Astron. Astrophys. 333, 583-590 (1998)

5. The number of galactic low mass BHB systems

These results are now used to estimate the number of low mass interacting BHB in the galaxy. The philosophy will be similar to that used in early estimates of the population of millisecond pulsars (Kulkarni and Narayan 1988; Kulkarni, Narayan & Romani 1989); one assumes that the observed sample of X/O novae is representative of the underlying population and estimates the volume (in ) covered by the major surveys that discovered these objects. For each observed source i one integrates over the Galactic plane distribution (3.1) to find the fraction of the Galaxy in which the source could have been detected (and confirmed optically, including the Galactic extinction Sect. 3) in the survey j. The extrapolated number of similar binaries of type i in the Galaxy as a whole is then

which can then be summed to obtain the total Galactic population.

Note that inclusion of low mass BHB in this study requires both X-ray detection and an optical radial velocity study. The estimated number of Galactic systems thus depends on the completeness of this optical confirmation. These radial velocity studies are generally done in the R band, to mitigate against Galactic extinction and to increase sensitivity to the cool late type secondaries. For the standard estimate, it is assumed that the X-ray selected X/O novae with quiescence magnitudes brighter than have been well studied. Some fainter low mass BHB have been observed, but other excellent candidates at brighter magnitudes remain to be dynamically confirmed (e.g. 1009-45=Vel93 at - note that Beekman et al. 1997 list this as a highly likely BH system). The sensitivity to this assumption is noted below. For the standard assumptions ( from Eq (3.1), optical limit ) the Galactic population of systems like each of the known objects are listed in Table 4.

Table 4. BHB numbers

Assuming that none of the other X-ray detected X/O novae is a low mass BHB, and ignoring the constraints imposed by optical confirmation, one gets the estimates . Including the (more severe) requirement that an optical mass function has been measured raises the numbers to . The estimates indicate that roughly X/O novae should have been detected over the past 20y by the various sky surveys and should be awaiting dynamical (optical) confirmation. The actual number of good candidates listed by Tanaka & Lewin (1995) and Tanaka & Shibazaki (1996) from X-ray spectra and light curves is roughly 1/2 this value. This suggests that either the optical confirmation cut is too restrictive or that a good X-ray spectral study and BH candidate ID of an X/O nova requires even larger threshold fluxes (esp. for Ariel 5 and Ginga detections) than assumed in Sect. 4. Thus the accuracy of the completeness limits and temporal coverage in Sect. 4 are major remaining uncertainties in the population estimate; further study of the ASM survey data may reduce this. Interestingly, BATSE detection rates can also provide a check on the population estimates: at present sensitivities BATSE detects candidate BH X/O transients at a rate of /y (Tanaka and Shibazaki 1996). Thus, estimation of the rate of BATSE detections predicated on the larger numbers (column 4) provides a useful check on the population numbers. This estimated BATSE rate is independent of the adopted recurrence times, but since P sources (especially 0422+32-like BHB) dominate the BATSE detections, the computation of this rate is rather sensitive to the estimated luminosities of these binaries. If the BATSE transient peak sensitivity can be reduced to mCrab for the entire plane as discussed by Grindlay et al. (1996), one would expect a discovery rate of /y.

Population estimates of this sort are most useful when the dependence on various assumptions are tested. If, for example, the optical confirmation limit is the total number of systems is then 1220; however the predicted BATSE rate is somewhat low at 1.5 transients per year. On the other hand if the confirmation limit is as bright as , the total galactic population rises to . If the classical nova Galactic distribution with a bulge population (3.2) is adopted for the BH novae the observed systems extrapolate to a galactic population of 2420 low mass BHB. Thus the estimate of the total number of Galactic low mass BHB systems may be uncertain by as much as a factor of .

© European Southern Observatory (ESO) 1998

Online publication: April 20, 1998
helpdesk.link@springer.de