{"id":"54bdd759-6f88-46eb-a9db-96af058ec968","arxiv_id":"2501.13998","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Isotope ratios in a Palomares submarine canyon sediment core confirm that weapon-grade plutonium from the 1966 accident reached the sea, while AMS and alpha-spectrometry agree on total plutonium levels.","lead":"This paper measures plutonium isotopes in a seafloor sediment core from the Spanish Mediterranean coast and finds that plutonium from the 1966 Palomares bomb accident reached the marine environment. It also shows that a compact, low-energy accelerator mass spectrometer agrees well with alpha-spectrometry for total plutonium concentration.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Isotope-ratio accuracy is the load-bearing assumption; a certified AMS standard check would settle it.","rationale":"The reader's verdict of CONDITIONAL is appropriate: the paper's qualitative conclusion is plausible and consistent with prior Palomares studies, but a conditional element exists. The reader highlighted both the endmember uncertainty and the unvalidated AMS ratio. I agree with the latter's importance, but I identify it as the single most load-bearing concern because a systematic ratio bias would invalidate not just the quantitative apportionment but the fundamental detection of weapon-grade plutonium. The endmember extrapolation (15% from Seville soils) is a legitimate concern for the Pwg percentages and the 84% total, but even if the true local fallout were 12% instead of 15%, the measured ratios (mostly 6-10%) would still lie below it, preserving the qualitative conclusion. In contrast, if the AMS ratio were biased downward by, say, 30%, the entire observed ratio profile would be an artifact. The alpha-spectrometry validation in Fig. 2 is a useful check of total concentration but cannot detect such a ratio-specific bias. The proposed test with a certified reference material is a single, decisive experiment that directly probes the load-bearing assumption. Until that measurement is reported, CONDITIONAL is the correct verdict, not ACCEPT and not REJECT, because the method paper (Ref. [15]) may already have established ratio accuracy; this manuscript simply does not demonstrate it. I therefore recommend no change to the reader's verdict.","tokens_in":6402,"tokens_out":12294,"duration_ms":113112,"concrete_test":"Measure the 240Pu/239Pu atomic ratio of a certified plutonium isotope reference material (e.g., NIST SRM 947 or IRMM-085, which have certified 240Pu/239Pu values) using the identical AMS protocol described in Section 2 (670 kV terminal, same injection/counting sequence, same data reduction). If the measured ratio agrees with the certified value within the combined uncertainties (the stated 2% instrument error plus the certified uncertainty), then the Table 1 ratios are traceable and the conclusion is supported. If the deviation exceeds those uncertainties, the low 240Pu/239Pu ratios in the core samples must be re-examined, and the central claim cannot be sustained without a correction or another validation.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that weapon-grade plutonium reached the marine environment rests entirely on the AMS 240Pu/239Pu ratios in Table 1 and Fig. 3. The paper's only validation of the AMS technique is the comparison with alpha-spectrometry of total 239+240Pu activities (Fig. 2, slope 0.984±0.034, r²=0.9828). That comparison checks chemical yield and gross activity, but it is blind to the isotopic ratio that drives the conclusion. No certified reference material with independently known 240Pu/239Pu (e.g., NIST SRM 947 or IRMM-085) is reported in this manuscript, nor is any cross-check against a second mass-spectrometry method (e.g., ICP-MS). If the sequential-injection AMS at 670 kV has a mass-dependent transmission difference between 239Pu and 240Pu, or a detector energy-window bias, the measured ratios could be systematically biased low, producing apparent deviations from the fallout value even in the absence of Palomares material. The stated 2% instrumental precision is a measure of repeatability, not accuracy, and does not constrain such a bias. The qualitative interpretation and the reported Pwg percentages are both conditional on the truth of the AMS ratio. This is therefore the most load-bearing assumption in the paper, ahead of the endmember extrapolation, which mainly affects the quantitative apportionment rather than the qualitative detection.