Since the gradient in the temperature field is known before the calculation of the fringe temperature the factor must be calculated from a guessed temperature field. Thus, the final calculation of relies on a series of iterative steps with improved of the temperature gradients.

The refraction errors in the present set of experiments can be shown to be quite small. As a conservative estimate, the temperature gradient term can be replaced by its value at the wall. This would give an upper bound on the refraction error. For the following numerical values: for air at and and refractive index of air at the contribution of the refractive index term in Equation (14) can be evaluated as This value is small as compared to the wavelength

The number of fringes expected in a projection can be estimated directly from the relation

This equality was seen to be satisfied in all the experiments reported in the work. This also indicates that refraction errors in the present experiments are negligible.