Richard E. Swarbrick

Origin of overpressure and pore-pressure prediction in the Baram province, Brunei

Accurate pore-pressure prediction is critical in hydrocarbon exploration and is especially important in the rapidly deposited Tertiary Baram Delta province where all economic fields exhibit overpressures, commonly of high magnitude and with narrow transition zones. A pore-pressure database was compiled using wireline formation interval tests, drillstem tests, and mud weights from 157 wells in 61 fields throughout Brunei. Overpressures are observed in 54 fields both in the inner-shelf deltaic sequences and in the underlying prodelta shales. Porosity vs. vertical effective stress plots from 31 fields reveal that overpressures are primarily generated by disequilibrium compaction in the prodelta shales but have been generated by fluid expansion in the inner-shelf deltaic sequences. However, the geology of Brunei precludes overpressures in the inner-shelf deltaics being generated by any conventional fluid expansion mechanism (e.g., kerogen-to-gas maturation), and we propose that these overpressures have been vertically transferred into reservoir units, via faults, from the prodelta shales. Sediments overpressured by disequilibrium compaction exhibit different physical properties to those overpressured by vertical transfer, and hence, different pore-pressure prediction strategies need to be applied in the prodelta shales and inner-shelf deltaic sequences. Sonic and density log data detect overpressures generated by disequilibrium compaction, and pore pressures are accurately predicted using an Eaton exponent of 3.0. Sonic log data detect vertically transferred overpressures even in the absence of a porosity anomaly, and pore pressures are reasonably predicted using an Eaton exponent of 6.5.

Lateral transfer : a source of additional overpressure ?

Abstract Overpressure generated by disequilibrium compaction can be enhanced at structural crests by the lateral transfer of fluids from deep, overpressured parts of a basin along laterally extensive inclined aquifers. Fluid flow modelling shows that the effect of an inclined aquifer ranges from being minimal to being a significant contributor (>10 MPa) to crestal pore pressures depending on the style of structural evolution, the burial rate and the aquifer relief. Dynamic fluid flow models show that the Centroid Concept, a method for estimating crestal pressures based on a static view of subsurface fluid pressures, does not in general predict the pressure distribution associated with a dipping aquifer. A case study is presented from the Central North Sea where the Pre-Cretaceous reservoir forms part of an aquifer with 2.5 km of relief from the structural crest to the low point. The inclined aquifer is covered by a ubiquitous low permeability layer and has undergone extremely rapid recent burial (>500 m/Ma). As expected from the generic modelling analysis, this combination of factors means that lateral transfer has a minor effect on the pressure development in the case study.

Present-day stress orientation in Brunei: a snapshot of ‘prograding tectonics’ in a Tertiary delta

The Baram Delta province of NW Borneo is unusual when compared with most other Tertiary deltas, as it has built up upon an active margin. Hence, structures observed in the Baram Delta province are the result of both margin-parallel gravity-driven deltaic tectonics and approximately margin-normal transpressive tectonics associated with the active margin. Image and dipmeter logs have been examined for breakouts and drilling-induced tensile fractures (DITFs) in 47 wells throughout Brunei. Breakouts and DITFs observed in 19 wells suggest that the maximum horizontal stress is oriented margin-normal (NW–SE) in the proximal parts of the basin and margin-parallel (NE–SW) in the outer shelf region. The margin-parallel outer shelf stress field is interpreted as a local ‘deltaic’ stress field caused by the shape of the clastic wedge. The margin-normal maximum horizontal stress in the inner shelf is interpreted to reflect basement stresses associated with the active margin. However, the maximum horizontal stress in the inner shelf is approximately perpendicular to the strike of Miocene–Pliocene normal growth faults, suggesting that maximum horizontal stress in the inner shelf has rotated from margin-parallel (‘deltaic’) to margin-normal (‘basement-associated’) over time. Hence, approximately the same stress rotation has occurred over time in the inner shelf as is currently observed spatially from the outer to inner shelf. The spatial and temporal stress rotations in Brunei are thus interpreted to be the result of ‘deltaic’ and ‘basement-associated’ tectonic regimes that are ‘prograding’ basin-wards. The proximity of the active margin has resulted in progressive uplift and inversion of the hinterland that has ‘forced’ the delta system to prograde rapidly. The zone of active deltaic growth faulting (and margin-parallel maximum horizontal stress) has shifted basin-wards (‘prograded’) as the delta system has rapidly prograded across the shelf. After uplift and delta progradation, the old growth faults of the inner shelf ceased being active and have then been successively reactivated by a similarly ‘prograding’ margin-normal inversion front.

