Interstellar Extinction Law toward the Galactic Center IV: J, H, and Ks Bands from VVV Red Clump Stars

Schun Nagatomo; Shogo Nishiyama; Tetsuya Nagata

REVIEW 3 major objections 3 minor 19 references

Interstellar Extinction Law toward the Galactic Center IV: J, H, and Ks Bands from VVV Red Clump Stars

T0 review · 3 major / 3 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Near-infrared extinction toward the Galactic center falls steeply with wavelength, with a power-law index near 2.0 to 2.3.

desk verdict A careful independent VVV confirmation of the steep GC extinction law, but the quoted J-H ratio fails reddening closure and the triplet should not be treated as a precise calibration. read the letter →

arxiv 1908.01148 v2 pith:FKFV6JBZ submitted 2019-08-03 astro-ph.GA

Schun Nagatomo , Tetsuya Nagata , Shogo Nishiyama This is my paper · ORCID

classification astro-ph.GA

keywords interstellarextinctionGalacticcenterredclumpmethodnear-infraredlawVVVsurveytotal-to-selectivepowerdistancedetermination

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper sets out to measure how strongly dust dims starlight at near-infrared wavelengths toward the center of the Milky Way, using red clump stars in the VVV survey's aperture photometry. It claims the total-to-selective extinction ratios are $A(K_{\mathrm S})/E(H-K_{\mathrm S})=1.44\pm0.04$, $A(K_{\mathrm S})/E(J-K_{\mathrm S})=0.423\pm0.024$, and $A(H)/E(J-H)=1.25\pm0.04$, which together imply a steep power law $A(\lambda)\propto\lambda^{-\alpha}$ with $\alpha\approx2.0$--$2.3$. If true, the near-infrared extinction law toward the inner Galaxy is much steeper than the classical curve, and any distance derived from near-infrared reddening must be revised: a given observed color yields less extinction and hence a larger distance. A sympathetic reader cares because the extinction law is a basic conversion factor used to map the bulge and to place reddened Cepheids, red clump giants, and other tracers in three dimensions.

What carries the argument

The red-clump method is the load-bearing machinery. Red clump giants are roughly standard: a dense population of helium-burning giants with nearly constant intrinsic luminosity and color, so the Gaussian-fitted position of the red-clump peak in a color-magnitude diagram shifts from one sight line to another in proportion to the extinction and reddening. Plotting the peak magnitude against the peak color over many subfields gives a line whose slope is the extinction-to-reddening ratio, and the paper evaluates these slopes tile by tile to avoid the zero-point scatter that corrupts a global fit. The second critical piece is the assertion that in fields with $|b|\approx1^\circ$ the slope equals $A(\lambda)/E(\lambda_1-\lambda_2)$ rather than merely a differential $dA/dE$, because the bulk of the dust lies in the distant part of each line of sight at similar distances; this is the condition needed to convert measured slopes into a total-to-selective extinction law.

What would settle it

Take the paper's own two ratios $A(K_{\mathrm S})/E(H-K_{\mathrm S})=1.44$ and $A(K_{\mathrm S})/E(J-K_{\mathrm S})=0.423$; they imply $A(H)/E(J-H)\approx1.01$, whereas the paper reports $1.25$. Measuring extinction toward stars with independent distance anchors in the same fields would show which value is real, and a persistent $1.25$ would mean the red-clump slopes are not all true total-to-selective ratios.

Watch

Extended reading notes

Core claim

The paper's central claim is that the near-infrared extinction law toward the center of the Milky Way is steep, $A(\lambda)\propto\lambda^{-\alpha}$ with $\alpha\approx2.0$--$2.3$, not the gentler $\alpha\approx1.6$ of the classical curve. This is established by locating red clump stars in VVV aperture photometry for 12 tiles covering $|l|\lesssim2^\circ$ and $0.5^\circ\lesssim|b|\lesssim1^\circ$, fitting Gaussian peaks in small subfields, and plotting peak magnitude against peak color; the line slopes give $A(K_{\mathrm S})/E(H-K_{\mathrm S})=1.44\pm0.04$, $A(K_{\mathrm S})/E(J-K_{\mathrm S})=0.423\pm0.024$, and $A(H)/E(J-H)=1.25\pm0.04$. The paper shows that fitting all subfields together yields misleading slopes because of tile-to-tile zero-point differences, while tile-weighted means recover the values above and agree with an earlier red-clump measurement from a different camera system. It also argues that in these $|b|\approx1^\circ$ fields the red-clump slope is the true total-to-selective ratio, because the dust producing the differential reddening sits in the distant parts of the sight lines at comparable distances.

