Thermal winds in stellar mass black hole and neutron star binary systems

Abstract

Black hole binaries show equatorial disc winds at high luminosities, which apparently disappear during the spectral transition to the low/hard state. This is also where the radio jet appears, motivating speculation that both wind and jet are driven by different configurations of the same magnetic field. However, these systems must also have thermal winds, as the outer disc is clearly irradiated. We develop a predictive model of the absorption features from thermal winds, based on pioneering work of Begelman, McKee & Shields. We couple this to a realistic model of the irradiating spectrum as a function of luminosity to predict the entire wind evolution during outbursts. We show that the column density of the thermal wind scales roughly with luminosity, and does not shut off at the spectral transition, though its visibility will be affected by the abrupt change in ionizing spectrum. We re-analyse the data from H1743−322, which most constrains the difference in wind across the spectral transition, and show that these are consistent with the thermal wind models. We include simple corrections for radiation pressure, which allows stronger winds to be launched from smaller radii. These winds become optically thick around Eddington, which may even explain the exceptional wind seen in one observation of GRO J1655−40. These data can instead be fit by magnetic wind models, but similar winds are not seen in this or other systems at similar luminosities. Hence, we conclude that the majority (perhaps all) of current data can be explained by thermal or thermal-radiative winds.