Basaltics rocks harbor diverse and active microbial populations when associated temperatures are consistent with their development. However, they are rarely considered in engineering operations as carbon capture and storage technology (CCS), while they can potentially alter the storage conditions and hence the fate at depth of the injected gas. During this PhD work the first monitoring of the microbial diversity hosted in a deep basalt aquifer (400-800 m depth) was carried out on a CCS pilot site at Hellisheidi in Iceland. Microbial communities have been described before gas injection and their evolution was monitored during two successive injections of 174.6t of pure CO2 and 65t of CO2+H2S+H2. Before the injections, phylogenetic affiliations suggest that most of the detected microorganisms are consistent with subsurface environments and therefore endemic to the aquifer. Metabolisms inferred to the OTUs identified are consistent with the geochemistry of water, a result reinforced by statistical analysis, pointing out a potential link between the carbonated system and the structure of the microbial populations. Over the injections, analyses highlight variations in the composition of microbial communities that seem to be impacted only by the injection of pure CO2. Particularly, it records a "bloom" of a sequence affiliated to an autotrophic bacterium oxidazing Fe(II), Sideroxydans lithotrophicus. The stimulation of autotrophic metabolisms during the injection of pure CO2 updates the question of whether some of the CO2 was converted into biomass in the subsurface, hence providing an alternative way for CO2 storage.