Load-bearing premise

The load-bearing premise is that the slope of the red-clump line equals the ordinary total-to-selective extinction ratio and not merely the differential ratio between sight lines, which would fail if the reddening dust sits at very different distances along the sight lines.

Editorial extensions

If this is right

For a fixed observed near-infrared color, adopting the steep law gives a smaller $A(K_{\mathrm S})$ and therefore a larger distance modulus, moving heavily reddened inner-Galaxy objects outward.
The VVV-derived ratios agree with an earlier ground-based red-clump measurement using a different camera and filter set, so the steep law does not appear to be an artifact of the VVV photometric system.
The contrast with the PSF-based measurement of $A(K_{\mathrm S})/E(H-K_{\mathrm S})\approx1.1$ suggests the extinction law may depend on Galactic latitude, with the near-plane region possibly following a different relation.
The residual tile-to-tile variation means future extinction-law maps of this region should quote sight-line-dependent values rather than a single global curve.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The three quoted ratios are not mutually consistent: the first two imply $A(H)/E(J-H)\approx1.01$, while the paper measures $1.25$, so either a photometric zero-point offset or a partial composite-extinction bias remains in at least one color combination.
If the steep law extends to the lower latitudes sampled by deeper photometry, distances of classical Cepheids and other plane tracers derived from VVV reddenings would shift outward, potentially removing evidence for a young inner thin disk.
A direct test would compare the red-clump slopes with extinction toward stars in the same fields whose distances are known independently, such as maser-bearing objects.
Using the overlapping-tile magnitude differences to re-calibrate tile zero points before slope fitting could reduce the tile-to-tile scatter and sharpen the measured power-law index.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Dear [colleague],

The thing to know: this is a solid, honest re-derivation of the near-IR extinction law toward the GC using public VVV DR4 data and the red clump method. It broadly confirms the steep law (A(Ks)/E(H-Ks)=1.44, A(Ks)/E(J-Ks)=0.423, alpha~2) seen in previous work. But the three quoted ratios do not close: with the two Ks-based ratios, the implied A(H)/E(J-H) is about 1.02±0.09, not 1.25±0.04. At least one ratio is biased, and the paper should not be cited as a precise triplet calibration.

What is genuinely new and useful: it applies the RC method to independent VVV data in exactly the N06 fields, documents systematic tile-to-tile differences up to ~0.1 mag in overlapping VVV tiles, and checks that the newer PSF photometry would not change the conclusions. That documentation is real value. The central steep-law conclusion is probably correct and consistent with a large body of recent work; the H-K and J-K ratios alone imply a steep law.

The soft spots are concentrated in the J-H result and in the interpretation of the CMD slopes as total-to-selective ratios. The paper argues in Section 4 that the composite extinction bias of Nataf et al. is negligible at |b|~1, but the closure failure is a direct symptom that at least one slope is not the ratio the paper claims. The tile-to-tile zero-point offsets, which the authors themselves document, are a plausible culprit. Also telling is that their own all-subfield fits differ from the weighted means — for example, the H-K slope from all 566 subfields is 1.30 versus the weighted mean 1.44, and for J-H it is 1.72 versus 1.25. That internal inconsistency needs to be resolved, not just acknowledged.

The citation pattern is fine. N06/N09 are cited heavily, but as independent comparison data, not as inputs; there is no circularity. The paper is also appropriately cautious about sight-line variations.

Who this is for: anyone working on inner Galaxy distances, especially Cepheid work in the VVV survey. It deserves a serious referee, but with a demand that the authors either enforce reddening closure, re-derive the J-H ratio, or explicitly publish the slopes as differential reddening ratios with caveats. As a reader, I would treat the H-K and J-K ratios as evidence for the steep law, and set the J-H ratio aside until it is fixed.

