{"id":"624a2fa2-09ad-4a0b-bd4b-580422f1c119","arxiv_id":"2411.18735","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Synthetic DKIST/DL-NIRSP observations of the quiet Sun show that standard inversion plus flow tracking recovers only about 50 to 70 percent of the unsigned Poynting flux and badly underestimates the net flux, mostly through missing horizontal shearing motions.","lead":"The authors simulated what the new 4-meter DKIST telescope would see when measuring faint magnetic fields in the Sun's quiet regions, then tested how accurately standard analysis tools recover the energy flowing upward. They found current methods capture only about half to three-quarters of the true magnetic energy flux and miss most of the contribution from horizontal motions.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Abstract's recovery fractions (71.4%/52.6%) do not follow from Table 2's emulation rows (150.5/278.9=54.0%; 36.2/75.7=47.8%); the paper's central quantitative claim is not reproducible as written.","rationale":"The reader's designated weakest assumption was the noiseless emulation. That is a real limitation, but the paper explicitly acknowledges it and defers noise to future work, so it does not by itself invalidate the central claim under the paper's stated idealization. More immediate is an internal reproducibility failure: the abstract's quantitative headline percentages are inconsistent with Table 2 and with the conclusion. Since the claim is precisely a set of percentages, this is the single most load-bearing concern. The proposed check is simple arithmetic: recomputing the emulation recovery ratios from Table 2 and asking the authors to show the calculation behind 71.4%/52.6%. If the corrected fractions are about 54%/48%, the qualitative conclusion—existing schemes recover roughly half of the unsigned Poynting flux in the emulated DL-NIRSP data and miss the net flux—still stands, so a hard rejection would be too strong. The paper has useful independent structure: a realistic MURaM-based emulation and a clean decomposition into emergence and shearing terms, but those strengths do not resolve the number mismatch. The verdict should remain CONDITIONAL, with the condition being a full reconciliation of the abstract, conclusion, and Table 2 recovery fractions.","tokens_in":26491,"tokens_out":6191,"duration_ms":52233,"concrete_test":"Recompute the recovery fractions from Table 2's 'Emulation' rows: R0 = 150.5/278.9 and Rm1 = 36.2/75.7, and compare them with Section 4.2's stated 53.9%, the abstract's 71.4%/52.6%, and the conclusion's 72.5%/61.3%. Ask the authors to identify the exact derivation of 71.4%/52.6% (e.g., a different cadence, a subset of the field of view, or a different ground-truth definition). If the Table-2 ratios are 0.539 and 0.478 and no alternative calculation is shown, the abstract and conclusion must be corrected; if an alternative is shown, it should be added to Table 2 and the text reconciled with it.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The paper's headline deliverable is a quantitative capability estimate: the unsigned Poynting flux recoveries of 71.4% (log τ=0) and 52.6% (log τ=-1). These numbers are not supported by the paper's own Table 2. For the emulated DL-NIRSP observation, Table 2 lists ⟨|Sz|⟩ = 150.5 and 36.2 (in 10^6 erg cm^-2 s^-1) against ground-truth values of 278.9 and 75.7, giving recovery fractions of 53.9% and 47.8%. Section 4.2 states 53.9% for log τ=0, consistent with Table 2 but not with the abstract. The conclusion states 72.5% and 61.3%, also inconsistent with both Table 2 and the abstract. No row/column combination in Table 2 yields 71.4% or 52.6% as an unsigned-Poynting recovery; the nearest values are the net-shearing recoveries of 72.0% and 55.5% reported for the MHD-data (not emulated-observation) run in Section 3.3. The central claim therefore rests on an unreported calculation or a misstatement. This is more load-bearing than the acknowledged noiseless-emulation limitation because it fails under the paper's own noiseless idealization and cannot be checked from the presented evidence. The qualitative direction (underestimation, mainly in the shearing term) is plausible and is corroborated by the sign errors in the net flux in Table 2, but the precise abstract percentages must be reconciled with a stated calculation before the quantitative claim can be accepted.