{"id":"748c5f6b-4f16-4854-a4c9-c5a524167008","arxiv_id":"1908.04603","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A reduced PDE for clonal vegetation density, including a new |∇n|² term for rhizome growth, is derived from the detailed ABD model and reproduces its pattern-forming phase diagram qualitatively.","lead":"This paper derives a single partial differential equation that describes how clonal plants, such as seagrasses, spread through rhizome growth and form spatial patterns. The new term in the equation captures the signature of clonal reproduction, and the model reproduces the patterns of a more complex, previously published model.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The δ||∇n||² signature rests on a local apex-density closure (Eq. A6) verified on only one stationary pattern; its validity in the moving-front and traveling-pattern regimes where the term is claimed to matter is unsupported.","rationale":"The reader's weakest assumption is exactly the heterogeneous-space validity of the HSS relation N_a = η n_t (Eq. A6), and my reading identifies the same load-bearing point. The central claim is that a single PDE with the new |∇n|² term captures clonal-growth mechanisms and reproduces the ABD phase diagram. For that claim to hold, the reduction of the angular apex dynamics to a local flux must be accurate in the regimes where patterns and fronts actually move. The paper checks this only for one stationary pattern and does not test the closure under time-dependent fronts or traveling stripes, even though those are central applications of the δ term. This does not make me believe the central claim is false; the heuristic equation may well be qualitatively correct. But it is a genuine gap in the derivation, and the conditional verdict is appropriate until the closure is tested in dynamical regimes. I agree with the reader's assessment and recommend no change to the verdict.","tokens_in":12512,"tokens_out":7382,"duration_ms":85517,"concrete_test":"Simulate the full ABD model with the parameters of Ref. [5] in two regimes: (i) a radially expanding patch at the parameters of Fig. 3 (no modulational instability, front motion driven by δ-like effects), and (ii) a traveling stripe at low φ_b analogous to Fig. 4. Along cross-sections through the front or moving pattern, record N_a(r,t)/n_t(r,t) and the first-mode ratio A/||∇n_t||, and compare them with η = ρω_b/(ν + ρω_b) and c0 + c1 n_t where c0, c1 are given by Eqs. (7)-(8). If in the moving front or traveling stripe either local relation deviates by more than 10% in regions of large gradient, then the closure behind Eq. (A13) and the δ||∇n||² term fails exactly where the paper claims clonal-growth effects are strongest.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central derivation of Eq. (1) depends on the closure N_a = η n_t (Eq. A6), which is obtained from homogeneous steady states and then assumed 'valid for all r and t in any heterogeneous spatial distribution' (Appendix, after Eq. A7). The only numerical validation reported is a single stationary pattern, with error less than 10% of apex density. But the paper uses the δ||∇n||² term precisely to control front velocities (Fig. 3) and traveling patterns (Fig. 4). In those regimes the local steady-state balance ω_b = ω_d that produces Eq. A6 does not hold, and the additional closure that the first angular mode obeys a = -(η/π)(c0 + c1 nt)∇n_t (Eq. A12) is calibrated at the Maxwell point, not in propagating or time-dependent states. The angle equation (A10) gives only an asymptotic relaxation to θ = γ + π; whether that relaxation is fast compared with front motion is never checked. Thus the derivation of the flux term, including the new nonlinear term c1||∇n_t||² that is the paper's claimed 'distinct signature' of clonal reproduction, is not established in the very regimes where the signature is asserted to be observable. The 10% error bound is also stated relative to apex density, not to the flux term, so even the stationary check does not directly bound the error in δ.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a single scalar reaction-diffusion PDE, Eq. (1), for vegetation density in clonal-plant meadows, with the new term δ||∇n||² interpreted as the signature of clonal (rhizome) growth. The model is first motivated heuristically by symmetry and positivity requirements, then derived from the microscopic ABD model under a sequence of approximations: angular Fourier truncation, a local closure relating apex density to total density, a truncation of the moment expansion of the nonlocal interaction kernel, and a linear-in-n_t expansion of the saturating interaction. The authors analyze homogeneous steady states, the modulation instability, stationary and localized patterns, front velocities via a perturbative calculation, and traveling patterns for large δ. They claim the reduced model reproduces all qualitative features of the ABD phase diagram and that the δ term accelerates fronts and can induce parity-breaking traveling patterns.","tokens_in":12850,"tokens_out":4768,"duration_ms":47629,"significance":"If the derivation is sound, Eq. (1) provides a valuable minimal descriptor for pattern formation in clonal vegetation, connecting a biologically interpretable coefficient δ to rhizome branching dynamics and making a falsifiable prediction about clonal-growth effects on front propagation. The paper is strong in presenting an explicit derivation