The microbial community structure in petroleum-contaminated sediments corresponds to geophysical signatures

Abstract

The interdependence between geoelectrical signatures at underground petroleum plumes and the structures of subsurface microbial communities was investigated. For sediments contaminated with light non-aqueous-phase liquids, anomalous high conductivity values have been observed. Vertical changes in the geoelectrical properties of the sediments were concomitant with significant changes in the microbial community structures as determined by the construction and evaluation of 16S rRNA gene libraries. DNA sequencing of clones from four 16S rRNA gene libraries from different depths of a contaminated field site and two libraries from an uncontaminated background site revealed spatial heterogeneity in the microbial community structures. Correspondence analysis showed that the presence of distinct microbial populations, including the various hydrocarbon-degrading, syntrophic, sulfate-reducing, and dissimilatory-iron-reducing populations, was a contributing factor to the elevated geoelectrical measurements. Thus, through their growth and metabolic activities, microbial populations that have adapted to the use of petroleum as a carbon source can strongly influence their geophysical surroundings. Since changes in the geophysical properties of contaminated sediments parallel changes in the microbial community compositions, it is suggested that geoelectrical measurements can be a cost-efficient tool to guide microbiological sampling for microbial ecology studies during the monitoring of natural or engineered bioremediation processes.

FIG. 1.

Detailed map of the field study site within the Carson City Park, Carson City, MI. The positions of the VRPs at the contaminated (5) and background (9) locations are noted. Samples for microbial analysis were collected at the same locations, adjacent to the VRPs. To the north of the sampling site is the former petroleum refinery where the leaking underground storage tanks were located. The thicknesses (in meters) of the free-phase LNAPLs over the study area are given, as well as the boundary of residual LNAPL contamination. The location of a drainage ditch running through the site is also shown.

FIG. 2.

Soil lithology, soil conditions, and conductivity profiles of the petroleum-contaminated site (VRP5). (A) Percentages of silt and clay, sand, and gravel. (B) Residual, free, and dissolved LNAPL phases; water saturation; and sampling locations for 16S rRNA gene libraries 5V, 5F, 5C, and 5S. The inverted triangle denotes the level of the groundwater table. (C) Levels of conductivity expressed in millisiemens per meter.

FIG. 3.

Soil lithology, soil conditions, and bulk conductivity profiles of the background site (VRP9). (A) Percentages of silt and clay, sand, and gravel. (B) Water saturation and sampling locations for 16S rRNA gene libraries 9V and 9S. The inverted triangle denotes the level of the groundwater table. (C) Levels of conductivity expressed in millisiemens per meter.

FIG. 4.

Rarefaction curves generated for 16S rRNA gene libraries from VRP5 and VRP9 sediment samples. Error bars indicate 95% confidence intervals. Clones were grouped into phylotypes based on DNA sequence analysis, with members of phylotypes showing ≥97% sequence identity and the same RFLP patterns. Because of the omission of the chimeric clones, the total numbers of clones considered for the rarefaction analysis were 100, 100, 97, 95, 84, and 100 for the 5V, 5F, 5C, 5S, 9V, and 9S libraries, respectively.

FIG. 5.

Correspondence analysis of the conductivity values, the lithologies, and the fractions of the different phylotypes observed in association with the 16S rRNA gene libraries from the different depths at the contaminated (VRP5) and uncontaminated (VRP9) sites. Numbers along the axes represent the scores as calculated by correspondence analysis based on the percentages of gravel, sand, silt, and clay present at the elevations corresponding to each library, the conductivity data and the fraction of hydrocarbon degraders as shown in Table 1, and the percentage of each bacterial group present in each library as shown in Table 2. CA1, correspondence analysis axis 1; CA2, correspondence analysis axis 2.