1 Motivation

Open clusters are a gold mine for the development of many astrophysical topics. They offer a unique opportunity, for example, to compare theoretical studies with observations; they provide opportunities to develop models of chemical enrichment with respect to the center of the galaxy, and serious studies on stability can be tested only through an analysis of open clusters. However, despite the importance of these topics, research in these fields begins with the determination of the cluster member stars.

Membership determination in open clusters is done, canonically, with proper motion studies; but in practice, main-sequence fitting is used since it is easier, although less accurate and cannot be used on a star-by-star basis. However, for distant or faint clusters membership determination is not an easy task. uvby-β   photometry of open clusters provides an accurate method for determining distances to each star and, through the global behavior, throws light on the distance to the cluster and, hence, the membership of each star to the cluster.

In this paper we present our results on three clusters: two, IC 4665 and NGC 6871, are relatively well-studied and the other, Dzim 5, has very little published information.

The open cluster IC 4665 has been a subject of many studies. The membership in the cluster has been determined in many different ways. The classical proper motions studies were done by ^{Vasilevskis (1955)} and the spectral classification of its members was done through classical spectroscopy.

With respect to the photometric studies, there are some classical works like that of ^{Johnson (1954)}. With intermediate photometric bands there is the work of ^{Crawford & Barnes (1972)} who, with uvby-β   photometry, found an average cluster reddening E(b - y) of 0.14, and a cluster distance modulus of 7.5, corresponding to a distance of 316 pc. They obtained the same values from an analysis of the B-type stars and the A- and F- type stars.

Finding variables in the cluster has been a subject of some research. ^{Barannikov (1994)} confirmed that the star HD 161573 in the IC 4665 cluster has periodic (P = 19 d) variability.

With respect to NGC 6871 much research has been carried out. A good summary of its characteristics can be found in ^{Southworth et al. (2004)}. They stated that the open cluster NGC 6871 was a concentration of bright OB stars which form the nucleus of the Cyg OB3 association. For this reason it is an important object in the study of the evolution of high-mass stars. The cluster itself has been studied photometrically several times, but the scarce data on its nature mean that determination of its physical parameters is difficult. They further note that: “UBV photometry of the 30 brightest stars was published by ^{Hoag et al. (1961)}. ^{Crawford, Barnes & Warren (1974)} observed 11 stars in the Strömgren uvby system and 24 stars in the Crawford β system, finding significantly variable reddening and a distance modulus of 11.5 mag. This uvby-β   photometry was extended to 40 stars by ^{Reimann (1989)}, who found reddening E(b - y) with a mean value of 0.348 mag and an intercluster variation of about 0.1 mag. His derived distance modulus of 11.94 ± 0.08 and age of 12 Myr are both greater than previous literature values”.

^{Southworth et al. (2004)} carried out a study of the eclipsing binary V453 Cyg (W31) which, they claimed, is a member of NGC 6871. As we will see later, this is not the case.

The other cluster, Dzim 5 was reported by ^{Dolidze & Jimsheleishvili (1966)} but after this, there is only one reference to one of its members. WEBDA does not list distance, reddening age, metallicity or any other quantity except its coordinates. They refer only to the study of ^{Kazlauskas et al., (2013)} related to a new spectroscopic binary with which they establish that Dol-Dzim 5 is not a real open cluster.

Table 1 presents a summary of the most relevant findings of several papers for these clusters.

2 Observations

This article is a sequel to a paper on NGC 6633 that has already been published (^{Peña et al., 2017}, Paper I). The observations were carried out over a long season by two different observers, one from June 22th to 30th and the other from July 1st to 8th, 2016 (ARL and CVR, respectively) with different objects in each one although some were taken continuously (NGC 6633, ^{Peña et al., 2017}, and V 2455 Cyg, ^{Peña et al., 2019}). The open cluster IC 4665 was observed for four nights from June 22nd to June 25th. NGC 6871 was observed for two nights, July 2nd and 3rd and Dzim 5 from July 5th to the 8th. The observing and reduction procedures were described in detail in Paper I. The reduction was done considering both seasons together as one long season to increase the accuracy of the standard stars.

The observations were all taken at the Observatorio Astronómico Nacional de San Pedro Mártir, México. The 0.84 m telescope, to which a spectrophotometer was attached, was utilized at all times. The stars to be observed were selected randomly by drawing concentric circles on the ID charts provided by WEBDA and observing all the bright stars in each circle.

