Carbon dioxide sequestration into a deep aquifer is considered one of the most effective methods to solve the global warming problem. To realize the geological sequestration of CO₂, we need to reduce injection cost and increase the accuracy of characterizing reservoirs for CO₂ sequestration using inexpensive geophysical exploration methods. The interpretation of geophysical exploration records needs corresponding relationships between physical properties observed from geophysical exploration and CO₂ saturation and phase state. We monitored the behavior of gas, liquid, and supercritical CO₂ injected into a sandstone specimen saturated with brine by measuring resistivity variations. We made an experimental apparatus that reproduces the pressure of CO₂ sequestration reservoirs. A cylindrical sample of Berea Sandstone (5 cm [1.9 in.] in diameter and 12 cm [4.7 in.] in length) was used in this experiment. Two current electrodes were set on both ends of the specimen. Five electrodes for measuring electric potential were installed on the side of the rock specimen. The side of the specimen was coated with silicone rubber. Carbon dioxide (gas, liquid, and supercritical phases) was flooded through the sandstone specimen at three different flow rates. We monitored the behavior of the resistivity as it changed over time in the sandstone as affected by the CO₂. The saturations estimated from resistivity are nearly equal to the values calculated from actual outflow volumes. This result shows the high reliability of electric exploration to monitor CO₂ saturation. The replacement ratio and storage volume of liquid CO₂ are higher than those of supercritical CO₂.