Water: Monitoring & Assessment
13.0 Biological Process Measurements in Wetlands
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)
13.1 USE AS INDICATORS
This discussion addresses biological processes that are commonly monitored in inland wetlands. "Processes" here are considered to be synonymous with wetland "functions." Included are litterfall and decomposition, nutrient translocation, growth and production, and respiration. We have limited consideration mainly to studies where these processes have been monitored for an entire wetland, not just a dominant species or community within the wetland. Relatively few studies have monitored wetland biological processes in the context of evaluating a specific anthropogenic stressor, as has Mader et al. (1988).
Although an understanding of wetland processes and their vulnerability to anthropogenic stressors is fundamental for predicting future impacts, the limited evidence to date suggests that biological processes usually respond only weakly and slowly to stressors in wetlands. This may be because biological processes represent the net result of many potentially compensating mechanisms within biological communities (Schaeffer et al. 1988, Schindler 1987). In contrast to changes in community structure which tend to occur gradually, changes in processes, when they ultimately occur, may occur suddenly and catastrophically. Perhaps with further testing and development of new ways to measure and quantify biological processes, their utility to regulatory monitoring programs will increase. Advantages and disadvantages of use of wetland ecosystem processes as indicators of ecological condition are shown in Appendix A.
Enrichment/Eutrophication. The effects of enrichment on annual productivity, decomposition and denitrification have been studied primarily in cypress dome and northern bog wetlands. Responses are generally typical of what has been found in other aquatic systems--increased productivity with "moderate" enrichment and a decline in productivity with "severe" enrichment.
Effects of enrichment on decomposition rates are highly variable, with both increased decomposition and no effect reported (Almazan and Boyd 1978, Andersen 1979, Chamie 1976, Fairchild et al. 1984, Farrish and Grigal 1988, Meyer and Johnson 1983, Richardson et al. 1976). Differing conclusions may be due to differences in current velocity, leaf type, temperature, fertilizer type, ambient water quality, and other factors. Enrichment of wetlands with nitrogen-rich runoff may lead to an increased proportion of nitrous oxide release (vs. N2 release), which is of potential concern because even small changes in the production of nitrous oxide are potentially significant considering the role of this gas in destroying stratospheric ozone (Hahn and Crutzen 1982).
Enrichment commonly increases secondary production. For example, aquatic invertebrate production was correlated with enrichment (total phosphorus concentration) in Plante and Downing's (1989) analysis of aquatic bed community data from 51 lakes (164 samples) from temperate regions of the world.
Organic Loading/Reduced DO. The effects of severe organic loading, e.g., from wastewater outfalls, on annual productivity have been studied primarily in cypress dome and northern bog wetlands, and results were similar to the above. With regard to decomposition, Brinson et al. (1981) reviewed the available literature and concluded that decomposition in wetlands should occur most rapidly with aerobic conditions under some optimum regime of wetting and drying; alternating conditions of aerobic and anaerobic result in slower decomposition.
Contaminant Toxicity. The literature summary by Baath (1989) reports heavy metal-induced impairment of several microbial processes, such as respiration, phosphatase enzyme activity, denitrification (Grant and Payne 1982) and decomposition of leaf litter (Jackson and Watson 1977), in wetland soils. In one case enrichment has been demonstrated to mitigate toxicity effects (Fairchild et al. 1984). In general, the relative toxicity of metals to microbial processes decreases in the order Cd> Cu> Zn> Pb (Baath 1989). Cadmium in shrub wetlands can interfere with nitrogen fixation (Wickliff et al. 1980). The effects of metals on primary and secondary production, and the effects of other contaminants on other processes, have not been widely studied in wetlands.
Acidification. The effects of acidification on biological processes have generally not been studied in inland wetlands. Long-term decomposition rates, particularly of the most refractile litter components, are generally slower in acidic water bodies (Friberg et al. 1980), and few kinds of decomposer bacteria operate effectively below pH 4 (Doetsch and Cook 1973). Artificial acidification has been shown to decrease the decomposition rate of litter from an herbaceous wetland plant (Leuven and Wolfs 1988), but the degree of inhibition may depend on the buffering capacity of the litter (Gallagher et al. 1987). Increasing the pH by adding lime can speed decomposition in acidic wetlands (Ivarson 1977). Acidification can also affect nitrification rates in wetlands (Dierberg and Brezonik 1982), and secondary production. Aquatic invertebrate production was correlated inversely to pH in Plante and Downing's (1989) analysis of aquatic bed community data from 51 lakes (164 samples) from temperate regions of the world.
Salinization. The effects of salinization, e.g., from irrigation return water and oil drilling wastes, on biological processes have generally not been studied in inland wetlands.