Diagenesis in North Sea HPHT clastic reservoirs — consequences for porosity and overpressure prediction

Upper Jurassic clastic reservoirs of the Fulmar formation in the Central North Sea possess anomalously high porosities for their present day depth of burial. Reservoirs with the highest overpressures have the highest porosities, possess less macroquartz cement, and have significant secondary porosity. Quartz cementation in HPHT (High Pressure High Temperature) reservoirs has been inhibited by a combination of factors: high overpressure, limited fluid movement, presence of early grain coating cements, high pore fluid salinity, and possibly petroleum migration. Secondary porosity has contributed to reservoir quality, with an average of 4vol% extra porosity created. Quantitative prediction of porosity would require an improved depositional model for the Fulmar, accurate thermal and pressure modelling, and detailed knowledge of field filling and leakage histories. Theoretical calculations indicate that diagenetic reactions occurring in the Fulmar formation (smectite illitisation and quartz cementation), did not generate significant overpressure, because seal permeabilities were too high and the rate of volume increase associated with the reactions too small. Therefore diagenetic reactions can effectively be ignored when modelling overpressure generation in the Central North Sea, although cementation will affect rock permeability and rates of fluid dissipation.

Present-day stress and neotectonics of Brunei: Implications for petroleum exploration and production

The present-day state of stress in Tertiary deltas is poorly understood but vital for a range of applications such as wellbore stability and fracture stimulation. The Tertiary Baram Delta province, Brunei, exhibits a range of contemporary stress values that reflect the competing influence of the northwest Borneo active margin (situated underneath the basin) and local stresses generated within the delta. Vertical stress (v) gradients at 1500-m (4921-ft) depth range from 18.3 MPa/km (0.81 psi/ft) at the shelf edge to 24.3 MPa/km (1.07 psi/ft) in the hinterland, indicating a range in the shallow bulk density across the delta of 2.07–2.48 g/cm3. The maximum horizontal stress (Hmax) orientation rotates from margin parallel (northeast–southwest; deltaic) in the outer shelf to margin normal (northwest–southeast; basement associated) in the inner shelf. Minimum horizontal stress (hmin) gradients in normally pressured sequences range from 13.8 to 17.0 MPa/km (0.61–0.75 psi/ft) with higher gradients observed in older parts of the basin. The variation in contemporary stress across the basin reveals a delta system that is inverting and self-cannibalizing as the delta system rapidly progrades across the margin. The present-day stress in the delta system has implications for a range of exploration and production issues affecting Brunei. Underbalanced wells are more stable if deviated toward the hmin direction, whereas fracture stimulation in mature fields and tight reservoirs can be more easily conducted in wells deviated toward Hmax. Finally, faults near the shelf edge are optimally oriented for reactivation, and hence exploration targets in this region are at a high risk of fault seal breach.

Triggering of the Lusi mud eruption: Earthquake versus drilling initiation

ABSTRACTThe Lusi mud volcano in East Java has erupted unabated for almost 2 yr, fl ooding an area of 7 km 2 and displacing more than 25,000 people. Despite its disastrous impact, the mechanism for triggering the Lusi eruption remains highly controversial; two distinct mechanisms have been proposed. One hypothesis suggests that the eruption was triggered by the M w 6.3 earth-quake that struck Yogyakarta (250 km from Lusi) two days before the eruption. However, an examination of static and dynamic stress changes and stress transfer mechanisms indicates that the Yogyakarta earthquake was at least an order of magnitude too small to reactivate faults and open fl uid fl ow pathways under Lusi. An alternate theory suggests that Lusi was triggered by a blowout following drilling problems in the nearby Banjar Panji-1 well. Blowouts result from an inability to control pore fl uid intakes into the borehole and typically occur when the drilling win-dow (fracture pressure minus pore pressure) is approximately zero and when there is insuffi cient protective casing of the well bore. Pore and fracture pressure data from Banjar Panji-1 indicate that the well had a narrow drilling window of only 0–2.3 MPa. Furthermore, two planned casing points were skipped during drilling, resulting in 1742 m of unprotected borehole. The combina-tion of hazardously narrow drilling window and long uncased borehole would have made drill-ing problems in Banjar Panji-1 diffi cult to control, placing the well at high risk of blowing out. Furthermore, well-bore pressures following drilling problems in Banjar Panji-1 reached magni-tudes in excess of the fracture pressure and thus were suffi cient to create fl uid fl ow pathways in the subsurface. Therefore, we suggest that no viable method is known by which the Yogyakarta earthquake could have triggered the mudfl ow and that a blowout in the Banjar Panji-1 well was the most likely mechanism for triggering the Lusi eruption.Keywords:

