William H. Conner

ABOVEGROUND PRODUCTION IN SOUTHEASTERN FLOODPLAIN FORESTS: A TEST OF THE SUBSIDY–STRESS HYPOTHESIS

It has been hypothesized that periodically flooded forests have higher rates of aboveground net primary production than upland forests and near-continuously flooded forests, but a competing hypothesis holds that the benefits of periodic inputs of nutrients and water may be diminished by stresses associated with anaerobic soils or drought. To test these hypotheses, we measured groundwater table depths and aboveground productivity in floodplain forests of South Carolina and Louisiana. We established paired plots on locally dry, intermediate, and wet topographic positions across three hydrologic transects in each state. These plots encompassed upland hardwood, bottomland hardwood, and cypress swamp forests. Measurements of leaf litterfall, wood production, and groundwater table depth were made in 1987 and 1988. We then used mean growing-season water depth (MWD) to group the plots into three classes: wet (>0 cm), intermediate (0 to -60 cm), and dry ( 25% dead stems) the slope of this line was 5 times greater (-24 gm-2yr- -cm- ). We conclude that the subsidy-stress hypothesis does not adequately describe patterns of NPP across Southeastern U.S. floodplain forests. Conditions of periodic flooding and flowing water do not often lead to high rates of productivity compared with upland forests. However, extensive flooding is nearly always a significant stress on forest productivity, particularly when the flooding regime has been recently perturbed through levee construc- tion or impoundment. Our data support a more complex interaction between subsidy and stress factors.

Water level observations in mangrove swamps during two hurricanes in Florida

Little is known about the effectiveness of mangroves in suppressing water level heights during landfall of tropical storms and hurricanes. Recent hurricane strikes along the Gulf Coast of the United States have impacted wetland integrity in some areas and hastened the need to understand how and to what degree coastal forested wetlands confer protection by reducing the height of peak water level. In recent years, U.S. Geological Survey Gulf Coast research projects in Florida have instrumented mangrove sites with continuous water level recorders. Our ad hoc network of water level recorders documented the rise, peak, and fall of water levels (± 0.5 hr) from two hurricane events in 2004 and 2005. Reduction of peak water level heights from relatively in-line gages associated with one storm surge event indicated that mangrove wetlands can reduce water level height by as much as 9.4 cm/km inland over intact, relatively unchannelized expanses. During the other event, reductions were slightly less for mangroves along a river corridor. Estimates of water level attenuation were within the range reported in the literature but erred on the conservative side. These synoptic data from single storm events indicate that intact mangroves may support a protective role in reducing maximum water level height associated with surge.

Flooding and salinity effects on growth and survival of four common forested wetland species

The survival, growth, and biomass of baldcypress (Taxodium distichum (L.) Rich.), water tupelo (Nyssa aquatica L.), Chinese tallow (Sapium sebiferum (L.) Roxb.), and green ash (Fraxinus pennsylvanica Marsh.) seedlings were examined in an experiment varying water levels (watered, flooded) and salinity levels (0, 2, and 10 ppt, plus a simulated storm surge with 32 ppt saltwater). All seedlings, except for those flooded with 10 ppt saltwater, survived to the end of the experiment. In 10 ppt saltwater, flooded baldcypress, water tupelo, and green ash survived two weeks whereas Chinese tallow survived for 6 weeks. However, a second set of slightly older baldcypress, water tupelo, and Chinese tallow seedlings survived eight weeks of flooding with 10 ppt saltwater. When carried through the winter to the beginning of the second growing season, flooded baldcypress and Chinese tallow seedlings from the 0 and 2 ppt treatments leafed out, but only Chinese tallow recovered from the saltwater surge treatment. The diameter and growth (height) of each species was not affected when watered with 2 ppt saltwater, except for the effects of the height growth of baldcypress. Growth was reduced for all species when watered with 10 ppt saltwater. The diameter growth of green ash was reduced by freshwater flooding. The diameter growth of baldcypress and water tupelo was greater when flooded with fresh water. Flooding with 2 ppt saltwater caused a significant reduction in diameter growth in water tupelo, green ash, and Chinese tallow, but not in baldcypress. Root and stem biomass values were not significantly different for any species between the 0 and 2 ppt salinity watering treatments. However, seedlings watered with 10 ppt saltwater had significantly lower root and stem biomass values, except for baldcypress roots and green ash stems. Baldcypress was least affected by flooding with 0 and 2 ppt saltwater, although there were slight reductions in root biomass with increasing salinity. Because of the susceptibility of the seedlings of these four species to increases in flooding and salinity, their regeneration may be limited in the future, thereby causing shifts in species composition.

