Loading evaluation data…
Accelerating erosion of ice-rich Arctic coastlines reflects the coupled action of wave-driven heat transfer and mechanical removal of thawed sediment, yet most numerical models of Arctic coastal erosion either parameterize thermal processes or omit the spatio-temporal heterogeneity of convective forcing at the water-permafrost interface. This study presents thermalFOAM (v1.0), an open-source parallel OpenFOAM® solver. Its two novel features are (i) a face-resolved, wave-aware convective boundary condition that applies a water- or air-side heat transfer coefficient to individual boundary cell faces according to the instantaneous free-surface elevation, and (ii) a thaw-gated ablation scheme that removes computational cells only once a thaw threshold is reached. The model resolves transient heat conduction and phase change using an apparent heat capacity formulation and temperature-dependent constitutive relations for unfrozen water content, thermal conductivity, and heat capacity. The water-surface elevation and convective forcing were prescribed from experimental measurements rather than hydrodynamically resolved; thermalFOAM therefore predicts thaw-limited niche development within the tested wave regime, not the full hydrodynamic-mechanical erosion chain. The solver is validated against twelve laboratory experiments with different wave heights, periods, and ice contents. Thermal performance shows strong agreement with measured temperature evolution (R 2 ≈ 0.90 for thaw timing; mean bias < 1 min), while morphological validation yields Nash–Sutcliffe efficiencies of ∼0.77 for niche deepening rate. Sensitivity analyses demonstrate nonlinear amplification of thermal perturbations through the thaw threshold and highlight strong dependence on mesh resolution. The thermalFOAM model provides a mechanistically grounded, experimentally validated thermal module suitable for integration into broader Arctic coastal modelling frameworks.
The results show that repeated wave fluctuations induce long-term seepage erosion near the riverbank surface, resulting in progressive mass loss and degradation of soil strength.
hypothetical accelerated simulation without a validated physical timescale
Continuous erosion causes undercutting and stress deterioration within the bank, eventually leading to instability of the overhanging riverbank and subsequent block collapse.
qualitative sequence reproduced within a hypothetical simulation, not independently validated field timing or magnitude
The collapsed soil mass is further fluidized and eroded by cyclic wave action, producing repeated cycles of erosion, collapse, and retreat that progressively reshape the riverbank geometry.
hypothetical simulation with prescribed eroded-particle conversion rather than independent validation of fluidization
The numerical results showed good agreement with available analytical, experimental, and numerical reference solutions, demonstrating the robustness and accuracy of the proposed approach for coupled fluid-soil interaction and wave-induced erosion problems.
erosion agreement follows calibration to the same observations, without quantitative uncertainty or convergence assessment
Derived from the full evaluation — not a separate score.
Strengths
The framework adds explicit free-surface water and two-way interaction to the earlier single-continuum approach, allowing eroded bank regions and subsequent water–bank interaction to evolve.
↳ Section 1, Fig. 1; Section 2.1
The paper reports governing equations, constitutive response, discretisation, time integration, and boundary conditions, together with analytical and experimental benchmark comparisons.
↳ Sections 2–4, Tables 1–3, Figs. 3, 7 and 11
The discussion explicitly identifies the idealised geometry, homogeneous soil, prescribed waves, and limited quantitative validation of retreat, failure timing, and pore pressure.
↳ Sections 5.1–5.2
Limitations
The opening abstract reports thermalFOAM Arctic thermal-ablation modelling, while the full-text abstract and body describe hybrid SPH riverbank erosion. This material record-level discrepancy requires resolution before relying on the supplied summary.
↳ Supplied ABSTRACT versus full-text abstract, page 1
The erosion threshold and rate parameter are calibrated to observed profiles that are subsequently presented as validation. Without an independent erosion test, quantitative error assessment, or component ablations, agreement cannot establish the claimed coupled framework's predictive accuracy.
↳ Section 4.3, Table 3 and Fig. 14
The hypothetical field-scale case changes parameters and reports dimensionless time because erosion is artificially accelerated. It illustrates an erosion–collapse sequence but does not establish physical retreat rates or seepage erosion's controlling importance relative to alternative mechanisms.
↳ Sections 2.2, 5.1–5.2 and 6
The explicit free-surface and porous-medium coupling provides a meaningful extension of the earlier SPH framework described in Section 1 and Fig. 1. Detailed formulation and benchmark comparisons support the capability demonstration, but Section 4.3 fits erosion parameters to the observations subsequently used to claim validation. Sections 5.1–5.2 explicitly make the field-scale application hypothetical and accelerated, which limits the long-term controlling-mechanism interpretation in Section 6. The conflicting opening abstract is a separate record-level reliability concern, not evidence that the coherent SPH body demonstrates misconduct.
Nabu’s assessment, alongside the field’s view.
Are you an author of this paper?
Limited2.8
Confidence mediumThe framework meaningfully extends earlier one-way SPH modelling by representing explicit free-surface water, porous-medium response, and evolving eroded regions together. The advance is computational capability rather than independently established field-scale erosion mechanisms.
Equations, discretisation, and analytical and experimental comparisons support assessment, but the erosion benchmark reuses calibration observations and lacks component isolation and uncertainty assessment. The abstract mismatch is separately surfaced as a Red record-level concern and does not impose a reliability cap on this merit-based score.
The full-text progression from formulation to benchmarks and application is coherent, but the supplied opening abstract describes a different study. Within the SPH text, central long-term claims also exceed the accelerated demonstration's evidentiary scope.
The introduction distinguishes morphodynamic, finite-element, and SPH approaches, and the discussion identifies concrete predictive limitations. The conclusion nevertheless assigns long-term controlling importance to seepage erosion without resolving the timescale or alternative-mechanism limitations.
Concerns4 of 4 checks
The supplied opening abstract and the full-text article describe different studies, although the full-text SPH narrative has its own matching abstract. Calibration and accelerated simulation also constrain the interpretation of validation and long-term mechanism claims.
No specific research-conduct concern is established for this computational study. Competing interests are declared, data are offered on request, and no public code release is stated.
Flags: 2 declared / 5 total
58 of 59 checkable references verified
59 references in manuscript
No retraction notice found in Retraction Watch.
Sources: Crossref (incl. Retraction Watch) ✓
Where this paper’s evidence sits on the path from initial observation to real-world use.
Controlled benchmarks and a hypothetical field-scale demonstration establish a preliminary research-use pathway. Accelerated erosion and absent quantitative field validation prevent operational prediction.
AI-generated, human-governed. Something look off? Contact us to request a review.