{"id":"3aa3e86a-8fd7-4219-8c17-3319796ddd8b","arxiv_id":"2608.06199","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A finite-element model couples cooling, viscous lava with a growing elastic surface shell and shows that the elastic shell changes dome deformation from vertical uplift to lateral spreading.","lead":"A new computer model simulates lava flows and domes as a hot, fluid interior wrapped in a growing elastic shell, treating the solid crust as a mechanical layer rather than just a high-viscosity skin. The model shows that an elastic shell makes a dome spread sideways instead of growing upward, a difference that matters for reading deformation data from real volcanoes.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The elastic-versus-viscous dome comparison in Section 5.1 depends on the unphysical G = eta_shell/dt equivalence, so the abstract claim is not yet robust to alternative, physically motivated rheological choices.","rationale":"The paper's strongest claim is the Section 5.1 demonstration that an elastic shell yields more lateral expansion and less vertical uplift than a high-viscosity rind. The model is carefully verified in Sections 4.1-4.3 against manufactured and analytical solutions, and the code is publicly available, which is real independent support. However, the dome comparison is not a verification test; it is a physical demonstration whose meaning depends entirely on how 'elastic shell' is related to the viscous case. The stated relation G = eta_shell/dt makes the solid's Maxwell relaxation time exactly one time step, so the elastic shell is a viscoelastic material whose apparent stiffness changes if dt changes. The simulation time step is not reported in Section 5.1, so the reader cannot assess whether the result is in the elastic short-time regime or the viscous long-time regime. The additional assumption of zero strain in newly solidified material (Section 3.5) is also physically arbitrary; crust that forms from cooling lava likely retains some stress. Together, these two choices could determine the qualitative outcome, and no sensitivity analysis is provided. A different, physically grounded G (e.g., a measured dome lava modulus) or a Maxwell relaxation time longer than dt could plausibly change the deformation pattern. This is a standard 'parameterization affects the headline result' concern, not an internal inconsistency. Separately, Eq. 16a as printed uses exp(...) but the quoted viscosity values (100 Pa s at 1000 C and 1e12 Pa s at 500 C) are consistent only with 10^(...), so the equation is a reproducible-text bug that should be corrected; this is secondary to the central comparison. I agree with the reader's weakest-assumption identification and recommend keeping the verdict CONDITIONAL pending the sensitivity test.","tokens_in":22662,"tokens_out":7770,"duration_ms":74576,"concrete_test":"Re-run the Section 5.1 dome simulation with G fixed at a representative lava shear modulus (e.g., 1 GPa, or derived from measured elastic moduli of dome lavas) while keeping all other parameters equal, and repeat the same run with dt halved and doubled. If the lateral-expansion/vertical-uplift difference between the elastic-shell and high-viscosity-shell cases reverses or vanishes under any of these changes, the central claim is not robust. Also verify that the reported viscosity values in Section 4.3.3 are reproduced by Eq. 16a as printed; if they require interpreting 'exp' as 10^, the equation should be corrected and the affected results re-checked.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The demonstration that an elastic shell produces more lateral expansion and less vertical uplift than a high-viscosity rind is the central claim of the paper. The comparison in Section 5.1 sets the elastic shear modulus to G = eta_shell/dt, explicitly tying a material property to the numerical time step to 'maintain a consistent pressurization.' With this choice, the solid behaves as a Maxwell-like viscoelastic material whose relaxation time tau = eta/G equals the time step, so the difference between the elastic-shell and viscous-shell simulations is governed by dt rather than by any measured lava rheology. Because the time step is not reported for this simulation, the reader cannot determine whether the result reflects genuinely elastic behavior or a numerical artifact of the chosen dt. Compounding this, Section 3.5 assumes newly solidified material has zero elastic strain, so any stress frozen in during cooling is ignored. Neither the sensitivity to dt nor an alternative G assignment (for example, a fixed shear modulus from measured dome lava, or a Maxwell relaxation time much longer than dt) is tested. If either choice were changed, the predicted lateral-versus-vertical deformation pattern could change or disappear, which would invalidate the strongest claim in the abstract.