Eric A. Nelson
Westinghouse Savannah River Company
Aiken, SC 29808

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

This report has been reproduced directly from the best available copy.

Available for sale to the public, in paper, from:  U.S. Department of Commerce, National Technical Information Service, 5285 Port Royal Road, Springfield, VA 22161,  phone: (800) 553-6847,  fax: (703) 605-6900,  email:  orders@ntis.fedworld.gov   online ordering:  http://www.ntis.gov/support/index.html

Available electronically at  http://www.osti.gov/bridge/

Available for a processing fee to U.S. Department of Energy and its contractors, in paper, from: U.S. Department of Energy, Office of Scientific and Technical Information, P.O. Box 62, Oak Ridge, TN 37831-0062,  phone: (865 ) 576-8401,  fax: (865) 576-5728,  email:  reports@adonis.osti.gov

The F- and H-Area Seepage Basins received liquid waste from the F and H chemical separation facilities from 1955 through 1988. Movement of this waste into the seep-fed wetlands below the basins resulted in areas of complete vegetation mortality in the late 1970’s. Natural revegetation of the areas was first examined in 1993 by establishing a series of permanent plots in the F- and H-Area tree-kill zones. When the potential of the pump, treat, and inject treatment system near the seepage basin to change the hydrology of the seep-fed wetlands of Fourmile Branch was first expressed, the plots were re-assessed in 2000 to examine vegetation recovery and species shifts. Vegetative recolonization and wetland conditions of the F- and H-Area tree-kill zones were assessed again in 2001 and compared to prior years. The hydrology of both areas in 2001 was more similar to historic conditions than conditions in 2000. Species diversity, as measured by total species encountered, was lower than the early reports, and is reflective of several natural processes detailed in the report. The vegetative herbaceous cover in the F-Area tree-kill zone is increasing due to establishment of broomsedge, a grass that prefers dries sites. Vegetative stress in the F-Area, however, is still evident in the woody species in the area. The H-Area tree-kill zone continue to exhibit vegetative recovery noted in the previous report. Woody flora have formed a closed canopy in the drier areas. Areas previously dominated by cattails are beginning to deteriorate and allow a greater diversity of species to colonize the plots.

The F- and H-Area Seepage Basins received liquid waste from the F and H chemical separation facilities from 1955 through 1988. Tree mortality in seepline-fed wetlands downslope from the basins was observed in the late 1970’s, and investigations were conducted to determine the cause and source of the impacts (Loehle and Gladden, 1988; Mackey, 1988; Haselow et al., 1990; LeBlanc and Loehle, 1990; Greenwood et al., 1990). Analysis of the soil and water in the tree-kill zones demonstrated a strong chemical linkage with the F- and H-Area Seepage Basins (Killian et al., 1987a, 1987b). Although no single cause of the mortality was determined, it was believed to be the result of interactions of alterations in the hydrology and erosional deposition, along with lowering of pH and increased conductivity, sodium, aluminum, and nitrogen compounds (Looney et al, 1988). A mild drought during the growing season may also have increased the concentration of the chemical contaminants in the soil matrix. A recent sampling of porewater and sediments (Friday, 2001) indicated that, while levels are reduced, they are still significantly above undisturbed wetlands. This was especially the case in the F-Area tree-kill zone.

In 1988, the F- and H-Area Seepage Basins were closed and covered with a clay cap to reduce the rate of contaminant migration to groundwater. Subsequent studies of the chemical composition of the tree-kill zone groundwater (Dixon and Rogers, 1993) and toxicological characteristics of the seepline soil (Nelson and Westbury, 1994) have shown a reduced contaminant concentration. In 1993, an initial vegetation study was undertaken to determine the level of recovery by the plant communities in the tree-kill zones (Nelson and Irwin, 1994). A subsequent study repeated the initial vegetation investigation in 1994 (Westbury and Nelson, 1994) in order to further analyze and characterize the recovery of plant communities in the zones after an additional year of growth. With the installation of the groundwater cleanup program in the areas, a yearly reassessment of the vegetation in the wetland tree-kill areas was re-established in 2000 (Nelson, 2000) to examine both recovery of the plant community since 1994 and to begin assessment of impacts to the wetlands due to water extraction. The objective of this report is to summarize the 2001 sampling data and update trends and affects observed at the original sampling plots within the impacted wetlands.

