{"id":"d72ecc4d-9f62-4b62-8221-e884ef65e241","arxiv_id":"2607.14377","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Recalibrating the CHRONOS and Penza et al. (2024) solar irradiance models against satellite-era TSI removes their claimed 2.1–5.9 W/m² secular trend, leaving changes of roughly 0.1–0.4 W/m².","lead":"This paper re-examines two solar irradiance reconstructions that claimed the Sun brightened by 2.1–5.9 W/m² since the Maunder Minimum and finds those large trends are not supported by satellite observations. After recalibrating the secular amplitude to modern TSI composites, both models give a Maunder-to-1986 change below about 0.5 W/m², supporting the majority view of a modest secular increase.","discovery_kind":"replication","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Satellite-era regression does not pin down the Maunder-to-1986 amplitude: gamma/N is extrapolated far beyond the calibrated omega range, and Table 3's sign convention conflicts with the abstract's 0.1-0.44 W/m2.","rationale":"The reader's weakest-assumption identification is exactly the load-bearing point: the revised secular amplitudes are not independent predictions but regressions whose constant gamma/N is extrapolated from a short, low-dynamic-range calibration interval to the Maunder Minimum. I agree with the CONDITIONAL verdict. The paper's qualitative claim is credible - propagating the original gamma values with OW24/NMU17 gives poor agreement with modern TSI (Table 2), and the independent MHD/quiet-Sun-field constraints argue against multi-W/m2 trends. But the abstract's precise 0.1-0.44 W/m2 range is not supported by the CPMDF fit, whose Table 3 value is actually negative for CHRONOS; the quoted range appears to select the 'Mean' composite and to exclude the ACRIM and CPMDF rows after the fact. Because the reader already assigns CONDITIONAL with moderate confidence, this stress-test does not move the verdict; it sharpens the condition: the central numbers should be framed as model-dependent extrapolations, not measurements.","tokens_in":25285,"tokens_out":8894,"duration_ms":89446,"concrete_test":"Refit Eq. 11 and Eq. 15 twice. (1) Using only annual values from solar minima (1986, 1996, 2009, 2020) instead of all annual points, to remove contamination by the solar-cycle F0(t) term; (2) using a two-parameter secular model that allows gamma (or N) to change for omega values below the lowest value sampled in the satellite era (e.g., a piecewise-linear term activated below the minimum observed omega). If the fitted gamma/N, or the resulting 1700-1986 TSI difference, shifts by more than about 0.3 W/m2 relative to the single-constant fit, the linear extrapolation is not robust and the abstract numbers should be reported only as conditional upper limits.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The quantitative headline (0.1-0.44 W/m2 for CHRONOS, 0.20-0.25 W/m2 for PEA24) is an extrapolation, not a measurement. In Eq. 11, gamma is fitted by regressing the reconstructed TSI against satellite-era composites (Table 1/4), but the same constant gamma is then multiplied by omega(t) at the Maunder Minimum. The calibration interval covers only a narrow, comparatively flat part of omega; the Maunder value lies far outside it. The formal errors in Table 3 (e.g. -0.10 +/- 0.17 W/m2) quantify only the least-squares fit, not the validity of the linear/constant-parameter ansatz or of the OW24/OSF2 secular shape. The paper itself concedes in Section 5 that 'the satellite-era TSI record spans only a limited range of long-term solar variability and therefore may not fully constrain possible centennial-scale changes in the quiet Sun.' The Rempel (2020) and quiet-Sun field arguments independently bound very large trends, but they do not calibrate the residual 0.1-0.4 W/m2 difference. There is also an internal inconsistency: Table 3 lists CPMDF-based CHRONOS 1700-1986 differences as -0.08 to -0.10 W/m2 (i.e., 1700 brighter), while the abstract reports +0.1-0.44 W/m2; the abstract's range appears to rely on the 'Mean' composite and on excluding ACRIM/CPMDF rows post hoc.