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents VENUSS, a finite-element solver for cooling and solidifying free-surface lava flows and domes, coupling a viscous interior with an elastic shell whose thickness is set by an isotherm. The numerical formulation combines a unified viscous-elastic momentum equation, level-set representation of interfaces, XFEM enrichment, and a BDF2 time integration scheme. Verification is attempted against analytical solutions for lid-driven cavity flow, free-surface relaxation, diffusion, and one-dimensional solidification with temperature-dependent viscosity. The central demonstration compares a dome-like geometry with a high-viscosity shell versus an elastic shell, and reports that the elastic shell produces more lateral expansion and less vertical uplift. Software and input files are made available through GitHub and Zenodo.","tokens_in":23061,"tokens_out":5645,"duration_ms":53679,"significance":"If the central claim is established, the paper would make a useful contribution by showing that the elastic nature of a solidifying lava crust, not merely its high viscosity, changes predicted surface deformation patterns; this matters for interpreting geodetic and morphologic observations of lava domes. The strengths of the paper are the open availability of the code, the use of several externally defined analytical verification tests, and the unified treatment of viscous and elastic regions in a single momentum equation. However, the main physical conclusion currently rests on a rheological equivalence that is tied to the numerical time step and is not tested for sensitivity, so the paper's headline result is not yet robust.","major_comments":[{"comment":"The central dome comparison sets the elastic shell shear modulus by an equivalence with the shell viscosity and the time step; the text says \"product of the shell viscosity and the time step,\" but the dimensionally consistent form is G = eta_shell/dt, which makes the Maxwell relaxation time eta/G equal to the numerical time step. Because the time step is not reported for this simulation, the reader cannot determine whether the predicted difference between the elastic-shell and viscous-shell runs reflects elastic stress transmission or is controlled by dt. I ask the authors to report dt, test sensitivity to dt, and repeat the comparison with at least one physically motivated alternative, such as a fixed shear modulus for dome lava or a Maxwell relaxation time much longer than dt. Without such tests, the abstract claim that an elastic shell causes more lateral expansion and less vertical uplift is not yet supported.","section":"Section 5.1"},{"comment":"The assumption that newly solidified material has zero elastic strain is load-bearing for the dome demonstration because it sets the initial stress state of the entire solidified shell. The manuscript does not discuss whether residual stresses from cooling or from prior deformation of the solidifying front should be present, nor does it test sensitivity to this choice. A justification based on the relaxation time of the material relative to the solidification rate, or a sensitivity test with an alternative initial strain state, is needed before the predicted deformation pattern can be considered robust.","section":"Section 3.5"},{"comment":"The VFT equation is written with a natural exponential, exp(A + B/(T-C)), but the parameter values A=-2, B=2800, C=300 are stated to give 100 Pa s at 1000 C and 10^12 Pa s at 500 C. With the equation as written, the values are about 7.4 Pa s and 1.6e5 Pa s, respectively. If a base-10 logarithm was intended, Eq. (16a) should be mu = 10^(A + B/(T-C)). The same ambiguity affects the \"7 orders of magnitude\" viscosity contrast in the dome simulation of Section 5.1. Please state the intended form explicitly and check all reported viscosity values against it.","section":"Eq. (16a) and Section 4.3.3"},{"comment":"The convergence rates reported for the verification tests are well below the nominal accuracy of the Q2-Q1 elements: the lid-driven cavity test shows only O(h^1/2) to O(h) convergence, and the free-surface test reports between O(h^1/2) and O(h^2). For a smooth manufactured solution and Taylor-Hood elements, the expected rates should be higher, and the suboptimal rates are not explained. The verification section would be substantially stronger if the cause of the reduced order were identified (for example, the linear level-set representation or the XFEM integration) and if at least one test demonstrated the expected higher-order convergence.","section":"Sections 4.1.2 and 4.2.2"}],"minor_comments":[{"comment":"The caption says the error has a local minimum at a value of 0.01, but the x-axis and the text refer to values between 10^-7 and 10^-3; 0.01 lies outside the plotted range.","section":"Figure 8 caption"},{"comment":"The boundary conditions on y=0 and y=L_y contain the incomplete notation \"d u_y = 0\"; the derivative variable should be specified, for example d u_y/d y or d u_y/d x.","section":"Eq. (44)"},{"comment":"The integral notation in the denominator, with the integration variable shown as x-hat but the upper limit written after the integral sign, is hard to read; please define the integration variable and limits more clearly.","section":"Section 4.3.4, Eq. (48)"},{"comment":"The relative viscosity eta_r is listed with units Pa/Pa; since it is a ratio, it should be listed as dimensionless.","section":"Table 1"},{"comment":"The dome comparison would benefit from a table or explicit statement of the numerical setup, including mesh size, time step, total simulated time, the shell viscosity value, and the resulting shear modulus, so that the results can be reproduced and the time-step dependence assessed.","section":"Section 5.1"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is likely within scope for the journal and the code availability is a real strength. However, the central physical claim is currently tied to an arbitrary and untested time-step-based rheological equivalence, and the VFT equation contains an ambiguity that affects reported viscosities. These issues are fixable, but they are load-bearing and require additional simulations or a substantially revised demonstration before the paper can be accepted."