Click
to view article in PDF format.
Diagenetic Model: Deep Panuke Reservoir, Offshore Nova Scotia, Canada*
By
Rick Wierzbicki1, and Nancy Harland1
Search and Discovery Article #40136 (2004)
*Adapted from extended abstract for presentation at AAPG Annual Convention, Dallas, TX, April 18-21, 2004.
1EnCana, Calgary, Alberta, Canada ([email protected])
The Jurassic-aged Abenaki carbonate complex was deposited along the eastern margin of the North American continental shelf. The Deep Panuke reservoir was discovered in 1998 with the drilling of the PanCanadian PP-3C well. The PP-3C well encountered a thick vuggy limestone reservoir; subsequent wells encountered both vuggy limestone and dolomite.
Early work by Eliuk suggested that reservoir development on the carbonate margin was related to lowstand development of a meteoric lens, leaching of allochems, and mixing-zone dolomitization.
The reservoir has been subjected to multiple phases of dissolution, calcite cementation, fracturing, and dolomitization. Almost all of the diagenesis is post-burial. Shortly after deposition calcite cementation occludes much of the primary porosity. Aragonite and high-Mg calcite fossils were dissolved and infilled with blocky calcite. Subsequently matrix dolomitization developed along the platform margin. Later, saddle dolomitization occurred, lining vugs and fractures and recrystallizing some of the earlier dolomite. Dissolution of dolomites and limestone created vuggy limestone porosity. Fractures are an important component of the permeability network of the reservoir. Fractures consist of microfractures and swarms of short-length macrofractures, some of which have been enhanced by leaching. Reservoir distribution appears to be controlled by fault and fracture networks, creating conduits for diagenetic fluids.
|
IntroductionThe Abenaki carbonate complex was deposited along the eastern margin of the North American continental shelf in the Middle Jurassic to Early Cretaceous. The Deep Panuke reservoir was discovered in 1998 with the drilling of the PanCanadian PP-3C well (Figure 1). The PP-3C well encountered a thick, vuggy limestone reservoir charged with slightly sour dry gas; subsequent wells encountered both vuggy limestone and dolomite (Figure 2). Porosity types range from near cavernous vuggy porosity in limestone and dolomite to earthy microporosity in leached limestone. Dolomitic portions of the reservoir are dominated by intercrystalline and vuggy porosity. Early work by Eliuk and others suggested that reservoir development on the carbonate margin was related to lowstand development of a meteoric lens, leaching of allochems, and mixing zone dolomitization. The present work indicates some meteoric influence is present but diagenesis is dominated by burial dolomitization and dissolution. Jeff Dravis and Ihsan Al-Aasm conducted a petrographic, isotopic, and fluid inclusion study of the reservoir. The information and conclusions drawn are built upon their observations and interpretations and those of EnCana.
Stratigraphy/Sedimentology/Structural GeologyThe carbonate platform at Deep Panuke has been divided into seven third-order depositional cycles (Figure 3). The margin has been sculpted by gravitational collapse during deposition. The preserved margin area is made up of shallow to deeper water coral and stromatoporoid patch reefs and inter-reef and foreslope sediments while the stable carbonate shelf immediately behind it is dominated by shallow water oolitic shoals. The deeper foreslope area is dominated by coral sponge to sponge reefs, debris flows, and thrombolitic mudstones. Reservoir distribution appears to be controlled by fault and fracture networks creating conduits for diagenetic fluids. The fracture networks are not resolvable on the seismic data, and faults are difficult to map. The seismic data does show the areas where thick high porosity zones are present. Generally these occur as curvilinear trends near the edge of the margin behind faulted scallops (Figure 4). Porosity appears to be concentrated along these zones of partial failure or weakness where small fractures and faults would have allowed access to basinal diagenetic fluids. The areas where high-porosity, vuggy limestone has been encountered (wells H-08 and PP-3C) occur along linear trends back from the margin edge. These may represent areas of postulated strike-slip faulting. Fractures are an important component of the permeability network of the reservoir. Fractures consist of microfractures and swarms of short length macrofractures, some of which have been enhanced by leaching. Fractures are more common in the dolomitized margin areas. This is due to the combination of the more brittle nature of dolomite and the greater amount of flexure and faulting at the margin edge.