Site Condition, Structure, and Growth of Baldcypress Along Tidal/Non-Tidal Salinity Gradients

This report documents changes in forest structure and growth potential of dominant trees in salt-impacted tidal and non-tidal baldcypress wetlands of the southeastern United States. We inventoried basal area and tree height, and monitored incremental growth (in basal area) of codominant baldcypress (Taxodium distichum) trees monthly, for over four years, to examine the inter-relationships among growth, site fertility, and soil physico-chemical characteristics. We found that salinity, soil total nitrogen (TN), flood duration, and flood frequency affected forest structure and growth the greatest. While mean annual site salinity ranged from 0.1 to 3.4 ppt, sites with salinity concentrations of 1.3 ppt or greater supported a basal area of less than 40 m2/ha. Where salinity was < 0.7 ppt, basal area was as high as 87 m2/ha. Stand height was also negatively affected by higher salinity. However, salinity related only to soil TN concentrations or to the relative balance between soil TN and total phosphorus (TP), which reached a maximum concentration between 1.2 and 2.0 ppt salinity. As estuarine influence shifts inland with sea-level rise, forest growth may become more strongly linked to salinity, not only due to salt effects but also as a consequence of site nitrogen imbalance.

THE IMPACT OF WASTEWATER EFFLUENT ON ACCRETION AND DECOMPOSITION IN A SUBSIDING FORESTED WETLAND

Insufficient sedimentation, coupled with high rates of relative sea-level rise (subsidence plus eustatic sea-level rise), are two important factors contributing to wetland loss in coastal Louisiana, USA. We hypothesized that adding nutrient-rich, secondarily treated wastewater effluent to subsiding wetlands in Louisiana could promote vertical accretion in these systems through increased organic matter production and subsequent deposition and allow accretion to keep pace with estimated rates of relative sea-level rise (RSLR). However, we also hypothesized that nutrient enrichment could stimulate the decomposition of organic matter, thus negating any increase in accretion due to increased organic matter accumulation. To test these hypotheses, we measured leaf-litter decomposition, litter nutrient dynamics, and sediment accretion in a permanently flooded and subsiding forested wetland receiving wastewater effluent and in an adjacent control site, both before and after effluent applications began. We also measured organic and mineral matter accumulation in the treatment site before and after effluent applications began. A Before-After-Control-Impact (BACI) statistical analysis revealed that neither leaf-litter decomposition rates nor initial leaf-litter N and P concentration were affected by wastewater effluent. A similar analysis revealed that final N and P leaf-litter concentrations did significantly increase in the treatment site relative to the control after effluent was applied. Total pre-effluent accretion, measured 34 months after feldspar horizon markers were laid down, averaged (±SE) 22.3±3.2 mm and 14.9±4.6 mm in the treatment and control sites, respecitvely, and were not significantly different. However, total accretion measured 68 months after the markers were installed and 29 months after effluent additions began in the treatment site averaged 54.6±1.5 mm in the treatment site and 19.0±3.2 mm in the control site and were significantly different. Additionally, after wastewater applications began, the estimated rate of accretion in the treatment site (11.4 mm yr−1) approached the estimated rate of RSLR (12.3 mm yr−1). Most of this increase in accretion was attributed to organic matter inputs, as organic matter accumulation increased significantly in the treatment site after effluent application began, while mineral accumulation rates remained constant. These findings indicate that there is a potential for using wastewater to balance accretion deficits in subsiding wetland systems.

