Recent surges in machine learning applications and data-driven modeling are reshaping reactive transport simulations for subsurface systems, including carbon dioxide sequestration, enhanced oil and gas recovery, critical mineral extraction, geothermal energy storage, and hydrogen storage. Integrating machine learning with reactive transport modeling is unlocking new avenues for simulating and predicting geochemical processes in porous and fractured media. Traditional reactive transport models, while physically rigorous, are often computationally intensive and intractable for large-scale prediction. In contrast, machine learning offers pathways to accelerate simulation workflows, uncover hidden patterns, and enhance model explainability. Moreover, advanced image processing, data assimilation, and sensitivity analysis powered by machine learning enable more accurate predictions by extracting pore-scale information and integrating experimental observations into model calibration. This session highlights advances at the interface of data-driven modeling and machine learning for reactive transport. We welcome submissions on machine learning enhanced techniques for geochemical prediction across spatial and temporal scales.

• GC - Global Environmental Change