Referee Report

3 major / 3 minor

Summary. The paper determines near-infrared total-to-selective extinction ratios toward the Galactic center using red clump stars in VVV aperture photometry, reproducing the field selection of Nishiyama et al. (2006). The authors report A(Ks)/E(H−Ks)=1.44±0.04, A(Ks)/E(J−Ks)=0.423±0.024, and A(H)/E(J−H)=1.25±0.04 from weighted means of per-tile CMD slopes, and infer a steep power-law extinction A(λ)∝λ^{−α} with α≈2.0–2.3. They also document up to ~0.1 mag systematic magnitude offsets between overlapping VVV tiles and discuss possible tile-to-tile extinction-law variations.

Significance. If correct, the paper provides a direct, publicly-available-data determination of the NIR extinction law toward the Galactic center, supporting a steeper wavelength dependence than the classical Rieke–Lebofsky/Cardelli curves. This has direct consequences for distances and extinctions of heavily reddened inner-Galaxy objects, and the paper's use of the red clump method on VVV DR4 is a useful cross-check of earlier IRSF/SIRIUS results. The authors are appropriately cautious about tile-to-tile systematics and compare with independent PSF-photometry results. However, the central quantitative claim—a self-consistent triplet of total-to-selective ratios—fails an internal consistency check, which limits the paper's value as a precise calibration until the discrepancy is resolved.

major comments (3)

[Abstract and Table 1] The three reported ratios are not mutually consistent under standard extinction algebra. Using A(Ks)/E(H−Ks)=1.44±0.04 and A(Ks)/E(J−Ks)=0.423±0.024, the identity A(H)/E(J−H) = [A(Ks)/E(J−Ks)]·[1 + A(Ks)/E(H−Ks)] / [A(Ks)/E(H−Ks) − A(Ks)/E(J−Ks)] gives 1.02±0.09, which is 2.5σ away from the reported 1.25±0.04. At least one of the three quoted ratios is therefore biased, so the triplet cannot serve as a precise extinction-law calibration unless the closure failure is explained and corrected.
[Section 4, composite extinction bias discussion] The identification of the measured CMD slopes with total-to-selective ratios A(λ)/E(λ1−λ2) rests on the assertion that A(Ks)/E(H−Ks) ≃ dA(Ks)/dE(H−Ks) in the |b|≈1° fields. The paper supports this only by the qualitative statement that fitted lines pass within ~0.1 mag of the intrinsic RC points, and by an argument that foreground extinction is relatively unimportant. This is the load-bearing assumption that, if invalid, would explain the closure failure in the triplet. The authors should provide a quantitative test, for example by comparing the fitted intercepts with the Nataf et al. intrinsic values for all three CMDs and assessing whether the residual offsets are consistent with zero within the quoted uncertainties.
[Table 1 and Section 3, tile-to-tile systematics] The paper documents large differences between the all-subfield least-squares slopes and the weighted-mean slopes (e.g., A(Ks)/E(H−Ks) = 1.30±0.03 for all 566 subfields versus 1.44±0.04 for the weighted mean), and up to ~0.1 mag zero-point offsets between overlapping tiles. These systematics are not propagated into the quoted errors, so the reported uncertainties likely underestimate the true systematic error. The authors should either fold the zero-point offsets and the scatter between tiles into the final error bars, or justify why they are negligible for the weighted means.

minor comments (3)

[Table 1 and Table 2] Several per-tile slopes are unphysical or have very large formal errors (e.g., tile b332 H−Ks slope 3.45±3.00, Table 2 entries of 9.05 and −0.584, and NaN errors for some J−H entries). These should either be excluded with stated criteria or discussed explicitly, since their inclusion in weighted means could bias the results.
[Throughout] The notation for the Ks band is inconsistent: 'KS', 'Ks', and 'K_S' are used interchangeably. Please standardize, preferably to Ks throughout.
[Section 4, PSF photometry comparison] The comparison with Alonso-García et al. (2017) would be clearer if the authors stated whether the difference in A(Ks)/E(H−Ks) (1.44 vs 1.104) is statistically significant given the ±0.2 systematic error quoted for the comparison work, and whether the tile-to-tile variations reported here could account for that difference.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the extinction ratios are fit from VVV red-clump CMD peaks, with prior work used only for comparison and zero-point checks.