As part of the groundwater cleanup program, the pump, treat, and reinject system was put in place above the F- and H-Area seepline locations. Preliminary modeling of the effect on the lowered water table in the seepline-fed wetlands indicated that the water table would be reduced significantly. To follow this effect, a series of piezometers were installed through the areas that may be impacted (Halverson, 2000). Two of these locations (FPZ005A/B and HPZ001A) are within the plots detailed in this report. Results of 1999 monitoring indicate that the water table had been lowered, and that lowering was correlated to both volume of water being extracted by the wells and drought conditions at the location that year. The most recent report containing information on the two piezometers within the areas examined (Halverson, 2001), confirmed visual observation in last years report that water tables at the tree-kill sites were lower than historic levels. This lowering, as in the prior year, was probably due to both pumping activities and the continuing drought conditions during 2000. Rainfall in 2001 was more typical of historic averages, and both sample areas had water nearer to the soil surface than during the 2000 sampling. The hydrology of the sample areas were still drier than historic data indicates, but both areas currently possesses water levels that would allow their delineation as jurisdictional wetlands. Collection of data from a network of 23 piezometers will continue to monitor and long-term effects. This report discusses only two piezometer locations and may not be indicative of the entire sampling results. The objective of the following vegetation characterization was to examine the change from 2000 to the present at previously reported plots in the tree-kill zones of the F- and H-Areas. The data were examined to determine if they reflected the changes in hydrology that have occurred in the immediate vicinity of the seepline near the two tree-kill areas under study.

The two study sites that were examined and characterized in prior years were the most severely impacted wetlands in F- and H-Areas (Figure 1). These were defined as the complete mortality of the canopy trees. Prior to the occurrence of tree mortality, the study sites appeared to be swamp tupelo (Nyssa sylvatica var. biflora) wetlands located between naturally occurring seeplines and Fourmile Branch. All of the mature trees at these sites have died, completely eliminating the overstory.

A series of vegetation characterization plots were established in 1993 at uniform intervals along transects (Nelson and Irwin, 1994). The F-Area site was sampled along four transects (Figure 2) that crossed the zone laterally, and consisted of 40 plots, while the H-Area site was sampled along a single transect (Figure 3) that bisected the zone length-wise, and consisted of 56 plots. The transects were located in order to best characterize the variation present in the areas based on visual assessment and chemical data. Beginning and ending points of each transect are marked with PVC pipe and identified with metal tags. All flags marking plot centers were replaced as part of the current sampling.

Forty circular milacre plots (43.56 sq. ft./plot, 4.05 sq. m/plot) were examined at the F-Area site and 56 plots were examined at the H-Area site. These are the same plots as those established and measured in 1993. Within each plot, all individual were identified to the lowest possible taxonomic level, usually species. Vegetation was recorded in four strata based on characteristics. These were tree (greater than 3 meters tall), seedling/shrub (woody species less than 3 meters tall), herbaceous (including forbs, grasses, ferns, etc.), and vines. For each plot the height of tree and shrub seedlings, the number of vine stems, and the percent cover of herbaceous plants was recorded. Sampling was conducted on June 27 and July 12, 2001, whereas the 1994 sampling had occurred in June and the 2000 sampling occurred in Mid- to late-August. This years sampling was slightly later than was considered optimal for herbaceous characterization, but there was no indication of species loss and species diversity was greater than in 2000. Plots were installed subjectively to characterize the impacted areas, and therefore do not allow statistical analysis of the data. All data reported is, therefore, representative of a qualitative ecological description of the areas.