Fine Root Productivity and Dynamics on a Forested Floodplain in South Carolina

The highly dynamic, fine root component of forested wetland ecoi&g fine root dynamics is a challenging endeavor in any system, but the dilficulties are particularly evident in forested floodplains where frequent hydrologic fluctuations directly influence fine root dynamics. Fine root (53 mm) biomass, production, and turnover were estimated for three soils exhibiting different drainage patterns within a mixedoak community on the Coosawhatchie River floodplain, Jasper County, South Carolina. Within a 45cm-deep vertical profile, 74% of total fine root biomass was restricted to the upper 15 cm of the soil surface. Fine root biomass decreased as the soil became less well drained (e.g., tine root biomass in well-drained soil > intermediately drained soil > poorly drained soil). Fine root productivity was measured for 1 yr using minirhiitrons and in situ screens. Both methods suggested higher fine root production in better drained soils but showed frequent fluctuations in fine root growth and mortality, suggesting the need for frequent sampling at short intervals (e.g., monthly) to accurately assess fine root growth and turnover. Fine root production, estimated with in situ screens, was 1.5, l-g, and 0.9 Mg ha-’ yr-’ in the well-drained, intermediately drained, and poorly drained soils, respectively. Results from minirhizotrons indicated (hat fine roots in well-drained soils grew to greater depths while fine roots in poorly drained soils were restricted to surface soils. Minimizotrons also revealed that the distribution of fine roots among morphological dasses changed between well-drained and poorly drained soils. P WETLAND ECOSYSTEMS has been the focus of numerous studies. Most commonly, productivity is estimated using aboveground parameters such as litterfall and stemwood production (B&son et al., 1980; Conner and Day, 1992; Conner et al., 1993; Conner, 1994; Megonigal et al., 1997). Many investigators have acknowledged, however, that failure to include belowground data will seriously underestimate forest ecosystem productivity (Vogt et al., 1986a; Day and Megonigal, 1993). It has been suggested that fine root production accounts for up to 75% of total net primary production (NPP) in some forests (Nadelhoffer and Raich, 1992). Similar to aboveground foliage, large amounts of fine roots die annually and can contribute a quantity of litter similar in magnitude to foliar litter (McClaugherty et al., 1984). Fine root dynamics, therefore, represent a significant source of energy and nutrient flow through forested systems, particularly for those TerreU T. Baker III, College of Agriculture and Home Economics, New Mexico State Univ., Box 30003, MSC 3AE, Las Cruces, NM 88003-8003; William H. Conner, Baruch Forest Science Institute, P.O. Box 596, Georgetown, SC 29442; B. Graeme Lockaby, School of Forestry, Auburn Univ., 108 M.W. Smith Hall, Auburn, AL 36849. 5418; John A. Stanturf, USDA-Forest Service, Center for Bottomland Hardwoods Research, P.O. Box 227, Stoneville, MS 38776; Marianne K. Burke, USDA-Forest Service, Southern Research Station Center for Forested Wetlands Research, 2730 Savannah Hwy., Charleston, SC 29414. Received 11 Mar. 1999. *Corresponding author (ttbaker@

Effect of forest management practices on southern forested wetland productivity

In the interest of increasing productivity of forested wetlands for timber production and/or wildlife value, management schemes that deal mainly with water-level control have been developed. The three forest types in the southeastern U.S. most commonly affected are cypress/tupelo forests, bottomland hardwood forests, and wet pine sites (including pocosins). In forested wetlands, hydrology is the most important factor influencing productivity. In bottomland and cypress/tupelo forests, water-level control can have mixed results. Alterations in natural hydrologic patterns leading to increased flooding or drainage can cause decreased growth rates or even death of the forest. Bottomland hardwoods respond favorably in the short term to water-level management, but the long-term response is currently under study. In wet pine sites, timber volume can be increased significantly by water-level management, but the impact upon other ecological functions is less understood. It is difficult to adequately describe productivity relations in wetland forests because of the great diversity in habitat types and the lack of data on how structure and function might be affected by forestry operations. There is a definite need for more long-term, regional studies involving multidisciplinary efforts.

