Are changes in time‐lapse seismic data due to fluid substitution or rock dissolution? A CO2 sequestration feasibility study at the Pohokura Field, New Zealand. (28th June 2016)

• Record Type: Journal Article
• Authors: Sim, Cheng Yii; Adam, Ludmila
• Is Part Of: Geophysical prospecting. Volume 64:Number 4(2016)
• Page Start: 967
• Page End: 986
• Publication Date: 2016-06-28
• Subjects: Rock physics -- Petrophysics -- Time lapse -- Elastics; Prospecting -- Geophysical methods -- Periodicals; 622.15
• DOI: 10.1111/1365-2478.12405
• Languages: English
• ISSNs: 0016-8025

Abstract

ABSTRACT: Seismic methods are commonly used to monitor the subsurface when carbon dioxide (CO2 ) is injected into a reservoir. Besides fluid saturation and pressure changes, CO2 –water mixtures may cause rock alteration. In this petrophysical study, we compare the elastic property changes due to fluid replacement and those due to mineral dissolution for carbonate‐cemented sandstones at the Pohokura Field, New Zealand. We quantify the effects of fluid substitution from fully brine to fully supercritical CO2 saturation and carbonate cement dissolution on the seismic signatures of the reservoir rocks by combining laboratory results, petrographic analyses, and geophysical well log data. We conclude that elastic property changes due to mineral dissolution are significantly greater than those due to fluid substitution alone. The northern part of the Pohokura Field has coarser‐grained sandstones, which experience the largest changes in wave speeds. Our hypothesis is that these changes result from carbonate cement dissolution in the presence of CO2 –water–sandstone reactions. If time‐lapse seismic data were to be acquired in this field, the northern area could show P‐wave velocity reductions of up to 20% and a 131.7% increase in seismic amplitude from a brine‐saturated rock to an altered, fully CO2 ‐saturated rock. In comparison, the southern part of the field, where sandstones are mostly fine‐grained, we expect a P‐wave velocity decrease of 6% if such dissolution process took place. Finally, we show that the elastic properties of the reservoir rocks can be described with the constant‐cement model. The model is used to predict that the dissolution process reduces the volume of grain contact cement, on average, from 2.5% to 1.75% of the total rock mineral volume. Our analysis suggests that changes to the rock frame, which includes carbonate minerals, cannot be ignored for a CO2 injection scenario.