4.0 Microbial Communities

Impacts on Quality of Inland Wetlands of the United States:
A Survey of Indicators, Techniques, and Applications of Community Level Biomonitoring Data
Excerpts from Report #EPA/600/3-90/073
(now out of print)

4.1 USE AS INDICATORS

As used here, "microbes" includes bacteria, viruses, yeasts, and microscopic fungi. In wetlands, these have most often been measured indirectly, in the pursuit of estimates of microbe-related processes relevant to element cycling, such as decomposition and denitrification. Although microbial responses to contaminants have been summarized for other surface waters (e.g., Cairns et al. 1972) and upland soils (Baath 1989), few studies have looked at microbial community structure specifically in wetlands, or identified particular microbes as indicators of wetland ecological condition.

Following are discussions of community responses to various stressors. Although we have included some discussion of decomposition rates (an indirect measure of microbial biomass), that process is mainly discussed in Chapter 13.

Enrichment/eutrophication. Microbial abundance and community structure are profoundly affected by trophic status. Enrichment typically results in major increases in microbial abundance (e.g., Tate and Terry 1980) and sometimes richness (Pratt et al. 1989). Enrichment with nitrogen in particular may affect microbial communities, at least in riverine detritus-based systems. Adding nitrogen to streams increased leaf decomposition, microbial biomass, and microbial activity; added phosphorus alone had no effect (Fairchild et al. 1984). Photosynthetic protozoans appear to respond most immediately to nutrient additions (Pratt and Cairns 1985a). However, effects on species richness and community structure have not been extensively studied in most wetland types, and little is known of "indicator taxa" whose use might be most appropriate for signaling enrichment in wetlands.

Microbial colonization rates in a series of shallow Florida ponds was used by Henebry and Cairns (1984) to indicate trophic status. In pond systems (Schmider and Ottow 1985), enrichment increased microbial population densities and number of facultative-anaerobic bacteria (e.g., Streptococci, Enterobacteriaceae and aerobic spore forms, e.g., Bacillus spp., Pseudomonas alcaligenes, and Aeromonas spp.). Mesotrophic ponds had highest numbers of fluorescent pseudomonads. Oligotrophic water had more denitrifiers (Pseudomonas fluorescens and Vibrio spp.).

Organic loading/reduced DO. Given the naturally large organic concentrations in wetlands, it is probable that unique or adapted microbial communities are sometimes present (Felton et al. 1966). Indeed, microbial communities respond strongly to organic additions (Tate and Terry 1980). However, few studies have investigated the effects of increased organic loading and decreased dissolved oxygen on wetland microbial community structure. Low dissolved oxygen (DO) is tolerated or preferred by some taxa, so changes in DO probably trigger significant shifts in community composition.

In cypress domes that received wastewater, Dierberg and Ewel (1984) found faster rates of leaf decomposition (a mainly microbial process). In other surface waters, considerable attention has been focused on coliform bacteria and nuisance growths of Sphaerotilus spp. Large populations of these microbes characteristically develop where sewage has been introduced.

Contaminant Toxicity. The literature concerning response of microbial community structure to heavy metals is summarized by Baath (1989), who includes one study from presumably wetland (organic) soils. That study found a reduction in bacterial abundance at copper concentrations exceeding 275 ug/g. Evidence summarized from non-wetland soils indicates that heavy metal contamination reduces taxonomic richness of the microbial community and causes distinct shifts in taxa; some taxa with potential indicator value are identified by Baath (1989). A shift toward more fungal and gram-negative (vs. gram-positive) taxa may occur, but there is apparently little change in the overall ratio of mycorrhizal to decomposer fungi.

Addition of oils and synthetic organics may result in increased abundance of microbes, particularly species known to degrade and be sustained by petroleum (Walker and Colwell 1977). Microbes, particularly those associated with wetland plants, can be largely responsible for detoxifying some synthetic organic compounds (Hodson 1980) such as pentachlorophenol (Pignatello et al. 1985), and the herbicide glyphosate (brand name, Roundup or Rodeo) (Goldsborough and Beck 1989) as well as detergents (Federle and Schwab 1989).

Thermal Alteration. Although microbial communities are highly sensitive to temperature, few studies have directly examined the effects of thermal stress on community structure in wetlands. In other surface waters, Thermus aquaticus has been found only where heated effluents were introduced (Brock and Yoder 1971).

Acidification. Bogs and other acidic wetlands in many cases contain relatively low richness of microbial taxa (Stout and Heal 1967) and secondary production of microbes can be reduced under such conditions (Benner et al. 1985). However, naturally acidic bogs can have well-adapted, moderately diverse microbial communities (Henebry et al. 1981). Zooflagellate microbes and the ratio of bacteria to fungi can decline with acidification (Leivestad et al. 1976).

Fragmentation of Habitat. We found no explicit information on microbial community response to fragmentation of regional wetland resources. A study of microbial colonization at various distances from an intermittently flooded Virginia wetland (McCormick et al. 1987) found that fewer species colonized introduced substrates that were located a far distance from the wetland; similarity of microbial communities also decreased with increasing distance. One can surmise that as the distance between wetlands with potential microbial colonizers becomes greater, microbial taxa with narrow environmental tolerances and which do not disperse easily might disappear first.

