Search this site:

Alterations to Forested Wetlands due to Development, Agriculture, and Silviculture

Functions of forested wetlands and the concomitant goods and services they provide can be degraded or destroyed by human activities. Activities that affect forested wetlands fit into four broad categories: (1) urban development, (2) rural development, (3) agriculture, and (4) silviculture. Since each wetland impact carries a unique set of circumstances and responses, these categories are rather gross. Their use, however, helps to describe wetland status, trends, and impacts in the South.

NWI defines urban development as intensive use in which much of the land is covered by structures, including buildings, roads, commercial developments, power and communication facilities, city parks, ball fields, and golf courses. In rural development, land use is less intensive, and the density of structures is more sparse. Agriculture is defined as land use primarily for the production of food and fiber including horticultural, row, and close-grown crops as well as animal forage. Silviculture is defined here as management of land for production of wood (Dahl 2000).

The replacement of forested wetlands with urban and/or rural development constitutes an irreversible loss, since the wetland is replaced by upland. Developed areas lack wetland hydrology, soils, and vegetation, either singly or in any combination. Changing a forested wetland to an agricultural field typically changes its hydrology and vegetation and disturbs its soil. However, some of these agricultural activities, such as drainage and removal of native vegetation, can be reversed and wetlands restored. Silvicultural activities typically do not lead to a loss of wetland status but may temporarily affect wetland functions. In forested riverine wetlands, for example, the overstory vegetation is removed but hydrology is left largely intact. Like some agricultural effects, silvicultural effects can be reversed and the wetland functions restored. More specific aspects of these activities will be discussed next.

Urban and rural development—The effects of urban and rural development on riverine, flat, and depressional wetlands in the South are similar. Forest vegetation is cleared, areas are drained or filled to escape flooding, structures are built, and wetland vegetation is replaced. These activities eliminate the ability of forested wetlands to store and convey surface water and ground water. Water runs off these developed surfaces faster, reaching streams quicker and contributing to larger floods downstream. Development also eliminates the water-quality enhancement of forested wetlands. Development alters the hydrology and replaces the soils and vegetation with manmade structures which are not able to take up excess nutrients and other pollutants. The structures may actually contribute pollutants to adjacent aquatic ecosystems. Basnyat and others (1999) reported that urban land is the strongest contributor of nitrate to adjacent streams in Alabama. Alteration of hydrology and replacement of vegetation and soils with manmade structures also eliminate the forested wetland plant community and the wildlife associated with these areas. In other words, urban and rural development typically replace the wetland with upland and developed land with none of the functions of wetlands and little chance of restoration.

Agriculture—Generally, agricultural activities in forested wetlands manipulate hydrology, remove native vegetation, and disturb the soils for the purpose of crop production. Drainage, channelization, and levee construction impact the flow of water to and from a wetland site in an effort to dry out the area. When wetlands are drained for agricultural use, they no longer function as wetlands (Mitsch and Gosselink 2000).

In riverine wetlands, hydrology is the principal force for maintaining ecological processes and vegetation structure (Gosselink and others 1990). Drainage and channelization allowed water to reach the wetland but removed it from the site and/or watershed more quickly. Levees prevent floodwaters from reaching the wetland at natural intervals (once to several times per year). Thus, drainage, channelization, and levee construction result in changes in the timing of delivery of water (frequency), the amount of water delivered (magnitude), and the length of time the water remains in the wetland (duration). Duration of inundation is important in nutrient cycling, removal of pollutants and sediments, and export of organic carbon. Changes in hydroperiod also change the plant community, which alters the living and dead plant biomass components of nutrient cycling and organic carbon export. Construction of drainage ditches and channelization can affect the flow of subsurface water in a riverine wetland by changing the gradient of subsurface flow. Typically the result is a lower water table in the vicinity of the ditch or deepened channel. A shallower water table affects the ability of the riverine wetland to gradually contribute to stream flows during dry periods. Lowering the water table also affects biogeochemical processes and plant and animal communities that depend on the maintenance of a stable ground-water table (Ainslie and others 1999).

By impairing the ability of overbank flows to reach riverine wetland sites, levees prevent elements and compounds and sediments from reaching the wetland where they are deposited or removed. Levees prevent flood flows from transporting organic carbon to downstream aquatic ecosystems. They also act as barriers to aquatic species that use the floodplains for spawning and rearing (Baker and Kilgore 1994, Lambou 1990).

Clearing the native vegetation of a forested riverine wetland and replacing it with a crop dramatically reduces the site’s structural diversity, wildlife-food-producing capacity, and nesting and escape cover (Gosselink and others 1990). Clearing also affects forest patch dynamics by decreasing forest patch size, interrupting forest continuity, decreasing the percentage of regional forested wetland, and increasing edge between community types. Soil tilling is likely to decrease the amount of organic matter in the soil due to oxidation. It also reduces water infiltration by creating a plow pan (Drees and others 1994). Therefore, clearing of native vegetation and forest structure and repeated plowing and tilling have the aggregate effect of causing more water to run off farm fields, contributing greater flows and nonpoint-source pollutants (Basnyat and others 1999).

Many Carolina Bays have been significantly altered by agricultural practices, and some are being used for wastewater treatment (Richardson and Gibbons 1993). Managing forested depressions for agriculture involves clearing existing vegetation, installing drainage ditches through the rim of the Carolina Bay, tilling the soil, and planting the site in the desired crop species. Draining the depression alters the duration of ponding and the amount of water in the wetland. Plants, animals, and the biogeochemistry of the wetland are affected. Disrupting the surface of the soil by tilling affects the amount of organic material in the soil. As water is drained from the depression, soil organic material is exposed to the air, speeding its removal through oxidation. As soils are disturbed and more organic carbon is exposed from deeper in the soil and more is oxidized as a result, the balances among water, carbon, and other elements like nitrogen and phosphorous are disrupted. Accumulation of too much sediment in depressional wetlands, from erosion in nearby uplands, decreases wetland water storage volume, decreases the duration of water retention in wetlands, and changes plant community structure by burial of seed banks. As with riverine wetlands, clearing the existing vegetation in Carolina Bays alters the composition and structure of the native plant community and affects wildlife species that utilize the depression.

Sharitz and Gresham (1998) report that 97 percent of the Carolina Bays in South Carolina have been disturbed by agriculture (71 percent), logging (34 percent), or both. Agriculture is the oldest and predominant use of bays, having started in the 1940s. Soils in Carolina Bays are highly organic and have a high nutrient-holding capacity. They are attractive to farmers if drainage is accomplished; soil pH is raised by liming; minor nutrients tied up by the highly organic soils are supplied to the crops with spray; and weeds are controlled, primarily with herbicides. If these activities are completed, Carolina Bays are 10 to 15 percent more productive than upland soils, but these activities alter the structure and function of the Carolina bay.

Organic soil flats were cleared and drained for agriculture as early as the 1780s. Several large pocosins have been impacted by corporate agricultural operations, which have drained, limed, and fertilized these wetlands for corn and soybean production. Offsite effects of draining pocosins for agriculture included decreased salinity in adjacent estuaries; increased turbidity in adjacent streams immediately after development; and increased phosphate, nitrate, and ammonia inputs into adjacent streams and estuaries, particularly when runoff volumes are high (Sharitz and Gresham 1998). These problems can be minimized by managing the water levels in the drainage ditches with f risers, which maintain water tables and slow the delivery of water to adjacent streams and estuaries. In 1989 14 percent of pocosins in North Carolina were owned by corporate agriculture and 36 percent by major timber companies (Richardson and Gibbons 1993). Originally pocosins covered 2,244,000 acres in North Carolina, but by 1980 this had been reduced by 739,000 acres due to agriculture, silviculture, and development (Richardson and Gibbons 1993). Clearing pocosins for agriculture is no longer practiced due to restrictions placed on landowners by the Food Security Act and section 404 of the Clean Water Act.

Silviculture—Silvicultural activities in forested riverine wetlands typically consist of clearcutting overstory vegetation and allowing natural regeneration from sprouts (Kellison and Young 1997, Lockaby and others 1997b, Walbridge and Lockaby 1994). The stand then progresses from a thicket dominated by briars, vines, and tree seedlings and sprouts to a sapling stage after 10 to 20 years, to a pole timber stage after 20 to 30 years, to a small saw-log stage at 30 to 50 years, and finally to a mature forest stage beyond age 50 (Kellison and Young 1997). Hydrologic responses to this silvicultural regime typically are short-term elevations in the water table due to a reduction in evapotranspiration (Lockaby and others 1997b, Sun and others 2001). Removing the trees reduces the amount of the soil water transpired by plants, and the water then fills more soil pores, resulting in a water-table rise. However, this reduction in evapotranspiration is typically negated by the sprouting vegetation on the clearcut site within 2 years (Lockaby and others 1997a). Another hydrologic effect of harvesting riverine wetlands is soil compaction which interferes with the movement of water through the soil. Lockaby and others (1997) determined that the hydraulic conductivity of the saturated soil was reduced 50 to 90 percent in the ruts caused by skidding of logs. This effect can be temporary, depending on the soil type and hydrology of the wetland (Perison and others 1997, Rapp and others 2001).

