Integrated reservoir characterization workflow is indeed a great poster tagline, where ideally everybody talks to another and the collaborated efforts contribute some solution to a reservoir problem. What actually happen in most cases is that a sedimentology expert is hired to complete a study in a few months, facies model will then be handed over to a resident senior petrophysicist who works intensively for few weeks to deliver saturation model to the consultant geomodeler who works for few months to provide 3D models to the staff simulation engineer as illustrated in Figure1. In order to meet the year-end reserves estimation deadline, a robust workflow is then built to include shortcuts and assumptions because data are not readily available when needed. Disconnection between the stages, limited look back analysis and erroneous final reserves estimation are some of the implications of a robust workflow.
The ideal tagline of the integrated reservoir characterization workflow maybe intended as similar to a roundtable process as illustrated in Figure2, where continuous feedback always performed in every milestone, questions have always been asked and solutions are always iterated. But then it could take three years to complete single model realization.
Figure 3. Integrated reservoir characterization workflow where iterations are always performed
Take for example the interim petrophysical analysis carried out for a field in offshore East Java as illustrated in Figure3 water saturation model. A robust petrophysical analysis was required for the geomodeling input in order to calculate reserves and obtain a field certification. It has been reported that sedimentology study and core analysis were not available when the analysis is carried out. Capillary pressure data and NMR logs however had apparently been used in the saturation analysis workflow.
Figure 3. Sw in x-axis and height above free water level (HFWL) in y-axis cross plot coloured by lithology unit as points. Saturation height function models are plotted as solid lines
It appears that there have been attempts made to fit the height functions to the water saturation data points. It would have been a valid exercise in a normal circumstances.
Generally, large heterogeneity is the main concern of the discussion. The first part of the discussion has been focusing on the unit compartments with different colours, how they have been made and what are the constraints to limit the unit intervals. The second part comments on how the water saturation (Sw) model is generated, how accurate the well Sw compared to the entire dataset as well as the comparison with the capillary pressure as control points. There are potential issues raised by the geomodelers about J-function fitting inaccuracies. It is a compelling concern because such inaccuracy could lead to the erroneous hydrocarbon in place calculation. It has been suggested that different J-function should be applied to the different compartments (at least four compartments).
Specific geological review is then suggested for the orange color lithology unit where it has relatively flat Sw response above 100 ft, but contains transition zone data points tailing toward Sw=1. It could be different facies, different contact with hydrocarbon column thickness variation, or simply noise from the input data. The cause of such large range needs to be understood correctly. Large Sw range also appears consistently for the blue color lithology units as highlighted in Figure3.
The last part of the discussion addresses the J-function fitting. If the orange color lithology unit data spread is not noise, then fitting the data with the black solid line J-function is not accurate, e.g. the differential pressure should be modelled much higher to fit the tailing data points. The green color lithology unit, on the other hand, has data spread which crosses three J-function lines, and all of the J-functions are simply underestimating the blue data points. The geomodelers also raise the final point of their concerns regarding geomodeling misconception and underutilization of the final 3D water saturation model. Sw modelling is one of the complex exercises and yet the most important in hydrocarbon in place calculation. Continuous collaboration between the petrophysics, geomodeler, and simulator are critical in such exercises.
These are well-known issues but often disregarded within the time-constrained and resources-constrained project scheduling. The sedimentologist for the example above would probably not having enough time to answer questions about stratigraphical correlation and facies interpretation discrepancies for a particular well. The geomodeler in the meantime would be busy preparing the structural framework for the 3D model with the fellow geophysicist. The simulation engineer would either have been occupied with production priority or collecting the scattered PVT data to prepare the upcoming simulation.
Assumptions would then have to be made to fit the water saturation model, there are simply not enough time to develop the ideal 3D water saturation model which uses individual J-function and in the same time agrees with the petrophysical calculation to capture heterogeneity as depicted in the cross plot.
An external peer review team consists of functional experts hence might be a good support to facilitate an integrated workflow in a robust project scheduling, as illustrated in Figure 4.
Figure 4. Ad hoc external peer review team in the reservoir characterization workflow
The external peer review team should be able to provide insight over the front-end reservoir characterization activities, to describe quick sensitivity analysis, to explain how the simulation work and geomodeling outcomes would be with the assumptions which have been taken over data limitation, to provide missing data and information as needed.
There is always be a potential project distraction with external resources or there would be budget limitation for such team. We have seen this problem too often; our team have years of experiences in the reservoir characterization and very familiar with such setback situation. We could help to make a robust reservoir characterization workflow also be an integrated one.
Please contact DGS Peer Review Team for more information about our external review services.