REVIEW 4 major objections 7 minor 22 references
Correlations between the Variation of the Ionizing Continuum and Broad Absorption Lines in Individual Quasars
T0 review · 4 major / 7 minor · reviewed 2026-08-14 · deepseek-v4-flash
Pith's one-line read In each of 21 quasars, changes in the ionizing continuum track changes in broad absorption lines.
desk verdict Plausible result with overstated per-quasar significance; worth refereeing but needs re-analysis with epoch-level statistics. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The statistical machinery is the correlation between two fractional variations: $\Delta F_{\rm cont}/\langle F_{\rm cont}\rangle$ at 1350 Å in the rest frame, representing the ionizing continuum, and $\Delta EW/\langle EW\rangle$ for the C IV and/or Si IV BAL troughs, representing absorption strength. The paper pairs every pair of epochs within each quasar, computes Spearman rank correlation coefficients, and checks them with a Bayesian linear regression that includes measurement errors and intrinsic scatter. The physical mechanism invoked is photoionization response: continuum fluctuations change the ionization state of the absorbing gas, which changes the column densities of C IV and Si IV, and the sign of the response depends on whether the gas is in a low-ionization (positive response) or high-ionization (negative response) regime. Saturation of the BALs, especially C IV, is used to explain why Si IV often varies more than C IV and why the correlations carry substantial scatter.
What would settle it
Recompute the correlations using only one independent pair per observation, such as consecutive epochs, or a bootstrap resampling of epochs, and check whether all 21 quasars still show p-values below 0.01; if several drop below significance, the claim that the continuum drives most BAL variability in individual quasars is weakened.
Extended reading notes
Core claim
The paper's central claim is that the same quantity that measures continuum brightness variation also orders the equivalent-width variation of broad absorption lines within individual quasars. For each quasar, all pairs of epochs are used to compute fractional variations, and the Spearman rank correlation between continuum and BAL variation is significant at better than 99 percent confidence in 21 objects. The mix of signs, 17 anticorrelated and 4 correlated, is not noise but the expected behavior of photoionization: when the absorbing gas is at low ionization, more ionizing flux increases the ionic column of C IV or Si IV, while at high ionization the column decreases as the species is over-ionized. Eight of the objects with dense monitoring show the correlation on rest-frame timescales of days, implying that the absorbing gas responds to the incident continuum essentially promptly. The paper also finds that Si IV varies more than C IV in most of the 21 quasars, which it reads as a sign of widespread saturation, especially in C IV.
Load-bearing premise
The analysis assumes that every pair of observation epochs can be treated as an independent data point, even though each epoch is reused in many pairs, so the reported significance levels are probably optimistic.
Editorial extensions
If this is right
- BAL variability can be used as a monitor of the quasar's ionizing flux history, including epochs when the continuum itself is too faint or absorbed to measure.
- The absence of a sample-wide correlation no longer counts as evidence against photoionization, because individual objects can sit on either side of the ionization response.
- Day-timescale correlations imply that the absorbing gas lies close to the ionizing source, placing a practical upper limit on the distance of the BAL region.
- Saturated C IV troughs respond weakly, so unsaturated Si IV or higher-ionization lines should be the preferred tracers for future variability monitoring.
Reading between the lines
- If the correlation sign tracks the ionization state, a single monitoring campaign that measures both continuum and BAL changes could classify individual outflows as high- or low-ionization systems without detailed spectral modeling.
- The same per-object pair-correlation method applied to N V or O VI BALs, where saturation is weaker, would test whether the scatter shrinks and the correlations tighten.
- A bootstrap or independent-pair resampling of the epochs would give a direct check of how many of the 21 correlations survive when the non-independence of epoch pairs is removed.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. This paper analyzes multi-epoch SDSS spectra of 46 BAL quasars with at least five epochs, computes fractional variations of a 1350 Å power-law continuum and C IV/Si IV BAL equivalent widths from the He et al. (2017) catalog, and performs Spearman rank correlations on all C(m,2) pairs of epochs per quasar. The authors report significant correlations (p<0.01) in 21 of 46 quasars, 17 negative and 4 positive, and interpret this as evidence that ionizing continuum fluctuations drive BAL variability, consistent with photoionization model expectations. They also note that 8 of the 21 are SDSS-RM quasars with dense cadence, arguing for rapid photoionization response, and find larger Si IV than C IV variation amplitudes suggestive of C IV saturation.
Significance. If the claimed correlations are real, the paper would provide some of the first systematic per-object evidence that BAL variability in individual quasars tracks continuum variations, complementing earlier ensemble studies. The use of multi-epoch SDSS data and a consistency check with a Bayesian regression approach are strengths. However, the statistical analysis currently overstates confidence because the epoch pairs are not independent, so the significance of the central claim is not yet established; with appropriate corrections the paper could still be a valuable contribution.
major comments (4)
- [Section 2] The Spearman rank correlation is applied to all C(m,2) = m(m-1)/2 pairs of epochs, but these pairs are not independent observations: each epoch appears in m-1 pairs, so the effective sample size is far smaller than the number of pairs. The reported p-values (many listed as 0 in Table 1) are therefore overconfident, and the selection of 21 of 46 quasars using p<0.01 inherits this inflation. The paper should either use a statistical method that properly handles repeated measures (e.g., a bootstrap resampling whole epochs, a mixed-effects model, or a permutation test that randomizes epoch labels) or should report effective sample sizes and adjusted p-values. Without this, the central claim that "fluctuations of the ionizing continuum drive most BAL variations" is not yet supported at the stated significance level.