There is concern that harvesting and site preparation in wetlands cause or contribute to the generation of nonpoint-source pollutants, particularly sediment. Ensign and Mallin (2001) found that when compared to an upstream reference site, a stream in the Coastal Plain of North Carolina experienced higher levels of nutrients (nitrogen and phosphorous), higher fecal coliform levels, and recurrent algal blooms for up to 15 months after clearcut harvesting of adjacent forested wetlands. The authors speculated that these effects were due to the inability of the clearcut wetland site to retain and transform upstream agricultural pollutants. However, other studies indicate the magnitude of these effects is small and the longevity is brief (Lockaby and others 1997b, Messina and others 1997, Shepard 1994, Walbridge and Lockaby 1994). Studies indicate that after revegetation, sediment deposition in wetlands is actually greater on harvested sites because the amount of vegetation is greater, thus slowing floodwaters to a greater degree and allowing more sediment to drop from the water column (Aust and others 1997, Perison and others 1997).

The capacity of forested riverine wetlands to act as sinks, sources or transformers of nutrients and carbon, depends upon landscape position, the amounts of nutrients entering the wetland, and the time since disturbance. The degree to which silviculture affects a riverine wetland’s capacity to transform nutrients and sequester other pollutants is uncertain (Lockaby and others 1997b). Conceptually, riverine wetlands serve as sinks when they receive high inputs of nutrients. They may serve as sources when disturbed to the point where active oxidation of soil organic matter or export of mineral sediment is occurring, and they may serve as transformers in relatively undisturbed situations. However, Lockaby and others (1999) point out that few generalizations can be made about biogeochemical cycling and nutrient retention functions because of the variable nature of responses of riverine wetlands to harvests, and the inability of current scientific methods to detect subtle biogeochemical changes due to silvicultural activities. Thus, they conclude that the ability to predict whether long-term shifts in biogeochemical transformations occur due to silviculture is minimal and that there is a critical need to understand how silviculture affects the enhancement of water quality in riverine wetlands.

Perhaps the most apparent effect of silvicultural operations on forested riverine wetlands is the removal of the tree canopy. The ability of the forested wetland to recover from harvesting is of interest to both forest industry and conservation interests. Generalizations about the productivity of forested riverine wetlands and their ability to recover from harvests are difficult due to the diversity of forested wetlands. Different moisture regimes, hydrologic conditions, and soil types have resulted in the diversity of wetland types (Conner 1994). Comparisons between harvested sites and reference sites require long-term study. A study conducted 1 year after harvesting in a Texas riverine wetland showed little difference in the composition of tree species regenerating on the harvested site and the presence of those species on an unharvested site (Messina and others 1997). Another study conducted 7 years after harvest in a tupelo-cypress riverine wetland indicated that harvested stands were stocked with tree species similar to the reference. The stand harvested by helicopter had an even distribution of overstory species, while the stand harvested with ground-based methods was dominated by tupelo gum (Aust and others 1997). In a study conducted 8 years after harvesting a riverine wetland in South Carolina, no difference between the species composition of the overstory of harvested and unharvested stands was detected. However, midstory and understory vegetation differed between the two treatments (Rapp and others 2001). These authors concluded that the effects of harvesting are short-lived and that these stands will return to pretreatment species composition. Additional long-term research is needed to continue to track the development of the plant community and ecological functions in harvested stands compared with unharvested stands.

Wildlife species have a variety of ecological roles that contribute to the maintenance of the forested riverine wetland. Wildlife contributes to the dispersal of plants by caching and transporting seeds, and they alter forest structure and composition by eating vegetation and creating impoundments. They alter soil and forest productivity by burrowing and preying on macroinvertebrates. They support food webs, transport energy to surrounding ecosystems, and recolonize adjacent habitats (Wigley and Lancia 1998). Biotic and abiotic factors determine the inherent capacity of a forested wetland to support a community of wildlife species. Soils, topography, hydrology, disturbance, climate, stand vegetation, landscape pattern of habitats and land uses, wildlife community interactions, and human-related alteration of forest structure and composition affect the abundance of wildlife (Wigley and Lancia 1998). The contribution of wildlife to ecological processes and the factors influencing wildlife presence are complex. As a result, evaluating the effects of clearcutting with natural regeneration on riverine wetlands is difficult.

