nitude of contamination from each of these sources will be highly specific to the type of drilling and sampling operation.

Biological contamination of subglacial aquatic environments can come from a variety of sources that contain not only microorganisms but also the substrates they use for growth (Alekhina et al. 2007). Drilling fluids may contain both microbiota and growth substrates. The release of this inoculum plus growth medium into the relatively warm (by comparison with the ice hole and surface environment), liquid-water conditions of a subglacial lake could result in a short-term phase of increased metabolism and growth. It is of critical importance to minimize the introduction of living exogenous microbes, and even of exogenous nucleic acids, to prevent changes in native microbial composition and to allow for proper investigation of the native microbial community (and not contaminants).

A variety of allochthonous microbes (“contaminants”) could be introduced into subglacial aquatic environments from microorganisms native to the surface environment or from humans or equipment brought to the sampling area. Microbes could be transferred deeper into the system through sampling and could potentially grow (e.g., in meltwaters produced during hot-water drilling). Microbes immured in glacial ice for long periods of time might conceivably proliferate once they are released back into the modern-day biosphere through natural melting processes or research activities.

In subglacial lakes, large differences in chemical properties among liquid-water strata are possible, including the presence of anoxia at depth in the water column and in the sediments. Releases of water during sample removal could enrich or inhibit communities in the water column or overlying ice. As in lakes elsewhere, the sediments underlying subglacial lakes are likely to contain orders-of-magnitude higher concentrations of microbes, nutrients, and metals. The benthic microbial communities also are likely to have a very different phylogenetic composition than those in the water column, and transfers during sampling could compromise subsequent measurements of water and ice samples (e.g., DNA clone library analysis).

In addition, research activities targeting one component of the environment may have the potential to cause contamination or damage of another. Once contamination reaches the aquatic environment, a variety of hydrodynamic circulation processes are likely to operate in the subglacial waters, and these localized inputs of contamination could be widely dispersed. This is particularly important for subglacial aquatic environments that are connected via subglacial hydrological networks. However, protocols or codes of environmental conduct may be developed to include specific measures to minimize or avoid such effects on the aquatic microbial ecosystems of these environments. If downstream sites are the initial targets of investigation, this may reduce the potential risk of contamination to other environments along a specific flow path.

Molecular biological and genomics approaches have made it routine to sample the presence and diversity of microbial communities from many environments throughout the biosphere. Of particular importance to these sampling regimes, especially those taking place in areas where biomass is predicted to be low, is the prevention of contamination of the environment and subsequent samples with exogenous life or nucleic acids. These precautions should seek to preserve both the subglacial aquatic environment and the integrity of the scientific samples.

Most protocols, that focus on preventing contamination of the environments and protecting the integrity of samples extracted from the environment have employed advancements in drilling technology to obtain the necessary depths and allow the

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement