Microbial Community Analysis of Produced Water from Mid-Temperature Oil Reservoirs in North America

Abstract

Microbial enhanced oil recovery (MEOR) has recently garnered interest as a low-cost and environmentally-friendly alternative to thermal or chemical oil recovery techniques. The main targets of MEOR have been oil reservoirs with relatively low temperature (30-50°C). As preliminary investigation for a research project to develop MEOR technique for higher temperature (~80°C) oil reservoirs, we have investigated the indigenous microbial community structure of five different mid-temperature wells in Houston, TX, US (70-80°C) and Crossfield, AB, Canada (75°C). Produced water samples from Houston (4 spatially separated wells) and Crossfield (1 well, 1 oil-water separator) were analyzed for their community composition. Miseq sequencing of PCR-amplified V6–V8 region of 16S rRNA gene yielded 48,502-106,985 paired-end reads, which were processed with QIIME. These sequences were analyzed with additional 16S rRNA gene sequence data retrieved from NCBI’s SRA database. The additional data were Miseq amplicon sequencing data from 5 and 22 low-temperature oil reservoirs in Xinjiang, China (22-35°C) and Louisiana, US, respectively (28-45°C). Despite the geographical distance and difference in the chemical compositions, the OTU compositions of the two sites exhibited high similarity, as indicated by their close association in the NMDS plot. The major groups of recovered OTUs were associated with Clostridiaceae (18.4%), Methanosaetaceae (13.1%) and Pseudomonadaceae (16.8%), which have been recognized to be thermophilic, thermotolerant or spore-forming. The composition of the separator sample deviated greatly from the well samples. However, OTU network analysis showed that the separator sample was clustered with the sample directly from well in Crossfield, indicating the high similarity of microbial community between two sites was not due to contamination. The sequence data retrieved from the database exhibited different clustering patterns, as the samples from close geographical location clustered together in the NMDS plot and temperature apparently did not have significant role in shaping microbial community. The correlation network analyses (Spearman’s coefficient ρ < 0.5) revealed that the major OTUs in Houston and Crossfield samples formed a single tight network while OTUs from other sample formed total four separate networks. These observations suggest the importance of temperature as the dominant driving force for microbial selection in the thermophilic oil reservoirs.