This paper will investigate utilizing borehole acoustic logs to predict pore pressure at the borehole and how can we infer pore pressure from velocity logs. We will review common approaches to predict pore pressure at the borehole using acoustic logs. Two specific velocity to effective pressure equations will be the focus of this study: Eaton’s equation and Bower’s equation . We will discuss the normal compaction trend analysis and how it could help identify overpressure zones using Eaton’s equation. In conclusion, we aim to examine the advantages and disadvantages of those methods. This paper will provide a contribution in advocating the use of borehole acoustics to infer pressure at the borehole in the absence of direct pressure measurement.
I. Introduction [Narrow Azimuth OBC 3D Post Stack Time Migrated Seismic Survey on South Timbalier Block 54]: ￼We interpreted the D3 Sand within the South Timbalier 54 seismic volume to evaluate its potential for HC exploration. Various interpretation techniques were integrated to explain physical observations and anomalies. D3 Sand was deposited on the shelf in a steady to slightly dropping sea level resulting in increased in accommodation space; thus, a continuous aggradational to progradational deltaic heterogeneous sand . Seismic-wise, D3 appears as continuous layer across ST-54 with almost uniform thickness (Fig. 3 & 5), which fits with previous geological background. However, resolving its deltaic heterogeneities is beyond our seismic resolution; yet, implementing further seismic analysis could improve the detectability to resolve them. II. Methodology [Landmark Decision Space, GeoProbe]: We started by mimicking the seeded points on (200, 400, 600, 800 & 1000) across In-line & (50, 100, 150 & 200) along cross-line which provided us with the legacy 200x50 mesh for the top and the bottom of the D3 (Fig 1). Then, we refined the mesh down to 50x25 mesh. We altered the mesh size to minimize the effect of tracking along the low fold cross-line direction while keeping the mesh bigger within in-line where S/N is sufficient to be tracked by the software (Fig 1 & 5). Picks were made on min-phase peak assuming a min-phase data where we found coherent continuous reflection. Before tracking the mesh; we implemented user-controlled tracking process to account for the encountered irregularities within the mesh rather than using the unguided auto-tracker where we blocked tracking against polygons which were define based on anomalies driven from the attributes (Fig 1 & 2). Observed irregularities such as abrupt termination of reflectivity, mis-ties, vertical offsets & abrupt changed in seismic attributes (Amplitude, Frequency & Discontinuity) were later interpreted based on our structural & stratigraphic background of the region as faults. Results were quality controlled in 3D where we extracted the discontinuity along the fault plane (Fig 3). Structure Analysis [Previous existing topography, Graben, generated by deep faults originated at salt diaper]: To better illustrate our picks, let’s consider diagonal a NE-SW traverse line which orthogonally intersect most of our lateral irregularities as they commonly share a NW-SE trending. As we move from SW to NE; we observed 50 ms vertical offset within our section (Anomaly B); the offset is not unique only to our picked horizon at 1750 ms as it appears from 1100 ms all the way down to around 2400 ms or even beyond. Since data is offshore, sever near surface processing issues are excluded from our analysis and the continuity of such marginal offset in laterally and temporally indicate a major regional normal fault as those truncation does not reach the surface; instead, they gradually disperse upward. Moreover, we quality-control this interpretation ￼￼￼￼￼￼￼￼￼against a discontinuity volume where we extracted discontinuity RMS-Amplitude then transparently overlaid on the time horizon (Fig 5). Similar analogy was used to interpret anomalies B-H (Fig 5). In addition, Discontinuity volume-slicing, in 3D, showed an incoherency signature along slicing through our horizon and showed continuum discontinuities as we time-sliced our volume which supported our initial fault interpretation. Moving NE, we continued to observe these mis-tie anomalies with different magnitude and less truncation at the middle. The observed offset decreases toward the middle, then; increases sidewise. Using similar analogy to interpret those features as before, we reached a conclusion that they are mainly normal growth faults starting from zero-reflection zone beyond 2400 ms. We spot the D3 to be geometrically depressed block of sand broader by semi-parallel faults. Therefore, we interpreted the general structure to be graben. The graben is located above the center of the zero-reflectivity body, possibly the Louann Salt. Blending horizon and fault interpretation burial history, we observed that the sand could had been deposited comfortably on semi-syncline area on the top of active salt based on (Mov 1). Geological background, observed zero-reflectivity zone at the middle and the strike and dip direction of the faulting system alluded that ST54 is resides on top salt diaper/dome as the two regional major faults could be traced down to the salt; thus, originated there before the deposition of the D3 Sand . Kinematic modeling illustrated that lower strain on the sides of the graben; hence, more weight and compaction on the sides compared to the center which was the driving force