A Comparison of Travel-Time Inversions

A comparison is made between travel time inversion schemes based on the use of the delay time τ(p) = T(p) — pX(p) for a ray parameter p. Bounds on the delay time are used to generate extremal bounds on the velocity depth distribution using the method of Bessonova et al. The delay time values are also used in a linearized inversion scheme, and both techniques are applied to the results of a long range refraction profile in France. The combination of a number of linearized inversions with different starting models is shown to provide a useful control on the velocity structure outside low velocity zones. This approach is particularly useful in the zone of broadened extremal bounds below a low velocity zone.