Petrobras found almost 100 hydrocarbon accumulations in the Campos and Santos basins, between 50 and 300 km off the Brazilian coast (under water depths from 80 to 2,400 m), which produce from very different types of reservoirs, including mostly (1) pre-salt coquinas and microbialites, (2) post-salt calcarenites, and (3) post-salt siliciclastic turbidites. These different types of reservoirs, containing also different types of hydrocarbons and contaminants provided many challenges for their production development, related to distinct tools and workflows for reservoir (static/dynamic) characterization and management, seismic reservoir characterization and monitoring, recovery methods (water injection, WAG, etc.), well spacing, well types and geometries, subsea systems, and processing capacity of production units. Since the first oil and gas discoveries in the Campos (1974) and Santos (1979) basins, Petrobras continuously moved to aggressive exploration and production from shallow- to deep- and ultra-deep waters. During the last 40 years, the activities of reservoir characterization and management have also continuously evolved. Four major phases can be depicted: (1) shallow water fields developed with a large number of vertical or deviated wells (e.g. Namorado, and Pampo, Campos Basin); (2) deep water fields, still developed with a large number of wells, but now combining vertical/deviated and horizontal wells (e.g. Marlim and Albacora, Campos Basin); (3) deep to ultra-deep water, post-salt fields, containing light to heavy oil (13-31 °API) in siliciclastic turbidites and carbonates, developed with a relatively small number of mostly horizontal wells (e.g. Marlim Sul, and Barracuda, Campos Basin); (4) ultra-deep water, pre-salt fields with very thick (up to 400-500 m), light oil (27-30 °API) carbonate reservoirs, developed with largely-spaced vertical and deviated wells (e.g. Lula, and Buzios, Santos Basin).

The purpose of this work is to proceed uncertainties analysis with respectto the physical properties of a petroleum reservoir, in the process ofgeological characterization and flow simulation model. The approach is appliedin a real case of Campos Basin, in Brazil, and may be helpful to the decisionanalysis concerning the profitability of the exploitation project. The sequenceis performed in four parts: the quantification of the sources of uncertaintiesconcerning the reservoir volume; the quantification of the factors affectingthe oil recovery factor; the quantification of the uncertainty range of therecoverable volume; and the estimation of the production profiles. The firstpart is executed through geostatistic simulation of the net-to-gross variable, associated with variation on the position of the oil-water contact. The secondpart consists on the use of numerical simulation runs to verify the sensibilityof the recovery factor to the variation in the vertical permeability, therelative permeability to water, and the productivity index of the wells, themost important variables in this specific case. The effect of each factor issupposed to be independent of the others, and only two cases, representing thepessimist and the optimist case, are simulated for each variable. As thisindependence is not perfect, two additional simulation runs may be required toadjust the width of the distribution. Having the probability distributions ofthe oil volume and of the recovery factor, Monte Carlo simulations are used toobtain the distribution of the recoverable volume. Each value of reserve isrelated to a production profile, to obtain the cash flow of the project, andevaluate the uncertainty concerning to the economic parameters. Introduction When exploiting a petroleum field, we deal with a lot of uncertaintiesconcerning the internal variables, as the geological characteristics, and therock and fluids properties, and the external variables, as the oil price, thecapital expenditure and the operating costs, that impact the profitability ofthe project. The quantification of these uncertainties is helpful to supportthe decision making, by the evaluation of the risks. The economic analysis of a development system is based on the cash flow ofthe project, that is a direct function of the production profile of thereservoir. But, unfortunately, the estimate of the production curves cannot bea deterministic approach, because, in the moment of decision, we have too fewavailable data, and most parameters have a wide range of possible values, whatcause an uncertainty range in the cash flow. The most important factor affecting the recoverable volume is the volume ofoil "in place", which value is not accurately estimated, by the fact that theappraisal wells are often in low quantity, and the costs of data acquisitionare high. The uncertainties concerning the position of the reservoirboundaries, the depth of the contacts between fluids, and the fault system are, in general, the main factors affecting the uncertainty in the volume. Thedetermination of these ranges are not a straight forward methodology, as itinvolves the seismic interpretation, the depositional model, and thecharacterization of the variability and heterogeneity of the porous rock, executed through integrated studies. The application of the geostatistic toolsgenerates a wide range of possible images for the rock variables, as the netpay of the formation, or the rock porosity, incrementing the range of thevalues of the reservoir volume. Therefore, the uncertainty analysis in thevalue of reservoir volume, that combine "soft information" and objectivestudies, is the first step in the risk and decision analysis. The other parameters that mostly affect the recoverable oil are related tothe recovery factor of the formation, as the vertical and horizontalpermeability, the relative permeability of the fluids, the viscosity of theoil, and the productivity of the wells. The reservoir simulation model is theideal instrument to include all these properties together with accuracy. Themain obstacle of its use is how to study the sensibility of various parametersand calculate the probability of each combination.

