Exploring hydrocarbon play potential in new areas of the Triassic in the North Sea.
Hydrocarbon Potential of the North Sea Triassic Play
Figure 1: North Sea Exploration Track Record 2008-16. Figure 2: Play based creaming curves. Image source: Azinor Catalyst.
Although the UK North Sea is becoming increasingly mature as an exploration province, the search for hydrocarbons continues, albeit with new focuses. One of these in recent years has been the Triassic play. Statistics show that the Triassic is one of a few under-explored plays that is able to deliver substantial resources with a high commercial chance of success (Figures 1 and 2). Westwood Global highlight that the Triassic commercial success rate is 40% with an average discovery size of 95 MMboe (Figure 1). When we look at the number of exploration wells in the last 8 years (Figure 1) and the creaming curves broken down by plays (Figure 2) we see that the Triassic is very under-explored. So although not a frequent target for exploration, Triassic prospects are generally low risk and high reward.
This article highlights how the use of new data and best practice techniques can help companies unearth opportunities and value. While new seismic helps define traps previously too deep to be well imaged, it is the integration of this data with a regional basinwide understanding of the Triassic geology which provides detailed insight into reservoir quality, one of the key risks in the play. The big picture informs the local detail!
Stratigraphy of the Triassic, North Sea
Figure 3: Paleogeography of the Mid Triassic (modified from Geol Soc Memoir No.13, 1992).
The stratigraphic subdivision of the Triassic across the North Sea region varies both by basin and across international boundaries (Levrik, 2006). In the UK Central Graben, where the Triassic stratigraphy is relatively well understood, the Triassic Heron Group is divided into the mud-prone Smith Bank and sand-prone Skagerrak Formations. Triassic deposition is interpreted to have occurred in a dryland fluvio-lacustrine setting (McKie and Audretsch, 2005; McKie 2014; Cope, 1992, Figure 3). The modern-day Lake Eyre Basin, Australia is often used as a useful analogue to guide our understanding of the geomorphological elements of the Triassic Central North Sea. With the availability of technologies such as Google’s satellite imagery it is easier now for geologists to make comparisons of the scale and depositional elements of modern-day analogues to the basins in which they are exploring.
The Lower Triassic Smith Bank Formation
The overall setting for the Lower Triassic Smith Bank Formation was one of deposition under relatively arid conditions as ephemeral terminal fluvial systems, discharging into desiccated and evaporitic playas as terminal splay systems. In contrast, facies associations in the Middle to Upper Triassic Skagerrak Formation suggest a dryland depositional system which was typically wetter than that of the Lower Triassic, allowing formation of well-developed river courses which sourced perennial lakes and marshes (McKie, 2014). Fluvial channel facies are typically fine–medium grained and characterised by a low clay content, while lake margin terminal splay facies are finer grained, more argillaceous and more micaceous (Akbokodje et al., 2017). These depositional textures retain a primary control on porosity evolution through burial. Consequently, well-sorted, clean fluvial sandstones of the Skagerrak Formation provide the most important Triassic reservoir targets in the North Sea.
The Triassic Skagerrak Formation
The Triassic stratigraphy of the South Viking Graben is poorly understood compared to the Central Graben, owing to fewer well penetrations in the deeper, axial parts of the basin. Along its eastern margin numerous wells have been drilled in the Norwegian sector and through detailed chemostratigraphic, sedimentological and petrographic analysis we are able to recognise that the preservation of the younger Skagerrak Formation reservoirs is relatively patchy here. In legacy wells, some of the sand-bearing intervals assigned to the Skagerrak Formation actually correlate to the older Bunter Sandstone member of the Smith Bank Formation. This implies that they are likely to consist of unconfined terminal splay and poorly developed fluvial channel sandstones as opposed to the better developed fluvial channel sandstones typically found in the Skagerrak Formation. This has important implications for the expected reservoir quality of these units, particularly with increasing depth of burial, and may in part explain the poor reservoir properties observed in at least some of the Triassic wells along the flanks of the South Viking Graben. Seismic mapping away from well control on the flanks of the basin shows the potential for preservation of younger Skagerrak Formation reservoirs into the axis of the South Viking Graben and gives encouragement that, despite the greater depth of burial in the basin axis, better reservoirs can be expected in these more basinal locations.
New Broadband Seismic Data in the North Sea
In addition to new insights on the depositional settings and facies types, new broadband seismic data provides much better imaging of the deep trapping mechanisms. Figure 4a shows the large untested Boaz tilted fault block prospect on vintage data. Well 16/8a-10, drilled in 1988, reached total depth in the overlying Middle Jurassic Sleipner but did not reach Triassic Skagerrak sands. In the adjacent blocks to the east lies the Eirin field in Norway which flowed gas condensate from what were interpreted at the time to be Skagerrak Formation sandstones but are now recognised to belong to the Bunter. In the vintage seismic it is difficult to interpret below the Sleipner coals, which is why the industry has historically struggled to gain confidence in defining the Boaz trap.
Figure 4a.) Vintage seismic.
Figure 4b.) Recent broadband seismic.
With new broadband seismic data (Figure 4b) there is far greater clarity at depth and we can now correlate new prospects to known discoveries – Boaz to Eirin and to surrounding wells. We now have the ability to not only map the top of the Triassic directly but also to define important intra-Triassic reflectors and deeper stratigraphy, such as the Permian.
