{"id":"29e1e28c-f7e9-4e01-a485-ac7f9b62c32a","arxiv_id":"2411.15646","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Long-period (50-467 s) decayless kink oscillations are detected in 4-49 Mm active region loops, forming a separate branch in the loop length-period diagram.","lead":"Using Solar Orbiter EUI images, the authors find long-period (over 50 seconds) oscillations without visible damping in small coronal loops in two active regions, where only short-period oscillations were previously reported. The result suggests that these small loops follow a different period-length relation than large loops, which affects how magnetic fields are estimated from such oscillations.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Long-period detections with only 1.5–2 cycles in ~500 s windows may be fits to background trends; if the longest events fail a ≥2-cycle / red-noise significance test, the 82-event count and branch claim weaken.","rationale":"I read the paper as a detection and statistical characterization study. The strongest claim is that long-period decayless oscillations exist in short active-region loops and form a separate branch in the loop-length versus period diagram. The weakest point in the argument is the reliability of period and amplitude estimates for events with only 1–2 cycles, especially when the period is comparable to the observation window. The reader identified this same concern, and the manuscript itself confirms the prevalence of low-cycle events. This is not a disagreement with consensus; it is an internal detectability issue. The proposed test would settle whether the long-period events survive a minimal cycle count and a red-noise significance threshold. The paper does have independent supporting evidence: the phase-lag analysis in Figure 7 is a genuine check, and the public data and code are a credit. However, that phase-lag analysis is performed on only two loops and does not validate the majority of the long-period sample. If the concrete test removes most long-period events, the central claim weakens substantially; if the events survive, the conditional acceptance should stand. Because the reader already imposed a condition requiring robustness checks on the long-period detections, my read does not change the verdict. I am not asserting the oscillations are artifacts; I am identifying the specific test that would distinguish real oscillations from trend-fitting degeneracies.","tokens_in":19996,"tokens_out":4370,"duration_ms":44849,"concrete_test":"Using the time series behind Table 2, re-analyze all 82 long-period events: compute the number of full cycles in each fitted window and calculate a Lomb-Scargle periodogram significance against an AR(1) red-noise null (or compare Eq. 1 with a pure trend/quadratic model via BIC). Report how many events have at least 2 full cycles and a false-alarm probability below 1%; if the majority of P > 200 s events fail this test, the branch claim in Figure 5 is not supported by the current data.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is the existence of a population of long-period decayless oscillations in short active-region loops, with 82 events above 50 s. The load-bearing premise is that the fitted periods and amplitudes for events with very few cycles are genuine oscillations. Section 3 fits Eq. 1, a sinusoid plus a linear trend, to loop-centroid time series, and the paper states that 40% of the 105 oscillations have between 1 and 2 cycles. The most extreme cases make this concrete: Table 2 entry 71 has P = 467 ± 7 s in a 700 s window (≈1.5 cycles), entry 78 has P = 404 ± 9 s, entry 14 has P = 421 ± 16 s, and entry 37 has P = 434 ± 6 s. With only one full period plus a linear trend term, a long-period sinusoid is strongly degenerate with a secular drift or slow background variation, so the fit alone does not establish oscillation. Moreover, 'decayless' cannot be assessed for 1.5 cycles, and these longest-period events are exactly the ones that define the separate high-period branch in Figure 5. If they are artifacts, the main new population claim loses its support; if only a few survive, the claim reduces from a statistical population to a handful of candidates. The paper's own caveat in Section 4.3 that multiple wave modes and non-wave interpretations are possible does not resolve this, because the existence claim itself depends on the period estimates being reliable.