--> EXTENDED ABSTRACT: Charge and Leakage Analysis Integrating Different Scales: From Fluid Inclusions to Seismic Attributes. Loppa High, Barents Sea, Norway, by Pestman, Pieter; Tur, Noemí; Esteban, Mateu; Polo, Teresa; Sánchez, Alfredo; Tiwary, Devendra; Tocco, Rafael; Tritlla, Jordi; Vayssaire, André; #90135 (2011)

Datapages, Inc.Print this page

Charge and Leakage Analysis Integrating Different Scales: From Fluid Inclusions to Seismic Attributes. Loppa High, Barents Sea, Norway

Pestman, Pieter 1; Tur, Noemí 1; Esteban, Mateu 2; Polo, Teresa 2; Sánchez, Alfredo 2; Tiwary, Devendra 2; Tocco, Rafael 2; Tritlla, Jordi 2; Vayssaire, André 2
(1)Repsol Exploration Norge, Oslo, Norway. (2) Repsol Exploración, Madrid, Spain.

 

Introduction

The Loppa High (Norwegian Barents Sea) is a large (±150km long, ±100km wide) tilted fault block located north of the Hammerfest Basin in the western Barents Sea (Figures 1, 2). It consists of an eastern platform and a crestal western and northwestern margin. The crestal area is covered by a 3D seismic survey, SG9810, acquired in 1998 covering an area of 63.6 x 22.2 km. The study presented here focuses on the crestal area. To the west, the Loppa High is bounded by an important fault complex, called the Ringvassøy-Loppa Fault Complex or Polhem Fault.

The western crest of the Loppa High has been rejuvenated as a high at least four times since the Devonian, but the high, as defined now, is a result of Late Jurassic to Early Cretaceous and Late Creta-ceous-Tertiary tectonism. From Ladinian to Callovian times, the high was part of a regional cratonic platform. During much of the Cretaceous and Tertiary, the high was exposed subaerially. The Loppa High has no present-day topographical expression. It is, strictly speaking, a paleo-high.

The Loppa High contains Paleozoic sediments, and is covered unconformably by Triassic shale-dominated deposits. The Paleozoic includes reservoirs in Carboniferous to Permian carbonates of the Ørn Formation of the Gipsdalen Group (Figures 2, 3).

A well, 7220/6-1, was drilled in 2005 on the Obelix prospect close to the apex of this paleohigh (Figure 1). It encountered the Paleozoic uncon-formity, underlain by carbonates of the Ørn Formation, at 1138m. Basal Gipsdalen clastics (Falk Formation) were reached at 1436m, and fractured quartzitic basement at 1483m (Figures 3, 8). The well reached a total depth of 1540m. No economical amounts of hydrocarbons were encountered, but residual oil and good oil shows were obtained in carbonates of the Gipsdalen Group. Traces of oil (10 ml) were noted in an MDT water sample at 1184.5 m.

Similarly, three wells drilled in the southern portion of the paleohigh (Figure 1) only encoun-tered shows in the Paleozoic, no hydrocarbon accumulations. This raised the question whether the negative drilling results would condemn the whole structure. Consequently, the possible causes of these negative results were investigated.

The negative result of well 7220/6-1 might be ascribed to one or several of the following factors: (1) no reservoir, (2) no valid trap, (3) leakage, or (4) insufficient charge. In order to ascertain which of these explanations is correct, a study was carried out in which the hydrocarbon system was analyzed at a variety of scales: ranging from fluid inclusions, through cores and wireline logs, to seismic scale.

1. Reservoir

Sedimentology

The Paleozoic section penetrated by well 7220/6-1 consists mostly of carbonates of the Ørn Formation of the Gipsdalen Group (1138-1436m), underlain by shaly deposits of the Falk Formation (1436-1483m). This paper focuses on the Gipsdalen carbonates.

In well 7220/6-1, the Gipsdalen Group consists mostly of dolomitized micritic limestones. The lower portion of the section is horizontally bedded; no build-ups are apparent. Toward the top, the bedding becomes increasingly disrupted: cm to dm sized angular to subrounded carbonate blocks dominate the upper part of the interval. This is where cores have been cut (Figure 4). The carbonate blocks have been identified as “conglo-breccia”: a combination of karst breccia and clasts having undergone some transport by water.