The species lists for the F- and H-Area tree-kill zones indicated that total species richness, as measured by total number of species identified within the plots, was slightly lower than in 1994. Species lists of the current assessment are included in Table 1. The F-Area included 24 species this year, as compared to 30 species in 1994. This reduction may have been due to the later sampling date in 2001, which may have missed some of the early herbaceous species. Most of the reduction did occur in the herbaceous stratum. The H-Area assessment included 32 species, as compared to 49 species in 1994. Most of this reduction occurred in the herbaceous stratum of plots 1-15 due to overstory closure, and will be discussed later.

The F-Area site continues to exhibit minimal plant community development. While not saturated to the surface, the water level was considerably higher than during the 2000 sampling dates. A recording piezometer, FPZ005 A/B, was placed in the middle of the tree-kill zone in 1998 (Halverson, 2000). A baseline for the area was established in 1996 and 1997. This indicated that prior to extraction of the groundwater in F-Area in 1998 the tree-kill zone typically had two or more inches of standing water over the surface, and observations indicate that there was usually water flowing out of the area. The year 1996, when baselining was occurring, had a greater yearly rain deficit than did 1999. During the baseline period the level of standing water in the tree-kill zone remained fairly constant. Current data indicate that levels are up compared to last years sampling, although not to historic saturation levels. This has allowed the previously unconsolidated soil to solidify and begin to have structure. The soil has a very high proportion of organic material; this was previously noted as a component of the unconsolidated soil as a visual observation.

Species composition of the area has not changed markedly since the 2000 sampling. The number of individuals in the sapling/shrub category continued to be greatly reduced from the 1994 assessment. The most prevalent woody species, as measured by presence, continued to be red maple (Acer rubrum), loblolly pine (Pinus taeda), and swamp tupelo. Presence is calculated as the number of plots a species occurs in, relative to the total number of plots, and expressed as a percentage. A presence value of 100 would indicate that the species occurred in all plots, a value of 50 would indicate that it occurred in ½ the plots, etc. Presence values are included in Table 2. As part of the table, the presence of no individuals of a stratum is listed under NONE. General vigor of most woody individuals continues to appear stressed. Few plots contained members of the tree stratum, as in previous years. Number of plots examined that had no members of a stratum is listed in Table 3. There were more individuals in the 2001 sampling within the seedling/shrub stratum that seem to be establishing themselves, however, with 22 saplings measuring at least one meter in height. These were predominately pine and red maple. Nearly all plots had representatives of this stratum. Vigor and health of these larger individuals was considerably better than new germinants.

Greenbrier (Smilax spp.) continued to be one of the most common vine species encountered. Yellow jessamine (Gelsemium sempervirens) was also common, especially in the south end of the study area. Number of plots without vines was similar to the 2000 assessment, and both were lower than the 1994 characterization. The lower surface water levels encountered last year may have allowed the establishment of the species, which have persisted.

The herbaceous plant community in the F-Area tree-kill area has exhibited the greatest change from the prior condition. Herbaceous percent coverage for each transect and overall average for 1994, 2000, and 2001 are listed in Table 4. The grass species broomsedge (Andropogon virginicus) accounted for nearly all of the herbaceous cover in the plots. Average herbaceous cover went from approximately eight percent in 1994, to 19 percent in 2000, and over 31 percent in this year’s assessment. This herbaceous coverage was historically most pronounced in the southern end of the F-Area tree-kill zone. This area generally has shown lower metal and conductivity levels compared to the northern portion of the area. Current assessment of the northern transect has shown the highest increase in herbaceous cover since 1994. This was almost entirely due to invasion of broomsedge onto the drier soil surface that has resulted from the lower water table. Herbaceous cover of the plots at the start and the end of each transect are generally higher than the central plots. This is due to the proximity of these plots to the seepline and the upland vegetation as a seed source for movement into the wetlands. Broomsedge is listed as a facultative negative (FAC-) species in the list of indicator species for the Southeastern region (Sabine, 1992). This means that the species can occur in a wetland or upland situation, but that it prefers a drier habitat than would typically be found in a jurisdictional wetland. The shrub and sapling species occupying the site however, continue to show the presence of hydrophytic vegetation in the areas. The presence of hydrophytic vegetation is one of the three criteria required to classify an area as a wetland.

