Volume 528, April 2011
|Number of page(s)||22|
|Published online||11 March 2011|
Observations of standard star fields were carried out during the observing runs in September 2008 and November 2009 in order to provide an accurate calibration of the science observations. Fields were selected from the Landolt standard fields (Landolt 2007, 1992), to be used together with Stetson standard star photometry (Stetson 2000). Using the CTIO 4 m/Blanco telescope observations of standard star fields were taken during the night in between science observations, in order to monitor the photometric fidelity throughout the night. Standard fields were selected to cover a broad range of airmass and colour, to allow adequate determination of the airmass coefficient and colour term in the photometric solution. In order to ensure of the most accurate calibration possible additional observations of standard stars and Sculptor fields were taken using the CTIO 0.9 m telescope during 3 nights in 2008 under photometric conditions. Using these observations the calibration of the 4 m Sculptor fields using the 4 m standards can be checked and adjusted to the photometric solution obtained under photometric conditions on the 0.9 m telescope. Table A.1 lists the observations of standard fields taken with the 4 m Blanco telescope, during the runs in 2008 and 2009. The names of the standard fields are given, along with information on exposure times, seeing conditions (on image) and airmass. Table A.2 lists the observations of standard fields taken with the 0.9 m telescope, under photometric conditions during the 2008 run. Field names are listed, as well as exposure times, airmass and seeing conditions on image.
List of observed standard star fields with the 4 m CTIO Blanco telescope during two observing runs.
List of observed standard star fields with the 0.9 m CTIO telescope.
Obtaining an accurate photometric calibation is essential for a reliable interpretation of a CMD. For example the absolute luminosity of a main sequence turn-off star correlates with age. The observing strategy was chosen in such a way as to combine deep 4 m science images with 0.9 m calibration images taken under photometric conditions. These images are used to ensure the most accurate photometric calibration possible for the science images.
A comparison of the calibrated magnitude and true magnitude of the standard stars used to obtain the photometric solution versus observed magnitude (for both the 4 m and 0.9 m data) is shown in Fig. B.1. Similar plots showing the magnitude difference versus airmass and colour are shown in Figs. B.2 and B.3 respectively. A line indicating the zero level residual (black) and the mean of the residual (blue) is shown, along with errorbars denoting the average photometric error. These plots show the accuracy of the calibration applied to the science data. The standard deviation of the residuals is σB ≈ 0.033, σV ≈ 0.028, σI ≈ 0.032 for the 4 m data and σB ≈ 0.038, σV ≈ 0.028, σI ≈ 0.029 for the 0.9 m data. Figure B.1 shows that the zeropoint correction is consistent across all magnitudes used in the photometric solution, with the mean level of the residuals (blue line) being less than the average photometric error. Figures B.2 and B.3 show that the effects of colour and airmass are well taken care of, with
the mean level of the residuals (blue line) being less than the average photometric error. Residuals in the B band observations of the 0.9 m telescope are most offset from zero, due to the poorer seeing conditions and larger photometric errors in the B band. However, the B band residuals are consistent with zero in all plots, given the average photometric error in B, showing that an accurate calibration was obtained.
A second check of the photometry is done by comparing aperture and PSF magnitudes. We determined aperture magnitudes for small samples of stars in standard fields and in the central Sculptor pointing. A photometric solution was obtained from the aperture photometry, and used to calibrate the aperture magnitudes in the Sculptor central field. Then, a comparison of the PSF and aperture magnitudes is made (shown in Fig. B.4) for both the 0.9 m and 4 m data. The standard deviation of the residuals is σB ≈ 0.101, σV ≈ 0.040, σI ≈ 0.033 for the 4 m data and σB ≈ 0.108, σV ≈ 0.028, σI ≈ 0.029 for the 0.9 m data. Figure B.4 also shows that the differences between aperture and PSF magnitudes are negligible within the average photometric errors, giving confidence to the reliability of our PSF photometry. The B band residual is once again most offset from zero (by 0.05), but consistent with zero given the photometric errors (± 0.061).
