2 Observations and data reduction

The observations were performed for ten nights between February and March, 2000 at the BOAO. We used 1.8 m telescope and a SITe $2048 \times 2048$ CCD camera. The gain and readout noise of the CCD camera are 1.8e^-/ADU and 7.0e^-, respectively. The field of view is about $11\farcm6 \times 11\farcm6$ with the scale of 0 $\hbox{$.\!\!^{\prime\prime}$ }$34/pixel. In order to estimate physical parameters of NGC 2539, we carried out UBVI CCD photometry in a clear night on February 10, 2000. Johnson UBV and Cousins I filters were used. For the purpose of obtaining standard magnitudes and colors of stars in the cluster region, UBVRI standard star fields of SA 98 (RA $_{2000} = 6^{\rm h} 52^{\rm m} 22$ $.\!\!^{\rm s}$5, Dec ₂₀₀₀ = -0$^\circ$19$^\prime$6 $^{\prime\prime}$) and SA 99 (RA $_{2000} = 7^{\rm h} 56^{\rm m} 0$ $.\!\!^{\rm s}$5, Dec ₂₀₀₀ = -0$^\circ$18$^\prime$48 $^{\prime\prime}$) by Landolt (1973) were also observed.

We performed time-series photometry to search for variable stars in this cluster. A total of 581 time-series CCD frames were obtained with the V filter. The typical photometric seeing (FWHM) was about $2\hbox{$.\!\!^{\prime\prime}$ }0$. Exposure times were adjusted in a range from 18 to 30 s, depending on seeing and transparency. In order to minimize position-dependent external errors (Frandsen et al. 1989), we carefully controlled the telescope by locating a star at the same position in the CCD frames during our observing run. Figure 1 represents the observed CCD field of NGC 2539 and the detailed observation log is listed in Table 1.

Figure 1: Observed CCD field ( $11\farcm6 \times 11\farcm6$) of the open cluster NGC 2539. Seven new variable stars are represented by open circles with their IDs. The size of the symbol is proportional to the brightness.

   

Table 1: Observation log.

Date	Start	Run	No. of	Observation
(2000)	HJD	time	images	mode$^\dagger$
Feb.  9	2451584 $.\!\!^{\rm d}$02	2 $.\!\!^{\rm h}$0	55	TS(V)
Feb. 10	2451584 $.\!\!^{\rm d}$99	2 $.\!\!^{\rm h}$2	42	TS(V), Short$\cdot$Long
				STD( UBVI)
Feb. 16	2451590 $.\!\!^{\rm d}$99	2 $.\!\!^{\rm h}$0	44	TS(V)
Feb. 20	2451594 $.\!\!^{\rm d}$98	2 $.\!\!^{\rm h}$0	33	TS(V)
Feb. 21	2451596 $.\!\!^{\rm d}$02	3 $.\!\!^{\rm h}$0	47	TS(V)
Feb. 22	2451596 $.\!\!^{\rm d}$95	5 $.\!\!^{\rm h}$0	124	TS(V)
Mar. 14	2451617 $.\!\!^{\rm d}$94	4 $.\!\!^{\rm h}$5	116	TS(V)
Mar. 24	2451628 $.\!\!^{\rm d}$02	0 $.\!\!^{\rm h}$5	17	TS(V)
Mar. 25	2451629 $.\!\!^{\rm d}$00	2 $.\!\!^{\rm h}$0	41	TS(V)
Mar. 26	2451630 $.\!\!^{\rm d}$00	2 $.\!\!^{\rm h}$0	62	TS(V)
	10 nights	25 $.\!\!^{\rm h}$2	581

$^\dagger$ TS(V): V filter time-series.
Short: short-exposed images of NGC 2539
(U: 120 s, B: 30 s, V: 25 s, I: 5 s).
Long: long-exposed images of NGC 2539
(U: 480 s, B: 360 s, V: 180 s, I: 30 s).
STD: standard stars in the SA 98 and SA 99 regions.

All CCD images were preprocessed to correct overscan, trim unreliable and useless regions, subtract bias frames, correct flat fielding and reject cosmic rays using the IRAF/CCDRED package. Then, we obtained instrumental magnitudes of stars from the empirical point spread function fitting method of the IRAF/DAOPHOT package (Massey & Davis 1992). After taking into account the differences of aperture radii between standard stars ($\sim$7 $^{\prime\prime}$) and stars in the dense cluster area (1 FWHM), we derived the following standard transformation equations:

$$\begin{eqnarray*}\lefteqn{ U = u + 19.616 + 0.359 (B-V) - 0.488 X ,~\sigma_{U}=0... ...n{ I = i + 23.335 + 0.034 (V-I) - 0.057 X ,~\sigma_{I}=0.025}\\ \end{eqnarray*}$$

where U, B, V and I are standard magnitudes, and u, b, v and i are instrumental magnitudes. The airmass is denoted as X. Quantum efficiency of the BOAO CCD camera is sharply declined in U filter, resulting in large photometric errors and large dispersions near (U-B)=0.0. Therefore, we used the color term (B-V) instead of (U-B) in the transformation equation for U filter (Sung et al. 2001).

We obtained standard magnitudes and colors of stars in the observed field of NGC 2539 from the above equations, and compared our UBVI CCD photometric results with the previous photoelectric data; by Pesch (1961), $\Delta V_{\rm P1961-OURS} = +0.028 \pm 0.050$, $\Delta (B-V)_{\rm P1961-OURS} =+0.019 \pm 0.037$ and $\Delta (U-B)_{\rm P1961-OURS} = +0.004 \pm 0.033$, by Joshi & Sagar (1986), $\Delta V_{\rm JS1986-OURS} = +0.045 \pm 0.065$, $\Delta (B-V)_{\rm JS1986-OURS} =+0.000 \pm 0.055$ and $\Delta (U-B)_{\rm JS1986-OURS} = -0.034 \pm 0.021$, by Lapasset et al. (2000), $\Delta V_{\rm L2000-OURS} = +0.044 \pm 0.043$, $\Delta (B-V)_{\rm L2000-OURS} =+0.003 \pm 0.047$ and $\Delta (U-B)_{\rm L2000-OURS} = +0.020 \pm 0.062$ (Fig. 2).

Figure 2: Magnitude and color differences between our photometric results and the previous data by Lapasset et al. (2000).

Our V magnitudes are a little brighter by about -0.035 than the previous results, but our color indices are in good agreement with them.

For standardization of time-series images, we adopted an ensemble normalization technique (Gilliland & Brown 1988). To begin with, we chose tens of normalization stars which are bright and unsaturated stars from 11 mag to 13 mag, except for peculiar stars such as variables or stars near the CCD edge. The coefficients in the following equation were derived for each time-series frame:

V= v + a₁ + a₂(B-V) + a₃X + a₄Y

where V and v are the standard and the instrumental magnitudes of normalization stars, respectively. X and Ydenote the position of a star in the CCD frame. a₁ and a₂ are the zero point and the color coefficients, a₃ and a₄ the position coefficients of X- and Y-axis. The coefficients depend on the atmospheric extinction and the instrumental condition of each night. Then, we obtained standard magnitudes of all stars in each time-series frame using the above equation.