Continued observation of the hydrology of the area, through the piezometer location, and the vegetation development of this area is clearly needed to assess the impacts of the F-Area pumping program on these seep-fed Fourmile Branch wetlands and to monitor the recolonization of the mortality areas.

The H-Area tree-kill site was much more densely vegetated and its sediment was more consolidated and had a higher organic content than the F-Area site. The H-Area site had very little actual disruption of the soil structure during the tree-kill event. It is believed that this was due to the lower sodium load that the H-Area experienced relative to the F-Area (Vergil Rogers, pers. Comm.) Extensive development of a herbaceous wetland plant community and general plant vigor indicates that the conditions within the tree-kill area have improved considerably. The general character of the plants in the area did not show visible signs of stress and the high percentage coverage indicates minimal impairment of plant establishment and growth.

The transect along which the sample plots are located continues to consist of two distinct wetland types. The first part of the transect (Plots 1-15) is classified as a scrub-shrub, broad-leaved deciduous wetland (Cowardin et al., 1979). The remainder of the zone (Plots 16-56) is open-canopy, persistent emergent wetland dominated by cattails (Typha spp.) and cutgrass (Leersia oryzoides). In the current sampling, however, the cattail domination of the lower portion of the tree-kill area was greatly reduced, and is discussed later. A piezometer, HPZ001A, is located along the transect near Plot 26. Data from this location indicates that the water table has been lowered from several inches of standing water to just below the soil surface. During sampling, the soil was found to be saturated at or just below the surface. Areas of ponded water were once again present along the transect. These were not seen in the 2000 sampling. The species presence values for the H-Area are listed in Table 2.

Species composition was somewhat similar between the sampling periods 2000 and 2001, although the upper plots (1-15) generally had fewer total individual species identified than previously. This is probably indicative of the maturing canopy that is forming in this area. These plots have formed a closed canopy and are reducing ground level vegetation due to light competition. Herbaceous cover was also slightly reduced compared to the 2000 sampling (Table 4), indicating the reduced light availability on the ground surface. The lower plots (16-56) continued to be dominated by a limited number of herbaceous species, although new species are occurring as the cattail/cutgrass ecotype begins to breakdown.

The seedling and shrub layer of the upper series of plots (Plots 1-15) continued to establish itself since last year’s survey. The seedling and shrub species composition was the same in both years. Dominant species, measured by presence, included red maple, wax myrtle (Myrica cerifera), and red bay (Persea borbonia) (Table 2). Many of the individuals were greater than two meters tall and healthy. Swamp tupelo was rarely observed. It may be that there is not a seed source near enough to provide for the area of impact, while the lighter seeded species are represented on the slopes above the wetland and are easily blown to the soil surface. Rushes (Juncus sp.) and sedges (Carex sp.) dominated the herbaceous layer, although it was generally sparse due to the shading by the woody vegetation (Table 4). Herbaceous cover was reduced from 36 percent in 2000 to 5 percent this year. This minimal herbaceous coverage is characteristic of mature forested wetlands.

The vegetation of plots 16 through 56 is markedly different than the first 15. There is a short zone of several plots demarking the transition. These plots are dominated by herbaceous and early successional species; very few woody species were identified in the plot boundaries, although the number was greater than in 2000. This area has historically had several inches of standing water, and the nitrogen influx from the seepage basins makes this an ideal habitat for cattail (Typha latifolia) invasion. Also present in most of the plots was cutgrass. The percent cover by the herbaceous species has declined since the 1994 assessment, falling from over 90 percent to less than 63 percent in 2000. This year, herbaceous cover was again down to 39 percent in the lower plots (Table 4). The cattail stands have continued their decline and are allowing additional species to begin to colonize the habitat. This natural decline mentioned last year, has allowed additional light to get to the soil surface. New species observed this year were primarily herbaceous, although adequate seed sources for the introduction of woody species should be available, and should increase in subsequent years. The reduction of the plots without seedling/shrub representation from 41 in 2000 to 32 in 2001 was due largely to woody species occurrence in this lower set of plots. Exact species to appear will be determined by seed availability and future hydrology of the area.

