NGP-A STARS - DENSITY VARIATION, KINEMATICS AND THE TOTAL MASS DENSITY OF THE DISK

Authors
Citation
J. Knude, NGP-A STARS - DENSITY VARIATION, KINEMATICS AND THE TOTAL MASS DENSITY OF THE DISK, Astronomy and astrophysics, 327(1), 1997, pp. 90-106
Citations number
37
Categorie Soggetti
Astronomy & Astrophysics
Journal title
ISSN journal
00046361
Volume
327
Issue
1
Year of publication
1997
Pages
90 - 106
Database
ISI
SICI code
0004-6361(1997)327:1<90:NS-DVK>2.0.ZU;2-N
Abstract
From an uvby beta photometric survey of the north galactic pole where most A stars (A3 - A9) above b = 70 degrees and brighter than B = 11.5 were included we report the observed number density and age variation with z(pc). Stars between similar to 30 and 1300 pc are included. Our sample consists of 396 sharply defined main sequence/subgiant A stars following the Stromgren - Crawford definition, no B, imA or F type st ars are included, approximate color limits are 0.055 - 0.220 in (b-y)( 0). Proper motions from the PPM and CAMC catalogs are available for al l stars implying U and V velocities. Radial velocities for a subsample (179 stars with b greater than or equal to 75 degrees and z less than or equal to 450 pc) are available from the literature. From a luminos ity point of view the sample is volume complete within similar to 450 pc. For a fit to the observed density variation the choice of function is not obvious and it seems that these A stars may not be fitted by a single law but requires at least two and more likely up to four compo nents. Assuming exponential density variation two groups of stars seem present with scale heights around 65 and 165 pc respectively. The lat ter may be an underestimate due to volume incompleteness for the coole st unevolved A stars beyond 450 pc. The exponentials approximating the number density has a relative normalization 1.00: 0.03 for z = 0 pc. With these two components only the first of Perry's (1969) h = 40 and h = 600 pc groups seem represented in our volume complete sample. Adop ting instead a gaussian variation the data may indicate three componen ts with scale heights 45, 75 and 155 pc and relative normalization 1.0 :0.5:0.2, respectively. Finally a set of four sech(2)(z/h)s with scale heights 55, 85, 165 and 420 pc and relative normalizations 1.00:0.47: 0.18:0.005 for z = 0 seem to fit the number density data well. This co mbination is based on independent 75 pc bins. A sech(2) variation is p erhaps the most obvious choice being the expected variation for an iso thermal and selfgravitating disk. With a 100 pc running binning there are indications that the stars older than the sample median 0.75 Gyr r equire a component with a scale height 680 pc. It may also be present in the young sample, ages less than 0.75 Gyr, but only beyond similar to 400 pc. What perhaps is more surprising is the formal presence of g roups of stars diplaying a density maximum at z = 200 - 250 pc, presen t for both age groups, and most interestingly the angular momentum als o displays irregularities at the same z distances. The hump is not pro posed to represent a physical entity but something like it is required for fitting the observations better than 10%. A consequence of the hu mp is that derivatives of the distribution display large variations. T here seems to be significant differences between this distribution and the sum of four sech(2)s from the previous paragraph. W and U dispers ions seem to have a constant ratio within the z range where we have bo th kinds of data. sigma(U) shows a homogeneous increase from similar t o 20 to 45 km s(-1) within the completeness limit, with a 75 pc indepe ndent binning, and to similar to 70 km s(-1) at 800 pc. The complete U histogram may not be fitted with a single gaussian, three with differ ent dispersions and mean values provide a reasonable fit. A more detai led binning shows partial derivative sigma(U)/partial derivative z to change in a discontinous way at similar to 200 pc. An application of t he data might be a combination of the density and velocity data for a determination of the local, total disk density. But since velocity dis persions do obey some sort of age depence the age - z variation turns out to be of importance and there is a very clear systematic trend of mean ages with z. The average age increases almost linearly to similar to 0.75 Gyr, reached at 200 pc, and then it stays constant within the completeness limit. On the average the younger A stars show a relativ e absence beyond 200 pc implying that very different velocity dispersi ons might be seen on either side of 200 pc. The curve fitted to the de nsity data permits a study of its derivatives. Adopting a linear appro ximation to sigma(W)'s z variation, instead of isothermallity, partial derivative K-z/partial derivative z's variation within similar to 150 pc indicates that the combined Poisson - Jeans equation may only retu rn estimates of the local volume mass density that vary within a facto r of two but the fit by four sech(2)s turns out to be equivalent with a representation by two gaussians with scale heights 129 and 250 pc wh ich implies that between 100 and 200 pc rho(total) = 0.12+/-0.04 M. pc (-3) whereas partial derivative K-z/partial derivative z from the same fit and a linear approximation to the sigma(W) - z variation returns rho(total)(z approximate to 0) approximate to 0.05 M. pc(-3). The age, density and kinematic inhomogeneities revealed in the sample compel u s to the conclusion that the A stars do not trace the galactic potenti al in a simple way - probably because the age mixing varies with dista nce.