Available data suggest an advanced senile karst system, i.e. a system where karst processes have occurred for so long that the rate of porosity destruction (gravitational collapse, cementation, sedimen-tation, dismantling) exceeds the rate of porosity creation (corrosion, corrasion). This senile karst system affects the upper 80-120 m of the truncated Gipsdalen Group, with major development of breccias and conglobreccias of many different types and possible gravitational collapse. The karstification of top Gipsdalen can be recognized on seismic by a chaotic seismic facies (Figure 9).

While the senile karst may not have resulted in significant net porosity generation, vuggy porosity does occur in the well (Figure 4). Such porosity may also have been formed as a result of late burial diagenesis. Early diagenetic processes, such as the widespread dolomitization, also created porosity, albeit mostly intercrystalline. Overall, the Gipsdalen carbonates in the study well may have reasonable porosity, although permeabilities tend to be low.

Well 7220/6-1 is not representative of the Loppa High in general. This is shown by well 7120/2-1, drilled in the southern portion of the high, which encountered carbonate build-up facies and no conglobreccias. Furthermore, seismic data reveal a multitude of features associated to karst and to carbonate build-ups. Seismic attributes such as Spectral Decomposition, Semblance and Chaos are useful to highlight these features (Figures 5, 10). Carbonate build-ups or mounds are not isolated elements, but rather interconnected structures forming a polygonal network with internal depressions, which are well imaged by the seismic (Figure 6).

Consequently, it is also likely that the reservoir quality encountered in well 7220/6-1 is below average of what can be expected for the Loppa High as a whole. The best reservoir quality is expected not in the crestal part of the Loppa High, where this well was drilled, but more down-flank, where more build-ups are present and karst processes were less extreme. The eastward extension of the karstification is delimited by the oldest coastal onlap of the Tempelfjorden units (Figures 7, 9). In essence, the preservation of meteoric karst porosity results from the balance between generated and destroyed porosity. The best chances for this preservation occurs low in the flank and it is favoured by a rapid rate of transgression of the overlying unit.

The balance of porosity generation-destruction in the top Gipsdalen karst is also controlled by the type of lithologies affected by the karst system (Figure 9). The best porosity development would be expected in clean carbonates without marly intercalations, particularly mounds and skeletal sand shoals.

Petrophysics

The dominant lithotype in the Paleozoic section of well 7220/6-1 is dolomite with limestone as secondary mineral. Limestone and dolomite have similar porosities, typically around 11-12% (Figure 8), so porosity is independent of lithology.

The average water saturation in the interval from 1150 to 1450m is 69% (Figure 8). A net pay of 14m is distributed throughout the Gipsdalen group as thin laminar zones. This type of situation is commonly encountered when the seal is breached and hydrocarbon leakage occurred. The upper part of the Gipsdalen carbonate reservoir has a higher porosity than the lower part, but lower hydrocarbon saturation. This also suggests that leakage took place.

Core was cut from the very top of the Gipsdalen carbonates, with a length of 55m. Core analysis data show an average porosity of 13.3% and permeability of 1 mD (logarithmic average), with maximum values of 26.5% and 200 mD, respectively. However, the core analysis data are by no means representative of the reservoir: the cored interval corresponds to surface or near-surface infiltration zone of the profile. This is the worst part of the karst profile and even so, there is indication of incipient porosity development. Furthermore, the best reservoir properties in the cored interval correspond to the core rubble zone or there is no core recovery; no core analysis data could be obtained from these good-quality intervals, due to the impossibility to drill core plugs.

It is concluded that well 7220/6-1 is not located in a zone of buildups. The karst processes apparent in the cores and image logs point to low-porosity, senile karst. However, core and log analysis indicate a moderate porosity with, in places, fair permeabilities. Therefore, a poor reservoir can be discarded as the cause of the negative result of well 7220/6-1.