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reassesses the large secular TSI increases (2.1–5.9 W m^-2 between the 1700 and 1986 minima) claimed by the CHRONOS and PEA24 models. It reformulates CHRONOS to isolate the quiet-Sun secular component, replaces the original cosmogenic-isotope-based modulation potentials with the geomagnetic-data-based OW24 and open-solar-flux series OSF2, changes the smoothing from 22-year averages/EMD to SSA, and fits the secular amplitude parameters γ (CHRONOS) and N (PEA24) to satellite-era TSI composites. The paper finds that the original γ values, when combined with OW24/NMU17, reproduce direct TSI poorly (Table 2: R = 0.25–0.65, RMS > 0.57 W m^-2), while fits to composites yield γ/N near zero, leading to revised secular changes of 0.1–0.44 W m^-2 (CHRONOS) and ~0.2–0.25 W m^-2 (PEA24). It concludes that the original large secular trends arose primarily from improper linking of cosmogenic isotope and neutron monitor records and from the adopted smoothing approach.","tokens_in":25679,"tokens_out":6684,"duration_ms":65533,"significance":"If the central conclusion holds, the paper would help settle a long-standing discrepancy in solar and climate forcing by providing a falsifiable external check on the high-variability CHRONOS and PEA24 reconstructions. The explicit reformulation of CHRONOS into a component with an isolated quiet-Sun secular term (Eqs. 7–11) is a useful methodological contribution. The comparison of published, not refitted, γ values against direct TSI is a legitimate external test, and the paper considers multiple TSI composites, modulation potentials, and smoothing procedures, which strengthens the qualitative conclusion that the original secular amplitudes are too large. However, the quantitative updated secular values are not independent predictions: they are obtained by regressing γ/N to the satellite era and then extrapolating the same constant parameter to the Maunder Minimum. The paper's own Section 5 concedes that the satellite-era record may not fully constrain centennial-scale quiet-Sun changes. The significance of the headline numbers is therefore conditional on a linear, constant-parameter extrapolation that is not directly tested by the data.","major_comments":[{"comment":"The revised CHRONOS secular trend is obtained by fitting the constant γ in Eq. (11) to satellite-era TSI composites and then multiplying by ω(t) evaluated at the Maunder Minimum. The OW24 series covers only 1845–2020, and the satellite-era fit samples a narrow, comparatively flat part of ω(t); the 1700 value lies far outside this range. The formal error bars in Table 3 (e.g., −0.10 ± 0.17 W m^-2) reflect only the least-squares covariance, not the validity of the linear single-parameter ansatz or the uncertainty in the OW24/OSF2 secular shape. The manuscript itself states in Section 5 that 'the satellite-era TSI record spans only a limited range of long-term solar variability and therefore may not fully constrain possible centennial-scale changes in the quiet Sun.' To support the quantitative headline, the authors should either reframe it as explicitly conditional on the extrapolation or","section":"§3.1.3, Eq. (11); Table 3"},{"comment":"There is an internal inconsistency in the reported secular changes. Table 3 lists the CHRONOS 1700–1986 differences for the CPMDF-referenced fits as −0.08 to −0.10 W m^-2 (i.e., 1700 brighter than 1986), while the Abstract and §6 report a positive increase of '0.1–0.44 W m^-2.' The positive numbers appear to come from the 'Mean' composite (and possibly ACRIM) rather than from the full set of fits shown in Table 3. Similarly, for PEA24 with OSF2, Table 3 gives −0.25 and −0.20 W m^-2 for the CPMDF and Mean composites, yet the Abstract claims 'about 0.20–0.25 W m^-2' as a secular increase. The paper should report all composite-dependent results transparently and align the Abstract/§6 quantitative claims with Table 3, or explain explicitly which composite selection supports the stated range and why the others are excluded.","section":"Table 3 vs. Abstract and §6"},{"comment":"For the PEA24 model, most fits to direct TSI composites return negative or zero N (Table 4: e.g., −0.11 ± 0.03 for CPMDF/OSF1; 0.004 ± 0.030 for CPMDF/OSF2). The paper itself notes that negative N is not physically meaningful within the model. Yet §6 summarizes the PEA24 result as 'around 0.20–0.25 W m^-2' secular increase. This value appears to be the absolute value of the negative OSF2/CPMDF and OSF2/Mean entries in Table 3, which correspond to the Maunder Minimum being brighter than 1986. If those fits are unphysical, they cannot be used to support a positive secular increase. The quantitative PEA24 conclusion therefore needs to be either removed or substantially qualified.","section":"§4.2, Eq. (15), Table 