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper assesses the diagnostic capability of DKIST/DL-NIRSP for quiet-Sun magnetism by constructing an end-to-end pipeline: synthetic Stokes profiles from a MURaM simulation are degraded to emulate DL-NIRSP observations, inverted with SIR to obtain vector magnetic fields, processed with the DAVE4VMwDV algorithm to estimate photospheric velocities, and then used to compute vertical Poynting flux. The authors validate the velocity inference on high-resolution simulation data, investigate dependence on cadence, and evaluate the recovery of Poynting flux from the emulated observations. The central quantitative claim is that the unsigned Poynting flux can be recovered at about 71.4% (log τ=0.0) and 52.6% (log τ=-1.0) of the ground truth, with the net flux strongly underestimated and the shearing term identified as the main source of error. The paper also provides scale-based guidance for required velocity resolution and discusses limitations due to the constant-τ surface, azimuthal disambiguation, and the absence of photon noise.","tokens_in":26919,"tokens_out":7647,"duration_ms":60639,"significance":"If the quantitative results were internally consistent, this would be a valuable end-to-end assessment of what DKIST/DL-NIRSP can deliver for quiet-Sun energy transport studies. The pipeline uses an independent MHD simulation as ground truth, explicitly analyzes the limitations of constant-τ magnetograms for flow tracking, and offers concrete guidance on required spatial and temporal scales for recovering the emergence and shearing terms of the Poynting flux. The paper is also transparent about many known limitations, such as the noiseless assumption and the poor performance of the ME0 disambiguation. However, the central recovery percentages are mutually inconsistent across the abstract, main text, and conclusion, and they do not follow from Table 2. Until these numbers are reconciled, the paper's headline quantitative conclusion is not reproducible from the presented evidence.","major_comments":[{"comment":"The net Poynting flux values quoted in the conclusion differ from Table 2: the conclusion lists -1.3×10^7 and -8.5×10^6 erg cm^-2 s^-1, while Table 2 gives -10.5×10^6 and -8.6×10^6. This is another instance of the same numerical inconsistency.","section":"Abstract, Section 4.2, Section 6, Table 2"},{"comment":"The noiseless emulation is an acknowledged but understated limitation. Section 2.2 states 'We did not add noise to the synthetic spectra for this study,' and Section 5.4 defers noise effects to future work. For quiet-Sun Stokes Q, U, and V signals, photon noise is expected to be a dominant error source, so the quoted recovery percentages are optimistic upper limits rather than realistic capability estimates. The abstract and conclusion present these numbers without this caveat. The authors should either add a noise sensitivity analysis (e.g., a single representative noise level) or explicitly qualify the headline numbers as noiseless, idealized recoveries.","section":"Section 2.2, Section 5.4"},{"comment":"The DAVE4VMwDV free parameters (λ, d, dr, w) are optimized on the same emulated observation data that are then used to evaluate the Poynting flux recovery. This in-sample tuning means the reported recovery fractions are not out-of-sample estimates; the L-curve and L1-minimization procedures in Appendix A select parameters that best fit the specific realization. The paper should demonstrate robustness to parameter choices, for example by showing the range of recovery fractions across a plausible parameter grid or by fixing the parameters a priori based on the simulation-data experiments in Appendix A.1.","section":"Appendix A.2"}],"minor_comments":[{"comment":"The definition of E_rel appears malformed: the numerator should be the Euclidean norm of the difference vector and the denominator should be the product of the vector magnitudes, not the dot product as written. As printed, the denominator can be zero or negative, making the metric ill-defined.","section":"Equation (9)"},{"comment":"In the first paragraph, 'Goiet al. 2014, 2016' is missing a space; it should read 'Goi et al. 2014, 2016'.","section":"Section 1"},{"comment":"The text states 'The inferred unsigned Poynting flux is 2.0 × 10^8 ergs cm^-2 s^-1', but Table 2 gives 150.5 × 10^6 = 1.505 × 10^8 erg cm^-2 s^-1. The value 2.0 × 10^8 is inconsistent by about 33%.","section":"Section 4.2"},{"comment":"The percentage for the unsigned shearing term is quoted as 44.6% of the ground truth, while Table 2 implies 95.2/201.1 = 47.3%. Please check the calculation.","section":"Section 4.2"},{"comment":"The phrase 'Here and after' should be 'Hereafter' for clarity.","section":"Section 3.1 and Figure 4 caption"}],"recommendation":"major_revision","confidential_remarks":"The paper relies on DAVE4VMwDV, which is cited as 'P. W. Schuck 2024, in preparation.' This is a co-author's unpublished algorithm, and the reproducibility of the results depends on details that are not fully described in the manuscript. The authors should be encouraged to make the algorithm publicly available or provide a complete mathematical specification. The internal numerical inconsistencies are the primary barrier to acceptance; once those are resolved, the paper's scope and methodology are otherwise appropriate for the journal."