and the FODIBM one-sided approach genuinely improves on conventional DIBM at low reference heights. The main limitation is scope: everything is validated for attached flows only.","tokens_in":25166,"tokens_out":826,"would_cite":true,"duration_ms":106460,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["47.11.-j","47.27.E-","47.27.em"],"model":"glm-5.2","headline":"One-sided boundary forcing fixes wall-stress prediction in turbulent flow simulations","keywords":["immersed boundary method","wall-modeled large-eddy simulation","diffuse interface","wall shear stress","lattice Boltzmann method","turbulent channel flow","airfoil aerodynamics"],"falsifier":"If applying the WS+tau-model to a flow with massive separation (e.g., a stalled airfoil or backward-facing step) yields skin-friction or velocity-profile errors substantially larger than the sub-1% channel-flow benchmark, the linear stress-profile assumption would be shown to be regime-dependent rather than universal.","tokens_in":24315,"feed_emoji":"🌊","tokens_out":1053,"duration_ms":143597,"temperature":0.7,"pith_summary":"The paper addresses a specific failure mode in computational fluid dynamics: when diffuse-interface immersed boundary methods (DIBMs) are combined with wall-modeled large-eddy simulation (WMLES), the numerical diffusion inherent in DIBM contaminates the near-wall flow field and degrades wall-shear-stress prediction. The authors propose a fully one-sided diffuse-interface immersed boundary method (FODIBM) that confines all interpolation and spreading operations strictly inside the immersed body, eliminating cross-boundary diffusion. They then couple the wall-parallel immersed-boundary forcing directly to the wall shear stress predicted by an explicit wall model, without introducing any artificial parameters. To maintain the total shear-stress balance below the reference height where the grid cannot resolve turbulent fluctuations, they introduce a tau-model that adds a modeled turbulent shear-stress tensor with a linear profile from the wall to the reference height. The combined approach is validated in turbulent channel flow at Re_tau = 5200 (skin-friction error below 1%) and in flow over a NACA23012 airfoil (lift coefficient error below 5%).","feed_headline":"One-sided boundary forcing fixes wall-stress prediction in turbulent flow sims","feed_subtitle":"By confining all interpolation inside the solid body and coupling forcing directly to a wall model, skin-friction errors drop below 1% in ch","key_machinery":"FODIBM: one-sided interpolation/spreading confined to body interior; WS: wall-parallel IB force coupled to wall-model stress; tau-model: anisotropic modeled shear-stress tensor with linear profile wall-to-reference-height","core_discovery":"The central mechanism is the combination of three ingredients: (1) one-sided interpolation and spreading that removes the diffusion contaminating conventional DIBM near walls, (2) direct coupling of the wall-parallel IB force to the wall-model-predicted wall shear stress so that the forcing enforces the correct stress at the boundary, and (3) a tau-model that injects only the shear component of the modeled turbulent stress (rather than an isotropic eddy viscosity) to preserve the total shear-stress balance below the reference height without suppressing wall-normal turbulent mixing. Together these eliminate the log-layer mismatch that plagues conventional approaches and allow accurate WMLESon","pith_inferences":["The linear stress-profile assumption in the tau-model is untested in separated flows, adverse pressure gradients, and transitional regimes; if the actual profile is strongly nonlinear there, accuracy may degrade in exactly the industrial cases (stalled airfoils, bluff bodies) where IBM is most attractive.","The method is validated within a lattice Boltzmann solver, but the authors note it can be extended to Navier-Stokes solvers; whether the one-sided spreading and stress-coupling remain equally accurate on finite-volume or finite-difference discretizations with different numerical dissipation properties is an open question.","The airfoil test uses artificial tripping to trigger transition, which sidesteps the known difficulty of predicting laminar-to-turbulent transition in WMLES; the method's performance for natural transition or leading-edge laminar separation bubbles remains unassessed."],"forward_implications":["The method enables diffuse-interface IBM to achieve accuracy competitive with sharp-interface approaches for WMLES, while retaining the simplicity of Cartesian grids and avoiding complex boundary-cell treatments.","The parameter-free coupling between IB forcing and wall shear stress removes a source of tunable arbitrariness that has limited prior diffuse-interface wall-modeling efforts.","The tau-model's avoidance of log-layer mismatch suggests that anisotropic stress injection is preferable to isotropic eddy-viscosity enhancement whenever wall-normal mixing must be preserved.","The demonstrated robustness across inclination angles up to 45 degrees indicates the method can handle genuinely complex, non-axis-aligned geometries without body-fitted meshes."],"fun_headline_variants":["One-sided diffuse immersed boundary method removes near-wall contamination in WMLES","Keeping boundary forcing inside the solid body fixes wall-stress prediction in LES","Shear-coupled forcing and tau-model eliminate log-layer mismatch in wall-modeled LES","Fully one-sided DIBM improves skin-friction accuracy in turbulent airfoil simulations","Confined interpolation removes cross-boundary diffusion degrading wall-modeled LES"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The tau-model assumes the modeled turbulent shear stress varies linearly between the wall (where it equals the wall shear stress) and the reference height (where it vanishes). This linearity is not derived from first principles and has only been tested in attached channel and airfoil flows; if the true stress profile deviates significantly from linear in separated or strongly pressure-gradient-driven flows, the shear-stress balance enforcement would be incorrect.","fun_headline_variants_meta":{"raw":{"variants":["One-sided diffuse immersed boundary method removes near-wall contamination in WMLES","Keeping boundary forcing inside the solid body fixes wall-stress prediction in LES","Shear-coupled forcing and tau-model eliminate log-layer mismatch in wall-modeled LES","Fully one-sided DIBM improves skin-friction accuracy in turbulent airfoil simulations","Confined interpolation removes cross-boundary diffusion degrading wall-modeled LES"]},"model":"glm-5.2","effort":"low","cost_usd":0.0,"raw_usage":{"total_tokens":737,"prompt_tokens":638,"completion_tokens":99,"prompt_tokens_details":null},"tokens_in":638,"tokens_out":99,"duration_ms":18272,"temperature":1.0,"reasoning_tokens":null,"cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-09T10:35:09.419223+00:00","model_set":{"reader":"glm-5.2"},"falsifier":"If applying the WS+tau-model to a flow with massive separation (e.g., a stalled airfoil or backward-facing step) yields skin-friction or velocity-profile errors substantially larger than the sub-1% channel-flow benchmark, the linear stress-profile assumption would be shown to be regime-dependent rather than universal.","supporting_citations":[],"review_version":1}