NavList:

Marcel you wrote:

"The self-made program contains also a goographical database with the
coordinates of about half million locations and mountains and from the
mountains also their altitude. In addition I incorporated recently a
atmospheric model for approximate temperature and pressure values for a
given location at a given day and time... "

 

The numerical integration for this is neither especially difficult nor time-consuming: divide the atmosphere up into thin layers of varying density and apply the usual law of refraction at each step. Assuming your goal is to create accurate simulated views of landscapes (is it??), that's probably the best way to proceed. Then you can include as much detail as you want --temperature inversions, pressure gradients, anything-- and you can calculate extinction at the same time. As a check on your results, you can compare against the standard tabulated refraction values.