2. Trapping

The erosional truncation of the Gipsdalen Group under the mid-Permian unconformity and against the basement provides an erosional pinchout trap (Figure 9). Structural trapping against the Polhem Fault west of the Loppa High is not likely, since that fault extends upwards almost to the sea floor (Figure 2), and has been reactivated several times during Phanerozoic times. As a consequence, the sealing capacity of the fault is likely to be weak.

Although trapping of hydrocarbons on the Loppa High would be largely stratigraphic, a structural component will be required for the lateral seal. A set of east-west trending faults is cross-cutting the Loppa High, and could in principle act as lateral seals. However, like the Polhem Fault, these cross-cutting faults extend up to base Quaternary, which suggests reactivation and hence, a weak sealing capacity (Figure 11).

Seismic attributes

Seismic attributes were used not only for facies definition, but also in order to better imagine structural features. Attributes from which good results were obtained for structural definition include structural attributes (Dip, Azimuth, DipAziCombine, Chaos, Conformance) and edge attributes (Semblance).

Semblance is a seismic discontinuity volume attribute. It quantifies the degree to which neighbouring seismic traces vary from each other. This attribute was the one that gave the best results in order to support the structural interpretation (Figures 10, 11).

Stratigraphic trapping: risks

It is concluded that trapping of hydrocarbons in the Gipsdalen carbonates on the Loppa High must have been mostly stratigraphic, with east-west trending faults acting as – dubious – lateral seals. Triassic shales provide a good vertical seal, but the quality of the Bjarmeland carbonates that overlie the Gipsdalen carbonates east of the crest is uncertain. Likewise, the quality of the bottom seal is uncertain: where the clastics of the Falk Formation are thick and/or argillaceous, the bottom seal may be good. Where this is not the case, however, the Gipsdalen carbonates would be in contact with the underlying fractured basement, which is in contact with the Polhem fault zone. Obviously, there may not be a valid trap in the area where the study well was drilled, and that could be a key factor for the negative result of the well.

3. Leakage

Closely related to the trapping, is the issue of retention once hydrocarbons were trapped. The repeated reactivation of the fault system bordering the Loppa High to the west, and with it the fault systems cross-cutting the high (Figure 11), make it likely that hydrocarbons that reached the crest of the Loppa High, did not stay trapped.

Seismic interpretation indicated the presence of leakage, aided by faults. Shallow bodies of high amplitude may represent locations of the migrated hydrocarbons (Figure 12, right). The presence of "paleopockmarks" on the top Triassic unconformity indicates gas leakage before the Quaternary (Figure 12, left). Log analysis suggests that the oil present in the well is dead oil; this, together with the presence of a possible paleo-oil-water contact at approx. 1380m (Figure 8), points to initial trapping of oil and subsequent leakage due to fault reactivation. Consequently, the negative result of the study well may be due more to leakage after initial trapping, than to the absence of a trap in the first place.

4. Charge

Geochemistry

Several oil samples from well 7220/6-1 were available for this study: some oil from the MDT water sample taken at 1184.5m, and three sidewall cores (SWCs) from the Gipsdalen Group with oil impregnation. For well 7120/2-1, the results of a DST in the Falk Formation of the Gipsdalen Group (water with traces of oil) are available. The geochemical data mentioned for these two wells were retrieved from the FactPages of the Norwegian Petroleum Directorate for well 7220/6-1.

Biomarkers (m/z 191, m/z 217, m/z 218) and isotope values indicate a good genetic correlation between all samples. The biomarkers indicate that the oils were likely generated by the same, early-middle mature source rock with important contribution of marine organic matter. The isotopes, which are very light (d13C for saturate hydrocarbons varies from -31.3 to -31.4‰, and for aromatic hydrocarbons from -30.6 to -31.0‰), fall in the range of what Ohm et al. (2008) consider to be Permian/Carboniferous oils. SINTEF (2005), however, considers that biomarkers and isotope values of the Loppa High samples may be ascribed to Triassic source rocks.

Triassic samples are reported to have extended tricyclic terpanes ratios (ETRs) mostly greater than 2, as opposed to younger source rocks. The oil sample (DST-4) from well 7120/2-1 showed an ETR of 2.1, thus suggesting its source to be pre-Triassic considering the low ETR of the local Triassic organic extract and the high ETR of the local Permian and Carboniferous source-rock organic extracts.

