The paper presents a multi-physics investigation of the ground movements related to the cyclical and seasonal injection and withdrawal of natural gas in/from a depleted reservoir located in the Po Plain area, Italy. Interferometric Synthetic Aperture Radar (InSAr) data (from 2003) and Global Navigation Satellite System (GNSS) data (from 2008) provided a full and coherent panorama of almost two decades of ground movement in the monitored area (more extended than the field boundary). The analysis of the acquired millimetric-scale movements together with the detailed geological analysis, both at reservoir and at regional scale, represents the focal point for understanding the investigated phenomena. Based on this information, a fully integrated and multidisciplinary geological, fluid-flow and geomechanical numerical modeling approach was developed to reproduce the main geometrical and structural features of the involved formations together with the poromechanics processes induced by the storage operations. The main achievement of the adopted methodology is a deep knowledge of the system and the involved processes, which is mandatory for the safety of the urbanized areas and the effective management of the underground resources.

Machine learning-based Surrogate Reservoir Models (SRMs) can replace/augment multi-physics numerical simulations by replicating the reservoir simulation results with reduced computational effort while maintaining accuracy compared with numerical simulations. This research will demonstrate SRMs’ potential in long-term simulations and optimization of geological carbon storage in a real-world geological setting and address challenges in big data curation and model training. The present study focuses on CO₂ storage in the Smeaheia saline aquifer. Two SRMs were created using Deep Neural Networks (DNNs) to predict CO₂ saturation and pressure over all grid blocks for 50 years. 18 million samples and 31 features, including reservoir static and dynamic properties, build the input data. Models comprise 3–5 hidden layers with 128–512 units apiece. SRMs showed a runtime improvement of 300 times and an accuracy of 99% compared to the 3D numerical simulator. The genetic algorithm was then employed to determine the optimal rate and duration of CO₂ injection, which maximizes the volume of injected CO₂ while ensuring storage operations’ safety through constraints. The optimization continued for the reproduction of 100 generations, each containing 100 individuals, without any hyperparameter tuning. Finally, the optimization results confirm the significant potential of Smeaheia for storing 170 Mt CO₂.

Hydraulic fracturing, natural fracture reactivation and resulting induced microseismicity are interconnected phenomena involved in shale gas exploitation. Due to their multi-physics and their complexity, deep understanding of these phenomena as well as their mutual interaction require the adoption of coupled mechanical and fluid flow approaches. Modeling these systems is a challenging procedure as the involved processes take place on different scales of space and also require adequate multidisciplinary knowledge. An extensive literature review is presented here to provide knowledge on the modeling approaches adopted for these coupled problems. The review is intended as a guide to select effective modeling approaches for problems of different complexity.

This paper investigates the effect on compaction and subsidence induced by gas production using two techniques for coupling fluid-flow and geomechanics. To this end, a synthetic case representing a typical, shallow, weakly compacted, multi-layered, off-shore gas reservoir in the Adriatic Sea was set up and its dynamic and mechanic behavior during gas production was analyzed. Three numerical models (i.e., geological; fluid-flow and geomechanical) were built using high quality data set from an existing gas-bearing formation (off-shore Croatia). The laboratory analyses for deformation and strength parameters determination were conducted together with tests to define the coupling law required by the adopted coupling technique. Experimental data showed strong permeability stress-dependent behavior of core samples retrieved from gas bearing layers. Nevertheless, the results showed that the system stress-strain evolution always remains in the elastic domain and the deformation magnitude is extremely narrow (10⁻⁴ m/m) due to the limited net effective stress variation induced by the stressed production scenarios. The difference between the coupling techniques is negligible in terms of subsidence evolution at ground level but not in terms of compaction at reservoir level. Furthermore, the two-way coupled technique could be used for better development planning by integrating reservoir, drilling and completion management.

How to Approach Subsidence Evaluation for Marginal Fields: A Case History