The SDSS Early Data Release has 4480 stars on the Equator with 'good' spectra and radial velocities (good to about 15 km/s?, have to recheck this accuracy) of various stars of various types.
From the EDR equator data, from Paper I there are about 4100 stars with g' < 21 which have imaging photometry and are classified as A/BHB/BS stars. about 20% 777/4100 of these with g' < 20 have SDSS spectra.
It is interesting that in the region of Sagittarius North (200 < RA < 230) the dispersion of the faint A and F stars is quite small. Elsewhere it's a more uniform 'halo' type value sigma = 250? km/s --> 10^12 M_sol halo.
In the region of Sagittarius South (30 < RA < 50) on the Equator, the dispersion is larger.
For redder stars F-M, the fraction with spectra is lower, since BHB stars are specifically targetted for spectra by SDSS, while later types are not.
There's a cluster of stars at RA=180 degrees (very interesting as we see an F star excess there, however, part of the apparent spectral clustering is just sampling, as the SDSS took lots of stellar spectra (as opposed to Galaxies) on these few plates).
There are still lots of F stars with spectra to g' < 20, however, >750 with 18 < g' < 20.
Here they are split up by their g'-r' colors into more or less standard spectral types.
One can see that as one goes from the brighter (more distant) types to the fainter (nearer) types that the RV's of the stars go from following the -V_LSR red line to following the RV=0 (heliocentric) green line.
The coverage is not completely uniform, of course.
Estimated available targets:
F stars: in Paper II, 82000 with 0.1 < g-r < 0.3 and 18 < g < 20 in about 500 sq degrees --> 150 per square degree available and more if you want to go fainter than g=20. Going fainter, there are about 100 F stars per square degree with 20 < g < 21 and another 100 per square degree with 21 < g < 22. So in the Focas field of 6 arcmin x 6 arcmin = 0.01 deg sq (is this correct?) we would have only about 3.5 F stars per field.
A stars: in Paper I, 4100 with -0.3 < g-r < 0.0 with 16 < g < 21.5 in about 400 sq degrees --> 10 per sq. degree available.
We should also think if there might be a project which we could do easily with the data in hand. Perhaps get a Metalicity for some of the 4000 stars that the SDSS has spectra for and 'rerun' the Morrison Flynn and Freeman (1990) 'metal weak thick disk' experiment? We don't have the same lines of sight that they do, tuned to tangent directions, but we still may be able to get some kind of velocity ellipsoid or (velocity displacement - v_LSR) vs. metalicity or some such relation.
A stars A stars with Spectra
F stars F stars with Spectra
G stars G stars with Spectra
K stars K stars with Spectra
M stars M stars with Spectra