The vine layer was dominated by greenbrier in the upper plots, while the lower plots had fewer vines present, probably due to historic high water levels and competition from established vegetation. Fewer plots, however, contained no vine representatives this year (25 vs. 31 in 2000). Much of this reduction was due to blackberry (Rubus sp.) occurring in several lower plots.

The H-Area tree-kill site continues to develop an extensive post-disturbance plant community that is dominated by herbaceous wetland plants. After the swamp tupelo canopy was removed by the initial impact, wetland herbaceous species would be expected to colonize the saturated sediments as soon as the toxicity of the sediments became sub-lethal. Over time, shrub and then tree species would be expected to become dominant, eventually returning to the pre-disturbance composition. This will be a long-term transition towards a bottomland forest. The northern portion of the zone (Plots 1-15) has developed a sub-canopy of shrubs and saplings indicating a more advanced successional stage than the remainder of the zone. The main tree species present is red maple, which is an early successional invader. As the lower zone continues to become more open, movement of additional species into the area is anticipated. Continued monitoring of the water table and vegetation changes over time will help answer these questions.

General conditions at both tree-kill zones continued to show improving vegetation re-establishment. The F-Area site continued to exhibit toxicity or stress to the vegetation in the sediments, as was evident by stress on the newly germinated individuals. This was especially true of the northern portion of the site. Herbaceous species typical of drier habitats have begun to colonize the slightly elevated portions of the tree-kill area now that the water level is below historic elevations. This has exposed soil surfaces that were previously under water and allowed seed germination. The consolidation of the sediments, now that they are not constantly under water, may be a positive development. A new round of sediment and water sampling was conducted (Friday, 2001), and these may give a clearer picture of continuing toxicity candidates in the soil. The shifting of the vegetation away from the normally hydrophytic species is of some concern, and should be followed up in future years to ensure that this wetland criteria continues to be met. Hydrology of the site this year was more typical of a seep-fed wetland than in the previous year.

The H-Area site exhibited far greater plant community development and successional recovery than the F-Area site. The presence of a shrub sub-canopy in the northern portions of the zone, and the generally high percentage cover of herbaceous wetland plants, suggest lowered toxicity of the substrate to germination and establishment of vegetation. The opening up of the southern portion (cattail beds) is allowing establishment of additional woody and herbaceous species and beginning to develop towards a more complex wetland community. Hydrology of this site, as was the case with F-Area, was more typical of seep-fed wetlands.