It is concluded that the source rock for the Loppa High oil samples was either Paleozoic or Triassic (or both source rocks may have contributed). A younger source rock, such as the Upper Jurassic Hekkingen Formation, is ruled out. The most obvious candidate for the Paleozoic source rock is the Permian Ørret Formation, while the Kobbe Formation (or rather, its high-TOC marine equivalent, the Steinkobbe Formation) could be the Triassic source rock. The maturity these source rocks reached in the Loppa High area is discussed below.

Petroleum system

In order to gain an understanding of the timing of hydrocarbon migration, the type of hydrocarbon, and the most likely source rock, a basin modeling study was carried out. A 2D model was prepared (Figures 13, 14), along the seismic line shown in Figure 2.

We considered as source rocks the Steinkobbe/Kobbe (Middle Triassic) and Ørret (Late Permian) formations. The Steinkobbe kerogen was modeled as rich and oil-prone. The maximum burial depth was reached before the Eocene erosion 40 Ma ago. At present day, the Ørret source rock stands in the oil window except for the area on the western Loppa High which is immature, and the deepest part which reached the gas window. Kobbe is just at the limit of the mid-oil window, which could limit its effectiveness in expulsing hydrocarbons in great quantities (Figure 13).

The principal risk for the Loppa High is the efficiency of a carrier bed able to drain great quantities of oil produced tens of kilometers away from the structure. The seal efficiency may also be an issue (Figure 14). For the Kobbe source rock, which is not in direct contact with the Paleozoic strata of the Loppa High (Figures 2, 14), the presence of an efficient carrier bed is thought to be unlikely. Furthermore, its low maturity level in the vicinity of the structure may prevent it from playing an important role. The charge of the Loppa High is considered to have been derived virtually completely from the Ørret source rock, even though the hydrocarbons generated by the Ørret source rock must have migrated through the low-permeability Tempelfjorden heterolithics and Bjarmeland carbonates before reaching the Gipsdalen reservoir (Figure 14).

2D migration simulation shows that a commercial accumulation in the Gipsdalen carbonates of the Loppa High is possible if the following conditions are met: (1) Ørret Formation as source rock; (2) a very good seal that resists strong pressure changes; (3) a good and continuous carrier bed below the Ørret source rock. Even so, important migration losses are likely.

It is concluded that the Ørret source rock may have charged the Loppa High with first oil and later gas, but that the amount of hydrocarbons that reached the structure may have been limited due to the absence of a good carrier bed. Most oil migration from the Ørret source rock occurred prior to the regional uplift around 40 Ma ago.

Fluid inclusions

Fluid inclusions are small fluid aliquots (brine+gas+hydrocarbons) trapped during crystal growth that preserve information on fluid composition, temperature and pressure at the time of cement formation (Figure 15). For the study of the Loppa High, rock samples were taken from the cores of the 7220/6-1 and 7120/2-1 wells. These fluid inclusions were studied using microthermometric techniques and PVTx modeling.

Microthermometric freezing runs allowed the identification of brine electrolytes and the calculation of total salinity. In well 7220/6-1, fluid salinity was controlled by NaCl; in well 7120/2-1, CaCl2; is also present. Total calculated salinities were:
          Well 7220/6-1: 9.3-20.3 wt% eq. NaCl
          Well 7120/2-1: 20.3-25.8 wt% eq. NaCl
Salinities increase with depth.

Homogenization temperatures (Th) give an approximation of the temperature at which the fluid inclusions were trapped. Brine-bearing fluid inclusions present Th’s between 80 and 105ºC. Hydrocarbon-bearing fluid inclusions present lower Th values, between 58 and 93ºC.

Gas to liquid ratios (Fv) at 20ºC are a intrinsic characteristic of hydrocarbon compositions. Fv was calculated using the CLSM technique. Fv versus Th data plotted using a special abacus (Bourdet et al., 2008) allowed to classify the oil types, ranging from black oil to light oil/condensate. Since the lowest Fv values are considered not representative of the conditions during fluid inclusion formation (because of leakage detection), the oil at that time was likely light. This leakage can be linked with a general decompression, following uplift.