- [Section 3.5] The Bayesian consistency check using Kelly (2007) does not address the non-independence of pairs. The Kelly regression is applied to the same set of C(m,2) pairs, and its error estimates assume independent data points; the intrinsic-scatter parameter σ_int absorbs some scatter but does not correct the effective sample size. The statement in Section 3.5 that "most of the Bayesian results are consistent with the Spearman results" therefore does not resolve the concern about inflated significance.
- [Section 2] The paper tests two BAL lines (C IV and Si IV) per quasar and retains a quasar if either shows p<0.01, but no multiple-testing correction is applied across the 46 quasars and two lines. Even with a modest false-positive rate per test, the combination of dependence among pairs and selection on the same statistics used for inference makes it possible that some of the 21 quasars are spurious. The authors should report how many correlations would survive a control of the false discovery rate at the quasar or pair level, or validate the selection with a permutation test.
- [Section 3.3] The claim that the correlations reveal rapid BAL variability on rest-frame timescales of a few days is not directly tested. The analysis uses all pairs regardless of time separation; while the SDSS-RM light curves do contain short-cadence observations, the Spearman correlation is computed over the full set of pairs. To support the rapid-response claim, the authors should either restrict the analysis to pairs with short rest-frame time separations (e.g., < 10 days) or perform a lag or structure-function analysis. As written, the abstract's "timescales as short as a few days" overstates what the correlation analysis shows.
minor comments (7)
- [Introduction] In the second paragraph, "the the variability" should be "the variability".
- [Section 2] The notation "C2_m" is nonstandard; use C(m,2) or define it verbally in the text.
- [Section 2] "S/N >10 in one-epoch observation at least" is awkward; suggest "with at least one epoch having S/N > 10".
- [Table 1] The column headers are split across two physical lines, making it hard to know which p-values correspond to C IV versus continuum and which to Si IV versus continuum; reformat the table so each column header is self-contained.
- [Figure 1] The panels are small; consider enlarging the figure or splitting it into multiple figures, and ensure all panels have clearly labeled axes and units.
- [Section 3.4] "Visually check Figure 1" should be "A visual check of Figure 1 shows...".
- [Section 3.1] For SDSS J022844.09+000217.0, it would be informative to show the correlation restricted to the 18 epochs used by He et al. (2014) to demonstrate that the difference in conclusion is due to the additional epochs and/or the method, rather than simply to the pair-based approach.
Circularity Check
No significant circularity: the per-quasar correlations are empirical measurements built from independent continuum fits and an external BAL EW catalog, with interpretation checked against external photoionization models.
full rationale
The paper's derivation chain is not circular. The continuum fractional variation is measured from power-law fits to the SDSS spectra (using a procedure from Lu et al. 2018), while the BAL fractional variations are computed from equivalent widths taken from the independent He et al. (2017) catalog. The two quantities are therefore measured from separate data products, and no parameter is fitted to the target correlation. The Spearman correlations are direct empirical statistics over the epoch pairs, and the Bayesian consistency check (Kelly 2007) is an independent statistical tool, not a quantity constructed to match the claimed result. The interpretation that the ionizing continuum drives BAL variability is justified by comparison with external photoionization simulations and previous systematic studies (Wang et al. 2015; He et al. 2017), not by the paper's own equations. The self-citations to Lu et al. (2018) and Lu & Lin (2018) provide measurement conventions and context, but they are not load-bearing in the sense of defining the correlation outcome in advance. The known statistical limitation that the epoch pairs are non-independent affects the confidence of the reported p-values, and the sample selection by p<0.01 from 46 candidates is a selection effect, but these are statistical validity concerns, not circularity. The paper does not rename an existing result, import a uniqueness theorem, or hide a fitted parameter inside a prediction. Its central claim is an empirical correlation analysis whose inputs and outputs are distinct.
Assumptions & free parameters
free parameters (1)
- Power-law continuum normalization and slope (per epoch, per quasar) =
not reported
assumptions (4)
- domain assumption The 1350 Å continuum flux from a power-law fit represents the ionizing continuum strength.
- domain assumption The BAL EWs from the He et al. (2017) catalog are accurate and consistently measured across epochs.
- domain assumption Photoionization models predict that EW response to continuum changes can be positive or negative depending on ionization state.
- ad hoc to paper All epoch pairs are treated as independent observations in the Spearman rank correlation.
