3 The open cluster NGC 3114

NGC 3114 is a sparse open cluster projected onto the outskirts of the Carina complex, in a fairly rich Milky Way field. Its membership to the Carina complex is actually not clear. It is a difficult object to study, due to the heavy contamination of Galactic disk field stars which does not allow one to unambiguously separate possible members and define the cluster size.

3.1 Previous results

NGC 3114 was studied several times in the past. The first investigation was performed by Jankowitz & McCosh (1963), who obtained photographic UBV photometry for 171 stars and photoelectric UBV photometry of 52 stars down to V=13 mag. They estimated that the cluster is 910 pc distant from the Sun, has a mean visual extinction E(B-V)=0.27, and an age between $6 \times 10^{7}$ and $2 \times 10^{8}$ yrs.

Afterward Schneider & Weiss (1988) got Strömgrem photometry of 122 stars down to V=12 mag. This study strongly revises the cluster reddening, which the authors suggested to be E(B-V)=0.03.

More recently, Sagar & Sharpless (1991) enlarged the sample of the measured stars, obtaining BV CCD photometry of about 350 stars up to V=16 in seven $3\hbox{$.\mkern-4mu^\prime$ }6 \times 5\hbox{$.\mkern-4mu^\prime$ }4$ regions located quite far from the cluster center, where the contamination is expected to be more important. By assuming the reddening estimate suggested by Schneider & Weiss (1988), they found a cluster distance of $940\pm 60$ pc, in agreement with Jankowitz & McCosh (1963), and an age of $1{-}2 \times 10^{8}$ yrs.

Finally, Clarià et al. (1989) estimated the cluster mean chemical abundance from UBV, DDO and Washington photometry of an handful of giant stars, finding that NGC 3114 is basically as metal rich as the Sun ( $[{\rm Fe/H}]=-0.04\pm 0.04$).

 

 

Table 3: Photometry of the stars in the field of the open cluster NGC 3114 in common with Jankowitz & McCosh (1963). The suffix CP refers to the present study, whereas JM indicates Jankowitz & McCosh (1963) photographic photometry.

ID	JM	$V_{\rm CP}$	$(B-V)_{\rm CP}$	$V_{\rm JM}$	$(B-V)_{\rm JM}$
3	105	10.580	0.069	10.59	-0.08
4	122	10.562	0.092	10.57	-0.03
5	137	11.274	0.097	11.27	0.10
6	73	11.065	0.014	11.12	0.00
7	130	11.393	0.364	11.38	0.22
9	83	11.222	0.068	11.24	0.11
10	76	11.726	0.118	11.78	0.02
11	96	10.268	1.250	10.22	1.12
13	109	12.272	0.235	12.12	0.36
18	128	13.004	0.302	12.90	0.36

3.2 The present study

We provide UBVRI photometry for 2060 stars in a $3\hbox{$.\mkern-4mu^\prime$ }3 \times 6\hbox{$.\mkern-4mu^\prime$ }5$ region centered in NGC 3114, up to about V=22. The region we sampled is shown in Fig. 1, where a DSS map is presented.

According to Jankowitz & McCosh (1963) the cluster should have a diameter of $32^{\prime}$, although this estimate is rather uncertain, due to the difficulty to isolate the cluster from the field. Anyhow, the cluster seems to be rather extended, and our photometry covers only the central region, with no overlap with Sagar & Sharpless (1991) photometry.

The CMDs for all the measured stars in the planes V-(B-V), V-(V-I) and V-(V-R) are shown in Fig. 2. In the cluster center there are no stars brighter than V=10.0. Therefore, with respect to Sagar & Sharpless (1991), who provided the deepest photometry before our study, we do not find any indication of a Red Giant (RG) clump (see Fig. 9a in Sagar & Sharpless 1991).

The Main Sequence (MS) extends from V = 10 up to V = 22, and gets wider at increasing magnitudes. Several causes concur to broaden the MS: the presence of unresolved binary stars, the photometric errors and the contamination of fore-ground and back-ground stars.

A probe of the heavy contamination is the Galactic disk RG branch population, readily recognizable in the almost parallel sequence which departs from the MS at $V\approx 20$. This is a common feature in the CMDs of stellar fields in the direction of the Carina spiral arm (see Vallenari et al. 2000).

We have 10 stars in common with Jankowitz & McCosh (1963), which are listed in Table 3. The mean differences turn out to be:

$$V_{\rm CP} - V_{\rm JM} = 0.020\pm0.062$$

$$(B-V)_{\rm CP} - (B-V)_{\rm JM} = 0.068\pm0.077$$

where CP indicates our photometry, and JM stands for Jankowitz & McCosh (1963). Taking into account the different techniques used in extracting the photometry, the agreement is good for both magnitude and colors.

We do not report the difference between the (U-B) colors, since Jankowitz & McCosh (1963) measured the color $(U_{\rm c}-B)$, with the filter U defined in the Cousin system. This is quite different from the standard Johnson (U-B), and the authors do not provide the Johnson color for the stars listed in Table 3.

Figure 3: NGC 3114 stars brighter than V=17.0 in the (B-I)-(B-V) plane.

Figure 4: Two color diagram for the stars in the field of NGC 3114 with V $\leq$ 17.0. The arrow indicates the reddening vector. The solid line is the empirical un-reddened ZAMS from Schmidt-Kaler (1982), whereas the dashed and dotted lines are the same ZAMS, but shifted by E(B-V)=0.20and E(B-V)=0.60, respectively.

Figure 5: Reddening corrected CMDs for the stars in the field of NGC 3114 with $V\leq 17.0$ The dashed line is a ZAMS shifted by ( m-M)=12.50, whereas the solid one is the same ZAMS, but shifted by ( m-M) = 9.80. Filled circles indicate cluster candidate members, while empty circles indicated field stars. Finally, the dotted line is an isochrone for the age of 300 Myr.

3.3 Reddening

In order to obtain an estimate of the cluster mean reddening, we analyse the distribution of the stars in the (B-I)-(B-V) plane, which is shown in Fig. 3.

The linear fit to the main sequence in the (B-I)-(B-V) plane,

$$(B-I) = Q + 2.25 \times (B-V)$$ (1)

can be expressed in terms of E(B-V) for the R_V=3.1extinction law as

$$E(B-V) = \frac{Q-0.014}{0.159} ,$$ (2)

following the method proposed by Munari & Carraro (1996a,b).

This method provides a rough estimate of the mean reddening and, as amply discussed in Munari & Carraro (1996a), can be used only for certain color ranges. In particular Eq. (2) holds over the range $-0.23 \leq (B-V)_{0} \leq +1.30$. A least squares fit through the stars brighter than V=17 gives Q = 0.041, which, inserted in Eq. (2), provides $E(B-V)=0.17\pm 0.12.$

The uncertainty is rather large, and is due to the scatter of the stars in this plane, which indicates the presence of stars with different reddening, presumably a mixture of stars belonging to the cluster and to the field.

 

 

Table 4: List of the candidate members of NGC 3114 obtained in the present study. Two of them ($\char93 3$ and $\char93 8$) are newly discovered members.

ID	X	Y	V	(B-V)	(U-B)	(V-R)	(R-I)
2	289.93	159.96	10.216	0.017	-0.209	-0.142	0.070
3	60.20	129.03	10.580	0.069	0.046	-0.047	0.124
4	271.54	495.20	10.562	0.092	0.091	0.012	0.090
5	408.86	238.26	11.274	0.097	0.086	-0.011	0.087
6	-384.24	137.45	11.065	0.014	0.051	-0.077	0.196
8	213.23	346.53	11.746	0.131	0.229	0.045	0.013
9	-173.94	442.02	11.222	0.068	0.117	-0.014	0.118

To better derive the reddening distribution and identify cluster members, we plotted all the stars brighter than V=17 in the two color diagram of Fig. 4. With filled circles we indicated stars having a common low reddening $E(B-V)=0.07\pm 0.03$. They lie very close to the unreddened empirical ZAMS (solid line) taken from Schmidt-Kaler (1982).

Open triangles represent all the other stars, which exhibit a much larger reddening. These stars do not suffer from the same amount of reddening. To have an idea of the reddening of the field stars we have overimposed the same ZAMS, but shifted by E(B-V)=0.20(dashed line), and by E(B-V)=0.60 (dotted line), respectively.

In conclusion, two populations seem to exist: seven stars sharing a common low reddening, which are presumably cluster members, and all the other stars having larger reddening, which are field stars.

3.4 Age and distance

We estimate the age of NGC 3114 by studying the reddening corrected CMDs (see Fig. 5). In this plot filled circles are our candidate members, whereas empty circle are field stars. The bulk of this latter is fitted by a ZAMS shifted by (m-M) = 12.50 (dashed line), basically at the distance of the Carina spiral arm. Nevertheless, there seem to be stars located basically at any distance between us and the Carina spiral arm, confirming previous indications that the cluster is heavily contaminated by stars from the Galactic disk field.

The candidate members form a tight sequence, close to a ZAMS shifted by (m-M)=9.80 (solid line).

In order to estimate the cluster age, we over-imposed a solar metallicity isochrone (dotted line) from Girardi et al. (2000), for the age of $3 \times 10^{8}$ yrs. In fact, the brightest members lie off the ZAMS, and are clearly leaving the MS.

In conclusion, NGC 3114 is a rather poorly populated star cluster, heavily contaminated by field stars. From the study of the cluster candidate members (see Table 4), in our field we derive a reddening $E(B-V)=0.07\pm 0.03$and a distance of $920\pm 50$ pc from the Sun, in fair agreement with previous investigations.

Figure 6: DSS map of a region around Collinder 228. The box confines the field covered by our photometry.