No group of marine organisms has received as much attention as bacteria, and the advances in microbial ecology are indisputable (Epstein and Rossel, 1995). The importance of bacterial activity within surface sediments is now well established. Recent studies have confirmed the presence of activity of bacteria down to several hundred meters below the seafloor (mbsf) in deep-sea sediments (Cragg et al., 1992, 1995, 1996; Parkes et al., 1990). However, little is known about the factors that govern the detailed distribution of bacteria numbers in deep-sea sediments. One reason for this is that few geochemical laboratories possess the equipment and expertise required for enumeration of bacteria by classical microbiological techniques. Hence, there seems to be a demand for a quick and reliable method for estimation of bacterial biomass.
Measurement of Adenosine 5'-Triphosphate (ATP) has been used as a proxy for total living biomass in soil (e.g., Webster et al., 1984; Ciardi and Nannipieri, 1990; Pangburn et al., 1994) and as an estimate of bacteria biomass in coastal marine sediments (Bancroft et al., 1976; Bulleid, 1978), where living cell numbers typically exceed 1010 per gram (Alongi, 1992; Schallenberg et al., 1989) and the concentration of ATP reach several thousand nanograms per gram of dry sediment (Karl, 1980). However, to the authors knowledge, ATP measurements have not been used previously as a proxy for total bacterial biomass in deep-sea sediments where the total bacterial number may be 1-5 order of magnitude less (Cragg and Parkes, 1994; Cragg, 1994). All living cells contain ATP, and the use of ATP as a proxy for bacteria numbers in sediments is based on the assumption that other benthic organisms have been removed or are absent. Benthic organisms are seldom found below a few decimeters below the sediment-surface, and ATP is rapidly degraded by enzymes on an organism's death (Pangburn et al., 1994); hence only ATP measurements of surficial sediments may be affected by the presence of benthic organisms.
The objectives of this study were to modify existing methods for ATP analysis to enable analysis of deep-sea sediments, and to examine geochemical factors that may govern the detailed distribution of ATP (bacteria numbers by inference).