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports the first plutonium AMS measurements made on the 1 MV compact AMS system at the Centro Nacional de Aceleradores (CNA) using a 670 kV terminal voltage, applied to a 20 cm sediment core (URE) from the Aguas submarine canyon near Palomares, Spain. The authors measure 239+240Pu activity concentrations by both AMS and alpha-spectrometry on twin aliquots and find good agreement (slope 0.984±0.034, r²=0.9828), which they use to validate the AMS technique for total plutonium. The central scientific claim is that the measured 240Pu/239Pu atomic ratios, which average below 11%, are significantly lower than the ~15% expected for global fallout in the region, and thus confirm the presence of weapon-grade plutonium released during the 1966 Palomares accident. Using a two-source mixing model with a 5.8% weapon-grade endmember and a ~15% fallout endmember, the authors estimate that 40-95% of the plutonium in individual core sections, and about 84% of the total core inventory, originates from the accident. The deepest layer (URE-17) is an exception, with a ratio of 13.2±1.1% that yields only a 15.7% accident contribution.","tokens_in":6686,"tokens_out":5137,"duration_ms":45652,"significance":"If the reported isotope ratios are accurate, this is a valuable confirmation that weapon-grade plutonium from the 1966 Palomares accident has been transported to the marine environment, and it demonstrates the feasibility of low-energy AMS for plutonium isotope ratio measurements at environmental levels. The alpha-spectrometry comparison is a genuine independent check for total 239+240Pu activity, and the wide dynamic range of the agreement is a concrete strength. However, the central conclusion rests on the accuracy of the AMS 240Pu/239Pu ratios, and the paper provides no validation of this ratio accuracy against a certified reference material or a second mass-spectrometric technique. The quantitative source apportionment also depends on an extrapolated fallout endmember. These issues are addressable and do not invalidate the qualitative detection, but they must be resolved before the claims as stated can be accepted.","major_comments":[{"comment":"The validation of the AMS technique is based entirely on the comparison with alpha-spectrometry for total 239+240Pu activity (Fig. 2). This comparison is, as the authors state, a check of chemical yield and gross activity, but it is blind to the 240Pu/239Pu ratio that drives the central conclusion. No measurement of a certified reference material with independently known 240Pu/239Pu ratio (e.g., NIST SRM 947 or IRMM-085) is reported, nor is any cross-check against ICP-MS or another mass-spectrometry method. The stated 2% instrumental precision is a measure of repeatability, not accuracy, and does not constrain a possible mass-dependent bias in the sequential-injection, stripping, or detector energy-window stages of the 670 kV AMS system. Such a bias could shift all measured ratios systematically low and produce the same pattern as the presence of weapon-grade plutonium. I request that the authors report the results of a certified reference material measurement or a direct method intercomparison on at least a subset of samples, and temper the claim of definitive applicability accordingly.","section":"§2 (AMS method) and §3 (Table 1, Fig. 2)"},{"comment":"The Pwg percentages are computed from a two-source mixing model referenced to [9], but the equation is not given in the paper, so the reader cannot reproduce the values or propagate the endmember uncertainties. The assumed endmembers are 5.8% for Palomares weapon-grade plutonium (Ref. [1]) and about 15% for local global fallout, the latter extrapolated from Seville soils [11]. Neither endmember is measured in this sediment core. The qualitative conclusion that the core contains non-fallout plutonium is robust because most measured ratios (6-10%) are far below any reasonable fallout value, but the quantitative claim that the accident contributes 40-95% of the activity in individual layers and 84% of the total core inventory is directly conditional on these endmembers. Please state the mixing formula explicitly and provide a sensitivity analysis (e.g., Pwg for fallout ratios of 13% and 18%), or soften the quantitative claims accordingly.","section":"§3 (two-source mixing model, Table 1)"},{"comment":"The statement that 'in all samples this ratio is lower than that expected for fallout plutonium' is not well supported by the deepest sample URE-17, whose measured ratio is 13.2 ± 1.1%. This is within about 1.6σ of the assumed local fallout value of 15%, so the deepest layer does not, by itself, provide significant evidence of a non-fallout source. The aggregate claim, 'on average levels lower than 11%', is supported by the data, but the whole-core claim should be qualified to acknowledge that the lowermost section is statistically compatible with fallout.","section":"§3 (Table 1, Fig. 3; URE-17)"}],"minor_comments":[{"comment":"Line 3 of the Introduction: 'another are of investigation' should be 'another area of investigation'. The sentence beginning 'The first Pu AMS results on Palomares samples were obtained in 2006' would be clearer as 'The first Pu AMS results for Palomares samples were obtained in 2006'.","section":"§1 (Introduction)"},{"comment":"Table 1 does not