Up: IPP at the ROE

IPP Filename Structure

There are two parts to the filenames

• Observation
• Data reduction product
• In the form Observation.DataReductionProduct

Observation

Consider: o4997g0240o.77966

• o = OTIS (Observatory, Telescope, Instrument, and Software). Nothing else possible?
• 4997 = Julian Date - 2,450,000
• g = Gigapixel Camera. Let's us know this is a PS1 observation. Though what will it be for PS4?
• 0240 = Observation sequence number
• o = object, bais, dark, flat, etc.
• 77966 = Exposure ID

Data reduction product

Consider:

• Either:
  • o4997g0240o.77966.ch.41966.XY76.trace
  • o4997g0240o.77966.wrp.16984.skycell.1480.060.trace`
• 'ch' = chip or warp or ...
  • If ch then it's a chip (i.e. detector coordinates)
    • 41966 = chip analysis. Not clear what this means.
    • XY76 => chip in the GPC frame
      • i = 0 to 7, j = 0 to 7, with no 00, 07, 70, or 77)
    • trace
    • psf
    • pplmerge.mdc
    • mdl
    • log
    • cmf
    • ch.wt
    • ch.mk
    • ch
  • If wrp then it's a warp (i.e. mapped to sky coordinates)
    • 169841 = ?
    • skycell = invariant. Just lets you know the next two numbers are skycell references
    • 1480 = Skycell 1480. Doesn't tell you much if you don't know the particular choice of skycell geometry. Just that it's the 1480th cell in the sky.
    • 060 = the 60th subcell in the skycell.
      • trace
      • stats
      • pswarp.mdc
      • log

Skycells

Skycells are devised by tessellation of the 4pi sky. Equal-area rectangles are centred on the tessellation points such that no space is left covered and the borders of all rectangles overlap other rectangles.

Each skycell is further divided into subcells on a grid which (in future at least) overlap by a few arcsecs with neighbouring subcells such that no part of the sky is stuck only on a boundary.