Poster abstract details

Since the first discovery of narrow high excitation emission lines (mainly in the Bowen region) arising from the donor star in Sco X-1, narrow components have been detected in other observed sources in luminous states, leading to some of the first estimates of $K_2$ and mass functions of several LMXBs. In most cases, it is essential to adopt the Doppler tomography technique to invert all (phase-resolved) spectra into an image of emission distribution in velocity coordinates. The emission component from the companion can be mapped onto a compact spot at an expected position in the reconstructed image, from which we can derive a strict lower limit to the radial velocity semi-amplitude $K_2$.

Caveats are needed when applying the tomography-based method to datasets in the lowest signal-to-noise ratio regime, where the authenticity of a possible secondary detection is of most concern. Thanks to the vast increases in computational power, this problem can now be tackled by introducing a Monte-Carlo style bootstrapping test. I will discuss possible procedures to quantify the significance of features seen in a Doppler tomogram, and how this Monte-Carlo approach can lead to a reasonable estimate of the uncertainty in the measured velocity, which should then provide improved determination of the mass limits of the compact accretor.