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper analyzes 105 transverse oscillations in short (4-49 Mm) active-region loops observed by the Extreme Ultraviolet Imager (EUI) on Solar Orbiter in two datasets. The authors fit the loop-centroid time series with a sinusoid plus a linear trend (Eq. 1), measure loop lengths from manual footpoint identification, and report periods ranging from 23 to 467 s, of which 82 are classified as 'long-period' (>50 s). They find no significant correlation between loop length and period (cc = 0.07), identify a separate branch in the loop-length vs. period diagram for short loops, derive low kink speeds and magnetic field strengths using standing-kink seismology (Eqs. 2-3), and compare period distributions across active regions, quiet Sun, and coronal holes. The central claim is the existence of long-period decayless kink oscillations in short active-region loops, with implications for the wave mode and driving mechanism of decayless oscillations.","tokens_in":20244,"tokens_out":5948,"duration_ms":54480,"significance":"If the period and amplitude estimates are reliable, the paper makes a valuable contribution: it significantly enlarges the sample of decayless oscillations in short active-region loops, extends the observed period range beyond 200 s (which earlier work on the same dataset excluded), and challenges the simple standing-kink scaling by reporting a distinct branch in the L-P diagram. The high-cadence EUI data and the availability of the parameter table online (GitHub link in Section 4) are strengths, as is the explicit discussion of alternative wave modes and non-wave interpretations in Section 4.3. The principal risk is that the longest-period events, which define the new branch, are fitted with only 1-1.5 cycles in short EUI time series, so the existence claim rests on the reliability of those fits.","major_comments":[{"comment":"The long-period detections that anchor the central claim are fit with a sinusoid plus a linear trend over windows containing only 1-2 cycles. For example, Table 2 entries 71 (P=467±7 s), 78 (P=404±9 s), 14 (P=421±16 s), and 37 (P=434±6 s) have observation windows of roughly 500-700 s, i.e., about 1-1.5 periods. With only one full period, the sinusoid term in Eq. (1) is degenerate with a slow background drift, and the fit alone does not establish that the signal is a periodic oscillation. The paper itself states in Section 3 that 40% of the 105 events have between 1 and 2 cycles, so this concern is not limited to a few outliers. I request a quantitative robustness check: either a red-noise significance test (e.g., Monte Carlo generation of a linear trend plus noise on the same time sampling, comparing the fitted long-period power to the null distribution) or a minimum-cycle criterion (≥2, preferably ≥3) for an event to enter the long-period sample. Without such a test, the 82-event count and the separate branch in Figure 5 are not adequately supported.","section":"Section 3, Eq. (1), Table 2, Appendix Fig. 11"},{"comment":"The claimed separate branch in the loop-length vs. period diagram is defined largely by the long-period events with P>200 s and L<50 Mm (Figure 5, current-work points). If those events are manifestations of background variability or loop interactions rather than true oscillations, the branch inference collapses. The paper should also quantify the selection effects: the EUI time series are much shorter than the AIA sequences that produced the long-loop branch, so the number of observable cycles for a given period differs strongly between datasets. A comparison of detection biases (e.g., the number of cycles available as a function of period and window length for each instrument) would help demonstrate that the separate branch is not an artifact of combining heterogeneous datasets.","section":"Section 4.1, Figure 5"},{"comment":"The title and abstract characterize all 105 events as 'decayless,' but the paper only asserts in Section 3 that oscillations with more than two cycles do not show notable decay, without a quantitative measure. For the 40% of events with 1-2 cycles, a decayless status cannot in principle be assessed. Please provide a quantitative decay statistic (e.g., the amplitude ratio between the first and second half of the time series, or a fit with an exponential damping term compared to a constant-amplitude fit) for all events, or explicitly restrict the term 'decayless' to a subsample with ≥3 cycles and revise the title and abstract accordingly. This is important because the claim of 'decayless' behavior is a central part of the paper's contribution.","section":"Section 3, title and abstract"}],"minor_comments":[{"comment":"The text states that 'The period distribution in quiet Sun and coronal holes significantly differ from each other in the region of short periods (<50 s)', but Figure 10 compares the combined quiet Sun and coronal hole distribution with the active region distribution. Please clarify whether the statement refers to a direct QS vs. CH comparison or to the combined sample, and report the statistical test used (e.g., a KS test).","section":"Section 4.5, Figure 10"},{"comment":"The caption says that filled triangles represent oscillations in quiet Sun regions, while the text and the legend imply that the triangles also include coronal holes. Please make the caption consistent with the text.","section":"Figure 5 caption"},{"comment":"Several entries report zero uncertainty in the period (e.g., entry 10, '25 ± 0'), which likely reflects rounding to the nearest second. Please use a minimum uncertainty floor (e.g., 0.5 s) or report one