Sedimentation/Burial. The effects of excessive sedimentation on biological processes have generally not been studied in inland wetlands. Based on studies in other surface waters, respiration is likely to increase initially and decomposition rates may decrease.
Turbidity/Shade, Vegetation Removal. The impacts of increased turbidity on biological processes have generally not been studied in inland wetlands. Based on studies in other surface waters, primary production increases with increased solar energy, and secondary production may increase as well, depending on habitat availability and other factors. Decomposition in a southern forested wetland, as measured by the "cotton rate of rotting (CRR)" was 50 percent greater after removal of vegetation by herbicide than in an undisturbed forest (Mader et al. 1988).
Thermal Alteration. Decomposition may be enhanced by moderate temperature increases, but thermal effects are more likely to be overshadowed by effects of litter type, depth, consumer invertebrate density, and canopy cover (Hauer et al. 1986). Primary and secondary production generally increase with increasing temperature, but thresholds beyond which these processes start to decline are not known for any wetland type, and thermal loading may decrease the primary productivity of specific taxa and communities (e.g., Scott et al. 1985). Aquatic invertebrate production was correlated with water temperature in Plante and Downing's (1989) analysis of aquatic bed community data from 51 lakes (164 samples) from temperate regions of the world.
Dehydration/Inundation. In southern floodplains, production of woody vegetation was greater in forested wetlands that are flooded during some portion of the year but are well-drained (except for small, intermittent storms) during the growing season (Birch and Cooley 1983). Decomposition rates are generally slower in wetlands with longer duration flooding, anoxia, and greater water depths (Brinson 1981, Day et al. 1988), but dehydrated wetlands may experience considerable accretion of organic matter (Burton 1984, Elder and Cairns 1982). Effects of various inundation regimes on vegetation biomass have been reported by Knighton (1985), Fredrickson and Taylor (1982), Robel 1962, and others.
Fragmentation of Habitat. We found no studies that attributed a decline in individual wetland annual productivity, decomposition or denitrification rates to the regional declines in wetlands that have occurred. One can surmise that as the distance between wetlands becomes greater, and/or hydrologic connections become severed by dehydrated channels or dams, the simplified community structure of the remaining wetlands would support lower biological rates. However, this has not been tested.
13.2 SAMPLING METHODS AND EQUIPMENT
Some factors that could be important to measure and (if possible) standardize among wetlands when monitoring anthropogenic effects on the processes of annual productivity, decomposition and denitrification include:
age of wetland (successional status), water depth, temperature (site elevation, aspect), hydraulic residence time, conductivity and baseline chemistry of waters and sediments (especially pH, DO, organic carbon, and suspended sediment), current velocity, sediment type, stream order or ratio of discharge to watershed size (riverine wetlands), shade, ratio of open water to vegetated wetland, vegetation type, and the duration, frequency, and seasonal timing of regular inundation, as well as time elapsed since the last severe inundation or drought.
Methods for measuring productivity and other biological processes in aquatic environments are described in Edmondson and Winberg 1971, Kibby et al. 1980, Murkin and Murkin 1989, Smith and Kadlec 1985, Symbula and Day 1988, and others. A method for measuring whole-wetland respiration is described by Madenjian et al. (1990).
Because all biological processes are expressed as rates, they require data from at least two points in time. To measure annual productivity in wetlands, measurements of plant biomass are made at the onset of the growing season and at the time of peak live biomass. Measurements of decomposition are generally initiated during the mid to late growing season.
Methods that have been used in wetlands are described only briefly below. Measurement of biological processes in wetlands has generally been done with great innovation and adaptation, with few studies employing exactly the same procedures. Thus, only two measurements are described below--decomposition and tree growth. For other processes and parameters, methods used in other surface waters, e.g., for measurement of invertebrate production, might sometimes be applicable to wetlands.
Decomposition methods. Typically, several packs of biodegradable material are placed in surface water and subsets are removed over periods ranging from weeks (usually) to years. Decomposition is inferred by difference in weight over a specified period of time.
Organic matter decomposition rate in one wetland study was measured by as the tensile strength losses of soil burial cloth (93 percent cellulose) after 9 days (Mader et al. 1988). In another study, cypress leaves in mesh fiberglass screen bags were placed in the deepest spots (Dierberg and Ewel 1984); these authors cited the finding of Deghi et al. (1980) that there is no significant difference in decomposition rates between center and edges of cypress swamps. Five litter bags were collected at 15, 29, 58, 114, 205, 390, and 570 days. In a third study (in a stream), bags of air-dried leaves collected just before leaf-fall were placed in riffles in a control and a treatment stream. Bags were collected at 10, 30, 58, 87, and 115 days after placement (Meyer and Johnson 1983).