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this is a useful instrument-assessment paper with a headline that doesn't survive contact with its own Table 2. The abstract reports 71.4% and 52.6% recovery of unsigned Poynting flux at log tau 0 and -1; the conclusion says 72.5% and 61.3%; Table 2's emulation rows give 150.5/278.9 = 53.9% and 36.2/75.7 = 47.8%. No row or column in the table produces the abstract numbers; the nearest are the net-shearing recoveries from the MHD-data run (72.0% and 55.5%). So the central quantitative claim as written is not reproducible from the presented evidence. That is a fixable but load-bearing error.\n\nWhat is genuinely new: the end-to-end pipeline—MURaM ground truth, SIR synthesis/inversion, DL-NIRSP PSF and resampling, DAVE4VMwDV (the Doppler-constrained variant of DAVE4VM, which is unpublished), and ME0 disambiguation—applied specifically to assess quiet-Sun energy transport. The scale-resolved analysis is the most useful part: the paper shows that velocities at scales d > 150 km recover roughly 80% of the shearing-term Poynting flux, and that the vertical velocity needs d > 600-800 km for the emergence term. The decomposition into emergence vs. shearing terms and the honest discussion of why the shearing term is underestimated (constant-tau slicing, azimuth disambiguation, inversion smoothing) give the qualitative conclusion some teeth: standard schemes miss most of the horizontal-motion contribution and get the net flux sign wrong. That direction is consistent with earlier work on SUNRISE/IMaX data.\n\nThe soft spots are real but mostly acknowledged. No photon noise is added, which matters for the weak quiet-Sun Stokes signals; the authors say they will defer it. The DAVE4VMwDV free parameters are tuned on the same data used for evaluation via L-curves, which is a modest circularity. And the ME0 disambiguation is plainly inadequate for quiet-Sun scales, as they note. None of these undermines the qualitative finding; they just bound how much precision the quantitative claims can carry.\n\nWho this is for: DKIST/DL-NIRSP planning, quiet-Sun heating budgets, and anyone using flow-tracking on high-resolution magnetograms. It deserves a serious referee. Before acceptance, the abstract, conclusion, and Table 2 must be reconciled, and the authors should probably present the noiseless numbers as upper limits.","headline":"Solid forward-modeling assessment of DKIST/DL-NIRSP for quiet-Sun Poynting flux, but the headline recovery fractions don't match the paper's own Table 2.","tokens_in":27515,"tokens_out":3017,"would_cite":true,"duration_ms":26643,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["96.60.-j","95.30.Qd"],"model":"deepseek-v4-flash","headline":"The paper claims that emulated DKIST/DL-NIRSP observations recover only about 71.4% and 52.6% of the ground-truth unsigned vertical Poynting flux at two photospheric heights, and that the net flux is underestimated largely because…","keywords":["quiet Sun","DKIST","DL-NIRSP","Poynting flux","spectropolarimetry","Stokes inversion","flow tracking","solar magnetic heating"],"falsifier":"Rerun the same SIR plus DAVE4VMwDV pipeline on the same MURaM quiet-Sun frames after adding photon noise at the DL-NIRSP design level (e.g., $\\sigma\\approx 10^{-3}$ of continuum intensity) and compare the recovered unsigned and net Poynting flux fractions. If the fractions stay within a few percent of 71.4% and 52.6%, the missing flux is dominated by the underestimated horizontal velocity; if they drop substantially, the noiseless assumption is the controlling limitation.","tokens_in":26317,"feed_emoji":"☀️","tokens_out":14022,"duration_ms":216067,"temperature":0.7,"pith_summary":"The paper asks a practical question: once the 4-m DKIST telescope with its DL-NIRSP spectropolarimeter looks at the quiet Sun, how much of the magnetic energy flowing through the photosphere will the data actually show? To find out, the authors run a realistic MURaM magnetoconvection simulation through the full observational chain—synthesizing Fe I 630 nm Stokes profiles, degrading them to DL-NIRSP resolution, inverting them with SIR, and tracking velocities with DAVE4VMwDV—then compare the inferred vertical Poynting flux with the simulation's ground truth. Their headline result is that the unsigned vertical Poynting flux is recovered at about 71.4% at the photosphere ($\\log\\tau=0$) and 52.6% one scale height higher ($\\log\\tau=-1$), while the net flux is substantially underestimated and can even flip sign. The loss is traced mainly to the shearing term, which depends on horizontal magnetic-field components