TREE COMMUNITY STRUCTURE AND CHANGES FROM 1987 TO 1999 IN THREE LOUISIANA AND THREE SOUTH CAROLINA FORESTED WETLANDS

Paired plots were established across a soil moisture gradient (dry, periodically flooded, flooded) in three forested wetland watersheds in Louisiana and South Carolina, USA in 1986–1987. All trees greater than 10-cm diameter at breast height were tagged and measured annually through 1999 to determine density, basal area changes, ingrowth, and mortality. A greater number of tree species was found in South Carolina dry and periodically flooded sites than in Louisiana. Flooded sites in both states were dominated by water tupelo (Nyssa aquatica) and baldcypress (Taxodium distichum). The overall trend in density in both states was flooded > periodically flooded > dry. From 1987 to 1999, density decreased in all of the Louisiana sites except one, while in South Carolina, density remained the same or increased in four of the sites and decreased in the other four. The greatest changes in density occurred in those sites where water-level changes were occurring and in areas where storm winds struck. Basal area in 1987 was similar in both states, ranging from 20.8 to 55.2 m2/ha in Louisiana and 24.0 to 49.5 m2/ha in South Carolina. Flooded sites had the greatest basal area, and periodically flooded and dry sites had similar basal areas. Mortality rates in Louisiana and South Carolina forested wetlands are typically low (around 2%/year) in areas that have not been altered hydrologically. Annual mortality in Louisiana plots with increased water levels rose from 4% in 1987 to 16% in 1997. Wind storm events significantly increased mortality rates, and mortality rates remained high for years after the event, as damaged trees died. Of special concern are areas like the Verret Basin site where the exotic, invasive species Chinese tallow (Sapium sebiferum) invaded after Hurricane Andrew and has the potential to become a dominant canopy tree in the future.

Recognizing and Overcoming Difficult Site Conditions for Afforestation of Bottomland Hardwoods

SEPTEMBER 2004 183 In the last decade, about 370,000 acres (150,000 ha) of economically marginal farmland in the Lower Mississippi Alluvial Valley (LMAV) have been restored to bottomland hardwood forests (Stanturf and others 1998, King and Keeland 1999, Schoenholtz and others 2001). Planting of this considerable acreage is due to several federal programs, such as the Wetlands Reserve Program (WRP), that assist landowners by financing afforestation (Figure 1). Unfortunately, these operational plantings have not performed as well as smaller plantings or research plots (Stanturf and others 2001a). For example, a recent survey of WRP plantings in westcentral Mississippi revealed that more than 90 percent of the sites failed to meet the criteria of 100 woody stems per acre (247 stems per ha) three years after planting or direct seeding. While planting 1-0 bareroot seedlings of oak was more successful than direct-seeding acorns, only 23 percent of the land planted with seedlings met the criteria (C.J. Schweitzer unpublished data). Planting and direct seeding oak (Quercus spp.) on public land in the same area has been more successful. Meanwhile, Allen (1990) found 70 percent of the planted bottomland hardwood stands on the national wildlife refuges he evaluated had more than 200 trees per acre (494 stems per ha). We believe that the recurring problems in operational plantings on private lands are due in part to the failure of planters to recognize adverse site conditions and their failure to use appropriate methods for overcoming site limitations. Our objectives in this paper are to synthesize research and experience into guidelines for recognizing adverse site conditions due to hydroperiod, soil, competing vegetation, and herbivory. We describe techniques for overcoming these conditions and suggest promising research areas.

Leaf litter decomposition in three Louisiana freshwater forested wetland areas with different flooding regimes

Litter decomposition was studied for one year in three Louisiana forested wetland sites with different flooding regimes. Decomposition was significantly higher in a crayfish pond where flooding was manipulated by pumping. Only 20% of the original dry mass remained after 46 weeks versus over 40% in the natural and impounded wetland forests. Nitrogen was immobilized in the natural and impounded areas but was mineralized during the spring and summer in the managed area. Phosphorus was not immobilized in the natural and impounded areas like nitrogen but was mineralized much more slowly in these areas than it was in the managed area. Because of the slower decomposition in the natural and impounded areas, falling leaf litter buries the old, partially decomposed material. Thus, there tends to be a net accumulation of organic matter, nitrogen, and phosphorus in stagnant, more flooded areas and mineralization and/or export from free-flowing areas.