In the future petroleum exploration, it is presumed that technology for prediction of reservoir distribution become more important to reduce geological risks. Improved prediction of reservoir distribution is attempted by combination of 3-D seismic data and other exploration information. As the unique interpretation is not obtained from these data in most cases, it is necessary to make conceptual geological models intervene. Each interpretation method utilizes different state and degree of intervention of conceptual models. In the case of seismic attribute analyses, for example, conceptual geological models may be adopted with comparison on the basis of visible similarity. In the case of sequence stratigraphic analyses, in contrast, sequence stratigraphic model must be constructed and adopted as much complex models as concepts for stratal genesis. Furthermore, stratigraphic and sedimentary geological simulation by computer is deductive method in combination with several conceptual models of different types such as facies models and sequence models. Methods for prediction of reservoir distribution, based on the intervention of conceptual geological models, are applicable in the exploration of subtle traps such as stratigraphic traps which explorationists must face in the near future. Also such an approach must be significant in the extraction of new exploration play concepts within well-explored proven petroleum systems. In oil and gas fields of development and production stages with abundant petroleum geological information, heterogeneous distribution of reservoir properties and seal capacities can be discussed genetically based on the adoption of sedimentary geological conceptual models.
We are now executing a cooperative research project with UNOCAL in order to develop and apply geophysical technology for reservoir characterization and optimizing miscible flood enhanced oil recovery in carbonate reservoirs. This project aims to monitor the fluid behavior in carbonate reservoirs by time-lapse geophysical surveys such as 3-D seismic survey, VSP, cross-well technology and well logging. JNOC and UNOCAL decided to conduct this study at the Reinecke oil field in Texas, U.S.A., operated by UNOCAL, whose reservoir is Paleozoic dolomitic limestones.First, we interpreted the geometrical framework of the reservoir and described the reservoir in terms of P-P reflected wave volume and P-SV converted wave volume, both of which were generated from the data acquired in December, 1997. Based on these methods, we drew a detailed time structure map and found some new structural/sedimentological features of the reservoir. Second, for reservoir characterization, we calculated the Vp/Vs ratio from the isochron P-P and P-SV 3D volumes and created a distribution map of the Vp/Vs ratio. We evaluated the distribution map as a predictive tool of carbonate lithology by using the relationship between the Vp/Vs ratio and the dolomite content established through well log analysis. Our initial interpretation suggests the correlation between the the seismically deduced Vp/Vs ratio and the dolomite content. There has been, however, difficulties in the evaluation at its advanced stage. Believing that such 3D seismic volume information should include some useful attributes of reservoir, we are continuing to evaluate the data for further reservoir characterization. Third, we conducted a new 3D survey over the field in December, 1999, after Unocal started to inject CO2 to the reservoir in January, 1998. This is to monior the effect of the CO2 injection using the seismic attributes in those time-lapse data series. We successfully found some differences in and around the reservoir between the previous baseline sections and the rush sections of the monitoring survey. This evaluation will be completed with cross-equalizing the volume data sets calculated from the two, time-lapse surveys.
Premature water breakthrough was observed in a Lower Cretaceous carbonate reservoir under five spot pattern water injection in a Middle East oil field. Openhole log resistivity anomalies in recently drilled wells and time lapse pulsed neutron logs suggested preferential water movement in certain horizons. Reservoir heterogeneity depending on lithological variation is identified as a cause of the water breakthrough. Two high permeable lithofacies were defined by core and thin section observation. One is altered bioclast dominant coarse grainstone/packstone in the middle part of the reservoir. The other is very fine altered bioclastic peloidal grainstone/packstone in the upper part of the reservoir. These lithofacies are composed mainly of rarely micritized bioclastic grains with rigid framework preserving intergranuler porosity. Large molds and vugs associated with rudists in the upper part of the reservoir are locally connected and contribute to the fluid flow. Relative sea level change and paleo-topography are recognized as important factors to control lithofacies distribution. Areal distribution of the lithofacies shows aggradational stacking pattern and abrupt lateral lithofacies change particularly in the middle part of the reservoir. This suggests that simple carbonate ramp model currently applied is not appropriate for reservoir characterization purpose in this reservoir.
