Refraction is a very important aspect of geodesy and astronomy: it is one of the main sources for troubles, papers and dissertations, at least in geodesy. Particular important is the vertical refraction because of the up- and down-asymmetry in the atmosphere. However, we must not forget that gravity does not only cause the asymmetry of the atmosphere, also the asymmetry of the mechanical parts of the instrument (bending of the tube, deformation and excentricity of the axis, etc.) and even an asymmetry of the mind of the observer. See figure 1. If it becomes possible to measure the refraction these effects will often become limiting factors for the accuracy.

To geodesists and astronomers atmospheric refraction is a phenomenon, which makes it necessary to introduce corrections to the observations of carefully performed measuring series. Such corrections are generally obtained from tables giving values, which are applicable under “normal” or standard conditions. These conditions may be taken as more or less obtaining with a mean variation of temperature with height in the atmosphere - in the troposphere mainly a decrease of 0,6 – 0,7° per 100 m.

The precursor theory of an electromagnetic wavetrain propagating through a dispersive medium is briefly reviewed. The few successful experimental determinations of these precursors, which have been reported in the literature, are limited to transients in waveguides. The determination of precursors in the optical frequency range appears to be achievable in the laboratory. However, in the geodetic context of long wavepaths through the troposphere the utilisation of precursors for measuring distances and directions free of atmospheric effects is judged as not feasible at the present time.

The measurement of three-dimensional movements by geodetic techniques involves the determination of spatial co-ordinates by the observation of horizontal theodolite angles and EDM distances, with either vertical angles or spirit levelling providing the height control. For correct interpretation of the results it is essential that estimates of the accuracy of the position determinations are also provided.