|JADE FIELD: LATE-LIFE DEVELOPMENT CHALLENGES IN A TRIASSIC HPHT GAS CONDENSATE RESERVOIR|
|Field Developments and Case Studies|
1BG Group Plc
The Jade Field is an example of a high pressure high temperature heterogeneous fluvial reservoir which, at the time of development, appeared to be a straight-forward case of production by natural depletion. However, infill well results into the Joanne Sandstone have revealed complexities in reservoir architecture and production behaviour. This extended abstract highlights some of the uncertainties and technical challenges faced in characterising this HPHT field where learnings can be developed to increase understanding of these reservoirs for improving recovery.
|THE JADE FIELD|
The Jade Field is a high pressure/high temperature (HPHT) gas/condensate field located in the East Central Graben of the North Sea in UKCS Licence Block 30/2c, 270km east of Aberdeen (Figure 1). The field is operated by ConocoPhillips (32.5%) and the remaining ownership consists of BG (35%), Chevron (19.93%), Eni (7%) and OMV (5.57%). The field was discovered in 1996 by well 30/2c-3 and appraised in 1997 by well 30/2c-4. First production was achieved in 2002 through a normally unmanned production wellhead platform connected to the Judy processing facilities in Block 30/7a by a 17km pipeline (Jones et al, 2005).
This paper focuses on the development of the Joanne Sandstone reservoir in the Jade field.
At initial conditions, reservoir pressure was approximately 12,150 psia and temperature 331°F. No definitive fluid contacts had been established by exploration, appraisal or development wells in the Jade structure and there remains considerable uncertainty relating to fluid properties and distribution. The fluid samples from the 30/2c-4 DST intervals show a gas condensate with contrasting PVT properties and a CGR ranging from 66-192 stb/MMscf.
The conventional view is that the difference between the two Joanne sandstone condensates is explained by vertical reservoir compartmentalisation rather than gravity segregation. Due to the reservoir being produced in a co-mingled manner, it has not been possible to use production data to differentiate between different fluid regions. However, an alternative analysis presented in Figure 4 could support a view that a gravity segregated fluid is physically possible within the conditions seen in Jade.
|INITIAL DEVELOPMENT WELLS AND PRODUCTION BEHAVIOUR|
The initial development consisted of 4 sub-vertical production wells targeting the Joanne Sandstone (Figure 3), followed by 5 infill wells drilled between 2004-9 targetting both Judy and Joanne Sandstones. The field produced on plateau at 230 MMscf/d gas and 22 Mstb/d condensate until 2006. Current rates are 150 MMscf/d gas and 11 Mstb/d condensate. The initial development wells each encountered a succession of good quality stacked fluvial channel sands, silt and mud with net:gross in the range of 0.6–0.8 (Figure 8). The wells were drilled progressively down-dip with the intention of identifying the GWC, but even the deepest of these wells (J04Z) did not encounter a water leg (Figure 3). No formation water is being produced by the initial development wells.
Despite the stratigraphic heterogeneity observed in the well logs which is characteristic of the Skagerrak Formation, the initial 4 production wells demonstrate a very “tank-like” behaviour, with similar build up and common pressure profiles. The rate/cum behaviour of one of these wells is illustrated in Figure 5 and shows an approximate straight line behaviour. Further analysis using a Havelena Odeh F/Eg technique shows that initially the wells see a limited volume. This view is consistent with the “connected tank” model (Shibli et al, 2008), but over the longer term, the main part of the Jade field behaves like a large tank. It is considered unlikely that this analysis shows significant water influx into the main block.
|INFILL DRILLING RESULTS|
The results from infill wells have revealed additional complexity in reservoir and fluid distribution, faulting, compartmentalisation and depletion characteristics of the reservoir.
30/2c-J08 was drilled in 2004 and is the first development well on the eastern side of the structure. The well encountered a gas/condensate column with GDT. Although the net:gross is lower relative to the initial development wells, there is a reasonable succession of stacked channel sands, some of which were up to 900 psi “depleted” (Figure 8). This differential pressure appears to be limited to individual sandstone intervals, and gives a clear indication that relatively thin shaley sections are capable of providing effective barriers to vertical connectivity.