Petrographic, Isotope, Fluid Inclusions: ObservationsSampling is biased with most cores recovered from lower-porosity limestone and dolomites. Little indication of shallow diagenesis is observed. Most diagenesis is post-compaction, deep-burial related. The reservoir is dominated by secondary porosity related to deep-burial dissolution of limestone and dolomite. Early dolomite is present but has often been recrystallized or replaced by saddle dolomite. Dolomite grains cross-cut stylolites, and porosity is developed along stylolites and fractures, suggesting that dolomite diagenesis and dissolution was post-burial and post-fracturing. Microporosity is developed in both limestone and dolomite. The best porosity development observed occurs in limestone with a leached micritic matrix. Dravis proposed that the secondary porosity development was controlled by late acidic, possibly hydrothermal, fluids accessing the reservoir via faults and fractures. Fractures, brecciation, and vertical stylolites observed in thin section suggest local compression and faulting. Iron sulphides are common in the reservoir. Isotopic analysis indicates calcitic skeletal components generally fall close to the expected range of carbon and oxygen isotopes for early Jurassic marine carbonates or are somewhat depleted. (Figure 5). Early cements are similar but somewhat depleted while late cements are more depleted. These results indicate precipitation or recrystallization at elevated temperatures with initial shallow-burial diagenesis continuing into the deep burial zone. Matrix dolomite is not in the expected range of carbon and oxygen isotopes for primary Early Jurassic dolomite (Figure 6). Recrystallized matrix dolomite and saddle dolomites have a similar depleted isotopic composition. This also indicates precipitation at elevated temperatures and supports deep-burial diagenesis. Matrix dolomite that has not been recrystallized is slightly less depleted and occurred earlier and shallower. Sr ratios are generally enriched in calcite cements and fossils (Figure 7). Matrix dolomites are enriched in Sr ratio, but recrystallized matrix dolomite or saddlerised dolomites are mostly non-radiogenic (Figure 8). Results in both cases can be explained by recrystallization and cementation associated with two fluid types, a radiogenic juvenile meteoric source and a non-radiogenic basinal source. Fluid inclusion analysis was done on calcite cement and saddle and vug-lining dolomite. Homogenization temperatures for both ranged from warm to hot (Figure 9). It appears that saline and non-saline fluids were associated with calcite cement precipitation while the saddle and vuggy dolomites were precipitated from a moderately saline fluid.
Interpretation: Diagenetic ModelThe reservoir has been subjected to complex multiphase diagenesis (Figure 10). Most of the reservoir creating diagenesis is deep-burial related. After initial burial, chemically unstable fossils were dissolved and infilled with early calcite cement, which sometimes has a meteoric signature. Calcite cementation occludes much of the primary porosity. Burial dissolution and compaction created stylolites, which subsequently acted as barriers to fluid flow or were cut by later porosity. Most of the initial fracturing is post-compaction. Matrix dolomitization developed along the platform margin. Early matrix dolomite was likely created by precipitation from circulated seawater and Mg-rich water from dissolution of Mg-rich calcite skeletons. Early dolomite occurs in rock which had some permeability near the margin edge.This dolomite was partially recrystallized and fractured prior to saddle dolomitization. Saddle dolomite lines vugs and fractures. Dolomite recrystallization and later dolomite, including the saddle dolomite, appear to be hydrothermal related. Fluid inclusions in these dolomites have warm to hot temperatures of homogenization. Given that the timing of dolomitization was probably Early Cretaceous, the Abenaki 5 would have been buried to about 1000 meters and burial temperature would have been about 30 degrees C. The hotter fluid implies that the thermal cell extended several more kilometers into the underlying rock. That could mean that the Iroquois sabkha/carbonate ramp was involved; this would provide a convenient source for Mg and sulphate. Subsequently, aggressive dissolution of dolomites, and limestone to a lesser extent, created vuggy limestone and dolomite porosity. Dedolomitization occurred as well. An acidic sulphate-rich fluid will cause aggressive dolomite specific dissolution. This created large near-cavernous vuggy porosity in the linear trends behind the margin edge. It is unlikely that this dissolution was meteoric as there is little evidence of cave fill or spelothems and the late stage calcite cements do not carry a meteoric signature. Late blocky calcite was precipitated, partially to completely infilling some of the vugs and fractures. Fluid inclusions in the late-stage calcites also have hot temperatures of homogenization; this could have occurred later when the reservoir was more deeply buried. The diagenetic model explains the reservoir seen at Deep Panuke. The fracture and fault plumbing system is best developed along the margin edge due to dilation associated with partial collapse of the margin. The most productive part of the reservoir appears to be developed adjacent to the margin edge. The model created is not overly predictive as the hydrothermal fluids that control much of the late dissolution were delivered via a plumbing system that is difficult to image on seismic data and is fairly independent of lithofacies. |