In Beaumont and Berger (1975), strain observations from seven sites in the continental United States were analyzed and compared with the homogeneous tides predicted for a radially stratified earth including the effects of ocean-tide loading. In this work the effects of lateral inhomogeneities in structure are examined using two- and three-dimensional finite element models. The results demonstrate that the topographic and geologic corrections are as large as ±25 per cent of the homogeneous strain tide for some of the strainmenter sites examined. Cavity effects are normally less than 5 per cent unless the strainmeters are installed transversely in tunnels. Contour plots of the topographic effects show that they may exceed ±50 per cent of the homogeneous strain in extreme cases.
It is concluded from an error analysis that a theoretical model including local perturbations due to lateral inhomogeneities as well as the load and earth tides explains the observations. However, the errors in the observations, and uncertainties in our knowledge of the deep ocean tides in particular, prevent strain observations from providing useful measures of whole earth elastic properties (Love numbers) or anelasticity.
In addition to earth tides, the results demonstrate that secular strain, in situ stress, and seismic strain must all be corrected for site effects.