The limit was the faintness of the stars, since to reach the desired accuracy faint stars would require an exceedingly long time of observation. Hence, no astrophysical considerations, nor previous knowledge of the selected stars, was considered. For IC 4665 we measured thirty stars, sixteen for NGC 6871 and fourteen for Dzim 5. Although some of the stars had already been observed, a comparison between the sets gave us confidence in the data as shown from the standard deviations of the values for the same star from several studies, in some cases from three or four different measurements. In the case of Dzim 5 averaging the large discrepancy in the color indexes m ₁ and c ₁ could have led to possible misinterpretations of the physical characteristics of the stars and, hence, of the cluster.

3 Comparison with other photometries

A comparison with previous uvby-β   photoelectric photometry had to be done in order to test the goodness of our results and to enhance the sample by considering the previously measured stars in the direction of each cluster. A search, mostly from WEBDA, was done for references on uvby-β   photometry. These are presented in Table 6, in which the references and the number of the reported stars are listed. However, in our comparison, we only take into account those studies with a significant number of observed stars devoted to each cluster. Those stars with few points taken randomly in studies not devoted to the cluster were not included in the mean. Later, as we will see, a compilation of the stars in the direction of each cluster was done increasing the number of stars in the direction of the cluster, with the proven quality of their values.

Instead of considering the averaged values reported by WEBDA, we opted to include the original sources, because the mean value combined with our photometry would be biased; this saved us from cases like those presented for the open cluster IC 4665. The star W108, which we did not observe, but appears in the compilation of WEBDA, had two radically different values in magnitude reported: 7.508 and 9.820. Equally discrepant are the color indexes. To check which magnitude value was correct, we compared both with those reported in UBV in WEBDA which lists 7.460 and 7.490 mag inV. These systematic differences suggest a variable star. Nevertheless, the value corresponding to 9.820 was not further considered in our analysis.

All the analysis of the data of the three clusters is calculated in a linear fit in which the oldest source data are considered on the X axis and the newest, on the Y axis. A linear relation was calculated in a formula Ynew=A+BXold. The goodness of the fit is demonstrated by both the correlation coefficient R and the standard deviation of the points. For a good fit, A should be small and A close to 1. The correlation coefficient, R, has to be near unity and the standard deviation, small. All the values obtained for each cluster are presented in Table 7. The final column in the table presents N, the number of entries. Figures 2, 3 and 4 present the comparisons among the principal data sets for IC 4665 (^{Crawford and Barnes (1972)} vs present paper; for NGC 6871 ^{Crawford et al. (1974)} vs. ^{Reimann (1989)} and ^{Kazlauskas (2013)} vs. present paper, respectively.

For the cluster IC 4665 we considered the original sources of ^{Crawford & Barnes (1972)} with forty-five entries; of ^{Stetson (1991)} with only six measured stars; and our photometry (58 stars), to calculate the mean values of the stars. Of all the sets only star W32 showed anomalies. Its reported V magnitude in ^{Crawford & Barnes (1972)} is 10.188, whereas the V magnitude in ^{Stetson (1991)}, is 8.330. The large difference could be due to either intrinsic variability or eclipses. We ended up with a sample of fifty-six stars of which ten are presented for the first time, four measured in the three sets and thirteen in the intersection of ^{Crawford & Barnes (1972)} and the present paper’s photometry. Only two were measured in both data sets of the literature, ^{Crawford & Barnes (1972)} and ^{Stetson (1991)}. The mean value for each star is presented in Table 8 along with the standard deviation of each color index.

The column of photometric sources lists the authors whose values we consider in the mean; these are listed at the bottom of the table. The last column presents the spectral type determined from the uvby-β   photoelectric photometry in a procedure described below.

These coefficients are adequate despite the fact that the span of the color index limits is rather low, particularly in (b - y) from 0 to 0.4; m ₁ from 0 to 0.2. On the other hand, the magnitude V limits go from 7 to 11 mag. The linear fit in β was done without W88 which showed a relatively large difference (0.064).

The final list of compiled uvby-β   photoelectric photometry of IC 4665 is presented in Table 8. Column 1 lists the ID of WEBDA, subsequent columns present the mean magnitudes V and color indexes (b - y), m ₁ , c ₁, and β. The standard deviations are also presented, as well as the references utilized in the mean. The references are presented at the bottom of the table.