include rows for some core slices (URE-11 and URE-16, corresponding to approximately 10.5-11.5 cm and 16.5-17.5 cm depth intervals), and the text does not explain why these samples are missing. Please indicate whether these sections were not measured, lost, or omitted for another reason.","section":"Table 1"},{"comment":"The abstract's phrase 'first Pu AMS results for environmental samples' could be misread as the first such results worldwide; clarify that they are the first from the CNA 1 MV AMS system, as stated in the Introduction.","section":"Abstract and §2"},{"comment":"The regression line in Fig. 2 is reported with slope and intercept, but no confidence bands for the fit are shown. Adding them would make the validation plot more informative.","section":"Fig. 2"},{"comment":"The entries marked with * for 238Pu/(239+240)Pu are said to be unmeasurable due to traces of 228Th [14]; a brief sentence explaining the interference would help the reader understand the otherwise unexplained missing entries.","section":"Table 1 notes"}],"recommendation":"major_revision","confidential_remarks":"The paper is a plausible application report of a new low-energy AMS capability at CNA, and the qualitative finding of non-fallout plutonium in the upper core is likely correct. The main technical concern—lack of a certified standard or method intercomparison for the 240Pu/239Pu ratio—is testable and should be within the authors' reach, as such materials are standard in the AMS community. The endpoint extrapolation issue is also addressable with a sensitivity analysis. The paper would be strengthened considerably by these additions; in their absence, the 'definitive proof' language in the conclusions is overstated. The manuscript is within the scope of a nuclear instrumentation or environmental radioactivity journal, and I would encourage revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a competent applied paper with a genuine new dataset and a useful capability milestone. The first environmental Pu AMS results from CNA's 1 MV system are a real step, and the depth-resolved 240Pu/239Pu profile for a Palomares marine core is new. The AS-AMS agreement for total 239+240Pu activity (slope 0.984±0.034, r2=0.98) is solid and properly validated. I give credit where it is earned: the radiochemistry is careful, the mixing model is standard, and the hot-particle heterogeneity in the duplicate aliquots is acknowledged rather than hidden.\n\nThe qualitative conclusion, that weapon-grade Pu from the 1966 accident reached the marine environment, is well supported by the consistently low ratios (6.2–13.2%) and is consistent with earlier alpha-spectrometry work on coastal samples. I would not want to argue the accident signal is absent.\n\nWhere the paper is soft: there is no certified reference material with independently known 240Pu/239Pu reported for the AMS measurements. The alpha-spectrometry comparison only validates gross activity; it is blind to a possible mass-dependent transmission bias in the AMS system. That is a genuine gap. In practice, a 15% fallout ratio would need to be depressed by a third to mimic the core's top-layer values, which would be a large bias, but the paper gives no direct evidence of accuracy. The Pwg percentages inherit this uncertainty as well as uncertainty in the fallout endmember, which is extrapolated from Seville soils rather than measured locally. These are concrete, addressable limitations, not fatal flaws.\n\nIf this is going to peer review, I would send it out and ask the authors to add a reference ratio measurement (e.g., NIST SRM 947 or IRMM-085), ideally a cross-check by ICP-MS on a subset of samples, and a sensitivity analysis on the fallout endmember. The dataset is worth having, and the methodological lessons are directly useful to the AMS community.","headline":"A useful capability milestone and a new environmental Pu dataset, but the AMS 240Pu/239Pu accuracy lacks a direct standard check, so the qualitative finding is solid while the source-apportionment percentages carry more uncertainty than the paper lets on.","tokens_in":7260,"tokens_out":3094,"would_cite":true,"duration_ms":30656,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["29.30.Aj","29.20.Ba","89.60.Gg"],"model":"deepseek-v4-flash","headline":"Measured plutonium isotope ratios in a seafloor core off Palomares show that weapon-grade plutonium from the 1966 accident makes up about 84% of the core's plutonium activity, confirming that the land contamination reached the marine…","keywords":["accelerator mass spectrometry","plutonium isotopes","240Pu/239Pu ratio","Palomares accident","sediment core","alpha-spectrometry","global fallout","source apportionment"],"falsifier":"Analyse the same core slices for $^{241}\\mathrm{Pu}/^{239}\\mathrm{Pu}$ (or its decay product $^{241}\\mathrm{Am}$) while remeasuring $^{240}\\mathrm{Pu}/^{239}\\mathrm{Pu}$ by an independent mass-spectrometry