|IMAGING SUB-BASALT PALAEOCENE / MESOZOIC AND PALAEOZOIC TARGETS IN THE DEEP WATER UKCS ATLANTIC MARGIN: COMBINING EFFECTIVE REPROCESSING STRATEGIES WITH NEW DEEP TOWED ACQUISITION|
In 2009 TGS acquired 2,947 km of 2D long-offset multi-client data in the Atlantic Margin across the far northeastern part of the Faroe-Shetland Basin (FSB) crossing into the southwestern
Figure 1: Location map showing the NSR 2009 lines acquired with a streamer towed at 18 metres depth and the reprocessed FSB 1999 and 2000 surveys overlain on regional structural elements from Mudge et al., (2007). Oil and gas discoveries and concessional blocks are also shown. The region shown in pink approximates the extent of widespread Palaeogene volcanism. The area highlighted in orange refers to the more detailed structural elements map of the
Alongside the acquisition of the deep towed NSR dataset, 9,779 km of 2D long offset data were reprocessed from the FSB 1999 and 2000 surveys which overlap to the southwest and were acquired with a more conventional 9 metre depth streamer. The FSB surveys sample a series of southwest-northeast orientated intra-basinal highs in
The FSB reprocessing campaign, completed in January 2010, incorporated a number of processing approaches which demonstrated a dramatic improvement over the original time image. Applicable components of the reprocessing sequence were then applied to the new deep towed acquisition. The overlap of data allows the comparison of the original FSB dataset with the reprocessed data and the new deep towed NSR acquisition through the intersection of 2D lines showing a stepwise improvement in the sub-basalt image.
Figure 2: Tectonic elements map of the
|CHALLENGES IN ACQUISITION AND PROCESSING|
Imaging beneath thick basalt flows remains a challenge along the northwest European Atlantic Margin although many of the issues are now well understood. The strongly reflective top of the basalt and rugose nature of the flows, scatter much of the incident P-wave energy whilst interbed multiples generated within the basalt layers and surface multiples mask weaker sub-basalt reflections with similar moveout. The high velocity basalt layer absorbs and scatters the higher frequencies present in the source wavelet, not only limiting the effective resolution of the sub-basalt image, but large velocity discontinuities at top and base basalt interfaces result in significant ray-path distortion and multi-pathing. The key to improved imaging is therefore to generate and retain as much low frequency energy as possible in processing (e.g. Ziolkowski et al. 2001). Subsequently, more recent acquisition has seen the towing of cables and sources at increasingly greater depths using very large sources concentrating more of the available energy into the low frequency end of the amplitude spectrum through constructive interference of the free surface ghost.
Figure 3: (a) Data derived wavelet from the NSR 2009 18m streamer and (b) after resample to 4ms and 42Hz 36dB/octave high-cut filter; (c) data derived wavelet from the FSB 1999 survey and (d) and spectral manipulation.
Figure 3(a) shows the data derived source wavelet for the 18 metre towed cable with a 7 metre source and figure 3(b) after the application of a 42 Hz high cut filter and 36 dB/octave taper. This compares with the same data derived wavelet for the FSB 1999 survey (figure 3(c)) acquired with a cabled towed at 9 metre depth and 7 metre source depth. The equivalent amplitude spectra are shown in figure 4. The NSR deep tow dataset demonstrates a broad receiver ghost at ~ 42Hz but a much richer response in the low end of the amplitude spectra when compared to the shallow towed configuration with a notch at ~ 82 Hz.
Figure 4: Normalised amplitude spectra of the data derived wavelets shown in figure 3. Green: raw NSR deep tow (fig. 3(a)); Brown: after filter (fig. 3(b)).
|SUB-BASALT DATA PROCESSING STRATEGIES|
The FSB reprocessing utilised three key strategies which in combination produced a significant improvement in the sub-basalt image (Hardwick et al., 2010). These were (1) the enhancement of recorded low frequencies through spectral manipulation, (2) noise attenuation in multiple domains to maximise the signal-to-noise ratio and (3) ‘full-sequence migration multi-velocity analysis’ adopting a strategy analogous to that used in pre-stack depth migration for updating velocity models in areas of complex structure. In the processing of the NSR deep towed data we adopt the same strategies with the omission of the low frequency boosting operator. Figure 3(d) shows the FSB 1999 data derived wavelet after application of this operator and the amplitude spectra in figure 4, showing a similar response to the NSR deep tow at the low end.
Arbitrary intersecting lines allow the comparison of the NSR deep towed dataset with the original and reprocessed FSB dataset time image. Figure 5(a) shows such an intersection with the original time processing and figure 5(b) with the reprocessed image. Clearly the reprocessed FSB data shows a significant uplift in the sub-basalt image, but the deep towed data to a greater extent in the underlying Mesozoic and Palaeozoic section almost down to maximum recorded two-way-time (9 seconds). However the deep towed acquisition effectively limits the use of frequencies above the receiver ghost notch which compromises high frequencies in the overlying Tertiary section which were retained in the FSB reprocessing. The two datasets therefore need to be used in conjunction for regional evaluation.
Figure 5: (a) Original FSB time migration at an intersection with the NSR 2009 deep towed time migration and (b), the FSB reprocessed time migration at the same intersection.
We show that combining effective approaches from the reprocessing of an overlapping dataset in the northern UKCS with new deep towed acquisition allows improved imaging of deep Mesozoic and Palaeozoic structure below Palaeogene flood basalts. The deeper towed acquisition may not be suitable for the critical assessment of stratigraphic plays in the Tertiary.
The authors would like to thank TGS for permission to show the data examples and the hard work of colleagues involved in the reprocessing of the FSB datasets.
Hardwick, A., Travis, T., Stokes, S. and Hart, M. 2010. Lows and highs: using low frequencies and improved velocity tools to image complex ridges and basement highs in the
Moy, D.J. 2010. The Architecture, Growth and Tectono-Stratigraphic Significance of Rift-Oblique Lineaments on the
Mudge, D., Gail, M. and Holdoway, K. 2007. The
Ziolkowski, A., Hanssen, P., Galiff, R. and Li, X.Y. 2001. The Use of Low Frequencies for Sub-Basalt Imaging. SEG Extended Abstracts, 20, 74.