London Evening Lecture Review
Seismic 2.0: Is there a Broadband Seismic Res/volution?
with Floris Strijbos, Royal Dutch Shell plc
Review by Tim Sears, BP & Hassan Masoomzadeh, TGS
Following on from the PESGB Annual General Meeting, but no doubt looking forward to this lecture too(!), up to 150 members welcomed Floris Strijbos to talk about concepts and the impact of broadband seismic data.
After his Mining Engineering studies at Delft University, Floris has enjoyed a career with Shell, undertaking geophysical specialisms including multi-component and 4D seismic, rock physics and pore pressure, across postings in UK, USA, the Netherlands and Norway. Based in London, he is currently Shell’s Principal Technical Expert for seismic data acquisition, providing geophysical support to projects in Europe and CIS, and his technical calibre was apparent throughout the talk.
Floris began by quickly engaging the audience with a show of hands regarding wavelet shape and preference for low or high frequencies. The benefit of low frequency content for reducing side lobe energy, and role in improving resolution and interpretation, was highlighted from this outset.
The range of broadband acquisition and processing approaches and handling of the ghosts were noted, but a focus on concepts and impact of the technology given to the talk. He highlighted that industry is at the beginning of a learning curve and more needs to be understood, including suitability of processing and QC. Floris neatly illustrated the ghost concept through some finite-difference snapshots of wavefield propagation from both shallow and deep sources, and varying interference from the sea surface reflection. He further elaborated on complexities of low and high frequencies with connected themes of wavelet shape, spectral shaping, selected 1D and 2D wedge models, and band-limited reflectivity and impedance. Comparing two wavelets with similar low frequency content, Floris highlighted that a wavelet with a spectrum decaying towards the higher frequencies contains less energy than that of an almost flat spectrum, leading to a smaller peak yet much smaller side lobes. He called this a ‘broadband wavelet’, and the ‘peak to side lobe amplitude ratio’ an index of preference for the interpreter’s eye.
Floris continued beyond fundamentals with specific real data examples, explaining “2.0 broadband processing” would increase resolution both in time and space domains. To start, he described an interpolation and regularisation technique focusing on nearest useable offsets, to create High Definition (HD) volumes. Example HD 3D volumes showed striking improvements in resolution and interpretability over conventional, one case study in particular for drilling hazard assessment showing a compelling example of evident fluid migration upwards from a notable flat spot. Spectral decomposition on HD volumes further demonstrated the impact on overburden characterisation, and understanding of overpressure, faulting and fluid movement.
He moved on to the impact of broadband seismic data on band-limited inversion, again stressing the role of side lobe energy. Examples of inverted conventional and broadband data were shown together with band-limited synthetic impedance data, to help visualise the interference of side lobe energy.
Not limited to 3D, the value of 2D broadband data was demonstrated with complex structural and lithological mapping in an exploration setting; imaging of basalt and sub-basalt, complex faulting, (high angle) fault plane reflections and detachment faults all improved by data rich in low frequencies.
Moving to an appraisal setting Floris gave another articulate example of the impact of broadband, helping understand a reservoir where the structural spill point mapped from conventional seismic mismatched the oil-water contact. Broadband impedance data, however, revealed the details of a pinch-out responsible.
The talk gave a sensible balance of technical fundamentals and tangible examples of impact across E&P activities, showing the uplift and benefits of broadband over conventional seismic data, well delivered to satisfy a broad audience, with some good humour to boot.
Questions followed, from reflecting on how this ‘res/volution’ came about, to technical and commercial aspects of broadband. A member wondered “Shall we admit that we have screwed up for the last couple of decades, whose fault was it, and was it due to too much pressure from the operators demanding faster turnarounds?”! Floris recognised that there has been a blind spot in the industry. In relation, Gareth Williams, Dolphin Geophysical, reminded that not long ago we had low-cut filters in the acquisition system. A gentleman from CGG added that solid streamer technology has improved signal to noise in low frequencies. The source and producing more low frequencies using guns and bubble energy were raised, as were source generated noise. Floris emphasised the importance of clean low frequencies, attention to low-cut filters in both acquisition and processing, and care using window-based processes such as RMS scaling which might destroy the low frequencies.
Floris answered the title’s question, concluding that broadband seismic is a revolution which is here to stay, and suggested that interpreters could exercise and find different ways of looking at this new generation of seismic data. In answering a last question, there was a final word about the discomfort some interpreters feel in the appearance of broadband data. Floris reiterated the merits of reflectivity and impedance, and to improve our understanding of what an interpreter should be looking at. He went on to remark – to the satisfaction of many – that humans still have a role to play here…!