|
|
MIPS : Best Observing Practices |
|
These notes come from the SOM,
section 8.2.3.1 and surrounding sections, and the section references
here refer to sections in the SOM. For much more discussion,
please see the SOM.
General MIPS adviceLatent images at 24 microns created by bright sources could affect data, and are not currently removed by the SSC pipelines (see section 8.1.3.3.2). There is a short-wavelength light leak at 160 microns; extragalactic objects are unlikely to be affected, but stellar programs might be, especially debris disk programs looking for subtle excesses (see 8.1.3.6.6). A MIPS second is slightly longer than a true second (see section 8.1.4.1); plots and tables in this chapter and on the MIPS SSC website are in MIPS seconds unless otherwise specified, but times returned by Spot are true seconds. At least four images of a 160 micron target are recommended to ensure reliable photometry. Images of objects near the ecliptic plane are likely to include asteroids and/or Kuiper Belt objects; a second observation at a later time is necessary to ensure that moving targets have been removed. Observations at MIPS wavelengths may be limited by noise (including confusion noise; see section 8.2.2.1), but observers should ensure sufficient data redundancy for cosmic ray rejection, even if the confusion limit is reached with shorter exposures. It is best to build up long integrations with long exposure times, e.g. large total integration times built from 3 s exposures will not produce as high-quality a final product as fewer cycles of 10 s exposures.Observers should always visualize their observations using the latest version of Spot to be sure that the observation is doing what was intended.
Photometry-specific adviceSuper-resolution data need to be post-processed to take advantage of the sub-pixel sampling inherent in the AOT; the SSC pipelines do not currently provide this service.The AOTs each provide multiple images of the target; observers select the number of times that the 'standard cycle' is repeated. When defining raster maps, observers select an additional "map cycle" which indicates how many times the entire map is carried out; an AOT cycle is completed at each map location. In photometry observations, the dither pattern is such that the region of highest- quality S/N and best coverage is less than a full FOV; use Spot visualizations to determine the region of sky actually covered by all pointings for any given AOR. Because of the smaller effective size of the 70 micron array compared to previous expectations, to obtain the same coverage as pre-launch expectations for large- field photometry, some users will need to efficiently map the approximately 5x5 arcmin (2.5x2.5 arcmin fine scale) area covered by the original AOT. In order to do this, we suggest that the user create a cluster target with offsets of (0,+80), (0,-80) [for fine scale, (0,+60), (0,-60)] in array coordinates observing "offsets only." This will allow recovery of the full area of the original AOT while saving the slew tax from mapping with 2 AORs.
Scan-specific adviceStep by at most 1/2 array steps to obtain full sky coverage at all 3 wavelengths. For highest S/N coverage, step by at most 1/4 array steps. The AOT provides inherent redundancy in all wavelengths, but there are more images of a given source at 24 and 70 microns than at 160 microns. Fast scans provide only half coverage at 160 microns, so the scans must be repeated to approximately fill in the gaps. To achieve the recommended minimum 4 images of the same patch of sky, observers should plan additional scans with small cross- scan steps.Bright objects may leave latents in the data; to limit these effects, try to arrange for covering the same region of the sky in two directions (e.g., forward and back). Extremely bright objects might be isolated in a separate AOR. Note that the first several frames of a scan leg necessarily have extrapolated stim backgrounds (see Data Handbook for more information on this effect). In general, but especially in the case of fast scans over bright objects, be sure to have enough background measurements around your source that extrapolated stims are not an issue.
SED-specific adviceSince the saturation limits are so high for the 70 micron array, it's rare to saturate SED mode. Thus, except in situations where extremely bright objects are being observed, one should always use the 10 second exposure time.
|
help@spitzer.caltech.edu http://ssc.spitzer.caltech.edu/mips/bop.html
