The clearest limitation of snapshot observing is sensitivity (see Table 5); snapshot mode is suitable only for bright sources. Below 2 GHz and especially at P Band, the high sidelobe levels of snapshot beams exacerbate the problems created by confusing sources. Snapshots of fields near the galactic plane using the more compact VLA configurations at lower frequencies are frequently dominated by sidelobe clutter from confusing sources rather than by the noise that is quantified in Table 5. These problems are less severe above 2 GHz because of the smaller primary beam and the typical source spectrum (see Confusion).
The second limitation of snapshot observing is the restricted
angular size scale 
over which the u-v coverage of a
snapshot satisfies the sampling theorem and thus
permits reconstruction of the correct sky brightness distribution.
Table 6 estimates this limitation for snapshots on the meridian in the
standard VLA configurations and frequencies, averaged over all
declinations. Larger structures can be imaged by combining a few
snapshots taken at different hour angles.
| A | B | C | D | |
| 92 cm | 1.5' | 4.5' | 15' | 35' |
| 20 cm | 20'' | 1' | 3.5' | 7.5' |
| 6 cm | 5'' | 18'' | 1' | 2.2' |
| 3.6 cm | 3.5'' | 10'' | 30'' | 1.2' |
| 2 cm | 2.0'' | 6'' | 20'' | 45'' |
| 1.3 cm | 1.0'' | 3.5'' | 13'' | 30'' |
| 7 mm | 0.7'' | 2.2'' | 6.5'' | 22'' |
Figure: The u-v plane coverage at 4.9 GHz for instantaneous sampling of a
source at 
and hour-angle 0 
with the VLA
in the A configuration.
Polarization calibration may be difficult for short snapshot
programs; it is not easy to verify the instrumental polarization
calibration for a program whose total observing time is only a few
hours, as this calibration requires at least three observations of a
calibrator spanning a change in parallactic angle 
of
(see Lecture 7).
Snapshooters interested in polarimetry should ensure that suitable
polarization calibration is possible when designing their program, by
giving attention to its LST range.
Snapshots are most effective when all sources are observed within about 2 hours of the meridian. At larger hour-angles, foreshortening of the array leads to poorer sampling of the u-v plane, elliptical synthesized beams, etc.
The time taken to calibrate a snapshot data set is determined mainly by the total observing time. Snapshot programs require the same calibration effort as simple synthesis programs of the same total duration. The image construction, deconvolution and display steps of snapshot observing can require large amounts of computer time and your time, however. As a snapshot image of a given target may be as large as a full synthesis image of the same field, snapshot programs also make heavy demands on disk storage. This can be especially true for snapshots made in the more compact configurations below 2 GHz, which are particularly prone to degradation by sidelobe clutter from confusing sources.