more decimal place so that the uncertainties are not shown as exactly zero.","section":"Table 2"},{"comment":"The sentence 'The slit positions for the first dataset are approximately similar to the slit locations used in the study of Li & Long (2023)' is ambiguous: it is unclear whether the slits were placed independently on the same loops or at identical coordinates. Please specify the degree of overlap.","section":"Section 3"},{"comment":"Table 1 lists the field of view in Mm² and the plate scale in km; the units are inconsistent with the common practice of giving the plate scale in km/pixel. Please add the pixel size or specify the plate scale unit explicitly.","section":"Section 2, Table 1"},{"comment":"In Eq. (3), the notation 'ρ_i em' appears malformed; it should presumably be ρ_i (the loop density) and μ_0 (the vacuum permeability). Please correct the typesetting.","section":"Section 4.4, Eq. (3)"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses a timely and interesting question, and the EUI dataset is well suited to it. The main risk is the reliability of the longest-period detections; I would advise the editor that the authors should be asked to provide a red-noise significance test or a minimum-cycle criterion before the existence of the long-period population is accepted. The online availability of the parameter table is a good practice, but archiving the reduced centroid time series and the fitting code would further strengthen reproducibility. The paper's own caveats about alternative wave modes are appropriate and do not by themselves resolve the few-cycle concern."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know this paper reports something new: 105 decayless transverse oscillations in short active-region loops from EUI, 82 of them with periods above 50 s, including several above 300 s and one at 467 s. The same first dataset was previously analyzed by Li & Long (2023), who found only periods below 185 s and a strong loop-length–period correlation. This paper finds essentially no correlation (cc = 0.07 ± 0.10) and argues that short active-region loops follow a separate branch in the length–period diagram, similar to quiet-Sun and coronal-hole short loops. If right, that matters for wave-mode identification and for seismology, since the derived magnetic fields come out low.\n\nWhat the paper does well: the centering via Gaussian fitting with propagated intensity uncertainties is careful, the comparison with Li & Long is explicit and reasonable, and the phase-lag check on two loops is a good attempt to test the standing-wave interpretation. The authors are also honest about the limits. Section 4.3 lists several non-standing alternatives, and Section 4.4 explicitly says the kink-speed formula only applies if these are standing kink modes. The data table is on GitHub, which helps.\n\nThe soft spot is real and load-bearing. Forty percent of the fits have only 1–2 visible cycles, and the longest periods are the worst cases: 467 s in a ~700 s window, 404–442 s in similar windows. With a linear trend plus a sinusoid, a one-cycle oscillation is nearly degenerate with a slow drift. The paper does not report a minimum-cycle-count criterion or a significance test against red noise, so the long-period end of the distribution—the part that defines the branch—is the least certain. The kink-speed cutoff analysis (correlation rising to 0.7 above 400 km/s) is also post-hoc and should be flagged as exploratory or corrected for multiple cutoffs.\n\nThat said, I don't think the central claim is wrong. The authors already state the caveats, and several showcase events look genuine. But the field should not treat the 82-event population as established until the long-period detections survive a >=2-cycle cut and a red-noise test. The paper is worth a serious referee and likely a revision, not a desk rejection.\n\nFor you: if you work on decayless oscillations or coronal seismology, read it and cite it. I'd bring it to reading group and would send it to review with the request that the robustness analysis be added.","headline":"A genuinely new sample of long-period decayless oscillations in short active-region loops, with a plausible central claim that needs a robustness pass on the one-cycle detections before the population and the 'separate branch' interpretation are taken as established.","tokens_in":20869,"tokens_out":1966,"would_cite":true,"duration_ms":21692,"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":"This paper reports 105 decayless kink oscillations in short active-region loops (4–49 Mm), 82 with periods above 50 s and up to 467 s, and argues they form a separate branch in the loop-length–period relation.","keywords":["decayless kink oscillations","active region loops","coronal seismology","loop length-period relation","Solar Orbiter EUI","MHD waves","wave excitation mechanisms","coronal heating"],"falsifier":"Re-run the detection on the same EUI time series requiring at least three full cycles and a significance threshold against