Ray-theoretical seismic traveltime inversion: Modifications for a two-dimensional radially parametrized Earth

Abstract

Lithospheric-scale controlled-source seismic experiments are most often evaluated by modelling schemes that assume a flat Earth. The practice of modelling the Earth as flat to analyse such data continues because of the simplicity of the parametrization and the relative insignificance of the systematic error introduced by the approximation in many seismic surveys. Significant systematic errors introduced in larger-scale experiments can be compensated for by the application of the Earth-flattening transform; however, this compensation is not accurate when significant lateral velocity contrasts are present. To remedy this, a widely used ray-theoretical traveltime inversion routine has been modified such that rays are traced through a cylindrical coordinate system rather than a Cartesian coordinate system. Synthetic examples are presented to emphasize: (1) the nature of the systematic error present in a model if the Earth's curvature is ignored and (2) the systematic errors that remain if only the 1-D Earth-flattening transform is used to compensate for the Earth's curvature. The enhanced routine presented here improves 2-D interpretations without the need to apply approximations from the Earth-flattening transformation. Seismic refraction surveys greater than ~350 km in length should use this new method.