full rationale

The derivation chain is self-contained: RC peak magnitudes and colors are measured from VVV aperture photometry in 4'x4' subfields, Gaussian-fitted, and the slopes of the peak loci in KS vs H-KS, KS vs J-KS, and H vs J-H CMDs are identified with A(Ks)/E(H-Ks), A(Ks)/E(J-Ks), and A(H)/E(J-H). No fitted quantity is defined in terms of the target ratio, and no parameter is adjusted to force agreement. The intrinsic RC points from Nataf et al. (2013, 2016) are external calibrations used only to check that the fitted lines pass near the unreddened RC position (blue squares), not to define the slopes. The paper's own N06/N09 results are cited for field geometry and as comparison values (Table 1); they do not enter the least-squares fits. The paper explicitly documents systematic tile-to-tile zero-point differences up to about 0.1 mag (Section 3) and states the assumption A(Ks)/E(H-Ks) approximately equals dA(Ks)/dE(H-Ks) holds at |b| about 1 degree (Section 4); these are stated limitations and physical assumptions, not circular reductions. The internal inconsistency among the three quoted ratios is a robustness/correctness concern, but it does not make any step equivalent to its input by construction. Therefore no circularity is found.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

No free constants are hidden in the analysis; the quoted ratios are direct measurements. No new physical entities are introduced. The central claim rests on the standard-candle nature of red clump stars, the identification of CMD slopes with total-to-selective ratios, and the stability of VVV photometric zero points.

assumptions (4)

domain assumption Red clump stars have constant intrinsic color and absolute magnitude across the surveyed Galactic center fields.
The RC method identifies CMD peaks with extinction-modified standard candles; variations in RC metallicity, age, or distance distribution would bias the measured slopes. Invoked in Sections 2 and 4.
domain assumption The magnitude-versus-color slope of the RC peak locus equals the ratio A/E, not merely dA/dE; composite extinction bias is negligible at |b|≈1°.
Section 4 argues this by analogy to Nataf et al. 2013. The paper's own H versus J-H fit does not pass through the Nataf intrinsic point, so this assumption is questionable and load-bearing.
domain assumption VVV DR4 aperture photometry zero points are sufficiently stable after tile-by-tile averaging; overlap offsets up to 0.1 mag do not bias the color slopes.
Section 4 and Figure 7 show systematic offsets between overlapping tiles. The authors assert these do not alter in-tile slopes, but the final quoted errors do not include them.
domain assumption The extinction law can be approximated as a single power law A(λ) ∝ λ^{-α} across J, H, and Ks.
Used to convert the three ratios into α≈2.0-2.3. If the true law is not a power law, the quoted α is a convenient description rather than a measured physical index.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Interstellar Extinction Law toward the Galactic Center IV: J, H, and Ks Bands from VVV Red Clump Stars." pith.science (2026). https://pith.science/paper/FKFV6JBZ

@misc{pith2026190801148,
  author       = {Pith},
  title        = {Pith review of: Interstellar Extinction Law toward the Galactic Center IV: J, H, and Ks Bands from VVV Red Clump Stars},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FKFV6JBZ}},
  note         = {Machine review of arXiv:1908.01148}
}

abstract

We have determined the wavelength dependence of the extinction in the near-infrared bands ($J$, $H$, $K_{\mathrm S}$) toward the Galactic center from the VVV (VISTA Variables in the V\'ia L\'actea) aperture photometry of the stars in the region $|l|\lesssim2^\circ.0$ and $0^\circ.5\lesssim|b|\lesssim1^\circ.0$; this region consists of 12 VVV tiles. We have found significant systematic discrepancy up to $\sim0.1$ mag between the stellar magnitudes of the same stars in overlapping VVV tiles. However, by carefully using the positions of red clump stars in color-magnitude diagrams as a tracer of the extinction and reddening, we are able to determine the average of the ratios of total to selective extinction to be $A(K_{\mathrm S})/E(H-K_{\mathrm S})=1.44\pm0.04$, $A(K_{\mathrm S})/E(J-K_{\mathrm S})=0.423\pm0.024$, $A(H)/E(J-H)=1.25\pm0.04$; from these ratios, a steep power law $A(\lambda)\propto\lambda^{-\alpha}$ whose index $\alpha$ is $\sim2.0-2.3$ in the $J,H,K_{\mathrm S}$ wavelength range is estimated. The obtained wavelength dependence is consistent with those obtained with the MKO photometric system employed in SIRIUS camera attached to the IRSF telescope in previous studies. Such a steep decline of extinction toward the longer wavelengths is also in line with recent results based on deep imaging surveys. The determined extinction law seems to be variable in the VVV tile to tile, and it is not clear how much of this is due to real sight line variations and due to observational systematic effects. Thus, there might be room for improvement of the extinction law determination from the existing VVV data, but this steep extinction law tends to locate heavily reddened objects in the Galactic plane more distant from us when their distance moduli are calculated from the observed reddening values.

