{"id":"a6ddd568-f10f-47fa-b933-e18c266c440c","arxiv_id":"1908.07756","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"A directionally filtered, under-relaxed iteration on vector gravity data recovers shallow density fields accurately and deep density fields only slowly in synthetic tests.","lead":"This paper describes an iterative algorithm that reconstructs underground density from measured gravity components by applying a directional filter. It matters because it is a fresh attempt at an old inverse problem in geophysics, though the two demonstration cases are synthetic and the deep-field results are weak.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (7)'s convergence to the true density is unproven; the update rests on Eq. (6), which the authors state is 'usually false' when multiple density cells are in error, so the asymptotic exactness claim lacks support.","rationale":"The reader's weakest assumption identifies exactly the same load-bearing concern: Eq. (7)'s derivation relies on Eq. (6), which the authors admit is generally false when more than one density cell is inexact, and no convergence proof is supplied. This is the single most important gap because the paper's headline claim of 'asymptotic exact solutions' depends entirely on the iteration's convergence, not merely on the existence of a fixed point. The manuscript does provide some independent support: a concrete discrete forward model, reproducible synthetic experiments, and honest discussion of deep-layer limitations, including explicit statements that convergence rates were not studied. Those elements justify treating the method as a preliminary proposal rather than a proven algorithm. The proposed concrete test—a minimal multi-anomaly case with Jacobian spectral-radius analysis—would empirically and analytically settle whether the multi-cell contamination prevents convergence. Since the reader already conditioned acceptance on adding convergence analysis and comparison with baselines, and my concern is the same point rather than a new fatal flaw, the verdict should remain CONDITIONAL. No change to the reader's judgment is warranted beyond emphasizing that the asymptotic exactness claim should be explicitly retracted or reformulated as a conjecture until such a test is performed.","tokens_in":6751,"tokens_out":2280,"duration_ms":24987,"concrete_test":"Run a minimal synthetic test with, e.g., a 2×2×2 grid (M=N=8), assigning two nonzero density deviations at different depths and zero elsewhere, computing exact gravity by Eq. (4). Iterate Eq. (7) from zero and from near-true initializations with b=2 and the paper's adaptive under-relaxation; record max density error per iteration. Also compute the spectral radius of the Jacobian of the Eq. (7) update map at the true solution. If the density error plateaus above machine precision, or if the spectral radius is ≥ 1, the claimed asymptotic exactness fails for this simple multi-cell case.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that iterative application of Eq. (7) admits asymptotic exact solutions, i.e., lim_{it→∞} δρ_m^{it} = δρ_m^l when δg_n^c = δg_n^l. This requires the fixed point δρ^l to be attracting for the iteration map defined by Eq. (7). But Eq. (7)'s correction term is obtained from Eq. (6), which is a summation of Eq. (5) over all gravity points. Eq. (5) is exact only when a single density deviation differs from the true value; with multiple erroneous cells, the relation is not exact. The paper itself states (Section 2) that 'If however density corrections at different locations are responsible for the gravity corrections, Eq. (6) ceases to be equivalent to Eq. (5) and both of them will usually be false.' Thus the update applied to each cell is contaminated by the other cells' errors, and no argument is given that these contaminations vanish or cancel as the iteration proceeds. A fixed point at δρ^l is trivial, but local stability and global convergence are neither proven nor demonstrated beyond two hand-picked synthetic cases. The paper further concedes 'no study having been followed about convergence rates' and that the filter exponent b=2 was chosen 'apparently' to ensure convergence. The deep-layer density errors, despite tiny gravity residuals, are consistent with a null space or an alternative stable point, not necessarily with slow convergence to the exact solution. Consequently, the asymptotic exactness claim is not established by the manuscript's evidence.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes an iterative gravity inversion method that uses the three Cartesian components of the gravity deviation vector. Starting from a discretized form of Newton's integral, the authors sum over all observation points and apply a directional filter f_nm to obtain a scalar update equation for each density cell. An under-relaxed iteration (Eq. 7) is then used to update the density deviation field, with gravity fields recomputed by the forward model at each step. The method is tested on two synthetic 100-cell models with exact gravity data. In both cases, shallow density deviations are recovered very accurately, while the deepest layer (z = -8750 m) retains errors of tens of kg/m^3 even though the reconstructed gravity field is visually indistinguishable from the exact one. The paper claims that the equations 'admit asymptotic exact solutions' and that deep densities 'apparently converge to the correct values at very low convergence rates.'","tokens_in":7042,"tokens_out":3191,"duration_ms":34521,"significance":"If the central claim were established, the method would offer a