4"},{"comment":"The causal claim that the large original trends 'arose primarily from improper linking cosmogenic isotope and neutron monitor records, together with issues in the adopted smoothing approach' is stronger than what the analysis demonstrates. The paper changes three things simultaneously: the modulation potential series, the smoothing method, and (in the updated fits) the free amplitude parameter. The external test in Table 2 shows that the original γ values combined with OW24/NMU17 are inconsistent with direct TSI, but this does not isolate improper linking versus the adopted smoothing versus the physical assumption about the quiet-Sun atmosphere. The authors should soften the causal attribution to what is directly shown: that the original secular component, when combined with modern modulation-potential shapes, is not supported by satellite TSI, and that the best-fitting amplitude within","section":"§6 and Abstract"}],"minor_comments":[{"comment":"The 'Mean' TSI series is used in Tables 1, 3, and 4 but is only loosely defined in Section 2.3 ('the mean, minimum, and maximum TSI values across all the selected composites'). Please specify exactly which composites enter the 'Mean' series and whether ACRIM is included.","section":"§2.1/§2.3"},{"comment":"The text frequently refers to 'Fig. 2.1a)' etc., but the actual figure is labeled 'Fig. 1' with panels a–d. Please correct the internal cross-references.","section":"Figure 1"},{"comment":"The redefinition of φ_min using the 1880–1920 period is a major change relative to the original CHRONOS definition (the Spörer minimum). The implications for interpreting '1700–1986' differences should be stated more explicitly, since the normalization baseline is no longer the deepest minimum of the modulation-potential record.","section":"§3.1.3"},{"comment":"The discussion of SSA edge effects is useful, but the choice to retain 16 years rather than 22 years from the end of the series is justified only by two assumed future scenarios. Please make explicit that the fitted γ values would be slightly smaller under a weaker Cycle 25, and quantify this sensitivity.","section":"Appendix A"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses an important and contested question, and the external test of the original CHRONOS/PEA24 secular amplitudes against direct TSI is a valuable contribution. However, the quantitative headline numbers are not robust as currently presented: they depend on an extrapolation whose validity is conceded in Section 5, and they conflict with the sign of several entries in Table 3 and Table 4. I would be comfortable with publication once the authors either rescope the conclusions to the qualitative claim or substantially qualify the numerical ranges, and after the internal sign/composite-selection inconsistency is resolved. There is no concern about novelty or attribution in my reading."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this paper does a service by showing that the original CHRONOS and PEA24 models, with their published parameters, fit satellite TSI poorly, and that the large secular trends they claim are very likely overstated. I'm convinced of that qualitative conclusion. But the revised quantitative numbers in the abstract (0.1–0.44 W/m2 for CHRONOS, 0.2–0.25 W/m2 for PEA24) should not be treated as measurements; they are extrapolations of satellite-era fits to the Maunder Minimum.\n\nThe reformulation of CHRONOS in Eqs. 4–11 is clean and useful. The comparison in Table 2 is the paper's strongest evidence: applying the original gamma values to OW24/NMU17 and testing against direct TSI gives poor correlations and RMS >0.5 W/m2. That is a legitimate external test, not circular. The paper also deserves credit for being explicit about smoothing artifacts and for showing that the 22-year averaging and EMD are unstable at the edges.\n\nThe soft spots are mostly about the quantitative headline. Gamma is fit over a narrow range of omega(t) in the satellite era, then applied to Maunder-era omega, which is far outside the calibrated range. The same is true for N in PEA24. The paper acknowledges this in Section 5, but the abstract does not carry that caveat. Also, Table 3 shows negative secular changes for the CPMDF fits (-0.08 to -0.10 W/m2), while the abstract reports +0.1–0.44 W/m2; the positive range relies on the 'Mean' composite and ACRIM, which the paper elsewhere treats as problematic. That inconsistency needs fixing. The independent arguments from Rempel 2020 and quiet-Sun field observations are suggestive, not calibrating, and the paper says as much.