Critical review of subsurface processes governing CO2 leakage mechanisms.

Geological storage of carbon dioxide is widely regarded as an effective approach to reduce atmospheric CO2. Several geological formations have been investigated for CO2 sequestration, including depleted hydrocarbon reservoirs, unmineable coal seams, and basaltic formations that enable mineral trapping. Confined saline aquifers are generally viewed as a favorable option because of their global distribution. Storage safety in confined saline aquifers remains a major public concern due to significant hydrological challenges, particularly the potential leakage of brine and CO2 into overlying freshwater aquifers. The leakage processes of brine and CO2 have been extensively studied, yet their distinct mechanisms are often conflated in the literature. Brine leakage is primarily pressure-driven, whereas CO2 leakage is governed by saturation dynamics. Leakage may also arise from compromised aquiclude integrity, microfractures, or poorly sealed abandoned wells that act as conduits for vertical fluid migration. In response, numerous analytical, semi-analytical, and numerical models have been developed to describe and quantify leakage behavior. This review critically examines these modeling approaches, highlighting their underlying assumptions, applicability, and limitations, and identifies key knowledge gaps to guide future research on subsurface CO2 containment.