red noise using a wavelet or Lomb-Scargle periodogram; if most events with periods above about 200 s fail, the separate branch and the null length–period correlation would not survive. A secondary check would test whether the longest-period events repeat coherently in a second time window on the same loop.","tokens_in":19753,"feed_emoji":"☀️","tokens_out":9216,"duration_ms":77613,"temperature":0.7,"pith_summary":"Decayless kink oscillations are transverse motions of coronal loops that persist without losing amplitude, and they are a candidate channel for energy supply to the corona. This paper searches for them in short loops (4–49 Mm) inside two active regions using high-resolution Solar Orbiter EUI observations and reports 105 oscillations, 82 with periods longer than 50 s and several exceeding 200 s, up to 467 s. The key result is that loop length and period are essentially uncorrelated in this sample ($\\mathrm{cc}=0.07\\pm0.10$), which stands against the strong length–period correlation previously found both for long active-region loops and for short active-region loops sampled below 200 s. The authors interpret the long-period events as a separate branch of the loop-length–period diagram, similar to the branch already reported for short loops in the quiet Sun and coronal holes, and discuss driver-controlled periods rather than standing eigenmodes. A sympathetic reader would care because if the period is set by the driver rather than the loop, the energy flux and seismological inferences drawn from these oscillations change.","feed_headline":"Short coronal loops oscillate for up to 467 seconds without fading","feed_subtitle":"82 long-period events in 4–49 Mm loops break the loop-length–period scaling seen in longer loops, hinting the driver sets the rhythm.","key_machinery":"The carrying objects are decayless kink oscillations, defined here as transverse coronal-loop displacements that show no notable amplitude decay over more than two cycles. They are measured by placing artificial slits across a loop in EUI images, fitting a Gaussian to the loop's intensity profile at each time step to track its centroid, and fitting the centroid time series with a sinusoid plus a linear trend; loop length is estimated from footpoint positions under a semicircular assumption, with an assumed roughly 40% uncertainty. The theoretical anchor is the standing kink mode relation $C_k = 2L/P$, and its conversion to magnetic field via $B = C_k \\sqrt{\\frac{1+\\zeta}{2}}\\sqrt{\\mu_0 \\rho_i}$, which the paper applies cautiously. The decisive observational tool is the loop-length–period diagram: long loops define a strong correlation, while the short-loop branch reported here occupies long periods at small lengths. The paper also uses cross-correlation between oscillation signals at two slit positions to measure phase lag, finding $0\\pm0.01$ s and $0\\pm0.02$ s, and uses a kink-speed cutoff to show the length–period correlation strengthens when the slowest events are removed.","core_discovery":"The central claim is that long-period decayless kink oscillations exist in short active-region loops. Using time-distance maps from 3 s cadence EUI data, the authors measure 105 oscillations in loops with lengths 4.1–49 Mm; 82 of these have periods above 50 s, the average period is $151\\pm107$ s, and the longest detected period is 467 s. In the loop-length–period plane these short-loop points do not follow the linear scaling established for loops of hundreds of Mm; the correlation is $0.07\\pm0.10$, and a kink-speed cutoff analysis shows that only events with kink speeds above about 400 km/s recover a length–period correlation near 0.7. The zero phase lag measured at two loop positions leaves standing waves a viable interpretation, but the absence of a length–period correlation suggests that many of these oscillations are not fundamental standing kink modes and that the observed period may instead reflect the driver. This extends the earlier study of the same first dataset, which had only sampled periods up to 185 s and reported a correlation of 0.98. From the standing-kink assumption the authors derive kink speeds with a mean of 471 km/s and magnetic fields with a mean of 6.3 G, often lower than previous active-region estimates, and they caution that coronal seismology in short loops is unreliable until the wave mode is identified.","pith_inferences":["The paper's own kink-speed cutoff test (correlation rises to roughly 0.7 when only events with $C_k>400$ km/s are kept) implies a testable partition: the sample is likely a mixture of true standing kink modes and other motions, and the 'long-period branch' may be dominated by the non-kink component.","A coordinated observation that compares the oscillation period in each short loop with the local photospheric p-mode power would test the driver hypothesis directly; the paper does not perform this comparison.","The assumed 40% error in loop length, plus the 3 s cadence and sub-pixel Gaussian centroiding, set a floor on how strongly a