simple deterministic alternative to regularized potential-field inversion, with the notable feature of using full vector gravity data. The synthetic experiments are transparently described and the authors are honest about the method's limitations, including the lack of a convergence study. However, the asymptotic exactness claim is not supported: the derivation of the correction equation is explicitly admitted to be false when more than one density cell is simultaneously in error, and the two numerical examples do not demonstrate convergence to the exact deep density field, which is instead consistent with a null-space or alternative stable point. The paper would be a useful contribution if repositioned as a heuristic iterative scheme with documented behavior, but as written it overstates its theoretical grounding.","major_comments":[{"comment":"The derivation of Eq. (6) rests on Eq. (5), which is exact only when a single density deviation differs from its true value. The paper itself states that when density corrections at different locations are responsible for the gravity corrections, Eq. (6) 'ceases to be equivalent to Eq. (5) and both of them will usually be false.' Consequently, in the actual case of multiple erroneous cells, each update in Eq. (7) is contaminated by the other cells' errors. The statement that 'in the limit in which δg^c = δg^l, Eq. (7) accepts lim_{it→∞} δρ^{it+1} = δρ^l' only establishes that the true density field is a fixed point when the gravity residuals vanish; it does not establish that the iteration converges to that fixed point. No local stability or global convergence proof is supplied. This is a load-bearing gap for the abstract claim of asymptotic exact solutions, which therefore is not established by the manuscript.","section":"Section 2, Eq. (6)-(7)"},{"comment":"In both synthetic cases, density errors at the deepest layer remain large (tens of kg/m^3) while the reconstructed gravity errors are extraordinarily small. This behavior is equally consistent with a null-space or near-null-space of the forward operator as with slow convergence to the exact density. The text's conclusion that 'the evolution of the calculated results in the course of the iterative calculations suggests the possibility of exactness at the limit of an infinite number of iterations' is not supported by quantitative evidence. A convergence test is needed, for example a plot of the maximum or RMS density error versus iteration number on a log scale, showing monotone decay to machine precision for the deep layer, or a proof of unique invertibility of the discrete operator on the chosen grid.","section":"Section 3, Cases 1 and 2 (Figs. 3, 4, 7, 8)"},{"comment":"The convergence and uniqueness of the iteration are central to the paper's claims, yet the manuscript states that 'no study having been followed about convergence rates' and that the filter exponent b=2 was chosen because it 'apparently ensures consistent convergence.' Likewise, the under-relaxation factor R is said to vary automatically according to an unspecified error parameter, but the update rule is not defined. Without a sensitivity analysis over b and R, or at least a precise statement of the convergence conditions, the general applicability of the method beyond the two chosen synthetic cases remains unsubstantiated.","section":"Section 3, last paragraph before References"}],"minor_comments":[{"comment":"The sentence 'Eq. (7) is iteratively solved with R automatically varying according to one error parameter based on |δg^l - δg^c|' is too vague for reproducibility; please specify the exact formula used to adapt R.","section":"Section 3, Case 1"},{"comment":"The statement 'b=2, equivalent to use f_nm = α_nm' is inaccurate as written, because α_nm includes the factor -G δV_m in addition to the direction/distance dependence; the relationship is proportionality, not equality, and should be stated accordingly.","section":"Section 2, Eq. (8) discussion"},{"comment":"There are several typographical and style issues: 'Fig.s' appears in Section 2, 'undistinguishable' in Section 3, and equation references are often given as 'Eq.s (7a)' or 'eqs.'; these should be standardized.","section":"Throughout"},{"comment":"The comparison with Monte Carlo states only that 'observed overall convergence rates are similar, at least at the relatively small iteration numbers reached in MC calculations'; this is too vague to be informative and should either be removed or accompanied by quantitative results.","section":"Section 3, Monte Carlo comparison"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is honest about its limitations, which is a point in its favor, but the central claim of asymptotic exactness is not demonstrated. A revision can go in two directions: either add a rigorous convergence/uniqueness analysis and convincing numerical evidence of convergence for the deep layers, or substantially soften the abstract and conclusions to present the method as a heuristic iterative scheme worthy of further study. The current version is likely to be received skeptically by geophysicists familiar with the non-uniqueness of gravity inversion, so the authors should address the null-space interpretation of their deep-layer results explicitly."