EPC 2002 Tunis May 7-11, 2002 version 2.2

Accurate three-dimensional positioning from 2D vibrator DGPS. Simon ROBINSON [1], Jeff AUTY [2], Eric GILLOT [3], Alain DEPEYRAS [3]   Vibrator mounted Differential Global Positioning Systems (DGPS) have long been used in seismic surveying to verify surveyed source positions. The DGPS method described here can provide accurate final source co-ordinates and elevations for 2D vibrator surveys. The objective is to obtain an accurate three-dimensional position of the centre of gravity (COG) of the source array as it is vibrated ("as shot") with emphasis on obtaining accurate relative elevations. The elevation is normally the most difficult value to determine with GPS because it requires an extended occupation period relative to the co-ordinates. Typically, this extended occupation period is not available in high production 2D vibrator recording where the vibrators may be stationary for less than 20 seconds. The theoretical VP position was obtained by interpolation between the RTK Receiver Groups and applying a perpendicular This method of accurate source positioning starts with conventional layout and surveying of the COG of all receiver arrays using GPS in real time kinematic mode (RTK). As is common, bulldozers then clear and cut a track parallel to the surveyed receiver line. A vehicle, named a Fûret (or ferret), then drives along the track recording GPS in Kinematic mode. The theoretical VP position was obtained by interpolation between the RTK Receiver Groups and applying a perpendicular cross-line offset to the "Fûret" surveyed track. This theoretical VP position was then used to build the preplan SPS file given to the recorder. During acquisition, in the recorder, the vibrators DGPS "as-shot" COG position was checked according to a radius threshold of 5 m from the predicted VP position. For every VP not in this circle, the Observer checked the reason for the discrepancy with the Field Manager and a comment was inserted in the observer log. At the QC office, the VE432-APS file was checked in order to extract a list of VP's not matching the theoretical number of pad positions. Unreliable DGPS data was also edited out. This produced a provisional cleaned-APS file containing the x and y co-ordinates of every valid vibrator sweep position. Using these pad positions, a z (elevation) value was then extracted from the grid generated within FastQC® software. All final pad sweep positions were then averaged to create the final DGPS "as-shot" COG position. This new set of 3 dimensional co-ordinates was then used to build the SPS file for the Geovecteur® processing software. The following information for every source point was provided: Intra-array elevation difference. Number of vibrator positions averaged to calculate the centre of gravity. Length of the source array. The pad-to-pad separation between vibrators in each sweep position. The move-up between sweeps positions. [1].        Anadarko [2].        ECL [3].        CGG