Next, the calibrated 4 m and 0.9 m PSF mags and aperture mags are compared to check whether the 4 m data is consistently calibrated with the 0.9 m data taken under photometric conditions. A comparison of both data sets for the central Sculptor field is shown in Fig. B.5 for PSF and aperture magnitudes respectively. Solid black errorbars have been overplotted for the PSF magnitudes, indicating the average error on the magnitude difference at bright, intermediate and faint magnitudes. For all three selections the mean of the residual is smaller than the error on the magnitude difference (from bright to faint: B: 0.0043 < 0.0617, 0.0079 < 0.0880, 0.0095 < 0.1546, V: 0.0099 < 0.0176, 0.0154 < 0.0484, 0.0187 < 0.0890, I: 0.0002 < 0.0125, 0.0072 < 0.0272, 0.0185 < 0.0844). For the aperture magnitudes the mean of the residuals (B: 0.0170, V :0.0321, I: 0.0236) is comparable to the average error (B: 0.0759, V: 0.0292, I: 0.0191). The relatively high residuals are possibly due to the low number of stars with aperture magnitudes used in the determination of the photometric solution for the 0.9 m data.
The above figures show that the 4 m calibration properly takes into account zeropoint, airmass and colour effects. Furthermore, the calibration is consistent between aperture and PSF magnitudes, given the photometric errors. The absolute calibration of the 4 m data is consistent with that of the 0.9 m data for PSF magnitudes and to a lesser extent for aperture magnitudes. Thus, an accurate absolute calibration is achieved for the PSF magnitudes used in the final photometry catalog. The final catalog has an average accuracy due to random errors for all filters of ≈ ± 0.002 for the brightest stars, while at the faint end the accuracy is ≈ ± 0.2. The accuracy of the calibration varies for different magnitudes and filters, but the comparison of calibrated and true magnitudes shows an accuracy of ≈ 0.04 mag or better across the magnitude range used.
Comparison between calibrated and true magnitudes for the standard stars used to obtain the photometric solution. Each plot shows magnitude residuals vs. true magnitude for different filters for the 4 m data (left) and 0.9 m data (right). Stars classified as outlyers when determining the photometric solution are marked in green. The average standard deviation of the residuals over all filters is ≈ 0.033 for the 4 m data, and ≈ 0.03 for the 0.9 m data. A line indicating the zero level residual (black) and the mean of the residual (blue) is also shown (visible only if there is an offset from zero), along with errorbars denoting the average photometric error. The zeropoint correction residuals are consistent with zero across all magnitudes, given the average photometric error. The mean of the residual in the 0.9 m B (upper right) solution is most offset from zero (by 0.024) due to seeing conditions and large photometric error in B. However, the residuals are still within the average photometric error (± 0.032), indicating an accurate calibration is achieved.
|Open with DEXTER|
Magnitude residuals vs. airmass for photometric standard stars for the 4 m data (left) and 0.9 m data (right). See Fig. B.1 for details. No trend with increasing airmass is visible, showing that the effects of airmass are well taken care of in the photometric calibration.
|Open with DEXTER|
Magnitude residuals vs. colour for 4 m (left) and 0.9 m (right).standard stars. See Fig. B.1 for details. No trend with colour is visible, showing that the photometric calibration correctly takes care of the effects of the colour term.
|Open with DEXTER|
Comparison of aperture and PSF magnitudes in the central Sculptor pointing for the 4 m data (left) and 0.9 m data (right). See Fig. B.1 for details. The residuals are consistent with zero, given the photometric errors, showing that the calibrations using aperture and PSF magnitudes agree.
|Open with DEXTER|
Comparison between 4 m and 0.9 m photometry for the central Sculptor pointing for PSF (left) and aperture (right) magnitudes. A plume of short period RR Lyrae variable stars is visible in all bands in the PSF data at B ~ 20.5 − 21; V ~ 20 − 20.5; I ~ 19.5 − 20.5. Solid black errorbars overplotted in the left-hand figures indicate the average error on the magnitude difference at the corresponding magnitude. The residuals of the PSF magnitudes are consistently smaller than the error on the magnitude difference, while for the aperture magnitudes the residuals are similar to the error.
|Open with DEXTER|
© ESO, 2011
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.