Second EAGE Workshop on Pore Pressure Prediction

Prediction of pore pressures in large parts of a basin is very important to assure advancement of exploration, within safety limits, and to avoid risks related to anomalous pressures found in deep water frontiers. In Brazil, the Sergipe Sub-basin is located in the northeast of the country, corresponds to an Atlantic-type rifted basin with deep grabens controlled by large distensive rotational faults. The sub-basin represents one of pioneer and important areas of oil and gas production in this country with a renewed interest in the last decade for deep and ultra-deepwater opportunities. Despite that, the pore pressure in that area still little known. Thus, in these research we estimate the pore pressures throughout the stratigraphic units drilled by the P978-11 well, hoping in that way to contribute for future models of pore pressure in the basins of NE Brazil. For that, we applied on the sonic log of the P978-11 well the traditional Gardner and Eaton equations, but with own parameters values for the area, and then smoothed and calibrated the final results by loess regression. Four pore pressures zones were found ranging from 28 MPa to 89.4, suggesting an overpressure behavior in the Zonas 3 and 4.

Relationships between vertical effective stress (VES) and compressional sonic transit time (Δt) are widely used in the petroleum industry to predict pore pressures to ensure safe drilling operations and robust well designs. Whilst on most occasions trends can be established on VES-Δt plots to model mudrock pore pressures, there are increasingly more frequent examples where VES-Δt plots contain highly scattered data for which simple trends are difficult to establish.

Hydrocarbon trapping is the result of a contrast of a reservoir having a low entry pressure and a seal having a highest entry pressure. If there is an important overpressure or if the HC column is high the pressure can be close to the hydraulic fracturation criterion and there is a possible leakage.

This paper aims to demonstrate how the understanding of the reservoir extent, connectivity, facies as well as varying burial in a complex sub-salt context plays a role in determining the magnitude of overpressure in the inboard part of the Gulf of Mexico.

Uncertainties in pore pressure models arises due to a combination of various reasons: poor quality input data, challenging geological settings, dearth of sufficient predrill information, lack of geological understanding are some of the critical ones. Moreover, being a multi-disciplinary effort and in some cases due to a lack of sufficiently experienced resources - some of these challenges can get amplified. However, having an awareness regarding the effect of various geologic parameters on overpressure generation as well as its maintenance or dissipation through time plays a very critical role in influencing our overall interpretation of the ‘uncertainty envelope’. Many times it has been experienced that ‘knowing what we don’t know’ could be a key factor in how good we can quantify these uncertainties.

In Geomechanics for drilling engineering, there are two possible methods for predicting the FG: i) method based on the calculation of stress around the wellbore, ii) method using depth correlation established for a specific field from a set of leak-off pressure (LOP) data recorded on offset wells. In Total E&P, the method of calculating stress around the wellbore in order to predict the FG has been used and developed for a decade. It primarily consists of building and calibrating a ID MEM, using proprietary ID poro-mechanical earth model named PoroMEM. The calculation of the FG from the in situ stress model uses two conceptual models of fracturing: shear fracture and tensile fracture. From there, we define two limits for FG: one called FIPmax and the other FIPmin. Such conceptual models help to deal with the uncertainties of rock mechanical behavior because of vertical variability of lithology, help the preparation LOT/FIT during drilling operation, and also aid the interpretation of LOT/FIT. The in situ stress model from PoroMEM also provides the FCP gradient which is useful for well control and is sometimes as the FG. In this paper, we present the conceptual models and an example of FG prediction.

Pore pressure prediction in the Barents Sea is challenging for a variety of reasons including a complicated and variable burial history (leading to varying magnitudes and timings of burial, hiatus and uplift from well to well), variation in mudrock lithology and depositional environments. These factors can all create uncertainty when predicting pore pressure. Here are introduced two independent methodologies for present day shale pressure prediction across the Barents Sea, focussing in particular in the Southwestern Barents Sea where high overpressure is often recorded. Independent velocity log based and geological modelling based shale pressure prediction methods are used. Comparison of the two models generally shows good consistency in results, helping to validate the predictions and models and reduce uncertainty. This reinforces the need to integrate multiple techniques in order to reduce uncertainty in what is a challenging area for pore pressure prediction. The results also show that understanding the complex and variable burial history in each well is vital for pore pressure prediction, as this history defines the present day pressures. The models developed have clear implications for design of wells in the Barents Sea, and are also important for understanding reservoir overpressure variation (and potential reservoir fluid flow).

The oil industry is moving towards increasingly complex plays with challenging uncertainties and narrow drilling windows. To meet such a challenge Managed Pressure Drilling is increasingly being used to manage narrow pore pressure windows and drill deeper, higher pressure and higher temperature wells than was previously possible. At the same time the industry is facing stronger environmental and safety regulations that demand an improvement in risk characterization and pore pressure detection. While the new data from such techniques is valuable in constraining the pore pressure, it comes in new formats and often replaces or excludes some of traditional data such as connection gas that are normally used in real time pore pressure detection. Two cases are presented from two oil & gas exploratory wells in which the application of MPD (Managed pressure drilling) technique was critical for the success of the operation and where valuable pore pressure information was obtained to characterize and manage the drilling window. We conclude that MPD is an important tool in drilling wells with a narrow pore pressure and fracture gradient window. While MPD complicates some of the traditional methods of pore pressure estimation, it can be used proactively to get more accurate estimates.

Pressure transition zones are often expected when reservoir and cap rock pressures are thought to be in pressure disequilibrium. This could be either a pressure drainage effect on the cap rock due to drained underlying reservoirs or a pressure increase in the cap rock due to higher pressured underlying reservoirs. However, these transition zones are rarely seen nor verified with available data. As a result, such transition zones are modelled based on concepts from the 3D geological model and connectivity / lack of connectivity with calibration points. The case study aims to show a pre-drill prediction model where the transition zone within the shale dominated cap rock (due to drainage effect from the underlying reservoirs) was critical in order to ensure the optimal placement of the 13-3/8” casing string.

Only very few overpressure prediction accounting for hydrocarbon generation processes are actually based on real well results and data. This paper will show a case study example where both compaction disequilibrium and hydrocarbon generation participate to the generation of overpressures. This case study was built using standard log derived method as well as drilling observations and well events.