Volcanic rock units in the Minami-Nagaoka gas field are very heterogeneous because of their facies change and intense alteration. ln this study, a geological model of the Minami-Nagaoka gas field is constructed to explain such heterogeneity for optimizing further exploitation. This model emphasizes two aspects of this volcanic reservoir. The first is distribution of primary volcanic body, and the second is secondary alteration processes. Primary volcanic body is identified on the basis of the difference of the primary plagioclase type in phenocrysts. This volcanic rock is divided into 6 zones according to types of primary feldspar in phenocryst; zone 1, oligoclase; zone 2, oligoclase with labradorite and anorthite; zone3, albite; zone 4, oligoclase and andesine; zone 5, labradorite and anorthite; zone 6, albite. It is clarified that the very productive intervals are within the glassy rhyolite of “zone 3” which extends widely in this field. The glassy rhyolite in zone 3 of the southern area is characterized by perlitictexture, and dissolution pores are dominant in such glassy part. In the northern area, such glassy rhyolite underwent intense silicification and dissolution pores are underdeveloped compared with those in the southern area. It is inferred that such heterogeneity of porosity is due to the difference in recrystallization of glass by the influence of the overlying volcanic activity. Dissolution pores are generated by the process of montmorillonitization in which cations such as potassium and sodium leached out. Montmorillonite changes to sericite (illite) when provided with potassium and to chlorite with iron and/or magnesium. In the lower part of zone 3, chlorite occurs as fill of the pores. This means that the iron and magnesium leached from underlying mafic or intermediate rock units deteriorated reservoir quality with chlorite precipitation in the lowermost interval.
The primary reservoir of the Yurihara Oil and Gas Field, Akita, Japan, is basalticrock, named “the Yurihara Basalt” that erupted onto the deep sea floor during the Middle Miocene.The “ Yurihara Basalt” is composed of an olivine-augite basalt, consisting of pillow lava, sheet flowlava, hyaloclastite, and epiclastic-hyaloclastite, and intercalates shale beds. Vesicular pillow flowsand sheet flows are the most suitable as hydrocarbon reservoir rocks in the Yurihara area. Thepore space of the reservoir is classified into three types; vesicles, dissolution pores and fractures. Lithofacies such as pillow lava and sheet flow lava recognized in the core of the YuriharaBasalt are correlatable with those of the well log data. Well-to-well correlation has revealed thatthe facies change laterally. An intensive study over the field suggests that thicker sheet flows aredistributed near the vent and change gradually and laterally to thinner sheet flows, pillow flows, hyaloclastite and epiclastic-hyaloclastite with distance. Oil producing wells are located in a pillowflow-dominant area. The direction of the lava flows was able to be determined from inner structures of sheet flows.These flow structures were identified as the characteristic arrangement of the vesicles on boreholeimages such as Schlumberger's FMS (Formation Microscanner)TM. The facies distribution was predicted by using a model of lateral facies change and informationon the flow direction. Appraisal wells drilled in the predicted oil bearing pillow flows-dominantarea resulted in a great success and confirmed oil columns. These positive results strongly suggestthat the procedure presented in this paper conrtibutes a major break-through to exploration anddevelopment of basaltic hydrocarbon reservoirs.
In terms of generation and preservation of porosity, volcanic-rock reservoirs in Japan are studied. A model of generation and preservation of dissolution pores during interaction between volcanic rocks and sea- or formation-water is discussed. The volcanic rock reservoirs can be divided into the primary-porosity type, the dissolution-porosity type and the fracture-porosity type, based on the origin and preservation mechanism of pores. The first type was observed in Miocene pyroclastic rocks in the Yufutsu Oil and Gas field, Hokkaido. They are mainly composed of scoria and fine volcanic glasses which erupted on a volcanic island. Their pores are neither filled with clay minerals nor collapsed by compaction, but supported by unaltered brittle volcanic fragments. The second type was recognized in the Ayukawa Oil and Gas field and in the Akita area, northeastern Honshu. A reservoir in the Ayukawa Oil and Gas field occurs in dolerites which intruded into unconsolidated argillaceous sediments. Its pores were formed by dissolution of feldspar and clay minerals which once replaced mafic minerals. The framework that preserves pores is an initial texture which was strengthened by secondary K-feldspar precipitated around the original feldspar. Reservoirs in the Niigata area occur in Miocene acidic rocks which erupted on a seafloor. Their pores were formed by dissolution of groundmass and feldspar. The framework that preserves the pores is formed by aggregated authigenic quartz such as spherulite. The third type was observed in the Yurihara Oil and Gas field. One of its reservoirs was formed in highly altered basalt which erupted on a seafloor in the Miocene. Its pores were formed by fractures in porous lava flows which were strongly altered to become characteristically brittle. Microscopic observation of the dissolution pores and the experiments by the previous studies in water-rock interaction suggest that the dissolution pores and the framework could be formed with the formation of authigenic minerals during the water-rock interaction when igneous rocks erupted in the seawater or intruded into unconsolidated sediments. If this model is followed, the more water the igneous rock reacts with, the more dissolution pores and authigenic mineralization may occur. Accordingly, constant fluid flows may cause the generation of dense distribution of dissolution pores.
