Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 342, L5-L8 (1999)

Previous Section Next Section Title Page Table of Contents

2. Geminga

PSR 0633+17, or Geminga, provides one of the most ideal neutron stars for such an analysis, as HST observations have determined its distance by parallax to [FORMULA] pc (Caraveo et al. 1996). However, it is by no means clear precisely what the nature of its emission processes are, despite considerable observational efforts from the optical to [FORMULA]-rays. Original ROSAT observations suggested dominant thermal emission from the surface modulated by a hotter polar cap (Halpern & Ruderman 1993), although later observations especially those of ASCA indicated the emission was thermal and nonthermal in origin (Halpern & Wang 1997). Disagreement remains on the optimum SED fits to data from ROSAT, EUVE and ASCA with the pulsar's surface temperature within [FORMULA] K. There have been further suggestions that the pulsar's thermal emission is strongly affected by magnetospheric cyclotron resonance blanketing (Wang et al. 1998), which would severely compromise any estimate of the true neutron star surface temperature. Optical observations suggest an unusual SED, with deviations in the expected functional form, possibly as a result of ion cyclotron absorption/emission processes. The most recent of such observations ([FORMULA]), made using the Keck LRIS, spanned 370-800 nm and yielded a flat power law shape ([FORMULA] [FORMULA] [FORMULA]) and a noticeable broad dip at 630-650 nm with perhaps a slight modulation at V, B & I as advanced by Bignami et al. (1996). The composite power-law may be fitted by either a combined blackbody and power-law (i.e. nonthermal emission with [FORMULA] [FORMULA] [FORMULA]) or a blackbody plus global ion cyclotron emission. The former model is undoubtedly the most likely, following the discovery of a highly pulsed lightcurve in the B band (Shearer et al. 1998). Fig. 1. shows the model-fit spectrum of [FORMULA] with the results of Shearer et al. (1998).

[FIGURE] Fig. 1. Keck LRIS Optical spectrum of Geminga, with best-fit two component model of Martin et al. (1998) and TRIFFID/BTA pulsed and unpulsed flux estimates in B (Shearer et al. 1998)

A value of G = 4.4 is obtained by applying a solution in the form of (3) to the best-fit Keck observations of [FORMULA], setting A([FORMULA]) [FORMULA] 0 due to the pulsar's close proximity. Introducing our independently estimated unpulsed upper limit in the B band of [FORMULA], we conclude with G [FORMULA] [FORMULA], allowing errors in G for the spectral response of the MAMA/B-filter combination. The decrease can be inferred as either a reduction in the emission area or a drop in [FORMULA] (d [FORMULA] 160 pc). As a consequence, one expects the nonthermal component to increase proportionately in the original model fit, if this unpulsed estimate is representative of a general surface based thermal emission. If we assume that [FORMULA] remains unchanged, then the obvious conclusion is that R has been overestimated. In fact, using G [FORMULA] 3.6, [FORMULA] [FORMULA] [FORMULA] K (MHS98) and d = [FORMULA] pc suggests [FORMULA] [FORMULA] [FORMULA] km for a [FORMULA] blackbody source (indistinguishable from a Fe/Si atmosphere - see Pavlov et al. 1998). An alternative explanation for the change in G could be understood under the assumption of atmospheric opacity effects for the total surface thermal emission at X-ray energies. As [FORMULA] point out, there is no ideal way to reconcile the actual brightness temperature with the estimated X-ray [FORMULA]. Fits with magnetized ([FORMULA] [FORMULA] G) H atmospheres suggest that a spectrum in the optical regime can be fitted as a Rayleigh-Jeans tail with [FORMULA] ([FORMULA]). Applying this correction yields [FORMULA] [FORMULA] [FORMULA] km.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1999

Online publication: December 22, 1998