Figures

Figures reproduced from arXiv: 1908.01148 by the authors.

**Figure 1.** VVV observation area and fields used for data analysis. The pink rectangle lines are the edges of the 12 VVV tiles in our field analyzed. Following N06, each small square is a subfield of 4 ′ × 4 ′ . Only the colored subfields were used for the data analysis in this paper; the gray line squares are the regions where the magnitude and color of RC stars were not reliably determined due to large extinction [PITH_FULL_… view at source ↗

**Figure 2.** An example of KS vs. H − KS CMD of a subfield (left). In the rectangular region of the left panel, a color peak (middle) and a magnitude peak (right) are determined. 5 [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

**Figure 3.** For the whole data, each magnitude and color Gaussian-fit peaks of RC stars in each subfield KS vs. H − KS (left), KS vs. J − KS (middle), and H vs. J − H (right) CMDs are shown with the black dots, and the least-squares fits to them (pink lines). The blue squares at the upper left side represent the predicted apparent magnitudes and intrinsic colors of RC stars free from extinction, following Nataf et al. (2016). f… view at source ↗

Figures from the paper (4 more)

**Figure 4.** Figure 4: Each magnitude and color Gaussian-fit peaks of RC stars in the KS vs. H − KS CMDs for the VVV tiles b > 0.7 ◦ (top), 0.7 ◦ > b > 0 ◦ (middle), and 0 ◦ > b (bottom). In the bottom diagrams, black squares are 0 ◦ > b > −0.7 ◦ , and green dots are −0.7 ◦ > b data points. …

**Figure 5.** Figure 5: Same as [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

**Figure 6.** Figure 6: Same as [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]

**Figure 7.** Figure 7: Difference of H (left) and H − KS (right) of common stars between the adjacent b334 and b348 tiles. Each bin is 0.01 mag wide [PITH_FULL_IMAGE:figures/full_fig_p009_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

19 extracted references · 16 canonical work pages

[1]

Alonso-Garc´ ıa, J., et al., 2015, AJ, 149, 99 Alonso-Garc´ ıa, J., et al., 2017, ApJL, 849, L13 12 Alonso-Garc´ ıa, J., et al., 2018, A&A, 619, A4

work page 2015
[2]

Bhardwaj, A., et al., 2017, A&A, 605, A100

work page 2017
[3]

A., Clayton, G

Cardelli, J. A., Clayton, G. C., & Mathis, J. S. 1989, ApJ, 345 , 245

1989
[4]

A., Blum, R

Damineli, A., Almeida, L. A., Blum, R. D., Damineli, D. S. C., Navarete, F., Rubinho, M. S., Teodoro, M., 2016, MNRAS, 463, 2653 D´ ek´ any, I. et al. 2015a, ApJL, 799, L11 D´ ek´ any, I. et al. 2015b, ApJL, 812, L29 Girardi, 2016, Annu. Rev. Astron. Astrophys. 54, 95 Gonz´ alez-Fern´ andez, C., Ramos, A. A., Garz´ on, F., Cabrera-Lavers, A., and Hammersl...

work page 2016
[5]

J., Bandyopadhyay, R

Gosling, A. J., Bandyopadhyay, R. M., & Blundell, M. 2009, MN RAS, 394, 2247

work page 2009
[6]

