Question I created a model where all current carying conductors are at least 5 meters above the ground. I am interested in computing electro-magnetic fields between ground level and 1.5 meters. I have used an infinite soil model with a resistivity of 1e18 Ohm-m. Is that OK, or should I use the soil resistivity models at this particular substation to do the calculations?
Answer When computing electric fields at points whose elevation is relatively close to the air-soil interface, it is usually better to model the ground explicitly. The reason is that the air-soil interface acts almost like a perfect ground plane, which is at a very low potential and tends to bring the fields down. Another way to see this is to consider that for every above-ground conductor, there is (effectively) a buried "image" conductor that carries an opposite charge and that contributes to the electric fields: the sum of the two contributions is smaller than for a conductor in infinite space.
For magnetic fields, the dependence of the fields on the presence of a soil layer is much less pronounced. Therefore, if you are studying magnetic effects at points located above-ground caused by current sources also located above-ground, then using an infinite medium with a large resistivity (as you did) can yield a very good approximation of the results.
Note that you can also use an infinite medium if you are interested in the value of the fields close to the source, i.e. if the distance between the observation point and the source is much smaller (at least ten times) than the distance between the observation point and the ground plane.