Rosneft: Reservoir Interpretation Increases Drilling Efficiency
R.I. Abdrakhimov, R.R. Galiyev, D.D. Suleymanov
T.S. Usmanov, Chem.Dr. (LLC “RN-UfaNIPIneft”)
I.S. Afanasyev, Physics-Math. Dr. (OJSC “NC “Rosneft”)
Introduction
The Neocomian section in Western Siberia was formed through the gradual lateral infilling of debris into a relatively deep sea basin. For most of the territory, the source of the debris was the eastern margins of the plate. The infilling of the sedimentary basin happened against the background of continuous sinking in the region. All of this gave grounds for a clinoform construction of section [1].
The Priobskaya oil field, which is over 6,000 km2 in size, is a unique object to research geological structure of clinoforms.
The principal part of the field is being developed by three operators, namely, OJSC “Gasproneft-Khantos”, LLC “RN-Yuganskneftegaz”, and NJSC “Aki-Otyr”. Hydrocarbon reserves are concentrated in the АС group formations (АС7-АС12).. The principal areas for development are the АС10-АС12 formations, where the depositions change in lithology and oil-water contact is absent. Paleogeographic conditions of the rock formation, where the depositions are found, alternate westward from the littoral shelf and slope to deepwater [2].
Changes in the reservoir properties depend on the distance of various facies from the edge of paleoshelf. In the eastern part, the reservoir features relatively high permeability ((6-10).10-3 µm2), good adhesion and is a continuous area (shelf type). The western part is weakly bound, has low-permeability ((1-3).10-3 µm2) and has highly variable laterals (deep-water type).
Given the absence of free water in the depositions, the main objective of the geological support for the development of the deposit is forecasting the reservoir, both as an area and sectionally. And whereas the paleoshelf depositions present no difficulty due to the continuity of the oil-bearing intervals, the situation with the deep-water formations is completely different. The productive part of the section is concentrated in a small, poorly connected, lithologically screened lenses which at first glance appear to be “sporadically scattered” across the area. Due to this, the drilling of new wells is risky because significant saturated reservoir stratum has not been confirmed. During early drilling, the risks were compensated for by natural factors: above the deep-water depositions of the АС12 formation the continuous formations of АС10 and АС11 were always present. At the location of new wells, production intervals entirely correspond to the АС12 deepwater formation, and therefore required detailed analysis of all available data for reliable forecasting.
One such area is Gorshkovskaya (fig. 1), located in the northern and mostly undrilled part of the Priobskoye deposit, where over 1 bln. t of prospective oil reserves are concentrated and the drilling of over 2000 wells is planned. The territory occupies an area of 700 km2 and has had 48 exploration wells drilled, an entire complex of geophysical well surveys (GWS) has been performed. 9 core samples are available for lithological description, and high quality 3D seismic was shot, covering a total area of 1400 km2 in 2008 and 2009.
As a result of detailed correlation made using the 3D seismic data, within the АС10-АС12 horizons, six productive layers were established: АС100(2), АС101-3, АС110, АС111, АС120-1 and АС122-5 (protocol GKZ RF # 1989 dated 19.08.09).
A complex well survey, analysis and interpretation of the seismic data was made with the purpose of forecasting the sand bodies distribution.
1. As a result of analysis of the regional research data, overall features of the regional sedimentation were determined.
2. Facies zones were identified across the drilled part of the deposit based on an electrometric GWS data analysis with the application of a core macro description.
3. For areas with seismic data, seismic classes were specified based on a cluster analysis of the seismic trace wave patterns within the producing formation.
4. For the drilled areas, where 3D seismic data was available, facies zones were identified and based on the results of seismic and electro facies modeling analysis.
5. The selected seismic facies were combined on the map into larger zones with a similar pattern of sedimentation. For a more reliable identification of such zones, a detailed analysis of the seismic proportional sections was conducted.
6. For areas of the deposit that were undrilled, maps were drawn showing seismic attributes that most closely connected the reservoir thickness in the exploration wells. For each identified and mapped seismic facies their own characteristics were found.
Based on the well data and the specific characteristics from each of the seimic facie zones, maps were drawn featuring their net thickness, and were then “linked”.
As an example, below is a review of the modeling results for formation АС110.
Regional Research and Lithofacies Analysis
The generation of clinoform complexes took place during the lateral infilling of the paleobasin.
During the transgressions, under conditions of a relatively fast rising ocean level and a significant distance to the coastline, clay was accumulating and being deposited in the foundation of the regional clinocyclites (Pimsk, Sarman, Pokachev etc.). The principal sediments, however, were accumulated during the regressions accompanied by processes of avalanche sedimentation.
The river system would bring precipitation into the shallow water area of the paleobasin, which would later form sediment under the influence of long shore currents. Part of the material would move to a sink zone through downhill transport channels and form dejection cones (A.A. Nezhdanov, 2000).
