Core uncertainty management. An open discussion

The scope of this publication is to promote an open discussion rather to write a theoretical article or show complex calculation. Main topic will be how accuracy of core measurements could be managed. Let’s considerer that there is not easy way to evaluate these uncertainties. Please, take into account that many interprets think “Core measurements reflects accurately subsurface properties. This fact allows to evaluate our reservoir and most of the economical decisions are taken based on its measurements. Therefore this aspect plays a vital role in reservoir management.

Let’s challenge this paradigm in terms of:

Precision: An evaluation of its exactness

Accuracy: how close it’s a measurement with respect at its real value

Representativeness: Data sampling reflects each important part of the population

Meaning: Data sampling reflects each important part of the behavior of the population

Let’s analyze each individual aspect and hopefully we can build a roadmap altogether:

Precision: Most of the core measurement are precise and equipment are easy to calibrate; Nevertheless, in order to ensure this aspects labs should be visited in order to evaluate repeatability of the measurements. It is a good idea to collect some twin samples, and one should be send “a posteriori” and look for labs results. Also, a small amount could be send to another lab and results should be compared properly. Finally, different kind of measurements of the same properties should be compared and reconciliated. For example Pc measurements (centrifuge and porous plate), porosity measurements or mineralogical results.

Accuracy: In order to afford this aspect labs measurements should be compared with real values at subsurface conditions (Pressure, temperature, stress, wettability, fluids compositions, and mineral structure). Unfortunately, this is not possible, therefore we have to use our intuition and analyze each individual aspect since core cut, transport, cleaning, analysis, and final preservation. First at all drilling fluid invade core samples affecting clay minerals and other minerals structure, original fluids are displaced but mud filtrate, moreover wettability changes could be caused, emulsion, asphaltene or paraffin precipitation are rare but can occur. Secondly, As Holt explains "when the core is drilled in-situ stresses are released, during a certain interval of time the vertical stresses are less than the horizontal stress and the core is prone to mechanical damage, this damage is manifested physically by microcracks". When this type of phenomenon occurs it is no longer possible to make any correction and the measurements will not be representative. This phenomenon must be minimized by a correct design of drilling parameters.Thirdly, during transportation and storage a meticulous effort is required in order to avoid physical damage and/or depth alterations. Finally but no least, during cleaning, samples preparation and lab analysis a physical, mechanical and chemical alteration can occur affecting final results.

Unconsolidated rocks are most susceptible to be affected.

Please take into account that the smaller core diameter is, the more prone to suffer mechanical damage and fluid alteration will be. Another important aspect are operational ones, ROP and WOB are key factors. Drilling crew should be committed to achieve core high recovery percentage. Drilling, preservation, transport, analysis and storage should be granted to services companies committed with quality and integrity. Finally, improvisation is not allowed in order to reach our goals, each single stage should be rigorously planned.

Digital core analysis aims to be a very useful tool, it is necessary to continue delving into the research. Most of the computational limitations has been overcome and now it is possible a bigger resolution than 10 years ago. Now Porosity estimation associated with capillary bound water ( 0.5 μm -4μm) and clay bound porosity (10-500 nm) is possible. Therefore, nowadays anisotropic permeability and resistivity estimation is more robust. Multiphase flow properties estimation by digital rock techniques is improving its performance.

Representativeness: To guarantee this aspect team work is required. Sample selection is a key point in core analysis. Better result come across when no single opinion prevails but all geosciences contribute with this aspect. Another critical aspect is the number of sample required. When only economical aspects are taking into account usually goals are not achieved. Probably core analysis could looks expensive but they worth it. In fact they are very cheap compared with core drilling. Finally, some core intervals should be preserved in original conditions, please take into account that core analysis always going to evolve and some samples should be preserved in order to take advantage of this point.

Meaningful: it is closely linked with areal and vertical core distribution within our reservoir, based on depositional environment and its variance. Neighbor reservoir cores should be incorporated in our model in order to reflect properties variability. This is often ignored and only reservoir data under study is available. Once core to log calibration is achieved and the model is extrapolated to uncored wells, we have to keep on eye in anomaly trends indicating that a new core in some extra zones should be taken in order to reduce uncertainties.

Please leave your opinion. Discussion is open!