Cite this review
Pith. "Pith review of Correlations between the Variation of the Ionizing Continuum and Broad Absorption Lines in Individual Quasars." pith.science (2026). https://pith.science/paper/PUB7CDSK
@misc{pith2026190803844,
author = {Pith},
title = {Pith review of: Correlations between the Variation of the Ionizing Continuum and Broad Absorption Lines in Individual Quasars},
year = {2026},
howpublished = {\url{https://pith.science/paper/PUB7CDSK}},
note = {Machine review of arXiv:1908.03844}
}
read the original abstract
We discover the significant (significance level of >99%) correlations between the fractional variation of the ionizing continuum and that of the C IV and/or Si IV BALs in each of 21 BAL quasars that have at least five-epoch observations from the Sloan Digital Sky Survey-I/II/III. This result reveals that the fluctuation of the ionizing continuum is the driver of most of these BAL variations. Among them, 17 show negative correlations and the other 4 positive correlations, which agrees with the prediction of photoionization models that absorption line variability response to ionization changes is not monotonic. 8 quasars out of 21 examples have been observed at least 30 times on rest-frame timescales as short as a few days, which reveals that changes in the incident ionizing continuum can cause BAL variability even in such a short period of time. In addition, we find that most of the 21 quasars show larger variation amplitude in Si IV than C IV, which reveals the ubiquity of saturation in these BALs (at least for C IV BALs).
Reference graph
Works this paper leans on
-
[1]
Chen, Z.-F., Gu, Q.-S., Chen, Y.-M., & Cao, Y. 2015, MNRAS, 450, 3904 10
work page 2015
-
[2]
M., Kraemer, S
Crenshaw, D. M., Kraemer, S. B., & George, I. M. 2003, ARA&A, 41, 117
2003
-
[3]
Gabel, J. R., Kraemer, S. B., Crenshaw, D. M., et al. 2005, ApJ , 631, 741
work page 2005
-
[4]
Gibson, R. R., Brandt, W. N., Schneider, D. P., & Gallagher, S. C. 2008, ApJ, 675, 985
work page 2008
- [5]
-
[6]
E., Siegmund, W
Gunn, J. E., Siegmund, W. A., Mannery, E. J., et al. 2006, AJ, 131, 2332
2006
-
[7]
F., Rodr ´ ıguez Hidalgo, P., Prochaska, J
Hamann, F., Kaplan, K. F., Rodr ´ ıguez Hidalgo, P., Prochaska, J. X., & Herbert-Fort, S. 2008, MNRAS, 391, L39
work page 2008
-
[8]
2004, in Astronomical Society of the Pacific Conference Series, Vol
Hamann, F., & Sabra, B. 2004, in Astronomical Society of the Pacific Conference Series, Vol. 311, AGN Physics with the Sloan Digital Sky Survey, ed. G. T. Richards & P. B. Hall, 203
work page 2004
Show all 22 references
-
[9]
2017, ApJS, 229, 22
He, Z., W ang, T., Zhou, H., et al. 2017, ApJS, 229, 22
2017
-
[10]
2014, MNRAS, 443, 2532
He, Z.-C., Bian, W.-H., Jiang, X.-L., & W ang, Y.-F. 2014, MNRAS, 443, 2532
2014
-
[11]
S., Grier, C
Hemler, Z. S., Grier, C. J., Brandt, W. N., et al. 2019, ApJ, 87 2, 21
2019
-
[12]
2019, arXiv e-pri nts, arXiv:1905.02407
Huang, H.-Y., Pan, C.-J., Lu, W.-J., et al. 2019, arXiv e-pri nts, arXiv:1905.02407
2019 arXiv
-
[13]
Kelly, B. C. 2007, ApJ, 665, 1489
2007
-
[14]
2018, ApJ, 862, 46
Lu, W.-J., & Lin, Y.-R. 2018, ApJ, 862, 46
2018
-
[15]
2018, MNRAS, 473, L106
Lu, W.-J., Lin, Y.-R., & Qin, Y.-P. 2018, MNRAS, 473, L106
2018
-
[16]
2017, MNRAS, 468, L6
Lu, W.-J., Lin, Y.-R., Qin, Y.-P., et al. 2017, MNRAS, 468, L6
2017
-
[17]
N., Dawson, K
Shen, Y., Brandt, W. N., Dawson, K. S., et al. 2015, ApJS, 216, 4
2015
-
[18]
A., Gunn, J
Smee, S. A., Gunn, J. E., Uomoto, A., et al. 2013, AJ, 146, 32
2013
-
[19]
2019, MNRAS, 486, 2379
Vivek, M. 2019, MNRAS, 486, 2379
2019
-
[20]
2014, MNRAS, 4 40, 799 W ang, T., Yang, C., W ang, H., & Ferland, G
Vivek, M., Srianand, R., Petitjean, P., et al. 2014, MNRAS, 4 40, 799 W ang, T., Yang, C., W ang, H., & Ferland, G. 2015, ApJ, 814, 150 W eymann, R. J., Morris, S. L., Foltz, C. B., & Hewett, P. C. 1991, ApJ, 373, 23
2014
-
[21]
R., & Allen, J
Wildy, C., Goad, M. R., & Allen, J. T. 2014, MNRAS, 437, 1976
2014
-
[22]
G., Adelman, J., Anderson, Jr., J
York, D. G., Adelman, J., Anderson, Jr., J. E., et al. 2000, AJ , 120, 1579
2000
Reviewed August 14, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.