No abstract

Recently discovered oil fields in the Pre-salt area in the Santos Basin, offshore deepwater Brazil, contain a huge volume of hydrocarbons in carbonate rocks, mostly of microbial origin, with pronounced heterogeneity, without any analogues in the world. The fluid usually presents expressive levels of contaminants and a significant compositional grading, laterally and vertically. This complexity denotes how challenging is to define robust development plans and suitable recovery mechanisms. In order to adequately address these issues and optimize the hydrocarbon recovery, the development strategy was supported on three pillars: (i) extensive data acquisition; (ii) strong interpretative work on the geological characteristics which impact fluids displacement and (iii) adoption of comprehensive recovery mechanisms and strategies, adequate for the various scenarios generated in the studies. The data acquisition program includes, at first, a high resolution seismic data acquisition and interpretation. Well information comprises conventional logs, image logs, cores, downhole fluid samples, sidewall cores, pressure transient tests. Additionally, for each area candidate for a production system, an extended well test is performed, with on-line interference tests with neighboring appraisal wells, aiming to evaluate the dynamic properties in terms of areal and vertical connectivity, as well as to anticipate potential problems regarding formation damage, fluid composition variation and flow assurance. Even with this comprehensive data acquisition, the complexity of reservoir rock sustains a large range of uncertainty in their properties, which will only be reduced throughout the productive life of the field. Therefore, the second pillar of the field development is built through a deep dive on the data by the geoscientists, in order to extract the geological attributes that impact the displacement and sweep efficiency of the oil. This goal is not always possible to obtain in a straight and deterministic way, but often through multiple possible scenarios. This is referred mainly to features as horizontal permeability distribution, presence of faults and barriers, vertical communication, fracture corridors, high permeability channels, among others. For the construction of the third pillar, a strong synergy between the disciplines of geoscience and reservoir engineering is necessary, so that the geological interpretation is incorporated and applied to the flow simulation model and, the most important, support the design of robust strategies to optimize oil recovery, under uncertainty conditions. When information is considered as perfect, it is immediately incorporated into the plan. Otherwise, multiple scenarios will persist for a longer period, making it necessary to apply strategies that can be effective in different conditions, and can be matched as soon as the real scenario is revealed. This paper describes the approach adopted for the development of the Pre-salt fields, supported by the aforementioned three pillars, and detail...

At the time of selecting the development plan of a petroleum field, little information is available and the range of uncertainty is large. Once the main variables that most strongly affect the production behavior have been identified, the next step consists in generating the set of possible results for production forecast, and obtaining the probability distribution of the economical parameters, as the net present value (NPV). Depending on the range and nature of the uncertainty variables, the best way to protect the project, instead of collecting information, may consist in changing the exploitations strategy, in order to reduce reservoir risks. In next phase of exploitation of Marlim Sul field, Campos Basin, Brazil, the main source of uncertainty is an extensive fault, located in the middle of the reservoir, which breaks the oil zone into two blocks. According to the strategy of the original plan, the injection wells are located on one side of the fault, while the production wells on the other side. The sealing capability of this fault is a critical issue, since it could be an obstacle to oil sweeping by the water injection. To face this issue, we have two alternatives: either collect more information in order to reduce the level of uncertainty, or select an alternate plan, less sensible to transmissibility of that fault, and more adequate to the various scenarios. To compare these alternatives, we utilize uncertainty analyses, enriched with risk aversion considerations. The use of the expected monetary value (EMV) as parameter of comparison proved to be a good way to meet better all the scenarios, but not sufficient to traduce the economic risk of the project. The use of concepts of the utility theory indicated that the alternate plan has lower sensibility to critical uncertainty, is more protected against reservoir risks, and is the best development strategy to be implemented. Introduction The selection of the best development project of an oil field is mostly accomplished with base on limited data available, in a scenario of large uncertainty about reservoir properties, mainly those related to fluids volume and flow capacity. This is an intrinsic feature of the petroleum industry, where data collection, such as well drilling, formation tests or seismic acquisitions, requires quite high cost investments. The adopted exploitation strategy is the one that reach the maximization of economical parameters, like the net present value (NPV), when applied to the most likely reservoir scenario, which show the expected values of the properties linked to geometry and fluid flow in porous medium. When the range of uncertainty is high, this strategy reveals not being efficient, since the adopted project is the best for the moderate scenario, but may be inadequate for other scenarios, such as, for example, the pessimistic one. In this case, protective and risk reducing action should be undertaken. The use of expected monetary value (EVM), which represents the values of NPV of each scenario, weighted by its probability, is a more consistent parameter to help the selection. An alternate strategy, less sensible to sources of uncertainty, apparently is more adequate, even having a lower NPV in the most likely scenario. Nevertheless, even the utilization of the EMV parameter may not be sufficiently robust to indicate the best option to be pursued. While analyzing aspects of risk aversion, it can be noticed that, in a general sense, deficits resulting from a pessimistic scenario are fare more serious than simple economical losses. Further questions may be involved, such as company image, team motivation, and even company survival. To represent issues of this nature, the preference or utility theory and the concept of utility function are tools of great value and are applied here for the decision making process. The objective of this paper is to present a comparison between oil field development strategies, under conditions of high reservoir uncertainty, utilizing, as base the EMV, conjugated to concepts of utility function. The target field is Marlim Sul, located in the Campos Basin, Brazil. In the following sections, we describe the reservoir aspects, its critical sources of uncertainty and the strategies for oil recovery to be compared. Subsequently, the methodology for strategy comparison is presented and, lastly, the practical application and the results analysis.