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Read this one for the verification suite and the code, not for the abstract claim. VENUSS is a serious piece of modeling: XFEM-level-set coupling of a viscous interior to a growing elastic shell, with analytical tests for lid-driven cavity flow, free-surface relaxation, diffusion, and a Stefan-type solidification front. The convergence studies are honest, including the modest O(h^1/2) to O(h^1) rates on Q2-Q1 elements, and the code is available. That is real, reproducible work.\n\nThe soft spot is exactly where the stress-test note lands. The Section 5.1 dome comparison sets G = η_shell/dt, so the elastic displacement and the high-viscosity rind are compared at a Maxwell relaxation time equal to the numerical time step. That is not a measured lava property, and it means the \"more lateral, less vertical\" result is a statement about the discretization choice unless sensitivity is shown. No alternative G assignment or dt scan appears. The zero-strain assumption for newly solidified material in Section 3.5 is another unconstrained choice that could bias the same result. The central demonstration is therefore not yet robust, even though the model itself is well constructed.\n\nA separate, more elementary problem: the VFT parameters in Section 4.3.3 are A=-2, B=2800, C=300 °C. Plugging in T=1000 °C gives exp(2) ≈ 7.4 Pa·s, not 100 Pa·s, and T=500 °C gives exp(12) ≈ 1.6e5 Pa·s, not 1e12 Pa·s. The stated viscosity range is off by orders of magnitude. This should have been caught in proofreading, and it undermines confidence in the temperature-dependent verification unless the code uses different parameters than printed.\n\nThe citation pattern is fine: the unified formulation is properly attributed to Bordère–Caltagirone and Hübner, and the volcanology context is well referenced. Self-citation is not an issue here.\n\nVerdict: this deserves a serious referee. Send it out, but ask the referees to verify the VFT numbers and to require either a physics-based G or a sensitivity study on dt and the solidification strain assumption before the dome conclusion is accepted. The tool has value beyond that demonstration, but the abstract overstates what the current simulations establish.","headline":"A well-verified unified viscous-elastic FEM solver for solidifying lava, but the headline dome deformation claim rests on an ad hoc G = η/dt equivalence and the printed VFT parameters do not produce the stated viscosities.","tokens_in":23437,"tokens_out":2016,"would_cite":false,"duration_ms":20990,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A unified viscous-elastic model shows that a growing solid crust makes a lava dome spread laterally rather than inflate straight upward.","keywords":["lava domes","lava flows","solidification","viscoelastic crust","finite element method","level set method","XFEM","fluid-structure interaction"],"falsifier":"Repeat the dome experiment with a physically measured shear modulus for lava at the glass transition instead of $G=\\eta_{\\mathrm{shell}}\\Delta t$; if the surface velocity maximum no longer sits about 40 degrees from horizontal, the lateral-expansion result is an artifact of the equivalence, not of elastic stress transfer.","tokens_in":22492,"feed_emoji":"🌋","tokens_out":8008,"duration_ms":71349,"temperature":0.7,"pith_summary":"VENUSS is a finite-element model that treats a cooling lava body as a viscous interior capped by an elastic shell whose thickness is set by the glass-transition isotherm. The demonstration case feeds a hemispherical dome from below and compares two treatments of the solidified rind: a very high viscosity versus a true elastic shell. The elastic shell spreads deformation laterally, moving the locus of maximum surface velocity roughly 40 degrees off vertical, while the viscous rind deforms straight upward above the conduit. The paper's intended upshot is that lava-dome models which only raise viscosity near the surface cannot capture how a coherent crust redistributes stress, which matters for interpreting monitored surface deformation and forecasting breakouts.","feed_headline":"Lava domes spread sideways when their crust is elastic","feed_subtitle":"A cooling shell carries stress sideways, shifting a dome's fastest surface motion about 40 degrees off vertical.","key_machinery":"The load-bearing object is a single momentum equation that unifies viscous and elastic behavior: in fluid regions $\\mu^*=\\eta$ and $\\lambda^*=-\\frac{2}{3}\\eta$ with zero residual stress, while in solid regions $\\mu^*=\\Delta t G/\\alpha_0$ and $\\lambda^*=\\Delta t \\lambda/\\alpha_0$, with a residual stress tensor $\\sigma_0$ built from the previous elastic-displacement history through the BDF2 time scheme. Three level sets track the free surface, the glass-transition isotherm that marks the solidification front, and the basal topography, and an extended finite element method (XFEM, an enrichment that lets property jumps sit inside elements rather than on mesh lines) handles discontinuities across interfaces. Newly solidified material is assumed to start with zero elastic strain, so no residual stress is locked in, and elastic displacement is advanced in an Eulerian frame from the velocity field.","core_discovery":"The central claim is that mechanical coupling through a coherent elastic shell changes the deformation style of a solidifying lava dome, compared with a rind that is merely very viscous. In the model's dome experiment