Due to the absence of coeval brine-bearing and hydrocarbon-bearing fluid inclusions, full PVTx paleofluid modelling was not possible. Then, minimum PT trapping conditions were calculated on two samples from each well:

Well Sample Trapping temp. (°C) Trapping pressure (bar)
7220/6-1 1168.75m 90-105 230-300
1191.2m 80-100 230-350
7120/2-1 2074m 100-110 240-350
2154m 100-110 270-400

These values are in good agreement with the 2D basin model calculations (Figure 16), although the modelled pressures are slightly lower. Thus, it becomes apparent that the formation of the hydrocarbon-bearing fluid inclusions took place at the beginning of the Cretaceous and/or in the Paleogene. This means that hydrocarbons were present in the Loppa High before the Tertiary uplift. The reactivation of the fault systems in the area as a consequence of the Tertiary uplift may have caused leakage of much, if not all, of the trapped hydrocarbons.

Geochemical and basin analysis indicate that the oil in well 7220/6-1 has a Paleozoic source, possibly with a Triassic admixture, and that most of the oil migration occurred before the Tertiary uplift. This is confirmed by PVT analysis of fluid inclusions. However, migration pathways from these sources appear to be somewhat difficult, and insufficient charge may have contributed to the absence of an accumulation in well 7220/6-1.

Conclusions

It is concluded that leakage probably is the main cause of the negative result of well 7220/6-1.

Hydrocarbons are present, albeit at low saturations, and the presence of porous carbonates is likely. However, charge came mostly from the Ørret source rock. This had several consequences with a negative impact on the results of well 7220/6-1:

  • Hydrocarbon volumes that reached the Loppa High may have been small due to difficult migration route.
  • Oil reached the Loppa High before the Cretaceous, and has likely been leaked due to repeated fault reactivation.
  • From the Cretaceous onward, gas was generated, rather than oil.

Acknowledgements

The authors thank their colleagues of Det norske oljeselskap in Harstad, especially Geir Elvebakk and Kai Hogstad, for introducing them to the Loppa High and for many useful discussions.

References

A listing of the references cited in this paper can be obtained from the corresponding author ([email protected]).

Figure 1. Location map of Loppa High showing locations of wells that tested the Paleozoic, and area of 3D seismic survey. Base map based on data from NPD FactMaps.

Figure 2. Seismic cross-section through well 7220/6-1 on the Loppa High. Courtesy Fugro.

Figure 3. Paleozoic stratigraphy of Loppa High and adjacent Bjarmeland Platform. After Larssen et al., 2005.

Figure 4. Conglobreccia in well 7220/6-1. Left, middle: core photographs. Right: image log (dynamic normalization).

Figure 5. Example of the use of spectral decomposition in order to obtain a better image of karst features.

Figure 6. Carbonate build-ups at top Gipsdalen, displaying characteristic polygonal pattern.

Figure 7. Model for karst development on the Loppa High.

Figure 8. Wireline logs and petrophysical data (porosity and data saturation) for well 7220/6-1.

Figure 9. Gibsdalen carbonates: seals, and porosity enhancement due to karst and late diagenesis.

Figure 10. Time slices illustrating how Semblance attribute can enhance structural (left) and stratigraphic (right) features.

Figure 11. Left: integrated display using semblance to highlight how faults extend upward to base Quaternary. Right: top Gipsdalen time structure map, showing fault system crosscutting the Loppa High.

Figure 12. Left: paleopockmarks on top Triassic (base Quaternary). Right: time slice at Triassic level (semblance+amplitude) showing fault-related high amplitudes.

Figure 13. Present-day kerogen transformation ratios for the two source rocks.

Figure 14. Top: modelled layers. Bottom: Hydrocarbon saturation at present day, and migration pathways.

Figure 15. Examples of fluid inclusions, with different contents.

Figure 16. Combination of P and T histories of well 7220/6-1 with fluid-inclusion data.

 

AAPG Search and Discovery Article #90135©2011 AAPG International Conference and Exhibition, Milan, Italy, 23-26 October 2011.