At the stand scale, the vertical and horizontal dimensions of forest structure are important, because the more layers present from the forest floor to the canopy and the taller they are, the more opportunities for foraging, nesting, and escaping from predators (Wigley and Lancia 1998). As plant succession proceeds in forested wetlands, structural diversity tends to increase, but the frequency and duration of flooding may reduce the mid- and understory vegetation. Thus, some animals needing lower layers of the forest, such as the wood thrush, hooded warbler, and Swainson’s warbler, may not be present in natural forest stands (Howard and Allen 1989). However, flooding may contribute to vertical diversity by creating snags, which are important to some species like the prothonatary warbler, wood ducks, woodpeckers, and bats (Wigley and Lancia 1998). Horizontal diversity refers to the distribution of vegetation or other structural features in patches throughout the stand. This horizontal diversity can provide habitat for early successional species in a mature stand or mature stand species in an early successional stand. Diversity of mast-producing species can also ensure a consistent food supply. When production of one tree species is low, that of another species may be high.

Edges occur between wetland forest types, wetland and upland forest types, or between land uses. The effects of these edges vary. Edges can increase species diversity by providing habitat for the species in the abutting habitats plus those species that prefer edges. On the other hand, edges can increase predation and brood parasitism by brown-headed cowbirds and add exotic species (Wigley and Lancia 1998). Riverine wetlands can serve as regional migration corridors for black bear, neotropical songbirds, and waterfowl (Gosselink and others 1990). However, these corridors can aid in the conveyance of species from one habitat to another or, as with edges, can convey predators, diseases, and parasites. Forested wetlands also fit into a landscape mosaic of habitat types that may be important to species needing several habitats to fulfill life requirements. Species presence and productivity are sometimes viewed as functions of the size and shape of a wetland habitat patch, amount of edge, distance from patches of similar habitat (isolation), amount of time since isolation, and immigration and dispersal of animals from habitats (Wigley and Roberts 1997). However, much of the landscape-scale information on the effect of these wildlife habitat functions on the presence and productivity of wildlife populations is based on theory. Few data exist for managed forest landcapes to validate these theories (Wigley and Lancia 1998; Wigley and Roberts 1994, 1997).

Riverine forested wetlands have an abundance of detritus, hard and soft mast, snags, cavity trees, and large woody debris on the ground as well as multilayered vegetation, and these typically support conditions rich and diverse wildlife communities (Ainslie and others 1999, Gosselink and others 1990, Wigley and Lancia 1998). Forest management activities potentially influence wildlife habitat at site-specific and landscape scales. Clearcuts with natural regeneration temporarily reduce availability of hard mast and canopy and cavity trees (Wigley and Roberts 1994, 1997). However, regeneration of woody vegetation and ground vegetation growth typically increase after harvest, downed woody debris often increases due to harvesting (assuming it is not windrowed and burned), and early successional wildlife species may increase. Clawson and others (1997) found that amphibian population diversity and abundance were only temporarily affected by harvesting. Thus, many habitat alterations due to forest management are temporary.

From a landscape perspective there is a growing recognition that the lack of early successional forest, including but not exclusive to forested wetland, is limiting biodiversity in the Eastern United States (Hunter and others 2001, Litvaitis 2001, Thompson and Degraaf 2001, Trani and others 2001, Wigley and Roberts 1997). Thompson and Degraaf (2001) suggest that silvicultural operations can contribute to landscape diversity by creating early successional habitats in forested landscapes. Several studies have suggested that in largely forested landscapes, early successional patches increase wildlife diversity (Thompson and others 1992, Welsh and Healy 1993). However, as previously pointed out, little is known of the effects of forest management in landscapes permanently fragmented by conversion to agriculture or urban development.

Silviculture: depressions—Sharitz and Gresham (1998) note that managing Carolina Bays for timber requires clearing the existing vegetation, installing drainage ditches within the bay and through the rim, bedding the bay soil, and planting trees. Any of these activities greatly alters the structure and function of the bay ecosystem.

