GB1349496A - Determination of response and passband of seismic data gathering system - Google Patents

Abstract

1349496 Seismic signal processing MOBIL OIL CORP 15 March 1972 [23 March 1971] 11994/72 Heading H4D A measure of the efficiency of a given seismic source receiver spread configuration in attenuating multiple reflections is obtained by processing traces recorded using the spread. In the illustrated embodiment (Fig. 6) values of time TMC and velocity VCR of a travel time/ velocity curve to be used for moveout (NMO) corrections and stacking of a common depth point (CDP) set of traces are read in at 44. For a time-window of this curve, limits of uncertainty of the associated velocity value are set to define at 51 a velocity range for this window. At 52, trace portions relating to refraction effects are rejected and the effective traces to be used for stacking are determined. At 46 a wavelet 20 (Fig. 1) in the time-window on the effective traces is stacked in phase, using the correct velocity, to give a reference power/time response 23, and is then successively stacked using velocities with various assigned percentage errors to obtain for each velocity error a respective degraded power/time response such as 25. The ratio of the latter response to the reference response is computed to give a "relative response". By computing values of relative response for one particular velocity error but for other time-windows, contour lines linking points of equal relative response can be drawn on a map having co-ordinates of time and velocity (Fig. 2, not shown). Similar, respective maps can be drawn for other velocity errors. If it is known that, at a given record time, the velocity for multiples differs from the velocity for primary reflections by, say, 6% then by examining the relative response map associated with a 6% velocity error the relative response for that record time can be found. If the relative response is high, e.g. 75%, it is known that multiples will stack with 75% of the power of the primaries and hence re-designing the spread configuration may be necessary to attenuate the multiples satisfactorily. The relative response for one particular time-window may also be plotted as a function of velocity error to establish the velocity passband of the spread configuration, i.e. the velocity range over which the relative response is above 70%. Successive values of half the bandwidth, DV, are computed for respective successive time-windows. These values are used at 55, as a guide to velocity increments used in a velocity search to determine the correct velocity needed for obtaining the reference response 23 and for deriving at 56 NMO corrections to be used subsequently. The above procedure is repeated for other CDP sets of traces.