method: the two-source model predicts each slice should fall on the mixing line between the $5.8\\%$ Palomares endmember and the $\\sim 15\\%$ fallout endmember, so a slice off that line, or a ratio the independent method does not reproduce, would expose a third source or a systematic AMS bias and invalidate the $84\\%$ estimate.","tokens_in":6237,"feed_emoji":"☢️","tokens_out":20992,"duration_ms":165361,"temperature":0.7,"pith_summary":"This paper reports the first environmental plutonium measurements made at 670 kV on a compact accelerator mass spectrometry (AMS) system, and it uses them to trace the legacy of the 1966 Palomares nuclear accident into the sea. The key observation is the $^{240}\\mathrm{Pu}/^{239}\\mathrm{Pu}$ atomic ratio in a 20 cm sediment core from the Aguas submarine canyon: the ratios lie below the $\\sim 15\\%$ expected for local global fallout, often below $11\\%$, and approach the $5.8\\%$ fingerprint of the weapon-grade plutonium released in the accident. A two-source mixing model between Palomares plutonium and global fallout attributes roughly $84\\%$ of the core's total plutonium activity to the accident, with layer-by-layer accident fractions mostly between $40\\%$ and $95\\%$. Because twin aliquots measured by $\\alpha$-spectrometry agree with the AMS activity concentrations across a wide range, the paper takes this as validation that the low-energy AMS technique can support environmental plutonium studies. The result is a demonstration that the land contamination did not stay on land: weapon-grade plutonium has been transported into the marine environment.","feed_headline":"Weapon-grade 1966 plutonium dominates Palomares seafloor core","feed_subtitle":"Plutonium isotope ratios put roughly 84% of the core's plutonium activity on the 1966 accident.","key_machinery":"The central object is the $^{240}\\mathrm{Pu}/^{239}\\mathrm{Pu}$ atomic ratio, a source fingerprint that is about $5.8\\%$ for the Palomares weapons plutonium and about $15\\%$ for local global fallout. It is measured by low-energy AMS: a compact 1 MV accelerator run at 670 kV, with $^{242}\\mathrm{Pu}$ as the internal spike, plutonium injected as $\\mathrm{PuO}^-$, stripped to $\\mathrm{Pu}^{3+}$, and counted isotope-by-isotope in a gas ionisation detector. The second piece of machinery is the two-source mixing model, which solves each measured ratio for the accident-derived fraction of plutonium activity once the two endmember ratios are fixed. Agreement between duplicate aliquots and between AMS and $\\alpha$-spectrometry on total $^{239+240}\\mathrm{Pu}$ activity is the evidence that the ratios feeding that model are reliable.","core_discovery":"The paper establishes that plutonium can be measured in environmental sediments at 670 kV on a compact AMS system, and it uses that capability to show that weapon-grade plutonium from the 1966 Palomares accident has reached the marine environment. In the URE sediment core from the Aguas submarine canyon, every measured $^{240}\\mathrm{Pu}/^{239}\\mathrm{Pu}$ atomic ratio falls below the $\\sim 15\\%$ local fallout value, most lie below $11\\%$, and the lowest approach the $5.8\\%$ characteristic of the Palomares weapons material. A two-source mixing model assigns $40\\%$–$95\\%$ of the plutonium activity in most layers, about $16\\%$ in the deepest layer, and roughly $84\\%$ of the total core inventory to the accident. Twin aliquots measured by $\\alpha$-spectrometry agree with the AMS $^{239+240}\\mathrm{Pu}$ activity concentrations over a wide range, which the authors present as validation of the AMS technique; the depth profile itself is homogenised by mixing and punctuated by probable hot particles, so it cannot support a deposition chronology.","pith_inferences":["Extension: measuring $^{241}\\mathrm{Pu}/^{239}\\mathrm{Pu}$ in these slices would test the binary-mixing assumption directly; any sample that departs from the $5.8\\%$-to-$15\\%$ mixing line would uncover a third plutonium source and shift the $84\\%$ estimate.","Extension: the authors identify probable hot particles but do not image them; single-particle analysis of the 2 cm and 5–6 cm layers could show whether the weapon-grade signal arrived as intact fuel fragments or as a dispersed, leached component.","Extension: the same 670 kV AMS capability could measure other actinides, such as $^{236}\\mathrm{U}$, in the same aliquots, giving the land-to-sea transport story a test that does not depend on plutonium chemistry alone.","Extension: the $84\\%$ figure is an inventory for this single core; converting it into a regional statement about how much Palomares plutonium now resides on the Mediterranean shelf would require additional cores and a sediment budget."],"forward_implications":["Environmental plutonium surveys can now be done on a compact low-energy AMS system, with $^{240}\\mathrm{Pu}/^{239}\\mathrm{Pu}$ ratios available at sample sizes where