Pondcypress swamps are harvested for sawtimber and increasingly for landscape mulch. Typically, they are harvested by clearcutting. Clearcuts regenerate well (Ewel and others 1989), but leaving some mature trees to produce seed is advocated due to uncertainty of resprouting and seed production (Ewel 1998). After harvesting, water levels in pondcypress swamps typically rise, and amphibian and wading bird usage of the postharvest swamp increases. Mammal usage also changes, with fewer nest and den sites but more prey available (Ewel 1998).

Silviculture: mineral-soil pine flats—On mineral-soil flats, three parameters stand out as being essential for determining the degree to which ecosystem processes are altered by a given impact: (1) the alterations in the hydrologic regime, (2) alterations in fire regime, and (3) alterations in the soil. These changes in ecosystem processes on mineral-soil flats alter plant and animal habitats. Hydrologic fluctuations determine the composition of fire-tolerant vegetation, and soil conditions control the dynamics of biogeochemical transformations by soil microbes. Fires maintain open, sometimes treeless savannas by precluding species that would otherwise shade out characteristic savanna plants and provide nutrients in discrete pulses utilized by savanna plants (Rheinhardt and others 2002).

Silvicultural impacts on flat wetlands typically include surface and subsurface drainage, ditching, harvest and mechanical reduction of native vegetation, bedding, which alters microtopographic relief, and the construction of roads (Harms and others 1998). The objective of intensive management on these mineral-soil flat wetlands is to produce pine plantations. Most biogeochemical processes in wetlands depend on the distribution and timing of flooded and dry conditions. Draining a mineral-soil flat eliminates flooding and soil saturation, which in turn alters processes that depend on flooded conditions, including fermentation, and denitrification.

With the exception of artificial drainage, most alterations to hydrologic regime are localized in their effect on biogeochemical processes and habitat quality. For example, a dam (even a low one such as a road fill) can impede surface flow and back water up over a large area. One result is a longer period of inundation. Input of excess water from offsite can likewise increase the duration and depth of water levels. Alterations to water balance change the duration and timing of flooding and the saturation of soil in the upper horizons. In contrast, artificial drainage reduces inundation periods. Artificial drains transport water, nutrients, and dissolved organic matter into streams downstream, altering the water flow and chemistry for a period of 2 to 3 years. (Amatya and others 1997, Beasley and Granillo 1988, Lebo and Herrmann 1998). However, these studies also indicate that the hydrologic effects of ditches can be ameliorated with water-control structures such as flashboard risers (Sun and others 2001).

Soil condition on mineral-soil flats also can be affected by intensive silvicultural activities (Miwa and others 1997, 1999). Microbial organisms and plants are adapted to characteristic microtopographic structure, soil texture, and nutrient regime. Alterations to soils affect these conditions upon which soil microbes and plants depend. The result may be a change in biogeochemical cycling processes. For example, harvesting under wet conditions can affect water-holding capacity and available water for plant growth and slow internal soil drainage, causing higher water tables and slower site drainage in the immediate area of the harvest (Miwa and others 1997). Bedding is currently the best available technique to ameliorate these effects. However, bedding also may affect soil-bulk density both on the beds and in the trenches between, thus altering interstitial pore space and substrate conditions on which soil microbes and plants depend. In addition, microtopographic variation is changed by a regular distribution of small, low (10 to 20 cm high), regularly distributed hummocks to a parallel array of trenches and high ridges (15 to 30+ cm high). On bedded sites, duration and frequency of flooding are increased in trenches and decreased on beds relative to unaltered conditions, which result in altered rates, timing, and magnitudes of biogeochemical processes (Rheinhardt and others 2002).

Mechanical treatment of native vegetation and bedding a mineral-soil flat to produce pine plantations affects fire-maintained wildlife habitat of wet pine flats. For example, several amphibian species are associated with fire-maintained landscapes and travel across wet flats to breeding ponds in cypress depressions. There is evidence that intensive silviculture may detrimentally affect amphibian and reptile populations (Rheinhardt and others 2002), because intensive silviculture relies on a series of raised parallel-aligned beds on which pine seedlings are planted. Standing water in the troughs between beds may cue amphibians to lay their eggs in these troughs, where water sits for too short a time to support larval development, rather than in deeper, more permanent cypress depressions which are commonly scattered throughout wet pine flats.