and horizontal velocities, and which the flow-tracking step systematically under-recovers. This matters because the quiet Sun covers most of the solar surface and its magnetic fields are a candidate energy source for heating the chromosphere and corona, so the answer determines what can be concluded from DL-NIRSP data about the Sun's energy budget.","feed_headline":"DKIST recovers only half to three-quarters of quiet-Sun Poynting flux","feed_subtitle":"Horizontal shearing motions are underestimated, so the net magnetic energy budget is biased low and may even flip sign.","key_machinery":"The machinery is a forward-model pipeline with a built-in ground truth: the MURaM magnetoconvection simulation provides the atmosphere, SIR is used twice (forward synthesis of Stokes profiles and inverse retrieval of vector fields), DL-NIRSP's point spread function and pixel scale degrade the spectra, and DAVE4VMwDV estimates the three-dimensional photospheric velocity by minimizing the residual of the ideal induction equation $\\partial B_z/\\partial t = -\\nabla_h\\cdot(B_z v_h - B_h v_z)$ together with a Doppler-velocity penalty. The vertical Poynting flux is then computed as $S_z = (1/4\\pi)\\int (B_h^2 v_z - (B_h\\cdot v_h)B_z)\\,dS$ and split into an emergence term and a shearing term. The shearing term is the key diagnostic: because it is bilinear in the horizontal magnetic field and horizontal velocity, it is the term most sensitive to azimuth-disambiguation errors and to the systematic underestimation of horizontal flows, and the paper isolates it as the main source of missing flux.","core_discovery":"Working on a quiet-Sun magnetoconvection simulation with a mean flux density of about 120 G, the authors synthesize Stokes profiles of the Fe I 630 nm lines, convolve them with the DL-NIRSP point spread function, rebin to 0.03 arcsec pixels, and invert with SIR to obtain vector magnetic fields at constant optical depth. Velocities come from DAVE4VMwDV, a flow-tracker that fits the ideal induction equation for the vertical field while constraining the line-of-sight component to the measured Doppler velocity. The paper claims that this pipeline recovers 71.4% and 52.6% of the ground-truth unsigned vertical Poynting flux at $\\log\\tau=0.0$ and $\\log\\tau=-1.0$, but that the net vertical Poynting flux is strongly underestimated; in the emulated case at $\\log\\tau=0$ the inferred net flux has the wrong sign ($-1.1\\times10^7$ versus $+3.7\\times10^6\\,\\mathrm{erg\\,cm^{-2}\\,s^{-1}}$). Decomposition into an emergence term ($B_h^2 v_z/4\\pi$) and a shearing term ($-(B_h\\cdot v_h)B_z/4\\pi$) shows the emergence term is recovered almost exactly on simulation magnetograms (99.6% and 88.8% of the net values), whereas the shearing term is the weak link (72.0% and 55.5%), and on the emulated data the signed shearing term drops to roughly 17% of the reference at $\\log\\tau=0$. The horizontal velocity field is the main culprit: DAVE4VMwDV reproduces large-scale (above roughly 150–200 km) flows well, but systematically underestimates flow amplitudes, and the shortfall worsens as cadence degrades, with an effective constraint $\\Delta x/\\Delta t > \\langle v\\rangle$ for reliable velocity estimates. The authors emphasize that photon noise was deliberately omitted, so these fractions should be read as upper limits on what the instrument chain can deliver.","pith_inferences":["Inference: because the pipeline is noiseless, the 71.4% and 52.6% recovery fractions are optimistic ceilings; real DL-NIRSP data with photon noise on weak quiet-Sun polarization will likely recover a smaller fraction, and the net flux sign may be even less trustworthy.","Inference: the paper's scale analysis implies that a correction function could be calibrated from simulation—recovery fraction as a function of pixel size, cadence, and noise level—and applied to real DL-NIRSP observations in future quiet-Sun energy-budget studies.","Inference: the specific failure mode, azimuth ambiguity corrupting $v_h\\cdot B_h$, suggests that multi-height or multi-line inversions that resolve the 180-degree ambiguity without a smoothness prior would be the single highest-leverage upgrade over the current scheme.","Inference: the result generalizes beyond DKIST: any spectropolarimetric quiet-Sun Poynting-flux measurement using similar flow-tracking will be biased low, so comparisons between instruments such as Hinode, SUNRISE, and DKIST should be made on recovery-corrected quantities rather than raw fluxes."],"forward_implications":["For DKIST/DL-NIRSP quiet-Sun campaigns, unsigned vertical Poynting flux measured with SIR plus DAVE4VMwDV should be treated as a lower limit, with the missing fraction concentrated in the shearing contribution from horizontal motions at scales below a few hundred