EAGE GeoTech 2021 Third EAGE Workshop on Practical Reservoir Monitoring

The value of seismic history matching by converting flow simulation models to synthetic seismic (Sim2Seis) data is illustrated with a case study on Gannet E, a Palaeocene oil field in the UK North Sea. Rock physics models were built to allow converting the dynamic model reservoir properties to elastic properties, and to establish the sensitivity of the modelled seismic data to reservoir variation. A modified Hertz-Mindlin uncemented (soft) sand model was used for pressure and porosity modelling, and Gassmann fluid substitution was used for saturation modelling. Results showed that saturation change had a large impact on acoustic impedance, while the impact of pressure change was significantly smaller, showing that a clear 4D signal due to sweep would be expected in the field. Flow simulation models were used with the rock physics model to create synthetic seismic data to match the baseline and two monitor seismic surveys available. After two iterations of Sim2Seis modelling and reservoir model updates, a much-improved seismic match was obtained, resulting in a reservoir model that is a more reliable reservoir monitoring tool. An infill well is now being planned based on 4D analysis and Sim2Seis modelling, and further opportunities are being assessed.

Planning and managing operations of subsurface reservoir assets in terms of conformance is crucial for responsible CO2 storage. One important component of conformance management is the monitoring of the reservoir dynamics in response to the implemented operational strategies, particularly for early detection of deviations from intended behavior (i.e., non-conformance). In recent work, we have introduced a model-based quantitative workflow to objectively assess the usefulness of monitoring for conformance verification in CO2 storage and shown how to use state-of-the-art supervised learning techniques to achieve a more practical workflow. In the present work, we investigate the use of a semi-supervised anomaly detection approach based on auto-encoder neural networks as an alternative to circumvent limitations of the supervised classification approaches explored so far. The results of our case study of a real storage aquifer show that auto-encoders trained on simulated time-lapse seismic data from (only) conformance scenarios can be used to accurately detect scenarios where the migration of CO2 deviates from the desired range of behaviors. These promising results confirm that the proposed approach can be used to derive efficient conformance classification workflows without an explicit finite dataset representing non-conformance to be defined in advance.

In this work we tackle the challenge of estimating reservoir property variations during a period of production, directly from 4D seismic data. We employ a deep neural network to invert 4D seismic amplitude maps to the simultaneous changes in pressure, water and gas saturations. The method is applied to a real field data case, where, as is common in such applications, the data measured at the wells is insufficient for training neural networks, thus, the network is trained on synthetic data. Training on synthetic data offers much freedom in designing a training dataset, therefore, it is important to understand the impact of the distribution of data in training datasets on the inversion results. We perform a study testing four different approaches to populating a training dataset, showing the impact of including physics-based constraints on the reservoir property distribution. Using the results of a reservoir simulation model to populate our training datasets we demonstrate the benefits of constraining training samples to fluid flow consistent combinations in the dynamic reservoir property domain, uncovering the potential of deep neural networks for 4D seismic inversion, in a situation where no appropriate measured data is present.

Two case studies are examined where a multidisciplinary approach was necessary to understand and quantify the 4D seismic interpretations. The first example involved compaction-related saturation changes, where a 4D softening response along the oil water contact of a depleting reservoir was observed. It was interpreted that oil was being pushed into the water leg of the reservoir due to relative high rates of compaction and inefficient drainage of the oil in the anticline. Integrated work using the reservoir flow simulation model, Pressure Transient Analysis, and geomechanical modelling was used to understand this phenomenon. The second example highlights the 4D softening effect around an oil water contact in a waterflood field. The response was interpreted to be driven by a live water scenario, where the pressure dropped below the bubble point in a rock that had significant residual oil trapped in the pore space of a swept zone. 2D modelling of complex pressure and saturation effects throughout the reservoir were needed to quantify the changes and aid in assisted history matching of the field. In both case studies, the 4D seismic interpretation required integration with engineering disciplines to investigate and understand the mechanisms responsible for the 4D seismic signal.

Gravimetry and subsidence surveys are regularly conducted to monitor the gas-producing Snøhvit field. This technology provides insight into reservoir dynamics with lower costs and shorter turn-around time than widespread 4D seismic. Through two decades, improvements in survey procedures and instrumentation have led to significantly better sensitivity to both gravity changes and subsidence. This has resulted in turn into improved sensitivity to mass changes and compaction throughout the reservoir. In this abstract, we analyze the evolution of the noise level in the three surveys at Snøhvit in 2007, 2011 and 2019. We present in detail the results for the 2011–2019 time-lapse, and we describe how these measurements are incorporated into the history-matching routines and used to reduce the uncertainty in key reservoir parameters.

In order to correctly interpret the shale gas reservoir in the evaluation area, this paper discusses the main controlling factors of the low resistivity of the shale gas reservoir. Use a large number of core data such as polarized light microscope, fluorescence microscope, QEMSCAN and X whole rock diffraction to reveal the cause of the low resistivity of shale gas reservoirs; further use the statistical comparison of shale gas reservoir logging data with core analysis data to analyze the storage the main controlling factor of the cause of low resistance. Comprehensive analysis shows that there are two main controlling factors that lead to low resistance of shale gas reservoirs: over-matured organic matter completely graphitized, and organic texture distribution pattern.

High-precision field-wide seafloor subsidence measurements have been applied to nine hydrocarbon fields in Norway. These measurements provide lateral information on the product of pore compaction and pressure depletion in the reservoir. The method uses water pressure measurements at the seafloor and reaches survey precision of 2 – 5 mm. Cases demonstrate that the lateral distribution of subsidence can be used to identify reservoir compaction, undrained compartments, and to calibrate the geomechanical model. In this abstract we discuss precision limitations and particularly temperature effects. Temperature stabilization of the sensors has recently been claimed to eliminate temperature-induced effects and provide superior accuracy at the Ormen Lange field. We argue that this is not consistent with the data provided. Temperature stabilization does not improve the depth repeatability when using 20 minutes recording time. The subsidence results indicate significant unaccounted errors or effects, and it is premature to conclude on the impact of the temperature stabilization on the subsidence errors after one such survey only. For surveys on the shelf, the measurement time can probably be shortened for pressure-only surveys compared to the common practice of 20 minutes for gravity surveys, irrespective of any system for stabilizing sensor temperatures.

NA