Salinization; Sedimentation/Burial; Turbidity/Shade; Vegetation Removal; Dehydration; Inundation. We found no explicit information on microbial indicators or community response to these stressors in wetlands. From knowledge of microbial responses in other surface waters, it appears likely that microbes in wetlands could respond dramatically to many of these stressors. Undoubtedly data are available from non-wetland surface waters that identify indicator assemblages and document microbial community response to many of these stressors (e.g., Krueger et al. 1988). However, reviewing these was beyond the scope of the present effort, and the transferability of these data to wetlands remains uncertain.

4.2 SAMPLING METHODS AND EQUIPMENT

It is particularly important when using microbes as indicators of anthropogenic disturbance that the comparison wetlands are of about equal age and have similar sedimentary regimes and vegetation densities. This is because microbial communities respond strongly to changes in sediment organic matter, which usually accumulates with wetland age. For example, recently disturbed ponds were found to have fewer microbe species than did natural and older reclaimed ponds on a surface-mined site (Pratt and Cairns 1985b). However, microbial communities in ponds more than two years old were indistinguishable from those in older reclaimed, unreclaimed, and natural ponds despite differences in water quality. Other factors that could be important to standardize among collections of microbial communities include:

light penetration (water depth, turbidity, shade), temperature, sediment oxygen, baseline chemistry of waters (particularly pH and conductivity), detention time, current velocity, vegetation density, dominant vegetation species, and moisture (e.g., time elapsed since last runoff, inundation, or desiccation event).

Replication requirements for microbial collections are usually significant, due to extremely great spatial and temporal variability of microbial density and diversity. Protozoan "blooms" are more likely to occur in wetlands than in rapidly flowing surface waters, and entirely different communities may exist without apparent cause within millimeters of each other (Carlough 1989). Sampling can occur at any season, but microbial biomass is often greatest in late summer (e.g., Murray and Hodson 1985) and autumn. Standard protocols are available; one is the manual by Britton and Greeson (1988).

Bacterial and fungal abundance are usually estimated as colony forming units (CFU) using plate count techniques. However, concerns have been raised about the validity of this technique for monitoring fungi; use of low-nutrient culture media (rather than the typical enriched media) are also recommended (Baath 1989).

Microbial communities in wetlands are generally collected from sediment samples, water column samples, artificial substrates, or natural organic substrates (e.g., leaf packs). These are described as follows.

Sediment sampling. Sediment sampling of microbial communities can be conducted in all types of wetlands. Dierberg and Brezonik (1982), working in Florida cypress swamps, sampled microbial communities of surface sediments using a sterile piston corer and a plastic syringe with an attached tube.

Water column sampling. Any wetland types that have surface water permanently or seasonally can be sampled using sterile, volumetric containers.

Artificial substrates. Plexiglass plates, acrylic rods, polyurethane foam, or similar inert, sterile surfaces can be placed in any wetlands that have surface water permanently or seasonally, and allowed to be colonized by microbes over a period of several weeks (e.g., Goldsborough and Robinson, 1983, Pratt et al. 1985, Pratt and Cairns 1985b). Substrates are then retrieved and community structure is analyzed. The use of artificial substrates may be a more practical method of sampling protozoa in wetlands than is direct collecting, because of the diversity of microhabitats in wetlands (Henebry and Cairns 1984).

Natural substrates. Natural organic substrates typically contain great numbers of microbes. Consequently, microbial communities have often been collected directly from detrital material, or have been indirectly monitored through measurement of leaf litter decomposition rates. Microbial biomass can also be indirectly monitored by analyzing relative levels of adenosine triphosphate (ATP), e.g., the ratio nM ATP/g ash-free dry weight (Meyer and Johnson 1983). Activity of certain microbial communities was estimated by measuring relative rates of lipid biosynthesis (Fairchild et al. 1984). Adenylate (ATP, ADP, AMP) energy charge ratios in microbes also have been suggested as metrics of ecosystem stress (Witzel 1979).

4.3 SPATIAL AND TEMPORAL VARIABILITY, DATA GAPS

In no region of the country, and in no wetland type, have data on microbial community structure been uniformly collected from a series of statistically representative wetlands. Thus, it is currently impossible to state what are "normal" levels for parameters such as seasonal density, species richness, and their temporal and spatial variability. Even qualitatively, lists of "expected" wetland microbial taxa have not been compiled for any region or wetland type.

Limited data suggest that among-wetland variability in microbial community structure is less than variability in vascular plant community structure, but that clear differences exist in microbial communities of marshes, fens, and bogs (Henebry et al. 1981). Microbial density, species richness, and/or colonization rates can be higher in some wetlands than in other surface waters (Duarte et al. 1988, Henebry et al. 1981).

Studies that have compared microbial communities among wetlands (spatial variation) apparently include only Henebry et al. (1981, 1984) and Pratt et al. (1989). The former study, covering 13 Michigan wetlands over a 5-year period, found a range of 93 to 365 protozoan species; Sorenson's similarity index ranged from 0 to 40, with a mean of 21. The latter study, covering 28 Florida ponds, found a range of 112 to 410 species, with a mean of 338 species in non-artificial ponds. Functional group structure of the resident microbial fauna changed slightly from year to year, but wetlands in the same geographic region and experiencing the similar climatic patterns had similar proportions of species in each functional group (Pratt et al. 1989). Microbial densities can vary by 2 to 5 orders of magnitude between sediments, aquatic plants, and the water column (Kusnetsov 1970).

Another study, which examined only one wetland complex (Okefenokee Swamp, Georgia) reported that microbial biomass in sediment ranged from 1 to 28 micrograms gram (dry weight) (Murray and Hodson 1984). A third study, Felton et al.(1967), from Louisiana, reported microbial densities of up to 108.