The use of leak-off tests as means of predicting minimum in-situ stress

Leak-off tests (LOTs) or, preferably, extended leak-off tests (XLOTs), can be successfully used in minimum in-situ stress, S3, estimations. Selecting a point on the leak-off graph that represents the best proxy for S3 can reduce inaccuracies in the use of LOTs as a means of determining S3. If the testing procedure is well conducted and recorded, picking the leak-off pressure (LOP) or instantaneous shut-in pressure (ISIP) gives equally valid estimates of S3. During testing, most of the pressure applied in the deduction of S3 is exerted by the static mud column, particularly in overpressured settings where higher drilling mud weights are used. Since the mud column contributes such a large proportion of the applied pressure, estimating S3 from tests conducted at greater depth means the observed small difference between LOP and ISIP has even less of an effect on the deduced S3 value. The data used in this study show that LOP closely matches ISIP when considering multiple cycle XLOTs. It can therefore be inferred that the LOP is the fracture re-opening pressure and hence Sh given that the assumptions made by the Kirsch equation for wellbore failure are upheld. This study also considers the implications for calculating the magnitude of SH.

Integrated study of the Judy Field (Block 30/7a) : an overpressured Central North Sea oil/gas field

The Triassic reservoirs of the Judy Field, an overpressured petroleum accumulation in the Central North Sea, have been studied to determine their pressure and petroleum filling history. The magnitude of overpressure in the Pre-Cretaceous aquifer is similar across the field at about 24 MPa (3500 psi), but with some higher pressure laterally in areas closest to the major depocentres. Pressure modelling shows that the magnitude of the overpressure can be attributed almost entirely to disequilibrium compaction, and largely due to late rapid Tertiary burial, although modelling does require nanoDarcy (i.e. shale-like) permeability in the Cretaceous chalk section. The contributions from other mechanisms, which may be relevant in this setting, i.e. gas generation and lateral transfer, appear to be small. The existence of both aqueous and petroleum phases in secondary fluid inclusions in microfractures in quartz and feldspar grains allows estimation of palaeopressure. The data show fracture healing from 115 to 136°C during a time when reservoir palaeopressures were above hydrostatic. Reconstruction of the temperature history of the reservoir places the timing of the fluid inclusion formation during fracture healing at between 3 and 1 My ago, coincident with a period of rapid burial. The palaeopressure estimation using fluid inclusions validates the pressure modelling studies, which are derived independently from commercial basin modelling software. The fluid inclusions also revealed a low-GOR oil over the whole field, prior to late stage flooding by gas leading to the variable (low to high) GOR fluids plus gas condensates within the field today.

‘Vertically transferred’ overpressures in Brunei: Evidence for a new mechanism for the formation of high-magnitude overpressure

Overpressures in sedimentary basins are commonly assumed to be the result of two distinct and separate mechanisms: disequilibrium compaction and fluid expansion. However, the potential for overpressures to be redistributed or transferred to other pressure compartments over time has been seldom considered and rarely demonstrated. Pore-pressure data and velocity-effective stress plots from 61 fields across the Baram Delta province of Brunei (northwest Borneo) reveal that two different types of overpressure occur in stratigraphically defined sections: the basal pro-delta shales contain overpressures generated by disequilibrium compaction, whereas the overlying sand/shale deltaic sequence contains overpressures that appear to be generated by fluid expansion. However, the geology of the deltaic sequences and high magnitude of the pore pressures precludes the overpressures in the deltaic sequences being generated by any conventional fluid expansion mechanism, such as kerogen-to-gas maturation or clay diagenesis. The fluid expansion overpressures are located in fields that were inverted during the Pliocene, an event that resulted in large-scale fluid migration from the pro-delta shales into the deltaic sequences, including charging of the numerous oil fields in the inner shelf. Hence, we propose that the overpressures in Brunei provide the first evidence for a new overpressuring mechanism whereby overpressured fluids have been “vertically transferred” from the pro-delta shales into the deltaic sequences during basin inversion. Furthermore, vertical transfer may be a mechanism for explaining overpressures observed in basins that have been recently uplifted or inverted.

Probabilistic longevity estimate for the LUSI mud volcano, East Java

Abstract: A new method for estimating the duration of a mud volcano eruption is applied to the LUSI mud volcano in East Java. The estimate is based upon carbonates at depths in the range 2500–3500 m being the water source, with an estimated area of 100–600 km2, thickness of 0.2–1.0 km, porosity of 0.15–0.25, an initial pressure between 13.9 and 17.6 MPa, and a separate, shallower source of mud (c. 1200–1800 m depth). The resulting 50 percentile for the time it takes for flow to decline to <0.1 Ml day−1 is 26 years. By analogy with natural mud volcanoes it can be expected to continue to flow at lower rates for thousands of years. Assuming subsidence rates of between 1 and 5 cm day−1, land surface subsidence of between c. 95 and c. 475 m can be expected to develop within the 26 year time period.