Integration of Formation Evaluation Data Overcomes Offshore Coring Challenges
A well was drilled into a prospective unconventional mudstone play offshore Norway. Two of five coring runs were successful while the rest yielded little to no core recovery. Subsequent investigation of the core substantiated that the coring issues largely had natural causes.
A well was drilled into a prospective unconventional mudstone play offshore Norway. Two of five coring runs were successful while the rest yielded little to no core recovery. Subsequent investigation of the core substantiated that the coring issues largely had natural causes. This understanding is being applied to two imminent coring operations and has driven selection of drilling, coring, and wireline technology and procedures and is informing casing design.
The Valhall field is in Block 2/8 of the Norwegian North Sea. The primary reservoir is Cretaceous chalk, located at approximately 2400 m true vertical depth. Production of this reservoir began in 1982. Above this reservoir are other formations, including a shallower potential unconventional mudstone reservoir.
A production well was drilled targeting the Cretaceous chalk reservoir. As part of the drilling program, logging-while-drilling (LWD) and wireline logs—including LWD density, neutron, gamma ray and acoustic, wireline spectral gamma ray, cross-dipole acoustic, nuclear-magnetic-resonance, and formation-testing and -sampling—were run across the potential mudstone reservoir to appraise the standard petrophysical properties. A conventional core was also attempted; however, recovery was poor, with repeated jams.
During coring operations, significant problems were encountered when coring the mudstone intervals. These challenges are illustrated in Fig. 1, which shows a section of the core from xx14 m, demonstrating the typical condition of the recovered core when still in the core barrel. At the time of operation, poor recovery was attributed to the coring practices used and no further coring was attempted.
The mudstone interval was not a primary target for the well. Therefore, limited formation evaluation of these intervals was performed, mainly to estimate the petrophysical properties. However, at the time of initial logging, more-advanced formation evaluation data (e.g., images, full waveform acoustic, and nuclear-magnetic-resonance data) were not evaluated fully.
A few years after initial operations, the log data were revisited—specifically, detailed interpretation was conducted for all recorded log measurements. On examination of the results over much of the interval, the integration of the conventional and advanced volumetric measurements gave considerable extra value in assessing the petrophysical properties of the mudstone. However, detailed discussion of these results is outside the scope of this paper. In addition to the extra information gleaned from the advanced volumetric interpretation, over certain intervals, some unusual log responses and interpretation results were seen that were deemed worthy of further investigation. These include:
- Local variations from the overall trend in acoustic azimuthal anisotropy
- Local radial variation of shear slowness with distance from the borehole that is not present immediately above and below
- Localized disturbance in bedding seen on high-resolution bulk-density and gamma ray images
- Change in frequency and amplitude character of monopole and dipole acoustic waveform data
- A noticeable very high amplitude reflective feature present in the near-wellbore reflection image
Because these observations occurred at a similar depth, it is unlikely that they can be attributed to issues with the log responses (e.g., tool issues or borehole effects on the log measurements).
Over much of the interval, the fast-shear orientation consistently is oriented approximately northwest/southeast, which suggests a predominant northwest/southeast principal stress direction. However, over the interval from xx29 to xx35 m, a noticeable perturbation in the anisotropy orientations and magnitudes can be seen, which also corresponds with a change in the character of the caliper curve. These perturbations suggest that a localized feature resulting in a change in the azimuthal anisotropy orientation may be present.
With regard to the logged and enhanced density image over the interval of interest, over much of the image, indications exist that low-angle-to-borehole planar bedding is present. However, over the interval from xx33 to xx38 m, this trend changes. Because of the relatively low resolution of LWD bulk-density images compared with other imaging technologies, the individual beds cannot be seen, but the general image character indicates disturbance in the bedding compared with areas above and below. This corresponds with the change in the azimuthal anisotropy direction from the overall trend observed in the azimuthal anisotropy analysis and thus is further evidence of a localized structure. Further evaluation of the acoustic waveform data allows evaluation of the radial variation of shear slowness with distance to the borehole.
While over much of the interval the radial variation in shear is relatively consistent, once the large variations in shear slowness near to the wellbore (most likely associated with the near-wellbore damaged zone) are passed, few changes in the slowness distribution away from the borehole with depth are observed. However, over the interval xx33 to xx39 m, a noticeable change in the radial variation trend can be seen that does not follow the trend above and below. Again, this correspondence with the other data suggests the presence of a localized feature with different properties than those of the surrounding formation.
The wireline acoustic data were then analyzed further to generate a near-wellbore reflection image using shear waves reflected from planar features in the formation. In this formation, imaging features extending up to 5 m from the borehole axis are possible. A 3D image investigating the 360° volume around the borehole was generated to visualize the formation around the wellbore fully.
The images display reflections from planar features in the formation. Of particular interest is the strong reflection present intersecting the borehole at approximately xx35 m. Because this can be observed to a certain degree in all planar orientations, this suggests that a high-acoustic-impedence contrast exists between the reflecting feature and the surrounding formation.
High-acoustic-impedance contrasts commonly are associated with structural features such as fluid-filled fractures and faults. The amplitude of the reflection here suggests that the feature has a large aperture or has a significant damage zone on either side of the feature, therefore suggesting strongly that the reflecting feature is a fault.
Analysis of the advanced measurements indicate that, over the approximate interval from xx30 to xx40 m, a localized change in the formation is present, although initial indications from the available logs at the time of drilling did not reveal the cause clearly. Fig. 7 of the complete paper presents a composite plot of the analysis where the various results are combined and compared with the cored intervals.
Over much of the interval, where core recovery was better, the formation properties are relatively consistent with little variation. However, this is not the case in all intervals. Of interest is the interval from xx33 to xx39 m, which is where core recovery was poorest. Associated with this poor recovery, a significant change in all the various measurements can be observed that is not consistent with the formations above and below. When all measurements are compared with the core, it can be clearly seen that this corresponds to the interval where the cores are showing significant localized damage and poor recovery. The rugosity on the calliper alone cannot explain the observed log responses. This correspondence can be seen to be more than coincidental and suggests strongly that the coring issues cannot alone be attributed to poor coring practices and that localized changes in the formation properties can be identified as a probable cause.
Comparison With Seismic and Other Information
The well penetrates a mapped fault at approximately the location of a 19-m-long rubble zone (from approximately xx08 to xx27 m) observed in the core. Close inspection of the rubble shows that each piece or pod was originally surrounded by black hairline fractures. The coring process dislodged most of these pods, though some have remained intact. It is likely that the rubble zone extends beyond the cored section based upon logging evidence, potentially to at least xx40 m measured depth (MD). Deeper in the cored section (at xx82 m MD), an 8-m zone of discrete fractures with well-developed slip planes and slickenlines is present. This zone is considered to be a damage zone related to the fault identified in the seismic data.
These observations from the core are supported by observations from the log data. The depths of the unusual log responses show a close correlation with the considerable damage and brecciation observed in the core, suggesting that the reason for these responses and observations is the presence of a fault.
Further data from additional wells are required, together with testing the production potential of these structures. A relatively impermeable fault core is surrounded by a highly fractured damage zone in a fractured, low-permeability matrix reservoir. Oil and gas from the host rock will drain into the fault zone.
This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 195866, “Overcoming Coring Challenges in a New Unconventional Play Offshore by Integration of Formation Evaluation Data,” by Thomas Bradley, SPE, Baker Hughes, and Per Henrik Fjeld and Jonathan Scott, Aker BP, et al., prepared for the 2019 SPE Annual Technical Conference and Exhibition, Calgary, 30 September–2 October. The paper has not been peer reviewed.