Estimating refractivity from land clutter: another look at a simple approach

In a previous report, two methods of estimating refractivity from land clutter were discussed where one of the methods used a parabolic equation (PE) algorithm combined with a least squares technique, and the other used a ray trace algorithm combined with a rank correlation scheme (Barrios 2000). Based on simulations alone, both of these methods were fairly successful in estimating a tri-linear representation of a radiosonde-measured refractivity profile over two mixed land-sea paths. However, each method appears to favor a certain type of land topography. The PE/least-squares method performed well over land paths characterized by steep peaks and valleys and the ray trace/correlation method performed well over land paths which were not sharply varying in elevation. This latter method offers a more attractive alternative to the conventional least-squares technique because of its execution speed and simplicity, therefore we take a more extensive look at this technique