Multi-Seismic Attribute Interpretation and Machine Learning Based Analysis to Unlock Development of Stratigraphically Complex Secondary Reservoir Potential

Abstract

Abstract This case study in the Oriente Basin demonstrates how integrated and weighted seismic post-stack and pre-stack attributes enable identification of upside potential as well as de-risking of well planning and execution in reservoirs with complex stratigraphic architecture within a mature oil field. This approach analyzes the usability of Machine Learning (ML) algorithms to drastically improve the spatial estimation of rock properties related to oil bearing reservoir sandstones. Multiple seismic attributes interpretation by Artificial Neural Networks (ANN) is a robust approach to map thin sandstone reservoir layers embedded in contrasting lithologies and allows for improved predictability in new well locations. Multiple attributes are generated, in overlapping windows to address interference associated with the thin sandstone layers, then analyzed, selected, ranked, and combined by ANN. Seismic amplitude interpretation is run in parallel with other processes to reduce subsurface risk, including seismic pre-stack inversion, spectral decomposition, and waveform classification. In 2021, a new stage of field development was initiated, targeting secondary reservoirs in the Cretaceous Napo Formation with low recovery factors. These reservoirs are characterized by rapid lateral and vertical stratigraphic variations, deposited in a tidal fluvio-estuarine environment. The multi-attribute workflow was implemented to identify sweet spot areas, assess potential and optimize well locations and designs. The execution of this new field development strategy was first deployed in a selected drilling pad, previously identified as the best candidate for secondary reservoir development. This pad includes 4 horizontal and 2 deviated producers, and 5 workovers (WOs). The allocated production in the area and reservoir increased 8 times from the baseline oil production rate due to this recent activity. This marked the first horizontal well targeting T Superior sandstones in the Napo Formation in the entire Oriente Basin. In addition, a waterflooding project and surveillance was implemented to maintain average reservoir pressure, balance patterns to maximize oil production, and minimize the possibility of early breakthrough and high-water cuts in producers. Multi-domain analysis with seismic attributes is used for optimization of well planning and well placement, assessing future opportunities, and understanding reservoir continuity for waterflooding. This workflow established a new standard to enable high-grading secondary reservoirs potential and optimize future development plans.