chain: the heuristic symmetry argument is independent of the ABD model, the angular-mode truncation is stated as an approximation, and the linear stability and front-velocity calculations are internally consistent and checked against numerical simulations. The perturbative calculation of the δ-induced front-velocity shift is a useful contribution. However, the central new term rests on a closure assumption that is validated only in a single stationary pattern, while the term's claimed effects are demonstrated in moving-front and traveling-pattern regimes where that validation does not directly apply.","major_comments":[{"comment":"The local closure N_a = η n_t (Eq. A6) is derived from the populated homogeneous steady state and then assumed 'valid for all r and t in any heterogeneous spatial distribution.' The only numerical evidence offered is that the maximum error in a stationary pattern is less than 10% of the apex density. This is not sufficient to support the subsequent use of the resulting flux term, which drives the δ||∇n||² signature, in the front-velocity calculation of §V/Fig. 3 and the traveling-pattern regime of Fig. 4. In those regimes the balance ω_b = ω_d that yields Eq. (A6) does not hold. I request either direct numerical tests of the closure in a moving front and a traveling pattern, or a revised claim that restricts the δ signature to stationary configurations.","section":"Appendix, after Eq. (A7)"},{"comment":"The coefficients c0 and c1 are calibrated at the Maxwell point of a stationary front, and the amplitude of the first angular mode is assumed to be a linear function of n_t only, C = c0 + c1 n_t. The relaxation timescale of the angle θ to its fixed point θ = γ + π in Eq. (A10) is not compared with the timescale of front motion, so the quasi-static elimination of the angular dynamics is not established in the regimes where the δ term matters. Moreover, the reported 10% error bound is on the apex density N_a, not on the flux term -ν∇·a' or on δ; even in the stationary test, an error in N_a does not directly bound the error in the gradient-dependent flux that produces δ||∇n||². The quantitative content of the new term therefore remains unsupported away from the Maxwell point. Please provide an error estimate for the flux term or additional numerical checks of C(nt,∇nt) in moving and time-dependent solutions.","section":"Appendix, Eqs. (A12)-(A16)"}],"minor_comments":[{"comment":"The statement that b = 1 and β = -1 are set 'without loss of generality' implicitly requires β < 0. For the rescalings to be real, the sign of β must be fixed; please state the sign condition or discuss the β > 0 case.","section":"Section IV, first paragraph"},{"comment":"The claim that the phase diagram reproduces the ABD model is supported only by a reference to Fig. S3 of Ref. [5]. A direct overlay of the pattern-existence boundaries or a side-by-side comparison would make the claim verifiable within this paper.","section":"Section IV, Fig. 2"},{"comment":"The sentence 'A term containing ∇⁴n could in principle be added to Eq. (1), but it does not add any qualitatively new behavior' is confusing because Eq. (1) already contains β(∇⁴n)n. The authors presumably mean a linear ∇⁴n term; please clarify.","section":"Section II, after Eq. (1)"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe thing to know about this paper is that it does what the title says: it reduces the detailed ABD model of clonal-plant growth to a single PDE for vegetation density, and the reduction produces a genuinely new term, δ|∇n|^2, which the authors interpret as the signature of rhizome growth. That term is absent from earlier dryland vegetation models, and the derivation from the ABD model is explicit enough to be checked. If the reduction holds, it gives the seagrass and clonal-plant community a tractable model for pattern formation.\n\nWhat is good: the derivation is systematic. They Fourier-truncate the angular apex distribution, assume the homogeneous relation Na = η n_t holds locally, close the first angular mode with (c0 + c1 n_t)∇n_t, and expand the nonlocal kernel in moments. They then compute linear stability, front velocities via a perturbative solvability condition, and use continuation to map stationary and localized patterns. The authors are honest about the rough spots: they state that the moment expansion truncated at fourth order loses quantitative accuracy, and they mention the 10% error check on the closure.\n\nThe soft spots are real, and the stress-test note puts the finger where it hurts. The closure Na = η n_t is verified on one stationary pattern only, but the δ term is precisely the one that matters in moving-front and traveling-pattern regimes (Figs. 3 and 4). In those regimes the steady state balance that gives Eq. (A6) does not hold, and the paper never checks that the angular relaxation to θ = γ + π is fast compared to front motion. So the derivation of the signature term is not yet established in the regimes where the signature is asserted to be observable. Also, c0 and c1 are calibrated at the Maxwell point of the same ABD model, so the reproduction of the phase diagram is partly circular—a consistency check, not a parameter-free prediction. A side-by-side comparison with the ABD phase diagram (Fig. S3 of Ref. [5]) is claimed but not shown. These are addressable, not fatal, but they should be fixed before publication.