Heterotrophic bacteria within the water column of rivers obtain C from organic molecules that are part of the DOC pool. Bacterioplankton production is expected to be of importance mainly in lowland rivers, where bacterial doubling times potentially can exceed washout rates, provided that DOC is of sufficient quality and quantity and other environmental conditions are favorable. However, microbial production in the water column generally is much less than ben-thic production. In the case of the River Spree, mentioned earlier, bacterial production in the sediments was nearly three orders of magnitude higher than in the water column. Estimates of production by bacterioplankton vary over at least two orders of magnitude (Figure 7.15), ranging from 0.14-0.52|g C L 1 h 1 in black-water rivers in the southeastern United States, the Amazon and Orinoco Basins (Benner et al. 1995, Castillo et al. 2004) to 40-75 |g C L1 h1 in anthropogenically enriched rivers like the Maumee and the Ottawa in Ohio (Sinsabaugh et al. 1997). Values for planktonic bacterial abundance are less variable than bacterial production, ranging between 0.05 x 106 cells mL 1 in the Kuparuk River in Alaska and 5 x 106 cells mL 1 in more disturbed systems like the Hudson River and the River Rhine (Findlay et al. 1991, Admiraal et al. 1994).
Environmental variables likely to influence planktonic bacterial production, in addition to the amount and quality of DOC, include nutrients, temperature, and pH. Greater availability of nutrients can directly enhance bacterial production whenever the OM that serves as a C source has a low nutrient content, thus forcing bacteria to obtain nutrients from elsewhere (Findlay 2003). Nutrient supply can also benefit bacterial production indirectly, by enhancing primary production and thus the C supply. The addition of various combinations of nutrients to bacter-ioplankton samples cultured in situ has shown
P limitation in clearwater rivers of the Amazon basin (Farjalla et al. 2002) and in clear and black-water rivers draining the Guayana Shield in the Orinoco Basin (Castillo et al. 2003). In the Rio Negro, a blackwater tributary of the Amazon River, bacterial production was enhanced by the simultaneous addition of glucose, ammonium, and P (Benner et al. 1995), which suggests that C, N, and P were available in approximately the stoichiometric ratio needed by bacteria. The influence of temperature is reflected in seasonal variation in bacterial metabolism, with higher values during the warmer months (Edwards and Meyer 1986, Findlay et al. 1991). The effect of pH is ambiguous, as some studies suggest that bacterial extracellular enzymes can be negatively affected by acidification (Benner et al. 1989, Schoenberg et al. 1990); however, some exoen-zymes exhibit an optimum at low pH (Munster et al. 1992). Low pH may also affect the availability of DOC to bacteria because the cell membrane is more permeable to hydrophilic compounds, and at very low pH, humic compounds become more hydrophobic and thus are less permeable substrates (Edling and Tran-vik 1996).
DOC quality and composition likely are the most important factors limiting production of bacterioplankton. In fact, correlations of bacterial production with total DOC are rare (Findlay 2003), probably because bacterial production reflects the amount of labile DOC rather than the total (Findlay et al. 1998). The evidence suggests that some constituents of the heterogeneous mix of molecules that comprises DOC are more available than others, and preferential removal of LMW DOC has been reported in several studies (Kaplan and Bott 1982, Meyer et al. 1987, Kaplan and Bott 1989). Peptides and sugars support a large fraction of bacterial production, evidence of the high bioavailability of these compounds, which can be found free or forming complexes with humic molecules (Foreman et al. 1998, Findlay and Sinsabaugh 1999). Bioavail-ability is also related to the proportion of aliphat ic compounds, which are more abundant in algal and macrophyte leachate than in leachate from woody plants (Sun et al. 1997). However, there is evidence that bacteria can utilize humic substances, which are rich in aromatic components (Moran and Hodson 1990, Tranvik 1990, Wetzel 1995), but at a lower growth efficiency compared to LMW DOC (Amon and Benner 1996).
Direct measurement of the amount of labile DOC is difficult because DOC is composed of many different compounds, many of which are not well characterized (Volk et al. 1997, Seitzin-ger et al. 2005). DOC lability can vary with its sources and the type of molecules (Findlay and Sinsabaugh 1999) and be influenced by nutrient availability and the composition of the bacterial assemblage (Cottrel and Kirchman 2000, Findlay 2003). Moreover, the timescales of inputs can vary dramatically, ranging from highly episodic storm-induced transport to gradual leaching of organic material into stream channels or adjoining sediments. Although different methods have been used to directly measure the concentrations of individual compounds or sources of DOC, the amount of labile DOC routinely is estimated from bacterial production and respiration over extended incubations (Meyer et al. 1987, Amon and Benner 1996). DOC lability has also been measured as the decline in DOC concentrations due to bacterial uptake in batch incubations (Meyer et al. 1987) or in biofilm reactors where indigenous bacterial populations are maintained over time (Kaplan and Newbold 1995, Volk et al. 1997). One can also relate DOC composition to bacterial growth by measuring the activity of extracellular enzymes, because they are an indicator of substrate availability when C is limiting (Sinsabaugh et al. 1997).
Estimates of labile DOC availability based on bacterial growth in batch incubations suggest that, on average, 19% of riverine DOC is labile, although this value could be < 1% in blackwater rivers (Sondergaard and Middleboe 1995). Because this study included some large European rivers that receive wastewater inputs, the average value of 19% labile DOC may not be representative of less disturbed rivers. Amon and Benner (1996) estimated that 1.4-7.5% of DOC in the Amazon was labile. Based on DOC disappearance in batch incubations, del Giorgio and David (2003) estimated that on average 10% of riverine DOC was consumed during 1-3 day incubations; and this estimate was low compared with lakes and marine systems, possibly because the origin of DOC in the terrestrial landscape and its flow path into riverine systems provide ample opportunities for processing in soils. Using a biofilm reactor, Volk et al. (1997) estimated that on average 25% of DOC of a headwater stream was labile; the composition of the labile portion included humic substances (75%), carbohydrates (30%), amino acids (4%), and DOC > 100 kDa (39%). Carbohydrates were mostly polysaccharides and were bonded to humic compounds, as were most of the amino acids. Although only 25% of streamwater humic compounds were consumed in the bioreactor, they represented a large fraction of the labile DOC (75%), which contradicts the general view that humic substances are refractory.
Although the supply of organic C and nutrients in lotic ecosystems likely is limiting to bac-terioplankton growth much of the time, microbial respiration in the water column and sediments of higher-order rivers may nonetheless be sufficient to mineralize large amounts of DOC. The lower Hudson River is almost always supersaturated with CO2, evidence of an excess of respiration over production, and DOC concentrations decline as one proceeds downriver, evidence of its mineralization by microbial activity (Cole and Caraco 2001). In the lower sections of large rivers, significant quantities of organic C in transit are returned to the atmosphere by CO2 evasion. These results suggest that although some fraction of DOC likely is incorporated into aquatic food webs through bacterial production, a greater fraction of riverine DOC is converted to CO2 by instream respiration.
Was this article helpful?