alpha-spectrometry can only give a combined activity concentration.","The Palomares accident's marine footprint is established: sediment accumulating in the Aguas submarine canyon carries a majority of accident-derived plutonium, so land-to-sea transport is a confirmed pathway.","Isotopic fingerprinting makes the accident traceable: any future sample from the area with a $^{240}\\mathrm{Pu}/^{239}\\mathrm{Pu}$ ratio below the local fallout value can be attributed partly to Palomares material, regardless of the total plutonium level.","The homogenised depth profile and hot-particle layers mean this core cannot be used to date the 1966 accident or reconstruct a fallout chronology; its value is source identification and inventory, not stratigraphy."],"supporting_citations":[{"why":"Defines the 5.8% 240Pu/239Pu ratio used as the Palomares weapon-grade plutonium endmember in the mixing model.","marker":"[1]"},{"why":"Supplies the global-fallout 240Pu/239Pu reference value and the two-source mixing model used for the apportionment.","marker":"[9]"},{"why":"Earlier Palomares soil measurements whose about-15% fallout ratio is extrapolated as the local fallout endmember and that first showed ratios below 11%.","marker":"[11]"},{"why":"Describes the TEVA-resin separation chemistry used to prepare aliquots for both AMS and alpha-spectrometry.","marker":"[14]"},{"why":"Characterises the compact 1 MV AMS system and the 670 kV measurement conditions used here.","marker":"[15]"},{"why":"Provide the typical western Mediterranean margin sediment plutonium concentrations against which the core's activities are judged anomalous.","marker":"[17,18,19]"}],"fun_headline_variants":["Compact AMS reveals weapon-grade Pu from 1966 Palomares in seafloor","Palomares 1966 accident source of 84% of seafloor plutonium","AMS at 670 kV validates Pu measurement in marine sediment","Seafloor core ties most plutonium to 1966 Palomares accident","Weapon-grade Pu found in Mediterranean canyon from 1966 accident"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The apportionment rests on two linked assumptions: that every plutonium atom in the core is a mixture of only the Palomares $5.8\\%$ endmember and a $\\sim 15\\%$ local fallout endmember, and that the AMS-measured $^{240}\\mathrm{Pu}/^{239}\\mathrm{Pu}$ ratios are accurate, since the $\\alpha$-spectrometry comparison validates total activity concentrations but not the ratio itself.","fun_headline_variants_meta":{"raw":{"variants":["Compact AMS reveals weapon-grade Pu from 1966 Palomares in seafloor","Palomares 1966 accident source of 84% of seafloor plutonium","AMS at 670 kV validates Pu measurement in marine sediment","Seafloor core ties most plutonium to 1966 Palomares accident","Weapon-grade Pu found in Mediterranean canyon from 1966 accident"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000336,"raw_usage":{"total_tokens":1904,"prompt_tokens":1033,"completion_tokens":871,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":649,"completion_tokens_details":{"reasoning_tokens":772}},"tokens_in":649,"tokens_out":871,"duration_ms":7543,"temperature":1.0,"reasoning_tokens":772,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T15:48:59.958118+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Analyse the same core slices for $^{241}\\mathrm{Pu}/^{239}\\mathrm{Pu}$ (or its decay product $^{241}\\mathrm{Am}$) while remeasuring $^{240}\\mathrm{Pu}/^{239}\\mathrm{Pu}$ by an independent mass-spectrometry method: the two-source model predicts each slice should fall on the mixing line between the $5.8\\%$ Palomares endmember and the $\\sim 15\\%$ fallout endmember, so a slice off that line, or a ratio the independent method does not reproduce, would expose a third source or a systematic AMS bias and invalidate the $84\\%$ estimate.","supporting_citations":[{"cited_title":"Comportamiento ambiental de las partículas de combustible nuclear (fundamentalmente Pu) tras un accidente nuclear en un ecosistema de tipo mediterráneo","cited_arxiv_id":null,"evidence_quote":"Defines the 5.8% 240Pu/239Pu ratio used as the Palomares weapon-grade plutonium endmember in the mixing model."},{"cited_title":"Krey, E.P","cited_arxiv_id":null,"evidence_quote":"Supplies the global-fallout 240Pu/239Pu reference value and the two-source mixing model used for the apportionment."},{"cited_title":"Chamizo, M","cited_arxiv_id":null,"evidence_quote":"Earlier Palomares soil measurements whose about-15% fallout ratio is extrapolated as the local fallout endmember and that first showed ratios below 11%."},{"cited_title":"Chamizo, M.C","cited_arxiv_id":null,"evidence_quote":"Describes the TEVA-resin separation chemistry used to prepare aliquots for both AMS and alpha-spectrometry."},{"cited_title":"Chamizo, S.M","cited_arxiv_id":null,"evidence_quote":"Characterises the compact 1 MV AMS system and the 670 kV measurement conditions used here."}],"review_version":1}