6 Sources in the NVSS

The NRAO VLA Sky Survey (NVSS, Condon et al. 1998) is a survey of the sky above declination -40$^\circ$ at 1400 MHz. Its final resolution is 45 $^{\prime \prime }$ and is comparible with the WENSS resolution of 54 $^{\prime \prime }$ (in right ascension). The bandwidth was effectively 42 MHz. The noise in the final NVSS maps is about 0.45 mJy. These maps are constructed by taking the average of a number of snapshot observations. Each point on the sky is observed in about three snapshots.

Seventeen of the twenty-five pulsars with a WENSS counterpart are also detected in the NVSS (see Kaplan et al. 1998, who did not include PSR B0138+59, and Han & Tian 1999). Two pulsars (PSR B0353+52 and PSR J1518+4904) with a marginal WENSS counterpart coincide with a NVSS source. Kaplan et al. (1998) also lists the WENSS sources which coincide with their NVSS sources. Besides the three pulsars already mentioned, PSR B0809+74 is also not in their list, though it matches their coincidence criterion. Eight pulsars with a WENSS counterpart are not detected in the NVSS. This can be caused by scintillation or these pulsars must have a steep spectrum (spectral index between about -1.6 and -2.0). Two pulsars (PSRs J0218+4232 and B2319+60) may have a very steep spectrum (spectral index less than -2.6 and -2.3, respectively).

From the WENSS and NVSS flux density values the spectral index can be calculated. These can be compared with the spectral indices as determined from a large number of dedicated pulsar observations, as by LYLG. The latter indices will suffer less from scintillation effects. The uncertainty due to scintillation is included in the errors given by LYLG.

From Table 5 it is clear that there are some large differences between the spectral index as determined by the WENSS and NVSS flux densities and the long term averaged spectral index. The differences are displayed in Fig. 8. This histogram has a mean of -0.14 and a standard deviation of 0.48. There is a large spread and a WENSS-NVSS source flux density comparison is not a very precise way to determine a spectral index. This will limit the success of a pulsar candidate selection based on their WENSS-NVSS spectral index.

Figure 8: Histogram of the differences of the spectral index determined from the WENSS and NVSS source flux densities and of the pulsar spectral index, as determined by a large number of dedicated pulsar observations (Lorimer et al. 1995). The dashed line is a Gaussian distribution with the same mean and standard deviation as the histogram