\n\nThis paper deserves a serious referee. The qualitative conclusion is likely correct and important: the high-variability tail of TSI reconstructions is not supported by direct measurements. The quantitative secular values need more careful error treatment and a clearer presentation of which fits are physically meaningful. I'd send it to review, but I'd ask the authors to reconcile the abstract with Table 3 and to state more prominently that the Maunder projection is an extrapolation, not a fit.","headline":"A genuinely useful reassessment that makes the qualitative case against large CHRONOS/PEA24 secular trends convincing, but the quantitative secular numbers are extrapolations from satellite-era fits and should be read with that caveat in mind.","tokens_in":26224,"tokens_out":2108,"would_cite":true,"duration_ms":20541,"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 2–6 W/m² secular brightening of the Sun claimed by two irradiance models is an artifact; calibrated against satellite data, the real increase is about 0.1–0.44 W/m².","keywords":["total solar irradiance","secular trend","Maunder Minimum","cosmogenic isotopes","modulation potential","quiet Sun","irradiance reconstruction","solar activity"],"falsifier":"Resolve the disagreement among TSI composites: the regression gives γ ≈ −0.01 W/m² with the CPMDF composite but +0.38 W/m² with ACRIM — the inferred secular amplitude changes sign with the composite. A definitive reconciliation of the ACRIM-gap discontinuity using overlapping radiometer data would either confirm γ ≈ 0 or shift it by roughly the size of that discrepancy. A complementary test: inject a known large secular trend into the geomagnetic-based modulation-potential shape, add realistic cycle noise, and run the same regression; if it fails to recover the injected trend, the calibration","tokens_in":25127,"feed_emoji":"☀️","tokens_out":12030,"duration_ms":102141,"temperature":0.7,"pith_summary":"Most solar irradiance reconstructions put the Sun's brightening since the Maunder Minimum at under 1 W/m², but two of them — CHRONOS and PEA24 — claimed 2.1–5.9 W/m². This paper tries to establish that those large secular trends are artifacts, produced by how cosmogenic-isotope records were spliced to neutron-monitor data and by fragile smoothing choices. The paper reformulates both models so the amplitude of the quiet-Sun secular component is a free parameter, then calibrates that parameter against direct satellite measurements of total solar irradiance. Calibrated this way, CHRONOS yields a secular increase of roughly 0.1–0.44 W/m² between the 1700 and 1986 minima, and PEA24 about 0.2–0.25 W/m². If the paper is right, the pre-industrial Sun was not dramatically dimmer than today, and solar forcing of climate over the past three centuries is correspondingly modest.","feed_headline":"Sun's secular brightening is ~0.2 W/m², not 2–6","feed_subtitle":"Satellite data erase the 2–6 W/m² brightness rise two models claimed, shrinking estimates of solar climate forcing.","key_machinery":"The load-bearing device is a reformulation of CHRONOS, F(t) = F0(t) + γω(t)αQS(t), splitting the reconstruction into an active-region part and a quiet-Sun secular term; γ is then treated as a free parameter fixed by linear regression against satellite TSI composites, replacing the model-atmosphere guess of the original implementation. The temporal shape ω(t) comes from the modulation potential, and the key input is a geomagnetic-data-based modulation potential joined with a neutron-monitor-based record, which links cosmogenic-isotope and neutron-monitor data with less bias than before, together with an SSA low-pass filter that avoids the edge artifacts of 22-year binning and EMD smoothing. F","core_discovery":"The central claim is that the secular variability built into the original CHRONOS and PEA24 reconstructions is not supported by the satellite-era TSI record, regardless of which composite is used. The paper isolates CHRONOS's secular term F_QS − F_C = γω(t), where ω(t) is the normalized modulation-potential shape and γ the amplitude of the quiet-Sun irradiance change. Reproducing the original reconstructions requires γ ≈ −1.5 to −2.8 W/m²; regressing the same equation against direct TSI measurements gives γ ≈ −0.01 ± 0.04 