1. Cowardin, L.M., V. Carter, F.C. Golet and E.T. LaRoe. 1979. Classification of Wetland and Deepwater Habitats of the United States. FWS/OBS-79/31. United States, Fish and Wildlife Service, Biological Services Program.
2. Dixon, K.L. and V.R. Rogers. 1993. Results of the second quarter tritium surveys of the F- and H-Area seeplines: September 1992. WSRC-TR-93-129, Westinghouse Savannah River Co., Aiken, SC 29808.
3. Friday, G. P. 2001. Chemical properties of pore water and sediments at three wetland sites near the F- and H-Area seepage basins, Savannah River Site. WSRC-TR-2001-00241, Westinghouse Savannah River Co., Aiken, SC 29808.
4. Greenwood, K.P., M.E. Hane, A.L. Lander, C. Loehle, and C.J. Richardson. 1990. Assessment of tree toxicity near the F- and H-Area Seepage Basins of the Savannah River Site (U). WSRC-TR-90-253, Westinghouse Savannah River Co., Aiken, SC 29808.
5. Halverson, N.V. 2000. Monitoring of the water levels in the wetlands of Fourmile Branch near the F and H Areas of SRS: January 1999 to December 1999 (U). WSRC-TR-2000-00122, Westinghouse Savannah River Co., SRTC, Aiken, SC 29808.
6. Halverson, N.V. 2001. Monitoring of the water levels in the wetlands of Fourmile Branch near the F and H Areas of SRS: January to December 2000 (U). WSRC-TR-2001-00242, Westinghouse Savannah River Co., SRTC, Aiken, SC 29808.
7. Haselow, J.S., M. Harris, B.B. Looney, N.V. Halverson, and J.B. Gladden. 1990. Analysis of soil and water at the Four Mile Creek seepline near F & H Area of SRS (U). WSRC-RP-90-0591, Westinghouse Savannah River Co., SRL, Aiken, SC 29808.
8. Killian, T.H., N.L. Kolb, P. Corbo, and I.W. Marine. 1987a. Environmental Information Document: F-Area seepage basins. DSPT-85-704, E.I. duPont de Nemours & Co., SRL, Aiken, SC 29808.
9. Killian, T.H., N.L. Kolb, P. Corbo, and I.W. Marine. 1987b. Environmental Information Document: H-Area seepage basins. DSPT-85-706, E.I. duPont de Nemours & Co., SRL, Aiken, SC 29808.
10. LeBlanc, D.C. and C. Loehle. 1990. The effect of contaminated groundwater on tree growth: a tree-ring analysis (U). WSRC-RP-90-552, Westinghouse Savannah River Co., SRL, Aiken, SC 29808.
11. Loehle, C. and J. Gladden. 1988. Preliminary assessment of tree mortality near F- and H-Area seepage basins. DPST-88-260, E.I. duPont de Nemours & Co., SRL, Aiken, SC 29808.
12. Looney, B.B., J.E. Cantrell and J.R. Cook. 1988. Sampling and analysis of surface water in the vicinity of the F- and H-Area seepage basins. DSPT-88-229, E.I. duPont de Nemours & Co., SRL, Aiken, SC 29808.
13. Mackey, H.E. Jr. 1988. Initial evaluation of photographic data of F- and H-Area seepage basin outcrops. DSPT-88-314, E.I. duPont de Nemours & Co., SRL, Aiken, SC 29808.
14. Nelson, E.A. 2000. Plant community development within the F- and H-Area tree-kill zones – Changes from 1994 to 2000. WSRC-TR-2000-00346, Westinghouse Savannah River Co., SRTC, Aiken, SC 29808.
15. Nelson, E.A., and J.E. Irwin. 1994. Current vegetation characteristics within tree-kill zones of the F- and H-Areas (U). WSRC-TR-94-0203, Westinghouse Savannah River Co., SRTC, Aiken, SC 29808.
16. Nelson, E.A., and H.M. Westbury Jr. 1994. Lettuce seed germination and root elongation toxicity evaluation of the F-Area seepline soils (U). WSRC-TR-94-0510, Westinghouse Savannah River Co., SRTC, Aiken, SC 29808.
17. Sabine, B.J. (ed.). 1992. National list of plant species that occur in wetlands region 2 – Southeast. Resource Management Group, Grand Haven, MI, 128 p.
18. Westbury, H.M. Jr. and E.A. Nelson. 1994. Plant community development within the F- and H-Area tree-kill zones (U). WSRC-TR-94-0544, Westinghouse Savannah River Co., SRTC, Aiken, SC 29808.

Figure 1. Map of Study Area Showing Sampling Locations in Localized
Tree-Kill Zones (X’s). The Dash Line through F-Area is a Steam Pipeline.
Letters and Numbers within Boxes are Road Designations.
Cross-Hatched Areas are Seepage Basins.

Figure 2. F-Area Tree-Kill Vegetation Sampling Transects.

Figure 3. H-Area Tree-Kill Vegetation Sampling Transect.

Table 4. Change in Cumulative Herbaceous Percent Coverage in
Characterization Plots during Successive Sampling Years

F-Area							H-Area
	Year									Year
Transect	1994	2000	2001				Plots			1994	2000	2001
1	10.6	27.6	47.1				1-15			41.1	36.3	4.9
2	4.9	19.8	35.1				16-56			90.3	62.5	38.9
3	3.1	10.0	15.9
4	11.5	28.9	26.6				H-Area			77.1	55.5	29.7
							Average
F-Area	7.5	19.1	31.2
Average