Observations at 151 MHz are normally only possible during periods of minimum solar activity. At other times, due to variations in the ionosphere, the rate of change of phase on the longest baselines is often greater than a radian within the minimum integration period of 4 seconds, with a consequent irrecoverable smearing of the fringes. During the 151 MHz observations in April-June 1985 (near a sunspot minimum) it was found that throughout the observing session the phase did not change on even the longest baseline by more than a turn in ten to fifteen minutes. Not infrequently, it only changed by a turn in a period of a few hours. In stark contrast, in 1981, near sunspot maximum, there were occasional phase changes of more than a radian within an integration period at 408 MHz, at which frequency the effect should be almost a factor of three less than at 151 MHz.
At 151 and 408 MHz, the maps of individual sources could be significantly degraded unless the effects of confusing sources within the primary beam of the telescope are removed. In order to assess the problem and to ensure that the sources are strong enough to be mapped, fringe-frequency - delay plots are produced as described in §3.10.4 to separate out the individual sources within the telescope beam. Alternatively, NRAO provides an on-line catalogue based on NVSS which can be used to generate a list of confusing sources in a format suitable for direct subtraction during AIPS processing.
Although the radio astronomy band at 151 MHz is from 150.0 to
152.0 MHz, the bandwidth of the receivers is limited to 1 MHz centred
at 151 MHz, because of the presence of very strong signals near the
band edges such as those due to the national paging network at 153 MHz
(
48 dbm). At 408 MHz, the bandwidth of the receivers is restricted to
the defined radio astronomy band of 4 MHz. At L-band the front-end
filters define two separate sub-bands, 1370-1430 MHz and 1550-1730 MHz.
This is necessary to avoid strong interfering signals (including local
radar installations) at intermediate frequencies. The observing frequency
can be switched automatically between the sub-bands in a period of about
30 sec (note that frequency switching cycles should be substantially
longer than this since all data is lost during the switch-over period).
The primary adverse effect on observations at these low frequencies is the presence of human-generated narrow band signals (mostly data transmission) within the radio astronomy band, producing very high spurious visibility amplitudes, usually worst on short baselines. This is particularly bad at 151 MHz, where it is common to have to reject approximately 15% of the data. Spectral-line data, especially at wavelengths Doppler-shifted towards the edge or out of the protected radio astronomy bands, are also very vulnerable. It is sometimes possible to mitigate the effects of interference by small shifts in observing frequency, so it is important that the total velocity range over which emission is likely to be detected is specified as accurately as possible in spectral line observing proposals.
The standard technique of phase-calibration involves nodding the telescopes between target and phase-cal every few minutes. Prior to 2002 LO (and WA when in use) were not normally able to slew fast enough for this technique. At 1.4/1.7 GHz these telescopes only nodded to the phase-cal every 30 minutes or so. By choosing a reference telescope close to one or both of these telescopes (usually DA) the phase rates for these slow-moving telescopes are usually sufficiently small that this sampling is perfectly adequate. At 408 and 151 MHz there is often a phase-cal within the primary beam and this problem does not arise.