Time-lapse seismic analysis of focused fluid flow on the Vestnesa Ridge

Abstract

The Vestnesa Ridge is a large sediment drift at water depths of 1200-1300 meter and is the northernmost known gas hydrate province that exists along the Arctic continental margin. Several pockmarks connected to vertical fluid flow features are present at the crest of the Vestnesa Ridge. The fluid flow pierce through the gas hydrate stability zone and interrupt the bottom simulating reflection (BSR) originating from the transition zone between stable gas hydrates and free gas. Gas seeps into the water column from several pockmarks demonstrating that the fluid flow system is very active. The abundance of fluid-flow structures and the activity of this system make this an excellent area to study the genesis and mechanisms of focused fluid flow and their related geological processes. A unique approach to gain new knowledge about fluid flow systems in the subsurface is by use of 4D time-lapse seismic data that may help to better understand how these systems develop over relatively short periods of time. The feasibility of high-resolution 3D seismic data with a broad frequency bandwidth of up to 350 Hz for time-lapse studies has not yet been established. High resolution P-Cable 3D seismic data has been acquired over the eastern segment of the ridge in 2012 and repeated in 2013 and 2015. These three seismic surveys have been 4D processed side-by-side in order to highlight subsurface fluid-induced changes. In this thesis, a 4D processing workflow is developed in order to match the seismic data from the three surveys. The 4D processing steps included re-binning of geometry, time and phase matching, shaping filter, shallow statics correction and time-variant shifts. Several 4D attributes are used to quantify the repeatability of the 4D seismic data, the two main attributes being normalized root mean square (NRMS) and predictability (PRED). The NRMS value, for both of the repeats (2015-2012 and 2013-2012), improved significantly during the full processing workflow, while the PRED only changed minimal. PRED is more sensitive to noise and distortion than to time-shift, so the reason for the minimal change here is believed to be due to residual noise in the 4D data. The poorest NRMS values of up to 1.5 are observed over the areas of the gas chimneys, while outside of these structures the NRMS measures lower than 0.4 showing very good repeatability. The overall NRMS values obtained after 4D processing were not as low as anticipated. Several geological and non-geological factors contribute to this. The Vestnesa Ridge with its fluid flow structures presents a challenging and complex setting for the 4D processing in regards to achieving a reliable and confident 4D match and interpretation. The high-resolution 3D seismic data allows the mapping of small structures in the sub surfaces, but it is also very sensitive to noise and other non-geological factors like weather or acquisition effects. Nonetheless, the 4D seismic interpretation shows a clear brightening of amplitudes in the 4D data from 2012 to 2013 and from 2013 to 2015. The most evident brightening occurs in the zones beneath the BSR and may be interpreted as a consequence of increased gas accumulation beneath the BSR. Also chimney conduits show 4D changes with the larger changes associated with actively seeping chimneys. Gas migration pathways can be clearly imaged on the data and show a complex path both vertically and laterally and likely self-enhanced by hydraulic fracturing. However, the 4D interpretation is less confident because chimneys represent very inhomogeneous structures where seismic energy rapidly attenuates or is scattered away, which manifests itself in poorer repeatability measures. The time-span of 3 years, from 2012 to 2015, might prove to be too short to detect any major structural changes in the architecture of the chimney conduits. And it might also be too short to recognize and document any substantial fluid changes along the chimney conduits. Additional repeat surveys might help to further shed lights on the mechanisms of gas migration along focused fluid flow structures.