The present work deals with an experimental investigation on methane hydrate and carbon dioxide hydrate formation in presence of two different types of sand, which acted as seabed simulators. The first typology of sand consists in pure quartz and is commonly used for laboratory experiments on gas hydrate. The other type is named TS-2 and originated from the Tunisian seabed of the Mediterranean Sea. It is silica-based (≈99%), however it also contains other elements and its grains have different shape, size and porosity. Experiments were carried out in order to verify if the specific characteristic of the seabed, in which the hydrate reservoir is present, may intervene or not on the CO2/CH4 replacement process and if such contribution may improve or reduce the process efficiency. Results proved that physical and chemical properties of materials which composed the seabed, may strongly intervene on the replacement process. In particular, experiments revealed that sand TS-2 acted as kinetic and thermodynamic inhibitor for methane hydrate formation, while it promoted CO2 hydrate formation under the kinetic point of view. In this sense, sand TS-2 represents a strong ally for improving the replacement efficiency, due to its capability to both improving the kinetic of the process and reducing the methane hydrate re-formation phenomena. The present study revealed that, with a simple analysis of properties of sediments containing hydrate reservoirs, it would be possible to establish the convenience of intervening with a replacement strategy instead of a classical application for simple methane recovery.