Exploration Geophysics - Volume 20, Issue 4, 1989

An expanding reflection spread with a maximum shot-receiver offset of 25 km was recorded in the northern part of the Amadeus Basin, Central Australia, to obtain seismic velocity estimates throughout the crust. Arrival times of reflections indicate significant lateral variations in P-wave velocities superimposed on the small northerly dip of the sedimentary sequence. Nevertheless, a useful velocity profile that contains two pronounced low velocity zones has been derived for the basin sedimentary rocks to depths of 8.6 km. Below depths of 3 km, these interval velocities are a considerable improvement on those obtainable by standard velocity analysis techniques on the coincident near-vertical incidence reflection data; the results thus enable the determination of more reliable estimates of the thicknesses of the deeper sedimentary formations. The refracted arrivals from the expanding spread yield well-constrained P-wave velocities for shallow depths, but below about 1.5 km depth, the refraction interpretation is problematic owing to the absence of a refracted arrival through the high velocity lid in the interval velocities at depths between 1.8 and 2.7 km. The determination of useful velocity estimates for basement rocks is hampered by the presence of peg-leg multiples produced within the overlying sedimentary formations.

The technique presented in this article is an alternative approach to the established depth-conversion processes and is suitable for use in areas which lack well control, such as the dipping flanks of structures, where the reflectors are assumed to be of any arbitrary dip angles, but the change in dip is not very severe arid abrupt.

Consideration is given to the difficulties in predicting, away from well control, interval velocities, conversion factors compensating for the heterogeneity of the ground and for the conversion of normal incidence (Nl) raypath rms velocities (Vnir) to ‘true Vrms’, and a method is presented to check on their validity.

The checking procedure relies heavily on the preparation of a smooth, geologically representative, nmo velocity (Vnmo) map of each horizon by applying a suitable method of computer gridding and smoothing to the stacking velocities. This is then followed by the preparation of depth maps from which information on interval velocity, thickness and dip can be extracted to produce depth-interval velocity models. The application of forward raypath tracing to these models will generate mathematically computed Vnmo and To values.

The deviation between the model and original Vnmo and To attributes is related to inaccuracies in the estimation of the conversion factors which can be adjusted, and new models generated. The process is repeated until convergence is reached.

The paper discusses the nature of the two most important parameters related to the non-homogeneity of the ground, the Bias and the Heterogeneity Factor, Al-Chalabi (1973, 1974, 1979), which are utilised in the estimation of the conversion factors away from well control.

The results of applying the present technique to the mapping of the Snapper gasfield (Megallaa and Ashworth, 1987) are discussed. Fortran code for the forward and inverse raypath tracing and for converting Vnir velocities to average velocities for isovelocity plane-dipping models is provided.

A direct method for interpreting resistivity soundings acquired with any common electrode array has been developed. It is based on field resistivity transforms which are determined by the convolution of a linear filter with field apparent resistivity data. For a horizontally stratified earth model, the top layer is resolved in terms of its thickness and resistivity by means of two-layer curve matching and is then stripped off. Successive two-layer curve matching is applied to reduced field transforms until the whole section is uncovered. The quality of the interpretation is checked by computing model transforms by recursion for comparison with field transform data. Although designed for manual use the procedure can be semi-automated. Because the mechanics of the method highlight some important concepts in electrical geophysics the method has an educational role but it also has a place in the field situation whenever computers are unavailable.

We formulate the problem of a transient magnetic dipole located over a thin conductive sheet which may exhibit frequency dispersion or IP. (induced polarization). A first order analysis is carried through for the case when the source current is a step function. It is shown that the I.P. slow tail can be of opposite polarity to the principal electromagnetic response.

We formulate the problem of a transient magnetic dipole located over a thin conductive sheet which may exhibit frequency dispersion or IP. (induced polarization). A first order analysis is carried through for the case when the source current is a step function. It is shown that the I.P. slow tail can be of opposite polarity to the principal electromagnetic response.