For NGC 6871 ^{Crawford et al. (1974)} observed 8 stars in the complete uvby-β   system and 12 only in β; ^{Reimann (1989)} presented the majority of the observed stars, a sample of 18 stars in the uvby-β   system and 22 stars in uvby only. His star 101 is considered to be W31. One more source, ^{Cohen (1969)} observed eight stars of the cluster but only in β. The same procedure as in IC 4665 was done, and the coefficients of the cross fits are presented in Table 7. The mean values of the three sources are listed in Table 9, along with the standard deviation and the number of sources involved in the mean. The whole sample is constituted of 23 stars.

In this table all but three stars have standard deviations on the order of hundredths of magnitude. These three stars are W08, W10 and W25. W08 and W25 which were observed by us and are listed in the table, presented dispersions on the order of hundredths of magnitude, so the high dispersion found in Table 9 is due to either the photometry among the different sources ^{Crawford et al. (1974)}, ^{Reimann (1989)} and ^{Cohen (1969)} and ours for star W08 or ^{Crawford et al. (1974)}, ^{Reimann (1989)} and ours for star W25. For W10 we found a dispersion of 0.408, the highest of all the sample. This value was compared with that of ^{Reimann (1989)}. There is always the possibility that this might show a variable nature of the star.

For the open cluster Dzim 5 there are only two sources with Strömgren photometry, that of ^{Kazlauskas et al., (2013)} with only 13 measured stars and ours, with 14 stars. In both sets, basically the same stars were measured. Only two, one in each set, were observed separately. Star 5 was identified but no uvby-β   measurements are presented. The whole sample therefore contains fifteen stars. Studying the results of the coefficients derived from the linear regression of the uvby-β   color indexes (newest data vs. older data in the literature) for the three clusters, we observe that R, the correlation coefficient, is always larger than 0.9, and that the standard deviation is less than few hundredths of magnitude implying that all data sets are consistent, except for those in DZIM 5. In DZIM 5 the correlation coefficient R gave anomalous values in m ₁ and c ₁ implying, of course, no correlation between both sets. In view of this, we averaged only V and (b - y) in the sets of Dzim 5 between ^{Kazlauskas et al. (2013)} vs. present paper. Their photometry did not include H β. The averaged values are presented in Table 10, following ^{Kazlauskas et al., (2013)} ID numbers: In Figure 1, star 15 was added because it was not observed by ^{Kazlauskas et al., (2013)}. Because of the poor linear regression in m ₁ and c ₁ the average was done only for V and (b - y). Table 10 presents the mean of the uvby-β   photometry and the standard deviation of the sources.

4 Determination of cluster parameters

In order to determine the physical characteristics of the stars in the three clusters, IC 4665, NGC 6871 and Dzim 5, the same procedure as in Paper I for NGC 6633 was carried out. This procedure briefly, consists of the following steps:

To evaluate the reddening we first established to which spectral class the stars belong: early (B and early A) or late (late A and F stars) types; the later class stars (G or later) were not considered in the analysis since there is no reddening calibration for MS stars.

To determine the spectral type of each star we utilized the compiled uvby-β   photoelectric photometry of each open cluster calculating the unreddened indexes [m1], [c1] and compared their position with the stars of the open cluster Alpha Per (^{Peña & Sareyan, 2006}) for which the stars have well-defined spectral class. The results are presented schematically in Figures 5, 6 and 7 for IC 4665, NGC 6871 and Dzim 5, respectively. The last column of each compiled uvby-β   photoelectric photometry presents the assigned spectral type.

The photoelectrically classified spectral types of the stars are in very good agreement with those obtained by spectroscopy and reported by WEBDA. It can be seen that the observed stars, which are the brightest in the field, are of all spectral types in the case of NGC 6871 but all late type stars for Dzim 5.

The reddening was determined through Strömgren photometry once the spectral types were classified. The application of the calibrations for each spectral type, of ^{Balona & Shobbrook (1984)} and ^{Shobbrook (1984)} for O and early A type and of ^{Nissen (1988)} for late A and F stars, respectively, allowed us to determine their reddening and hence, their unreddened color indexes. As has been said, no determination of reddening was calculated for G or later spectral types. The procedure has been extensively described in ^{Peña & Martínez (2014)}. Once the reddening is calculated, the distances can be determined for each star.

The output for the three clusters is presented in Tables 11, 12, and 13 for IC 4665, NGC 6871 and Dzim 5, respectively. In each table Column 1 lists the ID of the star, Column 2 the reddening E(b - y); Columns 3 to 5 the unreddened indexes (b - y), m1, c1; Column six lists Hβ, the remaining columns list the V0 and the absolute magnitude MV. The next two columns present the distance modulus, in magnitudes, and the distance in parsecs. In the case of F type stars we present [Fe/H]. The last column provides the assigned membership, either M for the member stars or N for the non-members. Member stars within one sigma from the mean are considered members, the others, non-members. Figures 8, 10 and 11 present the histograms of the distance modulus for the stars.

As can be seen in Figure 8, in the case of IC 4665, the Gaussian peak is at 7.5 ± 0.6. Those stars within these limits are considered to be member stars and are denoted by M in Table 11. Those outside these limits are considered to be non-members and are denoted by NM in the same table. The last column of Table 11 lists the membership probabilities reported by WEBDA. For IC 4665 out of forty-four compiled stars within the spectral class limits, twenty five can be considered members of the cluster. Of these, eleven stars have high membership probability reported by WEBDA, larger than 0.7 and only four have very low membership probability. Of the nineteen stars that we considered out of the cluster limits, only three have been assigned a high membership probability in the literature. So, overall, the agreement is not bad and, hence, the membership that we assigned for those eight stars that did not have previously assigned probability is a new and important result. Among the F type stars that are within the distance limits there are five stars with determined [Fe/H]. The mean value gives -0.221±0.230 if the large value of -0.597 of W71 is considered; without this value the mean value of [Fe/H] is -0.127±0.107. At any rate, for our analysis we considered a solar value. WEBDA does not assign a metallicity value for IC 4665. Figure 9 presents the distance modulus histograms for each spectral type. In Figure 9 it can be seen the peaks for each spectral group, F, A, B and combined; they are all centered at the same distance modulus.

For NGC 6871 most of the observed stars could not be analyzed with the prescriptions to determine the reddening because many of them did not have Hβ measurements. With the compiled sample of Table 9, represented schematically in Figure 3, we realized that most of the stars lie in the early type stars branch. The applicable prescription for early type stars gives the results presented in Table 12. In this table we have separated the stars in three categories depending on their distance; the first group labelled as non-members, and for the second and third group, those below and above distance modulus of 11.5, were labelled as “close” and “far”, respectively. Mean values and standard deviation were calculated for the second and third groups. As can be seen, no overlap is possible, even considering the limits of the standard deviation, assuring us of a separate existence. The histogram of the distance is presented in Figure 10 in which the two groups are clearly discernible.

In this figure the stars are grouped in two peaks. A Gaussian fit determined one at a distance of 1750 ± 80 pc and the other at 2430 ± 194 pc. Given the uncertainties one cannot group all the stars in just one cluster because the spread would be too large. We encountered this situation before when we studied the cluster of NGC 6882/5 (^{Peña et al., 2008}) where we found two clusters at distances of 289 ± 92 pc and 1019 ± 134 pc. At the bottom of each group in Table 12 we present the mean values and the standard deviation of reddening E(b - y), distance modulus and distance for the stars that we consider to be members of each cluster. Unfortunately, the membership probability reported in the literature is for only five stars, none of which was measured in this paper.

In the case of the open cluster Dzim 5 it has been suggested in the literature that there is no clear evidence of the existence of the cluster. We have measured, as did ^{Kazlauskas (2013)}, the fourteen brightest stars in the field. Given the limitation imposed by the telescope-spectrophotometer system, no fainter stars could be observed. In the [m1] - [c1] diagram we can see that all the stars are of spectral types F and later, and there is no evidence of earlier stars. Hence, the analysis with the prescription of ^{Nissen (1988)} has to be considered and yields the results presented in Table 13 for the five F type stars shown schematically in Figure 11. They are all located at nearly the same distance, 215 ± 45 pc, and have a [Fe/H] value of -0.14 ± 0.30. This value is close to that determined from the values reported in Column nine of Table 2 of Kazlauskas (2013). The mean values and the standard deviation of reddening E(b - y), distance modulus, and distance for the stars that we consider to be members are presented at the bottom of Table 13. However, the question of the real existence of a cluster constituted by only four or five late type stars still remains.

To determine age one must first determine the temperature of the hottest main sequence stars. The effective temperature of these hottest stars was calculated by plotting the location of all stars on the theoretical grids of ^{Lester, Gray and Kurucz (1986}, hereinafter LGK86), after we calculated the unreddened colors (Figures 15, 16 and 17) for the correct chemical composition of the considered model.

For IC 4665 we have utilized the c ₀ vs. Hβ diagram of LGK86 which allows the determination of the temperatures of the hottest star with an accuracy of a few hundreds of degrees (Figure 15). This star is W62 at 16900 K.

For NGC 6871 we had to consider the existence of the two overlapped clusters. For the closest, the two hottest stars over the MS are stars W11 and W15 of 25000K and 23000K, whereas for the other cluster, the hottest star on the MS is star W13 at 35000K (Figure 16). For Dzim 5, Figure 17 represents the unreddened points in the (b - y) vs. c ₀ diagram. The hottest star, W15 has a temperature of 6,200 K.

Once the membership and effective temperature have been established, age is determined through the calibrations of ^{Meynet, Mermilliod & Maeder (1993)} as a function of the temperature:

• if the logTe is in the range [4.25, 4.56],

• if the logTe is in the range [3.98, 4.25],

• if the logTe is in the range [3.79, 3.98],

The location of the member stars in the isochrones provided by WEBDA has also been done. Since this figure is presented in the customary HR diagram (B - V) vs. V those stars that were determined to be members have been identified in the data set of ^{Hogg & Kron (1955)}. The chemical composition that best fits the data is 0.019. This model correctly describes the evolutive path. The other important parameters from the isochrone plot of Webda (Geneva) are a distance modulus of 7.73, E(B - V) of 0.174, Av of 0.158 and a log age of 7.65.

In the case of NGC 6871, WEBDA does not report metallicity values and all the stars measured turned out to be early type stars. There was one star of spectral type A which does not belong to the cluster and two more of spectral type G or later for which there is no calibration. In view of this we considered a solar composition. The location of the stars in the LGK86 grids for this cluster is presented in Figure 16. The value of log age is presented in Table 14.

5 Distances to the clusters determined through the GAIA data release 2 (GAIA DR2)

The above mentioned membership determined through the distance distribution on a histogram has been tested on the open cluster Alpha Per (^{Peña & Sareyan, 2006}). However, this determination can be verified using more modern and advanced techniques such as GAIA Data Release 2 for 2018, providing astrometric data from more than a billion sources. The distances cannot be determined merely by inverting the parallax since going from parallax to distance is not trivial. The way to obtain the pure geometric distance is by considering a Bayesian statistical analysis (^{Luri et al. 2018}). Once the parallax has been obtained, the parallax data can be used to infer geocentric distance taking this correction to account for the non-linearity of the transformations and the asymmetry of the resulting probability distribution as mentioned by Bailer-Jones (2018). In their paper they present a set of data of 1.3 billion stars with corrected pure geometric distances from the GAIA DR2 sources.

The data of the 1.3 billion sources are now accessible in the GAIA archive (http://gea.esac.esa.int/archive/) and were used in the present paper to look for the existence of the studied clusters, IC 4665, NGC 6871 and Dzim 5. To do so, we performed a cone search centered on the coordinates (RA/Dec) of the cluster with a radius greater than that assumed by Webda for the cluster, using the whole sample and taking 20 as a magnitude limit.

In the case of NGC 6871 the radius was chosen through visual inspection. In Table 15, Column 1 lists the ID, Column 2, the coordinates from Webda, Column 3, the assumed cluster diameter in arcmin, Column 4, the considered radius which contains the whole cluster, Column 5 the number of stars contained in the cone and Column 6 the distance in parsecs in the histograms (Figures 12, 13 and 14). The uncertainties are the RMS error of the fit. These results are also presented in Table 15 for IC 4665, NGC 6871 and Dzim 5, respectively.

The comparison of our determined distance measures with those in the Gaia catalog DR2 were also done on a star-by-star basis. These are presented in Tables 16, 17 and 18 for IC 4665, NGC 6871 and Dzim 5, respectively; Column 1 lists the ID of Webda, Column 2, the ID of GAIA DR2 ordered by the distance obtained in the present paper (Column 3). Column 4 presents GAIA’s distance and the final column lists the assumed membership in the present paper.