W/m² for the CPMDF composite, about two orders of magnitude smaller. For PEA24 the equivalent amplitude parameter N drops from 0.55 to values consistent with zero. The infl","pith_inferences":["If the small secular trend is right, the Sun contributed even less to long-term climate change than current estimates assume, and paleoclimate analogies based on 'grand solar minima' lose most of their forcing basis — a consequence the paper hints at but does not develop.","The calibration may be partly blind: the satellite era covers only a small slice of the modulation-potential range, so a genuine slow quiet-Sun component that happened to be flat over 1978–2026 would escape the regression. A synthetic test — injecting a known large secular trend into the modulation-potential shape and checking whether the regression recovers it — would quantify this blind spot.","The fitted amplitude changes sign depending on the TSI composite (ACRIM gives a positive, physically unphysical γ), so a definitive resolution of the ACRIM-gap controversy would directly tighten or shift the secular estimate.","The isotope-to-neutron-monitor linking problem that inflated CHRONOS and PEA24 may affect other cosmogenic-isotope-based irradiance reconstructions; re-running them with geomagnetic-based modulation potentials would be a cheap consistency check."],"forward_implications":["The secular TSI increase since the Maunder Minimum is below about 1 W/m², too small for direct irradiance to be a major driver of 20th-century warming.","The high-variability family of reconstructions — CHRONOS and PEA24 — no longer supports solar-forcing scenarios that assume 2–6 W/m² secular changes.","Modulation potential and open solar flux should not be used directly as proxies for quiet-Sun irradiance variability, since both include active-region and heliospheric contributions; doing so overestimates the quiet Sun's secular effect.","The quiet-Sun magnetic-field changes of roughly 11–29 G implied by the original models contradict observations, which show no measurable secular quiet-Sun field change over the satellite era.","Long-term irradiance reconstructions should calibrate their secular amplitude against satellite-era TSI measurements rather than prescribing it from model atmospheres for an unobserved pre-industrial Sun."],"fun_headline_variants":["Corrected solar brightening: 0.2 W/m², not 2–6 W/m²","Satellite data slash Sun's secular brightening to ~0.2 W/m²","Flawed isotope-neutron link inflated solar irradiance trend","Maunder Minimum solar rise: <1 W/m²"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The load-bearing premise is that the quiet-Sun irradiance response is linear in the modulation potential with a single amplitude γ (or N) calibrated from the satellite era — which spans only a small fraction of the activity range — and holds unchanged back to the Maunder Minimum; as the paper itself acknowledges, the satellite record may not fully constrain centennial quiet-Sun changes, so if that linear extrapolation or the secular shape of the geomagnetic-based modulation p","fun_headline_variants_meta":{"raw":{"variants":["Corrected solar brightening: 0.2 W/m², not 2–6 W/m²","Satellite data slash Sun's secular brightening to ~0.2 W/m²","Flawed isotope-neutron link inflated solar irradiance trend","Maunder Minimum solar rise: <1 W/m²"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000699,"raw_usage":{"total_tokens":3090,"prompt_tokens":937,"completion_tokens":2153,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":681,"completion_tokens_details":{"reasoning_tokens":2079}},"tokens_in":681,"tokens_out":2153,"duration_ms":17284,"temperature":1.0,"reasoning_tokens":2079,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T02:15:45.891652+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Resolve the disagreement among TSI composites: the regression gives γ ≈ −0.01 W/m² with the CPMDF composite but +0.38 W/m² with ACRIM — the inferred secular amplitude changes sign with the composite. A definitive reconciliation of the ACRIM-gap discontinuity using overlapping radiometer data would either confirm γ ≈ 0 or shift it by roughly the size of that discrepancy. A complementary test: inject a known large secular trend into the geomagnetic-based modulation-potential shape, add realistic cycle noise, and run the same regression; if it fails to recover the injected trend, the calibration","supporting_citations":[],"review_version":1}