EAGE Workshop on Broadband Marine Seismic Data

We discuss the reasons for adaptive de-ghosting and its advantages. The success of broadband processing relies on de-ghosting accuracy, and because the parameters of the physical process determining the ghost reflection are not precisely known, de-ghosting cannot be purely deterministic. The physical entities determining the receiver ghost are the sea surface, the receiver depth and the water velocity. For de-ghosting purposes these can be represented by an effective ghost delay-time and sea surface reflection coefficient.

The de-ghosting optimisation metric we employ is based on the kurtosis of the data-autocorrelation. Kurtosis is a statistical measure often used to associate a measure of “peakedness” to a random variable. A synthetic example characterizes the main features of the adaptive procedure proposed, which can be sequenced in two steps. The optimisation of the ghost delay-time is robust to errors in the reflection coefficient and to noise, and therefore can be performed even in the presence of reflection coefficient errors. Therefore, it should be performed first. The reflection coefficient estimate is more sensitive to noise and requires the delay-time to be optimized in advance, and can therefore be carried out as a second step. A real data example demonstrates the applicability of the proposed method.

An improved, matching pursuit based high definition frequency decomposition method has been introduced to meet the challanges of broad bandwidth seismic data. It is largely sensitive to the finely separated thin events, yet it also has an enhanced frequency resolution and lateral consistency.

We acquired 3D multi-azimuth and dual-sensor broadband seismic data in such an area to achieve step-change quality in seismic images and subsequent exploration risk mitigation.

Legacy 3D data contained limited seismic bandwidth and poor seismic illumination insufficient to image the steeply dipping structures, and dual-sensor broadband technique and multi-azimuth shooting along three oblique orientations were thereby employed to boost the seismic description, based on illumination analysis. Among several options, the dual-sensor method was essential to acquire broadband pre-stack data preserving AV O characteristics.

Crisp resolution images from broadband data processing, including Kirchhoff PrSDM and MAZ stack, clearly visualized geological details of the deepwater deposits and folding structure, including channelized turbidite fans, chaotic Mass Transport Complex (MTC) packages, erosional unconformity at the structural crest, and imbricated thrust sheets. These seismic geomorphological patterns recognition, along with regional well data, led to turbidite sands fairway delineation.

Besides, in line with high porosity and rather compliant elastic rock properties in offset wells, rock physics modelling indicated seismic AVO attribute dependence on reservoir fluid. Seismic AVO anomalies extracted on the structural traps were hence utilized to mature prospects.

High definition 3D seismic data from new multi-azimuth and dual-sensor broadband technique delivered crucial inputs to exploration risk mitigation and prospect maturation, by revealing structural and sedimentary details and enabling advanced seismic analyses, in deepwater fold and thrust belts.

Variable-depth streamer acquisition has become a popular solution for marine seismic acquisition to obtain broad bandwidth data: the curved cable profile produces notch diversity that minimizes residual ghost and the deep towing segment provides high S/N at low frequencies. However, there have been discussions in the industry regarding the fidelity of AV O response from variable-depth streamer data due to the obvious changes in frequency and wavelet with offset. To answer this question, we need to focus on these key questions: can we remove the ghost successfully? Can we compensate the earth absorption effect to balance the spectrum from near to far offset?

Based on the de-ghosting algorithm that is termed Ghost Wavefield Elimination (GWE) which inversion in the tau-p domain ( Wang et al. 2013 , Poole 2013), we analyse the influence of GWE on the AV O response by using 2D synthetic datasets which were modelled by visco-elastic wave equation (Kjstansson, 1979) with two streamer towing configurations - conventional flat tow (8m) and depth-variable way (7~50m). GWE was applied on both datasets and the results show that it is very effective in attenuating the ghost energy. Furthermore, the AV O responses obtained in GWE results are almost identical. However, both AV O curves still decay rapidly in far offsets. To compensate this, pre-stack depth migration that incorporates the synthetic Q model (Q-PSDM) was applied. The resulting AV O curves then match well with the Aki Richards synthetic.

Our approach was then applied on a recent BroadSeis survey offshore Vietnam with source/receiver de-ghosting, shallow water de-multiple, advanced depth imaging and proper Q compensation. The high resolution common image gathers (CIGs) were then compared with the well synthetic gathers: AVO trends were picked along the top of sand layers and a good match was found within a frequency band of 5–60 Hz.

To summarise, we have demonstrated through both synthetic and real data examples that AV O fidelity is preserved for depth-variable streamer data with GWE pre-migration de-ghosting technology and ray path honouring pre-stack absorption compensation through Q-PSDM.

This paper summaries a testing result of various merging low frequency cutoff during the seismic inversion parameterisation process, subsequently to show the value of the extended low frequency from broadband seismic data towards seismic reservoir characterisation. Seismic inversion with 4Hz merge produced better image and reliable result. The broadband data has successfully added low frequency component and reduced the uncertainty on a complex geology.

The case study was carried out in the Babar Selaru PSC block in East Indonesia in 2014. The block is located in southern Outer Banda Arc which was developed by the Neogene collision between the Sunda and Australia plates. The southern Outer Banda Arc is characterized by complex structural elements consisting of a series of Paleozoic and Mesozoic basins on the Australian Continental Plate and the Neogene Banda Arc thrust belt.

3D broadband seismic data and appropriate PrSDM imaging technique provided us significant improvements in seismic imaging for both thrust complex and sub-thrust structures. Broadband 3D seismic imaging contributed better recovery of seismic energy from deep sections where conventional seismic imaging could not reveal any reflections. Pre-migration deghosting technology provided us the stable phase of wavelets and desired broadband frequency spectrum in the seismic image. The velocity model constructed from the combination of Kirchhoff PrSDM and Controlled Beam Migration provided better resolution and continuity of seismic images under thrust complexes by overcoming the poor illumination issue.

The seismic image with broadband frequency contributed to evaluating fluid effect in shallow section, reducing depth structure uncertainty in poor illumination area and interpreting lateral lithological change in deep reservoir section.

The objective of this paper is to illustrate both the qualitative and quantitative impact of interpreting dualsensor towed streamer seismic from the perspective of the reservoir geoscientist by focusing on a new 9224 km2 3D survey over the Caswell Sub-basin in the Northwest Shelf of Australia.

This case study investigates the broad bandwidth seismic improvements for structural and stratigraphic interpretation, including prospect delineation at the Mesozoic level. It will also highlight the AVO fidelity for future AVO and pre-stack simultaneous seismic inversion work. Furthermore, deeper Triassic plays have been clearly imaged for the first time in the survey area and reservoir property estimation is demonstrably improved due to the contribution of enhanced low frequency amplitude and phase stability across the offset/angles.

The main objective of this paper is to evaluate and demonstrate the ability of acquired pre-stack broadband seismic (towed dual-sensor streamer) to: reduce the uncertainty on the structural interpretation for prospect and lead identification and also to estimate reservoir properties directly from broadband seismic without using well information for calibration.

The volumes of broadband seismic data acquired and processed by the industry have grown rapidly. The spectral content of this new quality seismic is demonstrably superior to conventional seismic, both at the low and high frequency end of the spectrum. However, broadband seismic data will only deliver its full reservoir value if a quantitative interpretation is possible. Inversion to the impedance domain, assuming one has a good well-tie, is one of the main steps in quantitative interpretation.