true length–period correlation could be detected, so the null correlation by itself cannot rule out a steep underlying scaling.","If long periods are driven, the energy flux carried by these oscillations should be computed from the driver's velocity amplitude and the loop's inertia rather than from the standing-mode formula, which would change estimates of decayless oscillations' contribution to coronal heating in active regions."],"forward_implications":["Short active-region loops contain a population of long-period decayless oscillations (50–467 s), so censoring periods above 200 s—as earlier short-loop studies did—removes most of the events and distorts the perceived length–period relation.","The flat length–period relation implies that for short loops the period is not a reliable proxy for loop length, and vice versa; standing-kink seismology estimates of magnetic field and kink speed from these oscillations are therefore uncertain.","The period distribution in the short-period regime differs between active regions and quiet Sun/coronal holes, pointing to different excitation mechanisms in different coronal regions.","Zero phase lag between slit positions keeps standing waves viable, so the conclusion is not that these oscillations are definitely propagating or driven, but that a mixture of wave modes is likely.","If the long periods reflect p-mode-like footpoint driving, then decayless oscillations in short loops could be used to study how photospheric drivers couple to coronal structures."],"supporting_citations":[{"why":"Provided the earlier active-region short-loop sample on the same first dataset, with periods up to about 185 s and a strong length–period correlation; it is the baseline the present sample is compared against.","marker":"Li & Long 2023"},{"why":"Established the loop-length–period correlation for decayless oscillations in long active-region loops, the reference branch in the length–period diagram.","marker":"Anfinogentov et al. 2015"},{"why":"Reported decayless oscillations in short coronal bright points in the quiet Sun and found no significant length–period correlation, supporting the separate short-loop branch.","marker":"Gao et al. 2022"},{"why":"Studied short loops in quiet Sun and coronal holes with the same detection approach; supplies the comparison data for period distributions and the separate-branch interpretation.","marker":"Shrivastav et al. 2024b"},{"why":"Provided the EUI-based intensity-uncertainty relations and the coronal density value adopted in the analysis, as well as high-frequency-wave context.","marker":"Petrova et al. 2023"},{"why":"Numerical simulation showing an inclined p-mode driver can excite long-period oscillations in short loops; it is the explanation the observed long periods are compared with.","marker":"Gao et al. 2023"}],"fun_headline_variants":["Short loops defy period-length scaling with long-lasting oscillations","Short loops oscillate up to 467s, no length-period link","Long-period oscillations found in short active-region loops","Decayless oscillations with long periods break short-loop scaling","Short loops host 467-second decayless oscillations"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the sinusoid fits with only one to two visible cycles—about 40% of the sample, including the longest periods (200–467 s) in roughly 500 s windows—are genuine oscillations rather than artifacts of dynamic loops or background intensity fluctuations.","fun_headline_variants_meta":{"raw":{"variants":["Short loops defy period-length scaling with long-lasting oscillations","Short loops oscillate up to 467s, no length-period link","Long-period oscillations found in short active-region loops","Decayless oscillations with long periods break short-loop scaling","Short loops host 467-second decayless oscillations"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000922,"raw_usage":{"total_tokens":4045,"prompt_tokens":1129,"completion_tokens":2916,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":745,"completion_tokens_details":{"reasoning_tokens":2836}},"tokens_in":745,"tokens_out":2916,"duration_ms":19760,"temperature":1.0,"reasoning_tokens":2836,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T14:03:28.588869+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-run the detection on the same EUI time series requiring at least three full cycles and a significance threshold against red noise using a wavelet or Lomb-Scargle periodogram; if most events with periods above about 200 s fail, the separate branch and the null length–period correlation would not survive. A secondary check would test whether the longest-period events repeat coherently in a second time window on the same loop.","supporting_citations":[{"cited_title":"2022, , 930, 55, 10.3847/1538-4357/ac62cf","cited_arxiv_id":null,"evidence_quote":"Reported decayless oscillations in short coronal bright points in the quiet Sun and found no significant length–period correlation, supporting the separate short-loop branch."}],"review_version":1}