In the Priobskoye field, debris material drifted westward. The Paleoshelf edge, mapped based on 3D seismic interpretation results, has a north-east orientation. Results of the lithological core descriptions show that the area west of the paleoshelf edge features depositions from bottom currents, landslide units and sandstone from dejection cone flaps, all of which corresponds to the sunken area of the paleobasin. The zone east of the edge is formed with sandstone with a convoluted bedding, corresponding to areas of the paleoshelf.
The results of the GWS analysis confirmed the core analysis data. For the eastern part, the GWS curves have a distinct regressive shape, and for the western part they are intensely dissected (units with erosive boundaries).
Based on core analysis and GWS data, areas were mapped with similar signal amplitude features [3].
Seismic Facies Analysis & Mapping
Seismic trace shape analysis shows that it is not a consistent area and responds to the changes in facies zones. Selecting the amount of classes (typical seismic traces) presented a problem, which was resolved as follows: The lower limit of classes (not less than eight) was set based on wellbore analysis, as the depositional model had to be updated. Because of the wide well spacing for the area, the probability of not tapping the payzone is relatively high. To determine an optimal amount of classes, a statistical approach was used. For this purpose, seismic facies maps were calculated with class amounts ranging from 5 to 30 and correlation coefficients were derived between typical seismic traces. As the number of classes increased, the difference between the similar traces effectively disappeared, while at the same time the difference between the traces that are very different, decreased. Thus, the occurrence of this saturation effect is established when the number of classes used reach 15.
In the seismic classes map (fig. 2, a), the boundaries of the colored zones to the right are oriented north-easterly, on the left they are stretched eastward and in the center they have a complicated shape. Such orientations of boundaries correspond well to the perceptions of the reservoir sedimentation: debris material in the paleoshelf area was oriented along the edge due to the influence of long-shore currents. In the sunken part of the basin it is perpendicular to the edge due to the effect of gravitational currents. Comparison of the mapping with the GWS data interpretation allowed for the exclusion of zones where tight rocks were tapped, as well as areas which were not tapped with wells (fig. 2, b). As a result, the remaining zones can be used to map the reservoir substitution boundary, and the direction of the class boundaries when drawing forecast maps for net reservoir thicknesses allows the sand packages to be evaluated across the area.
Drawing Forecast Maps
The seismic attributes were calculated, taking into consideration, the facies zones. Ideally, each facies were assigned its own attributes.
However, because the wells are widely spaced in this area, performing a detailed analysis is impossible. Therefore, the characteristics were only calculated for two large facies zones: deep-water and shelf-type. Maximum correlations with the net oil reservoir thicknesses, for the deep-water area, were achieved when the using attributes “Average value for the envelope line” and “Minimal value for the cube of seismic amplitude” for the shelf area. Using the derived dependencies thickness-to-attribute, the attributes maps were recalculated into net reservoir thickness maps.
Using the same method, maps of the initial oil-filled thicknesses were drawn for all six formations: АС100(2), АС101-3, АС110, АС111 , АС120-1 and АС122-5. Maximum net reservoir thicknesses are off-set in relation to wells 617Р, 1013Р northwards, towards the area confined with wells 616Р, 1017Р and 1015Р (fig. 3). As a result the well cluster drilling priorities were adjusted for the next 5 years. Thus, the cluster near wells 617Р and 1013Р, which were earlier classified as promising (see fig. 3, a) fall within an unfavorable area. Tapping increased thicknesses is expected in a narrow strand along the line of wells 1015Р, 1010Р and 420Р.
Overall, the new map of forecasted net oil stratum differs significantly from the one previously used (see fig. 3) and under otherwise equal conditions are more reliable because during its construction 3D seismic data was used and the well data and reservoir sedimentation features were considered.
Conclusions
1. In order to accurately forecast the potential for reservoir development, it is important to analyze the geological model, deposit information, reservoir sedimentation, wellbore data and 3D seismic results.
2. By using the approach reviewed in this article, risks would be minimized during drilling and planning for additional field exploration becomes more effective.
List of Literature
1. Priob oil-bearing zone of Western Siberia: Systematic lithological aspect/Y.N. Karagodin, S.V. Yershov, V.S. Safonov [and others] – Novosibirsk: SD RAS SRC UIGGM, 1996 – 252 p.
2. Forecasting oil and gas occurrence in low-permeability reservoirs of clinoform sedimentary formations of lower Cretaceous in Kondinsky-Priob oil and gas bearing zone/T.V. Kryuchkova, V.P. Igoshkin, V.P. Kuklin, G.I. Davitashvily// SPE 116955. – 2008.
3. Muromtsev V.S. Electrometric geology of sand packages – lithological traps of oil and gas. – L.: Nedra, 1984. – 259 p.