\n\nWho should read this? Anyone working on vegetation pattern formation, especially seagrass or clonal plants, and anyone interested in coarse-graining integro-differential PDEs. The paper deserves a serious referee—the derivation is transparent and the new term is interesting enough to earn scrutiny.\n\nI would send it to peer review, with instructions to the authors to show the phase diagram comparison, extend the closure check to propagating fronts/traveling patterns or soften the claims, and discuss the calibration status of c0, c1.","headline":"Useful reduction of the ABD seagrass model to a single PDE, with a genuinely new |∇n|^2 term, but the signature term's derivation leans on a closure checked on only one stationary pattern.","tokens_in":13342,"tokens_out":2734,"would_cite":true,"duration_ms":27107,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper claims that one scalar PDE, with a gradient-squared term for rhizome elongation, reproduces the full phase diagram and pattern dynamics of a detailed model of clonal plants.","keywords":["vegetation patterns","clonal growth","rhizome","reaction-diffusion equation","pattern formation","modulation instability","seagrass meadows","traveling stripes"],"falsifier":"Run the full angle-resolved ABD model in the regime where the reduced equation predicts traveling stripes (low branching angle, large $\\delta$) and track the local ratio of apex density to total density over time. If that ratio departs systematically from the constant $\\eta$ used in the reduction, the derivation of the $\\delta\\|\\nabla n\\|^2$ term does not carry over to the moving-pattern regime, and the parity-breaking prediction of the reduced model loses its stated microscopic basis.","tokens_in":12283,"feed_emoji":"🌿","tokens_out":7985,"duration_ms":76628,"temperature":0.7,"pith_summary":"The paper tries to establish that a single reaction-diffusion equation for vegetation density can describe the spatial patterns of clonal plants, which spread by horizontal rhizomes rather than by seeds. The decisive new ingredient is the term $\\delta\\|\\nabla n\\|^2$, which the authors identify as the signature of clonal reproduction: it increases local density wherever the vegetation density is spatially nonuniform, mimicking outward rhizome elongation. The authors derive this equation from the fully angle-resolved ABD model through a sequence of approximations, and they show that it reproduces the phase diagram, the sequence of stationary patterns, and the front dynamics of that detailed model. If the claim holds, ecologists studying clonal meadows such as seagrasses can work with one scalar density field instead of tracking the full distribution of rhizome growth directions.","feed_headline":"Clonal plant patterns reduce to one PDE with a gradient-squared term","feed_subtitle":"The gradient-squared term captures rhizome spread, shifting invasion thresholds and making stripes travel.","key_machinery":"The machinery is the single PDE for total vegetation density $n(\\mathbf{r},t)$ together with the derivation chain that connects it to the angle-resolved ABD model. The new term $\\delta\\|\\nabla n\\|^2$ carries the clonal mechanism: unlike linear diffusion $\\epsilon\\nabla^2 n$, which spreads density down gradients, this term feeds growth from the existence of a gradient and therefore pushes vegetation into empty space. To obtain it, the authors truncate the angular Fourier series of apex density to the first mode, assume the homogeneous steady-state relation between apex and total density holds locally (the weakest step), let the phase relax so apex flux points opposite the density gradient, close the first-mode amplitude as $-(c_0+c_1 n)\\nabla n$, and expand the nonlocal interaction kernel in gradients up to fourth order. Each term of Eq. (1) is then expressed in terms of biologically measurable rates: mortality, branching rate and angle, rhizome velocity, shoot spacing, and kernel moments.","core_discovery":"The central claim is that Eq. (1), $\\partial_t n = -\\omega n + a n^2 - b n^3 + \\epsilon\\nabla^2 n + \\alpha(\\nabla^2 n)n + \\delta\\|\\nabla n\\|^2 + \\beta(\\nabla^4 n)n$, is the generic continuum description of clonal-plant pattern formation, with the $\\delta\\|\\nabla n\\|^2$ term being the distinctive mark of rhizome growth. The term emerges from the ABD model because clonal propagation creates a biomass flux directed against the density gradient; expanding that flux gives the gradient-squared contribution alongside ordinary and nonlinear diffusion. The paper verifies that the reduced equation reproduces the qualitative behavior of the full model: subcritical coexistence between vegetated and bare states, a modulation (Turing) instability, the same ordering of negative hexagons, stripes, positive hexagons, and localized structures as mortality varies, and, for sufficiently large $\\delta$, a parity-breaking bifurcation to traveling stripes that the detailed model also shows at low branching angles.","pith_inferences":["Because the derivation relies only on rotational symmetry and low-order expansions, the same equation with $\\delta>0$ might describe other expanding populations whose local growth is promoted by density gradients, such as fungal colonies or clonal animals, with parameters reinterpreted.","The closure $N_a=\\eta n_t$ is the untested hinge for nonstationary dynamics; a direct check in the traveling-stripe regime, which the paper does not perform, would either secure the reduction or force a state-dependent $\\eta$.","The paper's quantitative link to real seagrass meadows remains indirect: one could measure the front-velocity shift and the traveling-stripe threshold in a real meadow and compare the inferred $\\delta$ with the value predicted from branching angle and rhizome speed.","The model suggests that monitoring the motion of vegetation band edges, not just their shape, could reveal the strength of clonal propagation in the field."],"forward_implications":["Modelers of clonal vegetation can drop the angular coordinate of rhizome growth and still get the correct pattern selection and colonization dynamics, provided the coefficient $\\delta$ is set from rhizome speed and branching angle.","Larger $\\delta$ shifts the Maxwell point to higher mortality, so a clonal species should be able to invade bare ground under conditions where a seed-dispersal-only species with identical local dynamics cannot.","At low branching angles the effective $\\delta$ rises, and the model predicts that stationary striped meadows become traveling bands via a parity-breaking instability.","The parameter mapping gives a direct biological reading of each term: mortality minus branching sets $\\omega$, rhizome speed and branching angle set $\\epsilon$ and $\\delta$, and the kernel moments set $\\alpha$ and $\\beta$.","The same sequence of pattern types observed in the detailed model, from bare soil and isolated patches to stripes and labyrinthine patterns, is recovered as mortality is increased."],"supporting_citations":[{"why":"Supplies the fully angle-resolved ABD model that the reduced equation is derived from and must reproduce in phase diagram and dynamics.","marker":"[5]"},{"why":"Provides the discrete growth rules (constant rhizome speed, shoot spacing, branching angle) that underlie the apex and shoot equations.","marker":"[13]"},{"why":"Gives the moment-expansion technique used to turn the nonlocal interaction kernel into the gradient terms with coefficients α and β.","marker":"[14]"},{"why":"Provides the perturbative front-velocity method used to compute the δ-induced velocity shift and the Maxwell-point displacement.","marker":"[16]"},{"why":"Supplies the general pattern-selection theory used to argue that the observed sequence of pattern types matches the full model.","marker":"[2]"},{"why":"Justifies the low-order polynomial-and-gradient expansion that gives Eq. (1) its generic form.","marker":"[6]"}],"fun_headline_variants":["Rhizome growth folds into one PDE with gradient-squared term","Single PDE with gradient-squared term captures clonal plant patterns","Gradient-squared term unifies clonal plant pattern equations","One equation packs rhizome growth and vegetation patterns","A single PDE reproduces clonal plant pattern phase diagram"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument rests on assuming that the balance between rhizome tips (growing points) and total shoots, a constant measured in uniform steady meadows, stays the same locally even when vegetation is patchy and changing over time; the paper checks this only on a single static pattern, finding errors below 10% of the tip density.","fun_headline_variants_meta":{"raw":{"variants":["Rhizome growth folds into one PDE with gradient-squared term","Single PDE with gradient-squared term captures clonal plant patterns","Gradient-squared term unifies clonal plant pattern equations","One equation packs rhizome growth and vegetation patterns","A single PDE reproduces clonal plant pattern phase diagram"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000182,"raw_usage":{"total_tokens":1260,"prompt_tokens":842,"completion_tokens":418,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":458,"completion_tokens_details":{"reasoning_tokens":338}},"tokens_in":458,"tokens_out":418,"duration_ms":4406,"temperature":1.0,"reasoning_tokens":338,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:37:17.434034+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the full angle-resolved ABD model in the regime where the reduced equation predicts traveling stripes (low branching angle, large $\\delta$) and track the local ratio of apex density to total density over time. If that ratio departs systematically from the constant $\\eta$ used in the reduction, the derivation of the $\\delta\\|\\nabla n\\|^2$ term does not carry over to the moving-pattern regime, and the parity-breaking prediction of the reduced model loses its stated microscopic basis.","supporting_citations":[{"cited_title":"Ruiz-Reyn´ es, D","cited_arxiv_id":null,"evidence_quote":"Supplies the fully angle-resolved ABD model that the reduced equation is derived from and must reproduce in phase diagram and dynamics."},{"cited_title":"Lefever and O","cited_arxiv_id":null,"evidence_quote":"Gives the moment-expansion technique used to turn the nonlocal interaction kernel into the gradient terms with coefficients α and β."},{"cited_title":"Lejeune, M","cited_arxiv_id":null,"evidence_quote":"Provides the perturbative front-velocity method used to compute the δ-induced velocity shift and the Maxwell-point displacement."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the general pattern-selection theory used to argue that the observed sequence of pattern types matches the full model."},{"cited_title":"(A2) in the case of α = β = 0","cited_arxiv_id":null,"evidence_quote":"Justifies the low-order polynomial-and-gradient expansion that gives Eq. (1) its generic form."}],"review_version":1}