Reservoir Permeability
The next challenge is reservoir quality. Big Data such as regional 3D seismic allows for a detailed understanding of the basin history and architecture. This is fundamental to making the correct analogues when comparing new basins with better known critical wells and fields within the same play.
Figure 5: Top Triassic two-way-time map showing the South Viking Graben’s proximity to the Sleipner Terrace and how it is more aligned structurally and in depth to the Central Graben.
The closest Triassic fields to the South Viking Graben are those on the Sleipner Terrace in Norway (Figure 5).
Slater et al., (2016) show that if you take Norwegian core data from the Sleipner Terrace (Figure 6, right), and project to depths of known structures (~4,500m) within the South Viking Graben, sandstone porosities will be degraded and provide only poor conduits to fluid flow. The dotted line represents the mean porosity, indicating porosities of below 10% at a depth of around 4,500m.
However, if you take the UK trend (Figure 6, left), predominantly driven by wells in the Central Graben, the indication is that porosities and therefore permeabilities will provide good quality reservoirs that can flow hydrocarbons. Porosities on this trend average around 15-20% for depths around 4,500m, and could be expected to have a reasonable permeability.
Figure 6: Porosity-depth trends, UK versus Norway.
Further to the primary facies control, reservoir quality is controlled by opposing sub-surface processes that can either degrade the reservoir, or act to preserve porosity.
To understand the rock properties, we need to understand the regional forces acting on them. Degradation is caused by two main controls: mechanical compaction (controlled by vertical effective stress) and chemical compaction (pressure solution and quartz cementation, primarily temperature controlled).
Preservation is assisted by three key controls inhibiting quartz cementation:
presence of clay grain coatings;
early migration of hydrocarbons;
and overpressure.
Understanding the rates and timing of all these controls is the key to forward predicting new areas of effective reservoir quality.
Isochore Mapping
Figure 7: (a) North Sea Isochore (Jurassic to Late Cenozoic) with strat column; and (b) corresponding Triassic overburden thicknesses for the South Viking Graben, Central Graben and the Sleipner Terrace for a similar time range.
Using a combination of isochore mapping and an understanding of periods of uplift and erosion, we can get a flavour of the burial history to see what forces are at play through time (Figure 7). Taking and comparing the South Viking Graben, Sleipner Terrace and Central Graben, we see that the first area undergoes an early period of rapid burial, creating over-pressured reservoir bodies and slightly earlier entry into the hydrocarbon generation window than the other two areas. Both of these physical processes will assist in preserving porosities and permeabilities. In addition to high sedimentation rates in post Triassic times, the low thermal gradient in the South Viking Graben along with regional gravity maps indicate that the graben has been a longstanding depocentre prior to Triassic. The net effect of this is that while the rocks may have been buried quickly they have been exposed to lower temperatures than other rocks at equivalent depths in the Central Graben for example. This has positive connotations for reservoir preservation.
Burial History
Figure 8. Burial history curves from the South Viking Graben, Central Graben and Sleipner Terrace.
Another way of showing this is by comparing the burial history models of the three areas. Figure 8 illustrates that the South Viking Graben is more analogous to the Central Graben area (having undergone periods of rapid burial) than the Sleipner Terrace area, and in the Central Graben reservoir quality is known to be good at a considerable depth.
The regional basin modelling and loading of the South Viking Graben rocks would mean that the physical processes acting upon this area would allow for preservation of adequate reservoir qualities, unlike those indicated from the trends identified from the Norwegian Sleipner Terrace.
The Sleipner Terrace area is an inappropriate dataset for guiding trends within the Viking Graben.
Quantitative Interpretation of Triassic Sediments
Figure 9: Tracks 1 and 2 (left) indicate that porous rocks (sands) are continually softer than the surrounding shale, using data taken from 14 wells spread across the North Sea. Crossplots (right) show clear trends between impedances and rock porosity and permeability.
Observations from the Triassic show that from a rock physics viewpoint it is actually very well behaved, and sands are almost always softer, i.e. lower acoustic impedance (AI) than the encasing shales.
This is seen in a number of basins around the world, including the North West Shelf, Australia, and is likely due to the sandstone structure maintaining a relative consistent framework, while the shales through time have reacted, dewatered and entered a later stage of equilibrium irrespective of depth.
This is illustrated in the left two tracks in Figure 9, highly porous sands are soft (low AI) while tight low porosity shales and silts are hard (high AI). When a formation behaves in this way it is incredibly useful, the seismic can be inverted to acoustic impedance, and from this we can get an indication of porosity and then permeability through rock physics transforms built around the data and relationships presented (right hand side plots Figure 9).
Rock physics models and transforms give us the ability to convert seismic reflection data and seismic velocities into petrophysical properties (facies, porosity and permeability), as shown in Figure 10.
Figure 10: From seismic to petrophysical properties.
Correlating Petroleum Plays in the North Sea
By driving for Big Data and detailed understanding, the industry can support pushing plays further than previously recognised conventional boundaries, extending them into new sub-basins. Big regional datasets and high-powered computing allows us to build large regional maps and isochores crossing several basins, letting us bridge basins and draw on more distal analogues. New data and integrated studies provide clarity and new insight into the subsurface.
Conclusion
From this article we can clearly see the benefits of using the data to extend the Triassic play from the more conventional Central Graben area north into the Viking Graben and possibly beyond. The Triassic section, in particular, shows very few anomalous rock properties and this can, and should, be taken advantage of in both exploration and production predictions.
