A complete, multicolor survey of absolute proper motions to B of about 22.5 - Galactic structure and kinematics at the north Galactic pole

We present results from a new complete survey of stellar proper motions and multicolor (UBVI) photometry to B ~ 22. 5 in the north Galactic pole field SA 57. The random error of the proper motions is approximately 0. 10" century^-1^ to B ~ 22 and has been tied to the absolute reference frame of galaxies and quasars to better than 0. 01" century^-1^. We stress the importance of secular magnitude and color effects in astrometric surveys of field stars. Serious systematic errors are found in a similar survey by Chiu. Using our high-quality (random errors less than 0. 02 mag) photographic photometry, which has been calibrated with photoelectric sequences in the same field, we have determined ultraviolet excesses and photometric parallaxes for 250 stars in the color range 0. 3 <= (B - V) <= 1. 10 and with U <= 21. 5. Thus we have been able to probe the chemical and kinematical distributions of F, G, and early K dwarfs to distances of up to 25 kpc above the Galactic plane, allowing for the first time an unbiased view of the properties of the thick disk and halo from dwarf stars in situ. We find an unexpectedly sharp break in the distribution of ultraviolet excess at ~5. 5 kpc above the Galactic plane. Features in the kinematical distribution are also seen at this distance. Together, these results suggest that there is a correspondingly sharp change in the mix of stellar populations at a distance of z ~ 5. 5 kpc above the Galactic plane, which we associate with an "edge" of the Galactic thick disk. The halo field stars in the survey show no metallicity gradient and a constant speed relative to the local standard of rest of -275+/-16 km^-1^ at all distances. If the velocity of the local standard of rest (LSR) is 220 km s^-1^, our data imply that the halo is in retrograde rotation about the Galactic center. We contend that the globular cluster system does not accurately trace either the kinematical or the chemical properties of the halo field stars. Our data are consistent with no metallicity gradient for the thick disk as a function of distance above the Galactic plane, but the distribution of thick disk ultraviolet excesses in our survey are not well described by Gaussian functions. The thick disk asymmetric drift varies linearly with distance from the Galactic plane, so that it has its most extreme value, ~- 120 km s^-1^, at the 5. 5 kpc edge. The most straightforward interpretation of the separability of kinematical, chemical, and spatial distributions between the thick disk and halo populations is that the thick disk was not formed as a transitional phase of Galactic collapse between the formation of the halo and thin disk. If true, this would require substantial modifications to the collapse scenarios for the formation of the Milky Way as outlined by Eggen, Lynden-Bell, & Sandage or more recently by Sandage. The properties of the halo, including its discreteness from the thick disk, its lack of a metallicity gradient, and its apparent retrograde motion, seem more compatible with the picture of halo formation as presented by Searle & Zinn. The thick disk and halo are distinct in their kinematical and chemical properties. Previously found correlations between kinematics and abundance in surveys of high-velocity stars in the solar neighborhood are the result of the fact that the metallicity dispersions for the thick disk and halo are broad and overlapping. We present evidence for a candidate halo moving group at Z ~ 4550 pc at the north Galactic pole.