Although we have just emphasized that the sensitivity of MIPS depends strongly on where one is looking on the sky, we have included here the results in the best sky regions. Note that starting in 2005, regular MIPS operations changed to include ''cold'' (telescope at ~5.5 K) and ''warm'' (telescope at ~8.5 K) campaigns. The sensitivity at 24 and 70 μm remains the same for any MIPS campaign; the 160 μm array is completely saturated during MIPS-warm campaigns, so its sensitivity is not relevant.
The sensitivity of the 24 µm band is shown in Figure 2.16 and Figure 2.17. The detector array is well behaved and the values for S/N can be scaled as the inverse square root of the integration time for point and extended sources. The analogous plots for the other bands are available on the website. The conversion of point source sensitivity to extended source sensitivity can be found in Table 2.6.
Because of the multiple response times of the far infrared arrays, sensitivity estimation is more complex for them. The best values require high-pass filtering to remove the slow component of response; see Chapter 7: Data Features and Artifacts for additional information (including practical advice) on such filtering. The high-pass filtering also suppresses extended emission from sources. Good data can be obtained on such sources, but it requires careful hand processing to remove ''streaks'' left from slow response effects. Such removal can best be done on scan map data, and it is highly desirable to obtain scans over the source in both directions.
Real-life specific examples of extended source work with MIPS can be found in the ApJS special issue; see, e.g., the paper by Engelbracht et al. (2004, ApJS, 154, 248) on NGC 55, Helou et al. (2004, ApJS, 154, 253) on NGC 300, and/or Regan et al. (2004, ApJS, 154, 204) on NGC 7331.
The sensitivity of the far infrared arrays can also be degraded if the stimulator flash interval is too short. In particular, the 3 sec photometry mode has a sensitivity degraded by about a factor of 2 from the numbers indicated above. We have left the rapid stimulator rate in this mode because it should help obtain higher photometric accuracy on bright sources. For faint sources, the 10-second photometry mode should be used in preference to building up integration time through many 3-sec cycles.
Figure 2.16: Sensitivity in scan map mode at 24 µm. At the medium and slow scan rates, the effects of cosmic rays and increased read noise largely offset the expected gains from increased exposure time.
Figure 2.17: Sensitivity for photometry at 24 µm, compact source (i.e., one that can always be kept on the array during the measurement).
Table 2.6: Conversion from point source sensitivity to extended source sensitivity.*
Sensitivity (MJy/ster) = 1.04 x 10-3 x Point Source Sensitivity (micro-Jy)
70 µm default
Sensitivity (MJy/ster) = 8.81 x 10-5 x Point Source Sensitivity (micro-Jy)
70 µm fine scale
Sensitivity (MJy/ster) = 1.77 x 10-4 x Point Source Sensitivity (micro-Jy)
Sensitivity (MJy/ster) = 2.52 x 10-5 x Point Source Sensitivity (micro-Jy)
*Values are for default BCD pixel scale.
2.5.5 Spectroscopy: Spectral Energy Distribution Sensitivity
SED mode can achieve 5σ point source sensitivities of 82, 201, and 447 mJy at 60, 75, and 90 µm in 500 seconds.