The performance of MIPS in actual use depends on many factors. The performance of the arrays themselves is important, but telescope and optical throughput also play an important role. An important factor for users to consider when predicting the sensitivity of a given measurement is the sky background present in the field of interest. Sources of background flux vary with wavelength and position on the sky. The character of the far-infrared background as understood at the time of observation is described below.
2.5.1 Imaging: Sources of Noise and Confusion
At 70 micron, zodiacal light (thermal emission from warm dust within the Solar System) and Galactic cirrus (thermal emission from warm dust clouds in the Galaxy) dominate the background. Zodiacal light is smooth on the scale of a MIPS frame, and decreases by roughly a factor of 4 between sight-lines near the ecliptic plane and towards the ecliptic pole. Zodiacal emission at a particular RA and Dec also varies as a function of time due to the orbital motions of both Spitzer and the zodiacal dust. As a rule-of-thumb, the zodiacal emission can be expected to change by about 1% over a one-day period. Galactic cirrus is concentrated near the Galactic plane, and varies on many spatial scales, including scales that are not resolved by MIPS. Structures within the cirrus include wisps and small knots, and are thus of interest, not only because of the background flux they contribute, but also because they can potentially be confused with sources of interest. The level of cirrus varies greatly, and must be evaluated in detail for observations where high sensitivity at 70 micron is desired.
Galactic cirrus contributes somewhat to the confusing structure of the background, but especially at long wavelengths, this structure is dominated by emission from partially-resolved and unresolved distant galaxies. Observations that are subject to this background are frequently referred to as ''confusion limited,'' because the spatial structure resembles the superposition of many point sources, and determining whether a particular faint source is a part of the background or is a target can be difficult. The structure of the background effectively increases its noise contribution above what would be calculated based on the flux contributed by all of the sources in a region. The structure contributes directly to the variance of the background as it would be measured by aperture or PSF-fitting photometry extraction software, for example.
The accuracy of photometry at 70 and 160 micron will often be confusion-limited. Because MIPS provides much smaller effective beams and higher sensitivity than any previous mission, predicting the confusion limit set by such sources is difficult. Pre-launch estimates of the 1σ confusion limits ranged from about 0.5 to 1.3 mJy at 70 micron, and from about 7 to 19 mJy at 160 micron (Xu et al. 2001, ApJ, 562, 179; Franceschini et al., 2002, ESO conference proceedings, astro-ph/0202463; and Dole et al. 2003, ApJ, 585, 617). Mission data from MIPS produced the following source density criterion (SDC) limits for confusion, due to extragalactic sources (Dole et al. 2004, ApJS, 154, 93): 56 µJy at 24 micron, 3.2 mJy at 70 micron, and 40 mJy at 160 micron. The 5σ photometric confusion limit at 70 micron is 1.5 mJy (Frayer et al. 2006, ApJ, 647, L9). At 24 and 70 micron, the confusion mainly results from the high density of resolved sources; at 160 micron, the confusion arises from a population fainter than the sensitivity limit; see Dole et al. (2004, ApJS, 154, 93). There is a range of values expected, and moreover it can be a function of where exactly one looks in the sky. Note that confusion due to Galactic sources is a strong function of position.
Other factors may influence the effective confusion limit for a particular observation. In some instances it may have been reasonable to integrate somewhat below the level of the confusion, for example when the observer had a priori knowledge of a source position. On the other hand, the presence of a nearby bright source with its diffraction artifacts will increase the effective confusion limit. Moving targets offer the possibility of taking a second ''shadow'' observation, allowing the suppression of confusing sources by subtracting them away.
Observers were also warned that they needed to specify AORs with enough cycles to provide adequate rejection of cosmic rays and other artifacts, even if a very short integration would nominally be adequate to reach the confusion limit.
