Large-scale geological storage of CO₂ is likely to bring CO₂ plumes into contact with existing wellbores and faults, which can act as pathways for leakage of stored CO₂ Modeling the flux of CO₂ along a leaky pathway requires transport properties along the pathway. We provide an approach based on the analogy between the leakage pathway in wells that exhibit sustained casing pressure (SCP) and the rate-limiting part of the leakage pathway in any wellbore that CO₂ might encounter. By using field observations of SCP to estimate transport properties of a CO₂ leakage pathway, we obtain a range of CO₂ fluxes for the cases of buoyancy-driven (post-injection) and pressure-driven (during injection) leakage. The fluxes in example wells range from background levels to three orders of magnitude higher than flux at the natural CO₂ seep in Crystal Geyser, Utah. We estimate a plausible range of fault properties from field data in the Mahogany Field using a shale gouge ratio correlation. The estimated worst-case CO₂ flux is slightly above background range. The flux along fault could be attenuated to zero by permeable layers that intersect the fault. The attenuation is temporary if layers are sealed at other end. Counterintuitively, greater elevation in pressure at the base of the fault can result in less CO₂ leakage at the top of the fault, because the capillary entry pressure is exceeded for more permeable layers. Since non-negligible leakage rates are possible along wellbores, it is important to be able to diagnose whether leakage is occurring. Concurrent pressure and temperature measurements are especially valuable because they independently constrain the effective permeability of a leakage path along wellbore. We describe a simple set of coupled analytical models that enable diagnosis of above-zone monitoring data. Application to data from a monitoring well during two years of steady CO₂ injection shows that the observed pressure elevation requires a model with an extremely large leakage rate, while the temperature model shows that this rate would be large enough to raise the temperature in the monitoring zone significantly, which is not observed. The observation well is unlikely to be leaking. Extraction of brine from the aquifer offers advantage over standard storage procedure by greatly mitigating pressure elevation during CO₂ injection. A proper management of the injection process helps reduce the risk of leakage associated with wellbores and faults. We provide strategies that optimize the injection of CO₂ which involve extraction of brine in two scenarios, namely injecting dissolved CO₂ and supercritical CO₂. For surface dissolution case we are concerned with bubble point contour, while for supercritical CO₂ injection we are concerned with breakthrough of CO₂ at extractors. In a surface dissolution project, the CO₂ concentration front shape when it reaches the saturation pressure contour defines the maximum areal extent of CO₂-saturated brine and hence the aquifer utilization efficiency. We illustrate the reduction of utilization efficiency due to heterogeneity of the aquifer. We develop an optimal control strategy of the injection/extraction rates to maximize the utilization efficiency. We further propose an optimal well pattern orientation strategy. Results show that the approach nearly compensates the reduction of utilization efficiency due to heterogeneity. In a supercritical CO₂ injection that involves brine extraction, the problem of avoiding breakthrough of CO₂ at extraction wells can be addressed by optimizing flow rates at each extractor and injector to delay breakthrough as long as possible. We use the Capacitance-Resistive Model (CRM) to conduct the optimization. CRM runs rapidly and requires no prior geologic model. Fitting the model to data recorded during early stages of CO₂ injection characterizes the connectivities between injection and brine-extraction wells. The fitted model parameters are used to optimize subsequent CO₂ injection in the formation. Field illustration shows a significant improvement in CO₂ storage efficiency. / text