2009.05.04

I've placed a collection of data for release on our rsync server. This represents a couple of passes on the Three Pi data from Run 2 through just the single-image analysis (chip & camera). There are two DVO databases and a collection of 3263 smffiles (one for each exposure).

• rsync://ipp0002.ifa.hawaii.edu/catdirs/ThreePi.Run2.v0 (80GB) : this directory contains the DVO database generated from the analysis of the grizy images from the Three Pi observations from October 1 - December 2, 2008. All processed exposures have been inserted into this database, regardless of the photometric conditions or the measured transparency. One caveat relative to this dataset: we have recently discovered a bug in the reported parameters errors from the non-linear fitting. The magnitude errors are not very far off, but the reported position errors are very wrong: you will see X_CCD_ERR, Y_CCD_ERR values of tens and hundreds of pixels. We have since fixed the code and will do another processing run eventually.

• rsync://ipp0002.ifa.hawaii.edu/catdirs/catdir.20090327 (240 GB) : this directory contains the DVO database generated from the analysis of only the izy exposures from November 2008. In this analysis, aperture photometry was calculated only for bright sources (S/N > 50). The only reason this database may be of interest over the one above is that in this database, I have merged the detections with the 2MASS photometry. We will be making this merge as a regular IPP data product, and those catalogs should be similar to this one. this database also defines JHK average magnitudes, making it easier to generate color-color diagrams which include one of the 2MASS photometry bands. In fact, in this case, I have merged the detections into an existing 2MASS database of the *entire* sky. This is probably not ideal, in retrospect, simply because the DVO detection-to-object algorithm will not make the best choices if the lower-resolution 2MASS data is already in place.

• rsync://ipp0002.ifa.hawaii.edu/catdirs/smffiles.ThreePi.Run2.v0 (64GB) : this directory contains the raw smffiles generated by the processing and then used to build the ThreePi.Run2.v0 DVO database. One particular point of interest: these smffiles now include a table of the matched detections between the PS1 observations and the astrometry/photometry reference catalog (synth.grizy). I've noted recently that there are some surprising trends of the PS1 magnitude vs the synthetic magnitude as a function of magnitude. In fact, that result was based on a very limited set of data (a single exposure!). It would be helpful if someone could dig into these matched reference tables and explore the trends of PS1 magnitudes vs the reference color and/or the reported magnitude. (NOTE: as of 16:48 HST, the smffiles were stilll being transfered from MHPCC. if you rsync this directory before ~18:00HST, you may want to try again to be sure you get them all. there should be 3263 smffiles.

Also, these DVO databases have many directories. I'm attaching a script which can be used to launch a number of parallel rsync jobs (edit the file to choose the catdir.20090327 database). rsync does not do very well with optimizing throughput in a single thread, and we find this helps to take advantage of the available bandwidth. On the other hand, the bandwidth to some sites may be low enough that no advantage is gained this way.

Finally, note the sizes of the directories! Don't hurt yourself...

aloha gene