G., D´ ek´ any, I., Minniti, D., & Gieren, W

Majaess, D., Turner, D. G., D´ ek´ any, I., Minniti, D., & Gieren, W. 2016, A&A, 593, A124

work page 2016
[7]

W., Bono, G., et al

Matsunaga, N., Feast, M. W., Bono, G., et al. 2016, MNRAS, 462 , 414 Minniti D. et al., 2010, New Astron., 15, 433 Minniti D. et al., 2017, AJ, 153, 179

work page 2016
[8]

2003, Proc

Nagayama, T., et al. 2003, Proc. SPIE, 4841, 459

2003

Show all 19 references

[9]

Nataf, D. M.et al. 2013, ApJ, 769, 88

2013
[10]

Nataf, D. M.et al. 2016, MNRAS, 456, 2692

2016
[11]

2006, ApJ, 638, 839 (N06)

Nishiyama, S., et al. 2006, ApJ, 638, 839 (N06)

2006
[12]

2009, ApJ, 696, 1407 (N09)

Nishiyama, S., et al. 2009, ApJ, 696, 1407 (N09)

2009
[13]

2018, A&A, 610, A83

Nogueras-Lara, F., et al. 2018, A&A, 610, A83

2018
[14]

H., & Lebofsky, M

Rieke, G. H., & Lebofsky, M. J. 1985, ApJ, 288, 618

1985
[15]

Saito, R. K.et al. 2012, A&A, 537, A107 Schultheis M. et al., 2014, A&A, 566, A120

2012
[16]

2013, A&A, 552, A101 Stead J

Soto, M.et al. 2013, A&A, 552, A101 Stead J. J., Hoare M. G., 2009, MNRAS, 400, 731

2013
[17]

2017, ApJ, 842, 104

Tanioka, S., Matsunaga, N., Fukue, K,, Inno, L., Bono, G., & K obayashi, N. 2017, ApJ, 842, 104

2017
[18]

T., Simons, D

Tokunaga, A. T., Simons, D. A., & V acca, W. D. 2002, PASP, 114, 180

2002
[19]

2003, ApJ, 590, 284 Wo´ zniak, P

Udalski, A. 2003, ApJ, 590, 284 Wo´ zniak, P . R., & Stanek, K. Z. 1996, ApJ, 464, 233 13

2003

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

Alonso-Garc´ ıa, J., et al., 2015, AJ, 149, 99 Alonso-Garc´ ıa, J., et al., 2017, ApJL, 849, L13 12 Alonso-Garc´ ıa, J., et al., 2018, A&A, 619, A4

work page 2015

[2] [2]

Bhardwaj, A., et al., 2017, A&A, 605, A100

work page 2017

[3] [3]

A., Clayton, G

Cardelli, J. A., Clayton, G. C., & Mathis, J. S. 1989, ApJ, 345 , 245

1989

[4] [4]

A., Blum, R

Damineli, A., Almeida, L. A., Blum, R. D., Damineli, D. S. C., Navarete, F., Rubinho, M. S., Teodoro, M., 2016, MNRAS, 463, 2653 D´ ek´ any, I. et al. 2015a, ApJL, 799, L11 D´ ek´ any, I. et al. 2015b, ApJL, 812, L29 Girardi, 2016, Annu. Rev. Astron. Astrophys. 54, 95 Gonz´ alez-Fern´ andez, C., Ramos, A. A., Garz´ on, F., Cabrera-Lavers, A., and Hammersl...

work page 2016

[5] [5]

J., Bandyopadhyay, R

Gosling, A. J., Bandyopadhyay, R. M., & Blundell, M. 2009, MN RAS, 394, 2247

work page 2009

[6] [6]

G., D´ ek´ any, I., Minniti, D., & Gieren, W

Majaess, D., Turner, D. G., D´ ek´ any, I., Minniti, D., & Gieren, W. 2016, A&A, 593, A124

work page 2016

[7] [7]

W., Bono, G., et al

Matsunaga, N., Feast, M. W., Bono, G., et al. 2016, MNRAS, 462 , 414 Minniti D. et al., 2010, New Astron., 15, 433 Minniti D. et al., 2017, AJ, 153, 179

work page 2016

[8] [8]

2003, Proc

Nagayama, T., et al. 2003, Proc. SPIE, 4841, 459

2003