Constraining diagenetic timings, processes and reservoir quality in igneous-affected basins

Abstract

As the demand for hydrocarbons increases, more complex, non-conventional plays have been targeted in volcanic margin settings. Consequently, it is important to understand the role igneous rocks have in affecting hydrocarbon systems. This is particularly relevant to the Rosebank Field, offshore UK Continental Shelf, North Atlantic. The field was discovered in 2004 with the 213/27-1z well and encountered Palaeogene volcanic rocks interbedded with siliciclastic and volcaniclastic rocks, representing a brand new play type. Three appraisal wells were drilled and encountered good quality light oil (37° API) with drill steam tests achieving ~ 6000 STB per day (Duncan et al., 2009). Initially four companies had a stake in the field; Chevron (40%), Statoil (30%), OMV (20%) and Dong (10%) however in 2011, OMV bought out Statoil leaving them with a 50% stake. Chevron remain the operators.

The importance of volcaniclastic sediments within these sequences has previously been overlooked and so a better understanding of how these rocks behave during burial is essential. This research quantitatively characterises and assesses the reservoir potential of a range of volcaniclastic and siliciclastic sedimentary rocks within such basins.

The Rosebank Field comprises a range of volcaniclastic lithic arenites and lava flows interbedded with sublithic arenites and quartz arenites. However samples could only be taken from limited the limited cores. This made interpreting 3D architecture difficult and therefore it was decided that a onshore analogue was required. The Staffa Formation of the Palaeogene Mull Lava Field, NW Scotland, provides an excellent analogue to the Rosebank Field within the Faroe-Shetland Basin. It comprises basaltic lava flows interbedded with a variety of volcaniclastic and sedimentary rocks. A range of rock types occur within the Staffa Formation, including vent-proximal pyroclastic rocks, such as massive scoria rich tuffs to re-worked volcaniclastic lithic arenites to vent-distal facies where the siliciclastic component begins to dominate, such as quartz arenites. The Staffa Formation has therefore, been compared and contrasted to the Rosebank rocks to better understand composition, burial history and reservoir potential.

Detailed characterisation of the volcaniclastic rocks has been undertaken using a variety of analytical techniques (optical microscopy, SEM and XRD), and demonstrates that volcanic material in potential reservoir rocks may significantly reduce their reservoir quality. Proximal pyroclastic rocks, which can have reasonably good porosity and permeability at the surface, degrade to non-reservoir values at shallow depths (<1km), as reactive volcanic components alter to both grain-coating and pore-filling clays during diagenesis. This process dominates diagenesis meaning that it only requires small proportions of volcanic material to be incorporated within a sediment to destroy its porosity. In some cases, alteration is so severe that the original rock textures and clast outlines are lost, making the rock difficult to identify. Several generations of pore-filling and grain-coating clays, formed from the alteration of volcanic glass, including gel and fibrous palagonite, Fe-smectite and chlorite, have been identified in the volcaniclastic rocks samples from both the Rosebank core and the Staffa Formation.

The nature of the volcanic material plays an important role in a rock’s ability to retain reservoir quality. Factors such as the composition, shape and grain size of volcanic clasts in these rocks affect how the sediment behaved during diagenesis. Spatter bombs and scoria, for example, react differently clasts derived from lava. Labile volcanic ash shards that underwent minimal surface reworking altered to fibrous clay, and were then flattened and moulded around framework grains during diagenesis and burial. This reduced the local porosity and permeability in the rocks. Clay alteration of weathered basaltic clasts resulted in the development of pseudomorph grains that preserved the original grain structure and had only a minimal effect on surrounding pore throats. Clay formation in these rocks was extremely heterogeneous and highly dependent on the immediate grain-scale mineralogy. Clay phases developed during early diagenetic stages blocked or altered later fluid pathways, which led to an extremely patchy diagenetic mineralogy. Rocks inferred to be located at more distal locations from the vent have higher proportions of siliciclastic components and somewhat simpler paragenetic sequences that are characterised by carbonate and silicic cements and minimal authigenic clays. Oxygen and hydrogen isotopic data provide constraints on pore water chemistry and temperatures during diagenesis.

Together, these data have been used to determine the diagenetic histories of the Rosebank Field and Staffa Formation, and enabled the development of a conceptual model to determine the threshold at which volcaniclastic rocks are no longer viable as petroleum reservoir rocks. The model shows that volcaniclastic rocks containing more than 10% volcanic clasts are likely to have very poor reservoir quality at depth. However, this is dependent on a number of factors such as clast size, clast type, depositional setting, sorting, pore water composition and timing. These data will be of use in the assessment of potential plays in volcanic rifted margins worldwide.