This notion of limited Kv is supported by evidence from additional shallow perforations added to J08 in 2007, which show that after 3 years of production from J08 itself and 5 years from the initial development wells, some sands remained at almost virgin pressure.
Analysis of J08 production (Figure 6) shows that the initial lower perforations were in connection with a limited volume of around 30 Bscf. This contrasts significantly with the volumes observed for the western segment wells (Figure 5). After the J08 additional perforations in 2007, the well has seen an increase in drained volume, but despite the original pressure data showing that the upper sands were in communication with the original wells, the total drained volume for this well has only increased to around 50 Bscf.
It is also interesting to note that this well shows a strong build-up in pressure during shut-ins, which is not seen in the Jade western segment wells. This suggests that the degree of stratigraphic baffling is much more significant in the J08 area than in the western segment.
These data together suggest that J08 accesses a smaller connected volume compared to the initial development wells, and could indicate the eastern segment of the Jade field is either structurally and/or stratigraphically more complex than the western segment. On this basis, it was concluded that J08 would probably not efficiently recover the volume from the entire eastern segment, and that an additional well would be required down-dip.
30/2c-J11Z was drilled in early 2009 approximately 900m SSE of 30/2c-J08 targeting a potentially unconnected down-dip volume. The well encountered a thinner Joanne sandstone succession with much lower net:gross and fewer stacked amalgamated channel sands. Wireline pressure measurements showed high levels of differential depletion in some sandstones within the payzone of up to 6,700 psi.
J11Z was expected to encounter a hydrocarbon column to base reservoir in common with the western segment. However, a GDT
was intercepted 1,300 ft tvd shallower than the deepest lowest-known-hydrocarbon for the field established by J04Z. Furthermore, individual sands within the water zone were depleted by up to 2,000 psi.
The infill drilling results in Jade have revealed reservoir connectivity and fluid distribution to be far more complex than the initial development well behaviour might have implied. The surprise of a WUT so much shallower than previous wells across a 450m offset led to deliberation about the mechanism responsible for this outcome.
J11Z is stratigraphically equivalent to J08 although structurally down-dip (Figure 7). It is also depositionally down-dip in the direction of palaeo-channel flow. However, the deeper section in J11Z is much poorer quality, which could indicate distality to the main channel complex at this location during the early-stages of Joanne deposition. Given this geometrical arrangement, the off-take point at J08 could have produced both hydrocarbons and water along the channel facies from the J11Z area (Figure 7).
The levels of depletion clearly indicate some communication between J11Z and the existing well stock. The J04Z well is located 450m SW of J11Z, and is one of the initial development wells. In common with the other western sector wells, the location around J04Z is now likely to be depleted to a SIBHP of 3,500 psia. Given the difference in pressure between J04Z and J11Z and the fact J04Z is not producing connate water, it can be concluded that there is an effective barrier to flow between the two wells. Similarly, no connate water is yet observed in the J03 well which is located 700m WNW of J11Z.
However, analysis of the salinities in well-head water samples taken from J08 indicates there may be some connate water production in this well. With the benefit of hindsight, it is possible to suggest the calculated water saturation (Sw) for J08 shows an increasing trend with depth, although this could also be attributed to an increasing shale content and poorer reservoir quality in the lower stratigraphy. If the GDT encountered in J11Z is common to the eastern segment, the J08 well reaches TD in the water-leg, although this is not logged or perforated (Figure 8). Furthermore, whilst not to be considered in isolation, the wetness (Wh) and balance ratios (Bh) (Haworth et al, 1985) for J08 together with total gas, give an indication for the presence of water at the toe of J08 (Figure 8).
There is a relationship between J08 and J11 in terms of production characteristics and interactivity, which contrasts with the performance and behaviour of the initial development wells. On this basis, it can be argued that Jade can be split into two distinct segments. However, the boundary between the two segments is not easily resolvable.
Jade sits below the base-Cretaceous unconformity, and as a result the seismic data is poor and riddled with multiple energy. As a consequence, fault mapping is highly interpretive, and there is a high degree of variation between generations of mapping. There is no obvious large-scale mappable fault between the east and west segments, and an obvious large throw is not observed on seismic data.