6 Discussion and conclusions

Our study of the three clusters through uvby-β   photoelectric photometry has been done and led us to derive interesting results. The first cluster, IC 4665, is a well-known cluster which has served as a prototype for classical works. This allowed us to do two things with our photometry. First, to corroborate the goodness of our photometric values; and second, with an extended basis, to corroborate the previous findings.

Although the observational data of NGC 6871 are scarce, with the GAIA DR2 results we were able to confirm the existence of two accumulations of stars, which, as in the case of NGC 6882/5 (^{Peña et al. 2008}), show up as two distinct clusters in the line of sight. However, there are some differences in interpretation between the GAIA DR2 results and those we obtained through Strömgren photometry: W11, W15, W21, W24, and W31 are placed in the nearest cluster, while we put them in the farthest one; W7 appears in the farthest cluster with the GAIA DR2 results, but in the nearest with ours. There is also the case of W25. We discarded it as a member of either cluster, but GAIA places it in the nearest cluster. As was mentioned in § 3, in our analysis this star presents a high dispersion, as can be seen in Table 9, and this dispersion could have caused the differences in interpretation. There is one star, W101, which we did not observe. We listed it in Table 9 but different sources assigned very discordant values to its magnitude. In view of these discrepancies we did not consider it in the analysis. At any rate, there are two clusters in the same direction regardless of the distance determination technique.

The final cluster, Dzim 5, as here stated and according to previous works, might not be a cluster despite the claim of its existence by ^{Dolidze & Jimsheleishvili (1966)}. Our findings are puzzling. We found that there is a small group of five or six stars, basically at the same distance, and all are of late spectral type. There is no indication of early type stars, present or past. The comparison with GAIA DR2 confirms the validity of our results since of the five distances determined, four are of the same order of magnitude.

We have compared our findings with those of GAIA DR2. For IC 4665 the results are amazingly coincident, corroborating the goodness of the uvby-β   photometric calibration. For NGC 6871 we found two clusters in the same direction, such as for NGC 6882/5 (^{Peña at al. 2008}). Our comparison with GAIA DR2 is in accordance with the existence of two clusters. Star W25 gives very different results: we determined a distance of 168 pc whereas GAIA fixed its distance at 1682 pc. Stars W4, W5, W7 and W101 do not have distances reported by GAIA. Finally, for Dzim 5 there is an accumulation of a few stars at the same distance determined by GAIA DR2, within the uncertainties. These results, particularly IC 4665, prove, as we did for the open cluster Alpha Per, that the results inferred from uvby-β   photoelectric photometry are trustworthy.

Unveiling the truth of open clusters is not a simple task. The nitty gritty obviously rests on determining the membership of the stars to the cluster. uvby-β   photoelectric photometry is a well-known canonical method. Results like those of the open cluster Alpha Per (^{Peña & Sareyan, 2006}) ensure the credibility of the results. One of these open clusters presented here, IC 4665, which has been well studied, corroborates the goodness of our results since it gives the same results as those previously determined through different methods. There are other clusters which apparently are well-observed, like NGC 6871 but that, when carefully analyzed, reveal that there are few observations (only twenty-three stars with full uvby-β   photoelectric photometry). What was unexpected was the presence of not one, but of another cluster in the same line of sight; the clusters are at distance modulus of 11.2 ± 0.1 and 12.0 ± 0.1. The histogram of the distances to the cluster determined through the GAIA Data Release 2 (GAIA DR2) suggests the presence of another peak. Finally, with respect to Dzim 5 we only add little to the puzzle beyond the categorical statement of its existence by ^{Dolidze & Jimsheleishvili (1966)} to the denial of its existence by ^{Kazlauskas et al., (2013)}, we could not add much. The results of this study showed that the m ₁ and c ₁ indexes of provided by Kazlauskas et al. had serious errors that could have led the authors to erroneous conclusions. Our findings suggest the presence of a small group of only late type stars at the same distance, a cluster by definition, but we could not find any vestiges of early type stars either in the present or in the past. The GAIA DR2 results also show an accumulation of stars at 238 ± 13 pc, close to the value determined in the present work (152 ± 32 pc).