the shell's shear modulus is set equal to shell viscosity times the time step so the two simulations are pressurized consistently; under identical feeding, the elastic shell, coupled to basal topography, carries lateral stress, producing maximum surface velocity oriented about 40 degrees from horizontal and almost no deformation directly above the source, whereas the high-viscosity shell produces maximum vertical velocity above the source. The paper argues that this demonstrates the need for lateral transfer of stress in solid layers to accurately interpret and predict dome deformation.","pith_inferences":["If the lateral-versus-vertical pattern is a real consequence of shell rheology, dome monitoring should weight horizontal displacements and off-vent stations, because a dome fed from below may show almost no vertical motion directly above the conduit.","The comparison sets $G=\\eta_{\\mathrm{shell}}\\Delta t$, which makes the elastic relaxation time equal the numerical time step; testing a Maxwell viscoelastic shell with a physical relaxation time would reveal whether the sideways-spreading pattern survives a more realistic crust rheology.","The zero-residual-stress assumption for new crust neglects thermal contraction stress; including locked-in strain could shift predicted failure zones and the amount of lateral expansion.","Extending the formulation from 2D planar and axisymmetric geometries to full 3D with irregular topography may show even stronger lateral stress transfer."],"forward_implications":["Dome models that capture cooling only by raising surface viscosity will misplace the locus of surface deformation; a coherent elastic shell must be included to reproduce monitored displacement fields.","The von Mises stress field identifies shell regions most likely to fail (about 15 and 55 degrees above horizontal in the demonstration), giving a physics-based starting point for forecasting crust fracture and lava breakouts.","Because the model handles rapid cooling and a thickening crust, it can be applied to submarine, subglacial, and extraterrestrial lavas, where existing flow models are not well calibrated.","Recovering elastic stresses from the strain field lays groundwork for future fracture modeling, including phase-field or discrete-crack approaches and distributed plastic failure in the crust."],"supporting_citations":[{"why":"Supplies the unifying viscous-elastic momentum formulation that lets one momentum equation describe both fluid and solid regions.","marker":"Bordère and Caltagirone (2014)"},{"why":"Provides the monolithic space-time fluid-structure interaction approach from which the unified formulation is adapted.","marker":"Hübner et al. (2004)"},{"why":"Extended finite element method for solidification problems, forming the basis for the enriched interface treatment of the solidification front.","marker":"Chessa et al. (2002)"},{"why":"Fast level set method used to propagate the free surface and maintain it as a signed distance function.","marker":"Adalsteinsson and Sethian (1995)"},{"why":"Earlier model of lava domes as brittle shells enclosing pressurized magma, the physical picture the elastic-shell comparison tests.","marker":"Iverson (1990)"},{"why":"Laboratory-derived constraints on how surface crust controls dome morphology and rheology, motivating the crust treatment.","marker":"Fink and Griffiths (1998)"},{"why":"Linearized analytical solution for a relaxing free surface used to verify the level-set and free-surface solver.","marker":"Rose et al. (2017)"}],"fun_headline_variants":["Elastic lava crust spreads domes sideways","Lava domes grow outward when their crust is elastic","Elastic shell steers lava dome deformation","Cooling elastic crust changes lava dome shape","Lava crust elasticity drives lateral dome expansion"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The demonstration ties the elastic shell's shear modulus to the numerical time step and assumes newly formed crust starts out stress-free; if either choice is wrong, the sideways-spreading result could change.","fun_headline_variants_meta":{"raw":{"variants":["Elastic lava crust spreads domes sideways","Lava domes grow outward when their crust is elastic","Elastic shell steers lava dome deformation","Cooling elastic crust changes lava dome shape","Lava crust elasticity drives lateral dome expansion"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000184,"raw_usage":{"total_tokens":1285,"prompt_tokens":881,"completion_tokens":404,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":497,"completion_tokens_details":{"reasoning_tokens":335}},"tokens_in":497,"tokens_out":404,"duration_ms":4287,"temperature":1.0,"reasoning_tokens":335,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T12:42:51.148237+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Repeat the dome experiment with a physically measured shear modulus for lava at the glass transition instead of $G=\\eta_{\\mathrm{shell}}\\Delta t$; if the surface velocity maximum no longer sits about 40 degrees from horizontal, the lateral-expansion result is an artifact of the equivalence, not of elastic stress transfer.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Earlier model of lava domes as brittle shells enclosing pressurized magma, the physical picture the elastic-shell comparison tests."},{"cited_title":"and Griffiths, Ross W","cited_arxiv_id":null,"evidence_quote":"Laboratory-derived constraints on how surface crust controls dome morphology and rheology, motivating the crust treatment."}],"review_version":2}