There are several possible scenarios which could explain the lack of connectivity between J04Z and J11Z and the differential depths to fluid contacts:
(i) Large fault with shale gouge. This is possible, but it is difficult to map a fault with throws large enough to generate a clay smear strong enough to a hold pressure differential of almost 8,000 psi;
(ii) Fault with smaller throw which is mineralised or cemented. This is reasonably likely given the depth of burial of Jade.
(iii) Stratigraphic compartmentalisation combined with fault offset. It is possible that, given the relatively thin nature of the sandstones (maximum 40 ft) even relatively small fault throws would be sufficient to offset these sands and juxtapose them against poorer quality rock, creating a barrier to flow.
There are merits to each of these explanations, and there is potential for the behaviour of the reservoir being attributed to a combination of all. Given that some sands in J08 and J11Z are depleted more significantly relative to others, this may give more support to the concept of juxtaposition. However, there is little correlation between the thickness of the sand and the level of depletion. This could indicate there is an element of “chance” to any one sand connecting to another sand across a fault with a limited throw and enabling pressure communication between the two. The likelihood of this scenario is higher given the heterogeneous and inter-bedded nature of the succession.
Explaining the significantly different contacts between the two sectors is more difficult. It could be due to the initial charging of the Jade structure. There are, again, a number of possibilities:
(i) The fault which separates west Jade from east Jade was fully sealing at the time of hydrocarbon emplacement.
(ii) East Jade and west Jade have different charge histories;
(iii) East Jade may have been filled directly through the eastern bounding fault of the horst structure, whereas west Jade may have been filled from the crest down to spill.
Infill drilling on the HPHT Jade Field has revealed significant and perplexing complexities in reservoir connectivity and fluid distribution. The initial development well behaviour gave indications that Jade behaved as a “tank”, and that the whole structure was in communication. However, the infill wells, particularly J11Z, have progressively revealed a bi-modal nature to the field and it’s depletion which proves difficult to rationalise.
The contrast in production performance, poor connectivity and free-water-level between the two Jade segments together with the surprising lateral variation in net:gross demonstrate the significant challenges subsurface teams face when attempting to accurately characterise these reservoirs.
Jade is a good example of how fields can present significant surprises, even when considered to be reasonably well understood. However, these surprises enhance understanding of these reservoirs and the learnings can be incorporated into development planning to improve recovery going forward.
The authors would like to acknowledge the Operator, ConocoPhillips Petroleum Company U.K. Limited, and co-venturers Chevron North Sea Limited, Eni UK Limited and OMV (U.K.) Limited for their permission and support in delivering this presentation.
Goldsmith, P.J., Rich, B. & Standring, J (1995). Triassic Correlation and Stratigraphy in the South Central Graben, UK North Sea. In: Boldy, S.A.R., (ed) Permian and Triassic Rifting in Northwest Europe. Geological Society of London, Special Publications, 91, 123-143
Goldsmith, P.J., Hudson, G. & van Veen, P. (2003). Triassic. In: Evans, D., Graham, C., Armour, A. & Bathurst, P. (eds). The Millennium Atlas: Petroleum Geology of the Central and Northern North Sea. Geological Society, London, 105-127
Haworth, J.H., Sellens, M. & Whittaker, A (1985). Interpretation of Hydrocarbon Shows Using Light (C1-C5) Hydrocarbon Gases from Mud-Log Data. AAPG Bulletin, 69, pp1305-1310.
Jones, A.D., Auld, H.A., Carpenter, T.J., Fetkovich, E., Palmer, I.A., Rigatos, E.N. and Thompson, M.W. (2005). Jade Field – An Innovative Approach to High Pressure/High Temperature Field Development. In: Dore, A.G & Vining, B.A. (eds) Petroleum Geology: North-West Europe and Global Perspectives – Proceedings of the 6th Petroleum Geology Conference, 269-283.
Pedersen, K.S. & Lindeloff, N. (2003). Simulations of Compositional Gradients in Hydrocarbon Reservoirs under the influence of a temperature gradient. SPE 84364, SPE ATCE, Denver, CO.
Shibli, S., Ferguson, A., Bruce, A., MacMillan, A., & Blackburn, J. (2008). Reservoir Management of multi-layered HPHT reservoirs – the Jade Field experience. Presented at SPE workshop on Multi-layered reservoirs, London, February 2008.