Kem Cook wrote:
This sums to 2,000,000,000 stars in the database for single
exposure depths.
This seems too low by a factor of a few. The USNO-B1.0 catalog
lists 10^9 objects to V=21 (whole sky, about 10% are galaxies).
Dividing by 2 for the sky coverage, and multiplying by 10**(0.3*3.5)
(0.3 for a moderate counts slope, 3.5 for the depth) gives about
6 billion. It could be factor ~2 or so more if the counts are
steeper than 0.3.
I conservatively assume that with 10 years of co-additions, we will see
about twice as many stars.
As you said, this estimate may be too conservative. According to
the SRD, we'll reach r=27.8, or about 3 mag deeper.
Using the same slope as above (0.3), this means that the final
coadded catalog could include about 10 times as many stars as
the single epoch catalog. That is, we could easily end up with
a catalog containing several tens of billions of stars, i.e.
many more than 4 billion.
Unless, of course, we exclude the galactic plane. But why
would we ever do that??? The Milky Way mapping is one of the
four main science drivers!
> Now, there is a choice which needs to be made. We can release light
> curves for all detected stars, or we can only release light curves for
> stars which are detected in the difference images (ie variable stars).
> We will have to set some limit (3 sigma? 5 sigma?) for
> detections in the difference images which will be greater than a
> detectable signal in the folded light curve of a periodic variable.
It's not obvious how to do that. One can't simply set rms level
because there are many stars which are usually steady and then
spend a small fraction of time with a very different brightness
(e.g. eclipsing binaries, eruptive stars, etc). rms cut is not a
good way to fins them. I would argue that we can and should provide
at least photometry (with error bars!) and positions for _all_
observations of _all_ point sources (detected in any single visit).
Tim Axelrod wrote:
> I'm inclined to say that unless an object has a useful SNR in a single
> image, it will not have a lightcurve, but only static properties.
This sounds reasonable, since photometric errors will already be
0.20 mag at the detection limit in single visits.
> In the
> middle one can do photometry of sequences of stacks, each of which is
> made by N individual exposures, where N = 4 or 10 or whatever. I
> suppose there might be some interesting science there, but I do have
> my doubts.
It may help with objects that vary only on long time scales, but I agree
that at this time this is not a strong driver.
I hope this helps,
Zeljko
_______________________________________________
LSST-data mailing list
[email protected]
http://www.lsstmail.org/mailman/listinfo/lsst-data
