The calculation of the incident heat flux distribution external to a jet fire is a challenging task due to the ray effect. In this paper a new adaptive stochastic quadrature method called hot sampling is described. The new methodology is applied to the incident heat flux distribution surrounding a lifted natural gas jet fire. Hot sampling is demonstrated to be significantly more efficient than other reverse Monte Carlo methods. The new quadrature scheme is also shown to be more efficient than the discrete transfer method, a radiation methodology in common use for jet fire simulation. Hot sampling is particularly appealing for problems where the ray effect is significant as the bigger the ray effect the bigger the benefit of using an adaptive stochastic quadrature scheme. (C) 2011 Elsevier Ltd. All rights reserved.