kilometers.","Estimates of the net vertical Poynting flux, and any conclusion about net upward versus downward magnetic energy transport, are not reliable with current schemes: the sign itself can be wrong at photospheric heights.","Velocity products from DAVE4VMwDV are trustworthy only at scales larger than roughly 150–200 km; users should not interpret small-scale flow structure recovered by the algorithm as real.","Cadence planning matters: to keep velocity estimates accurate, magnetogram cadence should satisfy $\\Delta x/\\Delta t > \\langle v\\rangle$, corresponding to roughly 2.5 s or better for 16 km pixels, which DL-NIRSP's several-second cadence can approach but may not always reach.","Improvements that place inversions on constant geometric height, such as magnetohydrostatic inversions or deep-learning methods, can recover an additional 10–20 percentage points of the shearing term, so the missing flux is a recoverable systematic rather than an intrinsic ceiling."],"supporting_citations":[{"why":"Supplies the quiet-Sun magnetoconvection run (O16bM) that serves as the ground truth for magnetic fields, velocities, and Poynting flux.","marker":"Rempel (2014)"},{"why":"Is the MURaM code used to produce the simulation, providing the realistic atmospheric structure the emulation relies on.","marker":"Vögler et al. (2005)"},{"why":"Provides SIR, the Stokes synthesis and inversion engine used to generate and retrieve the emulated spectra.","marker":"Ruiz Cobo & del Toro Iniesta (1992)"},{"why":"Defines the DKIST/DL-NIRSP High-Res mode point spread function, pixel scale, and cadence assumed in the emulation.","marker":"Jaeggli et al. (2022)"},{"why":"Is the original DAVE4VM flow-tracker that is extended with a Doppler constraint to estimate the photospheric velocity field.","marker":"Schuck (2008)"},{"why":"Supplies the minimum-energy azimuth disambiguation whose imperfect solutions degrade the shearing term of the Poynting flux.","marker":"Metcalf (1994)"},{"why":"Provides the SIR node configuration and weights adopted for the inversion, along with context on how noise affects the retrieved magnetic fields.","marker":"Quintero Noda et al. (2023)"},{"why":"Is the real SUNRISE/IMaX quiet-Sun Poynting flux study whose finding of an underestimated shearing term matches this emulation's result.","marker":"Tilipman et al. (2023)"}],"fun_headline_variants":["DKIST sees only half the quiet-Sun energy flow","Quiet-Sun shearing motion hides DKIST's magnetic energy flow","Horizontal flows dominate error in DKIST quiet-Sun Poynting flux","DKIST underestimates quiet-Sun Poynting flux, may flip sign","Horizontal flows are the weak spot in DKIST quiet-Sun energy budget"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that photon-free synthetic Stokes profiles stand in for real DL-NIRSP observations; the paper explicitly omits noise, and quiet-Sun polarization signals are weak, so actual recovery will likely be worse than the quoted fractions.","fun_headline_variants_meta":{"raw":{"variants":["DKIST sees only half the quiet-Sun energy flow","Quiet-Sun shearing motion hides DKIST's magnetic energy flow","Horizontal flows dominate error in DKIST quiet-Sun Poynting flux","DKIST underestimates quiet-Sun Poynting flux, may flip sign","Horizontal flows are the weak spot in DKIST quiet-Sun energy budget"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00128,"raw_usage":{"total_tokens":5431,"prompt_tokens":1344,"completion_tokens":4087,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":960,"completion_tokens_details":{"reasoning_tokens":3990}},"tokens_in":960,"tokens_out":4087,"duration_ms":25874,"temperature":1.0,"reasoning_tokens":3990,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T10:56:18.684413+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Rerun the same SIR plus DAVE4VMwDV pipeline on the same MURaM quiet-Sun frames after adding photon noise at the DL-NIRSP design level (e.g., $\\sigma\\approx 10^{-3}$ of continuum intensity) and compare the recovered unsigned and net Poynting flux fractions. If the fractions stay within a few percent of 71.4% and 52.6%, the missing flux is dominated by the underestimated horizontal velocity; if they drop substantially, the noiseless assumption is the controlling limitation.","supporting_citations":[{"cited_title":"2023, , 675, A93, 10.1051/0004-6361/202345890","cited_arxiv_id":null,"evidence_quote":"Provides the SIR node configuration and weights adopted for the inversion, along with context on how noise affects the retrieved magnetic fields."},{"cited_title":"2023, , 956, 83, 10.3847/1538-4357/ace621","cited_arxiv_id":null,"evidence_quote":"Is the real SUNRISE/IMaX quiet-Sun Poynting flux study whose finding of an underestimated shearing term matches this emulation's result."}],"review_version":1}