The litter decomposition rate can integrate short-term indexes of microbial activity (such as ATP, CO2 evolution, and microfaunal counts) over periods of several years (Edmonds 1987). Tree leaf and grass litter is collected and air dried; litter bags are set out and collected at 1, 2, 5, and 7 years.
Decomposition of different sections of three plant species was studied by Hill (1985), who collected Nelumbo leaf laminae and petioles, Typha leaves, and whole Ludwigia plants in the fall when the leaves were beginning to turn yellow. The litter was air-dried and cut into 10-cm pieces, and 2-5 g samples were put into nylon mesh leaf bags (15 cm2, 3-mm octagonal openings). Three to five replicates of each type were put between wire mesh to hold them on the sediment below the water level of a reservoir. Samples were collected from an inundated site at 2, 4, 7, 14, 21, 28, 63, 91, 119, and 154 days and from a drawdown site at 2, 4, 7, 14, 35, and 63 days. Macroinvertebrates were removed from the samples before air-drying.
Tree Growth. Increment cores can be used to estimate tree ages and growth rates, as well as for shrubs (Ehrenfeld 1986). Data from ring counts can be checked against aerial photographs (Klimas 1987). Lemlich and Ewel (1984) took cores of pondcypress (Taxodium distichum var. nutans ), a difficult species to age because of the presence of false rings. They identified false rings by their gradual change in cell size, as contrasted with true rings, in which small latewood cells are readily distinguishable from large earlywood cells.
Leavitt and Long (1989), working with southwestern conifers, described a method of using tree ring analysis to reconstruct historic precipitation and drought patterns. Their method is based on ratios of 13C to 12C, using the principle that, under drought conditions when stomates are closed, the tree will use a greater proportion of carbon-13 in photosynthesis.
Repeated measurement of tree diameter also can be used to gauge growth. It is important to define precisely where on the trunk the measurement is to be taken. In Franklin and Frenkel's study (1987), tree data could not be compared between years because the heights on the boles at which diameters were measured were not standardized. Straub (1984) took diameters of cypress trees at 1.37 m above ground or above buttresses, if present. Small nails were hammered into the trunks so remeasurement would be done at the same point on the tree. Aluminum vernier tree bands calibrated to one-hundredth of an inch are also used to measure tree growth (Sklar and Conner 1983).
More detailed measurements of diameter were used by Scott et al. (1985) in a South Carolina floodplain swamp. Tree biomass was measured by taking five diameters at 5-cm intervals above and below breast height. A nail was driven into the stem at the topmost measuring point to facilitate subsequent measurements; a chain with measurement intervals marked on it can be hung from the nail.
13.3 SPATIAL AND TEMPORAL VARIABILITY, DATA GAPS
In general, data on community-level biological processes have not been uniformly collected from a series of statistically representative wetlands in any region of the country. Thus, it is currently impossible to state, for any wetland type, what are "normal" rates for processes such as annual productivity, decomposition and denitrification.
Only a few studies have compared biological processes among wetlands or aquatic environments in a region or among regions. These include Brinson et al. 1981, Cushing et al. 1983, and Plante and Downing (1989).
Apparently few studies have compared year-to-year or long-term variation in biological processes in wetlands. Such unpublished data may be available from sites of the U.S. Department of Energy's National Environmental Research Park system, and sites of the National Science Foundation's Long Term Ecological Research (LTER) program.
Existing data on wetland plant productivity, collected by a wide variety of methods, was reported by Adamus 1983, Kibby et al. 1980, and (for Carex wetlands only) by Bernard et al. 1988. Net annual primary productivity of some inland wetland emergent species can exceed 6000 g/m2/yr, but usually is less than about 2000 g/m2/yr. Biomass of submersed macrophytes spans four orders of magnitude (Moeller 1975). Decomposition of emergent macrophytes in lacustrine wetlands may take from about 200 to 1000 days for 90 percent weight loss (Hill 1985). Breakdown rates (per day) range from 0.0008 for woody plants in bogs to 0.0190 for non-woody plants in riparian wetlands (Webster and Benfield 1986).
Secondary production in wetlands has been measured much less often than primary production. For invertebrates, Smock et al. (1985) reported 3.09 g/m2 annual production from an acidic South Carolina forested wetland; Plante and Downing (1989) compile estimates of invertebrate production from lacustrine wetlands. Fish production in a 4-year study of the Okefenokee Swamp in Georgia ranged from 43 to 187 kg wet mass/ha (Freeman 1989).
Limited quantitative data on other biological processes is available by wetland type in some of the "community profile" publications of the USFWS (Appendix C).