Search for a command to run...
In an earlier paper (Pel et al. 1988, Paper I) photoelectric photometry in the Walraven VBLUW system was presented for magnitude limited samples of F-G stars in three high latitude fields of the Basel halo survey. The stars were selected from the photographic RGU survey using the color criterion (G - R) <1.15, i.e. spectral types earlier than about G7. The primary aim of this project is to study the stellar metallicity gradient perpendicular to the galactic plane for a kinematically unbiased in situ sample of main sequence turnoff stars in the disk-halo transition region, and to compare the observed metallicity distributions with multi-component Galaxy models. In this paper we interpret the VBLUW data for 112 stars in field SA 141 (South Galactic Pole). This sample is complete down to V = 14.8 mag. Detailed reddening corrections are derived from the IRAS 100 mu map of SA 141. The mean reddening in this field appears to be very low: E(B - V) = 0.007 mag; most program stars have color excesses below 0.02 mag. Values of T-eff, log g and [Fe/H] are determined by means of new calibrations of the VBLUW colors of F-G stars. The photospheric parameters are combined with theoretical isochrones to determine absolute magnitudes, distances and (for the brightest stars) ages. Most observed metallicities are in the range -1.0 less than or equal to [Fe/H] less than or equal to 0, confirming that the Basel halo survey contains mostly intermediate population stars. Up to z = 900 pc a rather steep mean metallicity gradient in the z direction is found: -0.55 +/- 0.1 dex/kpc. This tends to flatten out to -0.18 dex/kpc at z = 1-2 kpc. Including 160 additional fainter stars from the Basel survey for which (less accurate) [Fe/H] values can be estimated from the photographic RGU photometry, we determine an overall mean [Fe/H]-gradient for z <4 kpc of -0.23 +/- 0.04 dex/kpc. A comparison with multi-component models of the Galaxy shows that a simple disk-halo model does not fit the observed [Fe/H] distribution for SA 141. The agreement is very satisfactory, however, when a thick disk component is included, with ([Fe/H]) = -0.6 and sigma [Fe/H] = 0.3 dex. No attempt is made yet to determine a ''best solution'' for the three-component fit to the SA 141 data; this will be done in combination with the data for two additional Basel fields, SA 94 and SA 107. The results for SA 141 are consistent with Sandage (1987) who proposed a local density normalization of 10% and a vertical scale height of 1 kpc for the thick disk. The observed stellar number-distance distribution for SA 141 is in excellent agreement with the prediction by the three-component model. This is an important independent check on the reliability of our absolute magnitudes and photospheric parameters.