The reduction of greenhouse gases emission is a growing concern of many industries. The oil and gas industry has a long commercial practice of gas injection, enhanced oil recovery (EOR) and gas storage. Using a depleted oil or gas reservoir for CO2 storage has several interesting advantages. The long-term risk analysis of the CO2 behavior and its impact on the environment is a major concern. That is why the selection of an appropriate reservoir is crucial to the success of a sequestration operation. Our modeling study, based on a synthetic reservoir, quantifies uncertainties due to reservoir parameters in order to establish a set of guidelines to select the most appropriate depleted reservoirs. Several production and sequestration scenarios are investigated in order to quantify key parameter for CO2 storage. The influence of parameters such as API gravity, heterogeneity (Dykstra-Parson coefficient), pressure support (water injection) and cap rock integrity are analyzed. Estimation of sequestration capacity is proposed through a sequestration factor (SF) estimated for different reservoir production drives. Multiple regression relationships were developed, allowing SF estimation. CO2 sequestration optimization highlights the best clean oil recovery strategy (CO2 injection and/or oil production). (C) 2004 Elsevier Ltd. All rights reserved.