3.3. Pulsars

The power spectra of most of the -ray pulsars (Thompson 1997) are extremely flat with maximum power often coming in the GeV energy range (see Fig. 3). Because pulsar models often involve electrons with energies up to 10¹⁵ eV, it would come as no surprise if TeV -rays should emerge from the pulsar magnetosphere and be detected. Although there is, as yet, no established model for high-energy -ray emission from pulsars, it appears that, in general, the detection of pulsed TeV -rays would favor outer gap (Romani 1996) over polar cap (Daugherty & Harding 1982) models. This is because, in the latter models, the TeV -rays are attenuated by pair-production interactions with the intense magnetic fields near the pulsars.

Figure 3. Power spectrum of -ray pulsars detected with EGRET (Thompson 1997).

Sensitive upper limits have been obtained for emission by the Crab pulsar (Lessard et al. 1999), Geminga (Akerlof et al. 1993), and the Vela pulsar (Yoshikoshi et al. 1997); in general, these confirm the steepening of the spectra seen at 10 GeV energies. The radio pulsar PSR 1951+32 is particularly interesting because the power spectrum indicates that the maximum power occurs at energies of at least a few GeV (Fig. 3); in the EGRET measurements there is no evidence for a high-energy cutoff. In fact, the flux continues to rise with energy up to the highest energy observation. Outer gap models suggest that the pulsar should be detectable at higher energies. Observations of PSR 1951+32 by the Whipple group reported only an upper limit (Srinivasan et al. 1997). This upper limit to the pulsed flux is 2 orders of magnitude below the flux extrapolated from the EGRET measurements. This represents the most dramatic turnover in the spectrum of a -ray pulsar and hence puts the most severe constraints on the models.