Document: VAU-3-70-3

Bacterial information length as a predictor of stream health.

MCARTHUR, JVAUN* ¹ and R.C.TUCKFIELD ²

Savannah River Ecology Laboratory, University of Georgia, Aiken, SC 29802, USA ¹
Westinghouse Savannah River Co., Aiken, SC 29802, USA ²

Abstract:
The use of indicator organisms (both prokaryotic and eukaryotic) as evidence of stream health has been a common practice. The presence of certain bacteria, e.g., E. coli, is an immediate red flag. However, native bacteria control either directly or indirectly the rates of most ecosystem level processes and little is known concerning their distribution spatially or temporally. More specifically, the distribution of various bacterial genes should be indicative of various functions and these distributions may be different than those of specific bacteria because bacterial genes can be shared among diverse taxa. We demonstrate using the concept of information length that the distribution of bacteria and/or their genes does relate to aspects of stream health. Because bacteria are the most intimate organisms with the environment these patterns may be the most sensitive indicators of stream condition. We have measured the information length of antibiotic resistance and soluble methane monooxygenase (sMMO) in both pristine and disturbed coastal plain streams. sMMO genes code for the breakdown of methane and also the fortuitous breakdown of trichloroethylene (TCE). We demonstrate that the pattern of antibiotic resistance is tightly linked to heavy metal contamination and meets specific predictions of the information length hypothesis. These data suggest that the distribution of antibiotic resistance is a good indicator of stream health and the actual geographic distance downstream affected by metal pollution. The distribution of sMMO is directly related to the production of methane. In our disturbed streams less organic matter accumulates and thus fewer sMMO genes can be found. These data indicate the power of the information length concept and suggest that various predictions arising from the concept can be used to address stream condition, stream function and processes.

Keywords: streams, bacteria, distributions, information length

This abstract is being presented at: 1:15 PM in session:
Oral Session #11: Trophic Cascades.