AME 28:117-129 (2002)  -  doi:10.3354/ame028117

ABSTRACT: We used specially designed microcosms filled with natural substrate to study microbial colonization in a shallow aquifer. Sterilized sediments were exposed to 3 types of groundwater varying in physical, chemical and biological characteristics: (1) pristine groundwater (site PI 92); (2) groundwater in an observation well at a pristine site (OMV 11); and (3) contaminated groundwater at a landfill site (OMV 5). The number of suspended bacteria was always highest at the landfill site (4.0 ± 4.2 [standard deviation, SD] x 10⁶ cells cm^-3), i.e. on average 16 times higher than in the well water (2.5 ± 3.0 x 10⁵ cells cm^-3) and 96 times higher than in the pristine groundwater (4.1 ± 1.3 x 10⁴ cells cm^-3). Sediments in the microcosms were rapidly colonized and the total number of attached bacteria after 10 mo of exposure was highest at the landfill site (1.8 ± 0.4 x 10⁸ cells cm^-3) followed by the sediment incubated in well water (1.5 ± 0.5 x 10⁸ cells cm^-3) and in pristine groundwater (5.0 ± 1.5 × 10⁷ cells cm^-3). As estimated from image analysis, attached cells from the landfill site were on average characterized by higher cell carbon contents (28 ± 36 fgC cell^-1) than at the well water (24 ± 23 fgC cell^-1) and the pristine groundwater site (21 ± 23 fgC cell^-1). The ratio of attached to suspended bacteria after 10 mo of exposure was highest in the microcosm incubated in pristine groundwater (1657:1) and lowest at the contaminated site (59:1). On the basis of our results we emphasize the importance of attached microbial communities in porous subsurface systems and underline the need for groundwater as well as sediment samples for a serious microbiological characterization of the subsurface. Furthermore, the ratio of attached to suspended bacteria in shallow aquifer systems is suggested to be an indicator of prevailing nutrient concentrations.

KEY WORDS: Bacteria · Subsurface · Biofilms · Colonization · Microcosm · Groundwater · Method