Hydrocarbon indicators on seismic data: Insights from poroviscoelastic modeling, amplitude, and frequency variation with offsets from the Drake Point gas field, Western Arctic Islands, Canada

Abstract

The evaluation of drilling prospects is frequently based on seismic amplitude anomalies. To decipher “true” seismic prospects from “false” ones, we used poroviscoelastic (PVE) models, as opposed to other formalisms such as acoustic, elastic, viscoelastic, and poroelastic models, that provided a solution that takes into account solid and fluid attenuation mechanisms separately to model the earth’s response to the propagation of a seismic wavefield. Here, a PVE impedance modeling scheme was tested using seismic and well-log data collected on a conventional gas reservoir in the Canadian Arctic. Comparisons between seismic-to-well ties achieved using acoustic and PVE media indicated that the latter provided more realistic synthetic seismograms. Although prestack analysis revealed that the present lithological context was of class I amplitude variation with offset (AVO), the seismic signature observed was of class III AVO. Consequently, the increase in amplitude with offset was interpreted to be induced not by a lithological change (i.e., shale to sand) combined with a gas-charged interval, but rather by an increase in porosity within the sandstone reservoir itself where the gas has accumulated. Frequency variation with offset analysis using spectral decomposition, image low-frequency shadows on the far offsets attributed to the gas accumulation that were correlative with the AVO anomaly. This highlighted the importance of far offsets in anomalous amplitude and frequency events attributed to the occurrence of gas reservoirs observed on stacked data and that these events can be missed if seismic hydrocarbon indicators were solely investigated on stacked data. Finally, the method of analysis emphasized the importance of combining indirect arguments coming from the observation of prestack and stacked seismic data in the time and frequency domains for reducing risk to an acceptable level before a prospect can be drilled.