出版社：SISSA, Scuola Internazionale Superiore di Studi Avanzati
摘要：LS I +61303 is an HMXB discovered as a g -ray emitter 30 years ago, and well known as a radio,
X ray, and most recently a TeV source. Despite the wealth of data, the choice between pulsar and
microquasar emission models remained unclear. Here we report on AU-scale radio imaging with
the VLBA, that literally resolves the issue in favour of the pulsar mechanism.
Our sequence of 10 images were spaced 3 days apart and covered more than one full orbital
period of 26.5 days. We observed for 5 hours per day with simultaneous reception at l 13 cm
and 3.6 cm. We show that: (i) The radio emission is resolved on scales of 2AU, and appears
cometary, with the `tail' pointed away from the high-mass star. (ii) The morphology varies
dramatically near periastron, with outow velocities 7500 km/s, and gradually around the rest
of the orbit, v<1000 km/s at apastron. This variation is easily explained by the geometry of the
eccentric orbit, rather than a precessing jet. (iii) Comparing simultaneous 13 cm and 3.6 cm
images, a synchrotron opacity gradient exists along the cometary tail, with the highest-energy
particles located in the head. No bulk relativistic motion is seen during any phase of the orbit.
We conclude that the pulsar model is strongly supported, i.e., the radio synchrotron emission
arises not in a jet, but from particles shock-accelerated in the interaction of the pulsar wind with
the dense equatorial wind from the Be star.
Astrometric accuracy of <0.1 milliarcsec in these observations, combined with re-processed
archival data, constrains the long-term proper motion of the binary over 14 years. We show that it
has a low velocity 10 km/s relative to star-forming regions W3 (IC 1795) and W4 (IC 1805) in
the Perseus arm, and was not ejected at 27 km/s by the supernova event, as previously believed.
We also show that the centroid of the radio emission varies during the orbit, with both systematic
and random components of 1mas, obscuring the parallax (0.5mas), and the semi-major axis of
the binary (0.25mas).
Finally, as a general comment, we note that radio synchrotron emission from X ray binaries
may be manifest on AU scales as pulsar wind nebulae or shells, besides the often-invoked jet
morphology. The latter two are low-power analogs of SNR and GRB's respectively.