behind expelling the salt upward in a semi-vertical direction in the center of the graben where the model showed higher strain; hence, more deformation at the middle of graben (Mov 1). In addition, faults strain modeling showed a higher strain on the middle indicating that those smaller faults could not be generated by salt directly; rather, they originated from other faults due to salt collapsing which is common in rapid increase of sedimentation [3,4]. Such conclusions were possible to draw considering the fact that D3 was deposited within local sea level drop during or closing of the gulf in the early stage of the graben collapsing; then, faults increase the temporal stratigraphic offset between its compartments while overall hot climate played a vital role in increasing the supply of sediments [2,3 & 4]. III. Stratigraphic Analysis [On Shelf Deltaic sand with good vertical resolution in seismic]: Extracting the RMS-Frequency at horizons showed a dominate Frequency around 20 Hz for both yielding a vertical resolution around 30 m; assuming constant 2400 m/s at the AOI (Fig 4). Since, Frequency is inversely related to thickness; such frequency respond infer sediments diverge against the normal faults as we moved from the footwall to the hanging wall block which tie with earlier analysis. ￼IV. Hydrocarbon Potential [Yes, Bright spot in post stack seismic section with 4.12 BCF] : ￼￼￼￼￼ ￼￼￼￼￼￼Utilizing the most successfully DHI, RMS-Amplitude showed bright spots and 3 major gas pools were identify (Fig 6). Since, it is below seismic resolution to draw early conclusion about multi-producing layers within the D3; we can safely assume that the D3 is homogenous layer from reserve booking point of view only. Hence, attribute calculations are generated from the top to the bottom of the D3. Surprisingly, we did not focus our attention on the brightest area; rather, shifted our auto-polygon function cut-off value down to 4500-5200 for two main reasons (Fig 6). First, the nature of the amplitude distribution within the D3 is quite good despite contamination coming from striping effect (Fig 6). A 4500 (Highest %75-%50) is a good estimate considering that we had the same amplitude respond in A3 where we used a %35-cutoff on amplitude histogram. Moreover, D3 has been primarily in oil-production since 1979 and the tiny bright spots on the high topographic map should not correlate to leads; as they are quite small in size for associated gas and oddly fit within topographic high (Fig 4 & 6). Therefore, those tiny bright spots are secondary gas caps which were generated as result of rapid production of oil; as pressure drop during production, pushing the dissolved gas within oil upward toward topographic high while liquid is being produced. The 3 gas pools contain many leads; however, with no marginal structure change nor resolvable strata change between them, we could assume that they belong to the same gas pool but that does not necessarily mean that they are pressure-connected. On the other hand, the three pools are segregated structurally by faults; therefore, they cannot pressure communicate. In short, our pools/leads are not text-book example of 4-way closure as stratigraphic thinning beds altered the rock properties and helped defining trapping mechanism which ties with the nature of heterogeneous deltaic sand deposits in general. Using full cycle min-phase wavelet assumption, average thickness was estimated and OGIP estimate and probability were generated (Table 1). V. Conclusion & Recommendation (Re-Processing, AVO, Inversion & Pore-Pressure Model): Seismic interpretation and analysis provided a key role in understanding the structure style with ST-54. In addition, re- processing AVO-friendly gather should be a by-production if we decided to re-process the data as the DHI approach did not quietly represent the booked gas reserve within this field; underestimating it three time. Processing/acquisition striping artifact were observed along the cross line in amplitude attribute and interpreters should be careful as those stripes have masked the true background amplitude signature. We ought to recommend PSDM to better image the deep faulting system and the salt. Post stack seismic inversion; with enough well control, could also be beneficial to generate an accurate porosity map while Pre stack seismic inversion could also help us identify the saturated zone within the reservoir itself. Moreover, pore pressure prediction based on a seismic-velocity should be considered as it aid in optimizing the drilling program as well as the drilling mud weight design to lower the risk associated with drilling hazards. ￼￼￼￼￼ References: ￼ Bose, S., and Mietra, S., 2014, Structural analysis of a salt-cored transfer zone in the South Timbalier Block 54, offshore Gulf of Mexico: Implicationsfor restoration of salt-related extensional structures: AAPG Bulletin, v. 98, no. 4 (April 2014), pp. 825– 849.  Stude, G. R., 1978, Depositional Environments of the Gulf of Mexico South Timbalier Block 54 Salt Dome and Salt Dome Growth Models. Transactions-Gulf Coast Association of Geological Societies, v. 28, p. 627-646.  Hudec, M. R., and M. P. A. Jackson, 2006, Advance of salt sheets in passive margins and orogens: AAPG Bulletin, v. 90, p. 1535–1564.  Vendeville, B. C., and M. P. A. Jackson, 1992, The rise of diapirs during thin-skinned extension: Marine and Petroleum Geology, v. 9, p. 331–353.  Investor Relation, 2013, Annual Report Filing to SEC, Energy XXI.