The controlling factors and principles which govern the spatial and chronological distribution of turbidite sandstones have been investigated based on the results of sedimentological, petrographical, and mathematical analyses on the Neogene and Quaternary turbidite sandstones in the forearc and backarc basins, central Japan. Main conclusions are as follows: 1) The sedimentological research revealed that the three-dimensional forms of individual turbidite sandstone beds, which are part of a depositional lobe of submarine fan system and regionally correlated bed by bed by using many useful tuff key beds, are similar to each other. The similarity continues vertically within the interval of a single depositional lobe or of a certain stratigraphic unit. Their thickness and distribution area reflect the magnitude of turbidity currents, while their form greatly depends on the grain size distribution of the sediments. 2) The petrographical research proved that the heavy mineral composition can be used as a good indicator to recognize turbidite sandstone bodies which share a single source area, and are useful to check the original relationship of turbidite sandstone bodies isolated each other. 3) The mathematical research led to the conclusion that the development of turbidite sandstones is controlled by the global changes such as the eustaic sea level changes, because the periodicity in an order same as the Milankovitch's cycles was recognized in the vertical change of number and total thickness of turbidite sandstone beds within an interval delineated by a certain accumulated thickness of hemipelagite in several turbidite successions.
The Guarico 13 field is located in the western Greater Oficina area, Eastern Venezuela basin, and consists of a large number of hydrocarbon accumulation in sandstones deposited in a coastal-plain environment. Trapping mechanisms for exploration opportunities in the study area combine structural faulting and stratigraphic pinch out. Individual reservoir thickness is normally 20ft or less and therefore it is hard to detect the reservoir distribution and the limits of hydrocarbon trapping area by using seismic data even if a seismic inversion process is carried out. Although the size of each individual reservoir is small, multiple reservoir sections were observed in each well and the sum of these reserves will be worth developing. It is very important to predict the potential reservoir facies to increase oil production and to discover additional new reservoirs in this kind of sandstones deposited in a fluvial-deltaic system. The high-resolution stratigraphy in the area was achieved by correlating all the lignite beds that occur between the major marine and lacustrine flooding surfaces. Then the core intersections, log character, borehole image logs and palynology were utilized to estimate the depositional environment of each hydrocarbon-bearing stratigraphic unit. After this environmental analysis, modern analogs and a geological model were applied to predict the reservoir facies and its extent. In this report, the authors demonstrate two case studies of estimating the reservoir distribution; one uses a template model of a major bed-load fluvial system with crevasse splay sands on interfiuve area, and another employs a crosscutting tidal inlet and associated ebb-tidal delta system as an analog. This sort of approach that uses modern analogs as templates is worth applying to predict the potential distribution of reservoir sandstone if 3 D seismic is not available or can not detect reservoir extent.
The Middle Cretaceous Burgan Formation (Wasia Group) includes two main oil reservoirs, 1st and 2nd Bahrain Sand in the Khafji oil field, offshore of the ex-Neutral zone between Saudi Arabia and Kuwait. Detailed 3D reservoir models considering complicated reservoir heterogeneity for both reservoirs were constructed to manage these two reservoirs more efficiently using a reservoir simulator. The Burgan Formation is mainly composed of sand and shale layers with thin carbonate and coal beds. The formation can be characterized by three types of log facies which are defined from the correlation between shapes outlined by gamma-ray and resistivity log curves and core description. As a result of the correlation on the basis of these log facies, it is interpreted that the formation is basically composed of repetition of a parasequence unit represented by widely distributing funnel-type log facies, which is often cut horizoutally by narrow and meandering channel sands represented by bell- and cylindrical-type log facies. The mode of channel distribution, sedimentary facies and palynological analyses in each parasequence unit indicate that the formation is composed of three depositional sequences recording eustatic sea-level changes which caused various depositional settings from the middle reach of a huge fluvial system to shallow marine near the river mouth. To construct a 3D reservoir model, digital isopach maps for both sand and shale layers were constructed considering specific depositional environments and the direction of regional sediment supply, and then the isopach maps were digitally stacked beneath the top structure map. Finally, reservoir property maps, such as average porosity and net gross thickness ratio maps, were generated, which were adjusted to become consistent with the pattern of isopach trends, considering the sedimentary energy condition in the interpreted depositional environments.
Geological modeling is a process comparing many cases and extracting most general ones from various factors that caused complicated geological phenomena. Geological phenomena are highly effected by local factors such as tectonic situation, basinal and provenance topography, and climate. Therefore, the creation of geological model need to distinguish general and important factors from complicated and various factors. However, an actual distribution of reservoir rocks in an objective basin is considerably controlled by the local factors. The geological modeling for prediction of distribution of reservoir rocks need to recognize general and local factors and discuss the role of them respectively.