Variable Depth Streamer: Benefits for Rock Property Inversion

Abstract The lack of low frequencies in conventional seismic data means that a low frequency model must be incorporated in seismic inversion process in order to recover absolute elastic attribute values. Typically, low frequency models are obtained from low-pass filtered impedance logs. If well-logs are sparse and the geology complex, the derived low frequency model may be inaccurate and cause biased inversion results. One option to improve the low frequency model is to use seismic velocities. Seismic velocities provide information at very low frequencies, they are not providing information for all missing frequencies. Seismic data acquired using variable depth streamers (VDS) are suited for inversion as they provide directly the missing low frequencies, hence removing the need to build low frequency models from well data. In order to quantify the impact of the low frequency content on seismic inversion, comparative elastic inversion tests were conducted using 2-D seismic data from Constant Depth Streamer (CDS) and VDS acquisitions. Both datasets from offshore NW Australia were acquired simultaneously. Both data sets were processed using PSTM sequences and inverted. Two gas-bearing sand reservoirs are identified. Delineation of reservoirs is improved with VDS seismic inversion which shows lower impedance and Poisson ratio values in the gas-bearing interval. The inversion results were subsequently interpreted in terms of facies using a supervised Bayesian classification procedure. Three litho-classes were considered: gas-sand, water-sand and shale. Introduction Quantitative interpretation teams need to solve two problems when using model-based inversion: extracting meaningful wavelets and building accurate low frequency models. Conventional seismic data are band limited and a low frequency model must be incorporated in the inversion process in order to recover absolute impedance values. Typically, low frequency models are obtained from low-pass filtered impedance logs. If well-logs are sparse and the geology complex, the well-derived low frequency model will be inaccurate and cause biased inversion results. One commonly used method to improve the low frequency model is to use seismic velocities. However, while seismic velocities provide information at very low frequencies [0–4 Hz], they are not usually providing information for the missing frequencies in the range from 4 to 10 Hz. Seismic data acquired using variable depth streamers are ideally suited for inversion as they provide directly the missing low frequencies, hence removing the need to build low frequency models from well data. In order to quantify the impact of the low frequency content on seismic inversion, comparative elastic inversion tests have been conducted using 2-D seismic data from Constant Depth Streamer (CDS) and Variable Depth Streamer (VDS) acquisitions.