Gabor deconvolution, an extension to nonstationary Weiner deconvolution, was utilized as a post- depth migration processing filter to determine its benefit for improved resolution and multiple attenuation in the image domain. Tests were carried on pre-stack depth-migrated 2D gathers on deep offshore data. The geology of the region combines complex salt tectonics with layers of evaporative sequences (LES). The LES sequence generates short-period multiples which are difficult to attenuate using velocity-based algorithms. Traditional deconvolution compensates for absorption through its assumption of “white” reflectivity spectra, with most of the implementation therefore implying an infinite “Q” attenuation function. In contrast, the nonstationary approach to the deconvolution process approximates the values associated with the attenuation function “Q”. We expect, therefore, that the nonstationary approach should be more suitable in the presence of the complex velocity where the assumption of an infinite “Q” would provide suboptimal results (i.e., failure of the underlying assumption of Stationary Deconvolution). We performed a series of tests which applied Gabor deconvolution in the image domain in order to suppress multiples and to balance reflection amplitudes. We established a systemic approach to test this method by applying Gabor either before or after a velocity-based multiples attenuator. Results were then compared as controlled group against each other through spectral analyses Application of the Gabor deconvolution in the image domain resulted in the recovery of frequencies between 20-40 Hz and the slight suppression of low frequency noise between 0-10 Hz. As a result we obtained a laterally well-focused stacked section with preferable amplitude balancing. A validation study was undertaken in order to compare our results with those obtained from the conventional predictive deconvolution. Our final results showed an overall improvement in resolution, better continuity of the reflections and suppression of multiples in several zones.
Building the low frequency model for seismic inversion plays a key role in the inversion process as it initially as it initially establishes the spatial distribution of reservoir parameters the spatial distribution of the reservoir parameters which could produce misleading inversion results from the seismic data. Commonly used, the model-based inversion algorithm depends heavily on the accuracy of the input low frequency model as it merges it with the measured seismic. Using a standard workflow, we interpolate the well log acoustic impedance spatially across our control points and then assign distance-based weight to each. Such a workflow would not honor an anomalous signal, such as a change in geological depositional environment, between those wells. The low frequency model can be derived using different techniques that honor the seismic amplitude away from the well, especially in the absence of sufficient control wells during exploration stage. In our work flow, two models are generated: the first one, “plain vanilla”, which uses simple interpolation between well logs, and the second one which is constrained using seismic attributes to guide the interpolation away from the wells in order to assist in qualifying the interpretation of the subsurface. This dual-interpolation approach was tested on both synthetic and 3D field seismic data, using control wells on field data. We aimed to resolve reservoir parameters and to reduce risk in the exploration stage, in which the interpreter seeks to identify sand bodies in varying alluvial environments. We will show that in these exploration settings, a more representative low frequency model enhances our inversion product by resolving tuning and wavelet effects. In conclusion, we determined that using the seismic amplitude to constrain the low frequency model building not only helped to improve the inversion process at well locations, but it also yielded lower error residual and, hence, produced a better representation of the subsurface sand distribution.