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a real method paper, not a toy, and the directional-filter idea is new as far as I can tell from the cited literature. But the core claim—that iterating Eq. (7) asymptotically recovers the exact density field—is not supported by the paper's own math. In Section 2 the authors state that Eq. (6) 'will usually be false' when more than one density cell is in error, and Eq. (7) is built directly on Eq. (6). No fixed-point stability or convergence argument is given. The stress-test note is right: the observed deep-layer errors with tiny gravity residuals are just as consistent with a null space or an alternative stable point as with slow convergence to the true field.\n\nWhat the paper does well: it takes full vector gravity data seriously, introduces a smooth direction-and-distance filter f_nm, and shows two synthetic inversions where shallow densities come out very accurate. The authors are admirably honest about the deep-layer weakness and about not having studied convergence rates. The writing is clear enough to follow the derivation even where it is flawed.\n\nThe soft spots are major but addressable. The update equation's invalidity in the realistic multi-cell case is the biggest one; a convergence proof is missing, and the Monte Carlo 'comparison' is just a statement that convergence rates look similar, with no details. There is no test with noise, no real data, and no comparison against a standard method like Li-Oldenburg or a regularized least-squares inversion. The exponent b=2 is hand-picked and equivalent to setting f=alpha, which makes the iteration a diagonally scaled Landweber/SART scheme; the paper does not acknowledge that connection. The 'asymptotic exact solutions' language in the abstract overstates what is shown. The deep-layer resolving power is poor, and the paper says as much, but then still claims the equations admit exact solutions—that tension needs to be resolved.\n\nCitation pattern is fine: standard references, no self-citation problem. The synthetic test design is standard forward-model-then-invert, which is not circular.\n\nWho is this for? Someone working on gravity or potential-field inversion who wants a new iterative update that uses vector components. It deserves a serious referee—not a desk reject—because the idea has merit and the limitations are stated rather than hidden. But as it stands I would not accept it: the convergence claim needs proof or at least a much stronger numerical study, and the method needs to be tested against noise and compared with existing methods.\n\nI'd send it to review with a request for major revision, and I'd be skeptical until the convergence question is settled.","headline":"Clever filter-based gravity inversion with appealing shallow-layer results, but the advertised asymptotic exactness is not established because the key update equation is admittedly wrong in the multi-cell case.","tokens_in":7574,"tokens_out":2237,"would_cite":false,"duration_ms":23432,"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":"Directional filtering makes gravity inversion asymptotically exact","keywords":["gravity inversion","directional filtering","vector gravity data","density deviation inversion","iterative under-relaxation","asymptotic convergence","potential-field inverse problem","synthetic gravity models"],"falsifier":"Generate a synthetic model with a shallow anomaly and a deep anomaly that both contribute to the same gravity observation surface, run Eq. (7) on exact vector gravity until the gravity residual is at machine precision, and continue iterating; if the deep-cell density errors do not approach zero while the residual stays flat, the asymptotic-exactness claim is falsified.","tokens_in":6525,"feed_emoji":"🌍","tokens_out":8179,"duration_ms":75382,"temperature":0.7,"pith_summary":"The paper tries to turn the inverse gravity problem into an iterative correction loop that, given exact vector gravity data, can in principle recover the underground density deviation field exactly in the limit of infinitely many iterations. The key move is to convert the discrete vector equation linking density to gravity into scalar correction equations using a smooth directional filter, so that every gravity component influences every density update. Because the update rule is built so that the density can stop changing only when the calculated gravity field matches the exact one, the method claims asymptotic exact solutions to the inversion problem. Two synthetic tests show that shallow density layers are recovered with very small errors while deep layers converge only very slowly, even though the reconstructed gravity field is visually indistinguishable from the exact field.","feed_headline":"Directional filtering makes gravity inversion asymptotically exact","feed_subtitle":"Vector gravity components drive every density update; shallow layers come out accurate, deep layers converge slowly.","key_machinery":"The central object is the directional filter vector $\\mathbf{f}_{nm}=\\mathbf{r}_{umn}/r_{nm}^b$, chosen with $b=2$, which makes the filter proportional to the forward coupling vector $\\boldsymbol{\\alpha}_{nm}$ between density cell $m$ and gravity point $n$. Its scalar product with the gravity correction vector isolates the component of the gravity-correction field along the line from the cell to the observation point, while the $r^{-b}$ decay gives more weight to nearby cells. This filter converts vector gravity equations into scalar update equations (Eq. (7)) in which every gravity datum contributes to every density correction, and it is the device that lets the method claim convergence of density only when gravity has converged.","core_discovery":"The central claim is that the density deviation field can be computed from presumed exact vector gravity deviations by iterating the under-relaxed scalar update $\\delta\\rho_m^{it+1} = \\delta\\rho_m^c + R\\, \\sum_n \\mathbf{f}_{nm}\\cdot(\\delta\\mathbf{g}_n^l-\\delta\\mathbf{g}_n^c) \\big/ \\sum_n (\\boldsymbol{\\alpha}_{nm}\\cdot\\mathbf{f}_{nm})$, with the gravity field updated afterward by the forward discrete Newton law. The derivation starts from Eq. (5), which is exact when only one density value is wrong; summing over all gravity points with a directional filter $\\mathbf{f}_{nm}$ gives Eq. (6), which the paper acknowledges is usually false when several densities are wrong, and Eq. (7) then treats the resulting scalar as a correction. The claimed property is that $\\delta\\mathbf{g}_n^c=\\delta\\mathbf{g}_n^l$ is the only fixed point of the iteration in the relevant sense, so the method admits asymptotically exact solutions; the synthetic results show very accurate shallow densities and apparent, very slow convergence at depth.","pith_inferences":["Editorial inference: A formal convergence proof would most naturally show that the update is a contraction in a norm where the forward map is coercive; the acknowledged inexactness of Eq. (6) for multiple density errors is the main obstacle.","Editorial inference: A testable extension is to make the filter exponent $b$ and relaxation factor $R$ adaptive per depth or per iteration, since the paper identifies deep-cell convergence as the bottleneck and did not optimize $b$.","Editorial inference: Because the method's failure mode is specifically deep-layer insensitivity, it could be combined with depth weighting or model regularization at depth, although the paper itself does not propose this.","Editorial inference: The same directional-filter scalarization could be applied to gravity-gradient tensor components instead of vector gravity; the cited comparative studies of gradient-tensor components would provide a direct benchmark."],"forward_implications":["With exact vector gravity data and enough observation points, the iteration is intended to recover the density deviation field exactly, but only in the limit of infinitely many iterations.","Shallow density anomalies can be recovered very accurately in practice, while deep density values converge at rates so low that practical runs leave tens of kg/m$^3$ errors.","A gravity field reconstructed from the inverted density can be visually indistinguishable from the exact measured field even when deep densities are substantially wrong, so small gravity residuals do not by themselves certify a correct density model.","The method as presented requires vector gravity data (or gravity-gradient data converted to vectors) and equal numbers of gravity and density unknowns, with interpolation and extrapolation suggested when data are scarce.","The convergence rate depends on the filter-decay exponent $b$ and on the under-relaxation factor $R$; $b=2$ was used because it gives consistent convergence without further intervention."],"supporting_citations":[{"why":"Shows how gravity-gradient tensor components can be used in inversion, the vector-data context this method extends.","marker":"Zhdanov (2004)"},{"why":"Compares inversion results from different gradient-tensor components and vertical gravity, providing the resolution baseline the paper cites.","marker":"Paoletti (2016)"},{"why":"Finds differences among tensor-component inversions and favors combining components, supporting the case for using all vector gravity data.","marker":"Pilkington (2014)"},{"why":"Supplies the Monte Carlo sampling method used as an independent comparison for convergence behavior.","marker":"Mosegaard (1995)"}],"fun_headline_variants":["Vector gravity with directional filter yields asymptotically exact inversion","Directional filtering achieves asymptotic exactness in gravity inversion","New method inverts gravity via directional filter, shallow accurate, deep slow","Gravity inversion with vector filters: asymptotically exact, deep convergence slow","Under-relaxed iterative scheme with directional filter is asymptotically exact"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that repeatedly applying the under-relaxed update (Eq. (7)) drives the density field to the true one even though the underlying correction equation (Eq. (6)) is derived from a single-error identity and is usually false when many density values are wrong; no convergence proof is given.","fun_headline_variants_meta":{"raw":{"variants":["Vector gravity with directional filter yields asymptotically exact inversion","Directional filtering achieves asymptotic exactness in gravity inversion","New method inverts gravity via directional filter, shallow accurate, deep slow","Gravity inversion with vector filters: asymptotically exact, deep convergence slow","Under-relaxed iterative scheme with directional filter is asymptotically exact"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000736,"raw_usage":{"total_tokens":3293,"prompt_tokens":951,"completion_tokens":2342,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":567,"completion_tokens_details":{"reasoning_tokens":2255}},"tokens_in":567,"tokens_out":2342,"duration_ms":113919,"temperature":1.0,"reasoning_tokens":2255,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:57:12.729997+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Generate a synthetic model with a shallow anomaly and a deep anomaly that both contribute to the same gravity observation surface, run Eq. (7) on exact vector gravity until the gravity residual is at machine precision, and continue iterating; if the deep-cell density errors do not approach zero while the residual stays flat, the asymptotic-exactness claim is falsified.","supporting_citations":[],"review_version":1}