Southeastern Naturalist K. Stephens, J. Sencindiver, and J. Skousen 2015 Vol. 14, Special Issue 7 40 Canaan Valley & Environs 2015 Southeastern Naturalist 14(Special Issue 7):40–57 Characteristics of Wetland Soils Impacted by Acid Mine Drainage Kyle Stephens1, John Sencindiver1, and Jeff Skousen1,* Abstract - A proposed section of Appalachian Corridor H, an interstate highway that begins at I-79 near Weston, WV, and will continue east to I-81 at Strasburg, VA, will pass through an area of the Beaver Creek watershed that was previously mined for the acid-producing Upper Freeport coal. Beaver Creek flows into the Blackwater River after flowing out of Canaan Valley. Partially reclaimed spoils from past mining activities are generating acid mine drainage. Wetlands adjacent to the spoils support plant communities that appear to be naturally treating the drainage. To better understand the chemical and physical functions within the wetlands and to assist the West Virginia Division of Highways in constructing wetlands for mitigating environmental damage, we described the soils of the mine-drainage-impacted wetlands (Narrow Wetland, Iron Pond, and Railroad Grade) and took samples for subsequent laboratory analyses. For comparison, we also described and sampled unimpacted soils in Elder Swamp, which is an adjacent wetland that receives no mine drainage. The impacted wetland soils had thinner organic and mineral horizons and were lower in C and N than unimpacted soils. The electrical conductivity was low for all wetland soils, and pH ranged from 3.2–6.1, with both low and high pH values in impacted and unimpacted soils. These results were reflected in the overall lower quality of vegetation that we noticed in the impacted wetlands. Introduction Section 404 of the 1977 Clean Water Act defines jurisdictional wetlands as “those areas that are inundated or saturated by surface or ground water at a frequency and duration sufficient to support, and that under normal circumstances do support, a prevalence of vegetation typically adapted for life in saturated soil conditions” (Environmental Laboratory 1987). Wetlands perform unique ecological functions: they filter water, recharge aquifers, mitigate flood events, and may support rare and diverse flora and fauna (Mitsch and Gosselink 2000). For an area to be identified as a jurisdictional wetland, three key features must be present: 1) hydric soils, 2) hydrophytic vegetation, and 3) wetland hydrology. Hydric soils form through unique pedogenic processes such as horizon development, carbon (C) accumulation and distribution, low redox conditions, and water saturation. These processes allow for the development of easily identifiable morphological features that are distinctly different from the features of nonhydric soils. A hydric soil, by definition, is “a soil which is saturated, flooded, or ponded long enough during the growing season to develop anaerobic conditions in the upper part” (USDA-NRCS 1995). Anaerobic conditions develop when water fills the soil pores, which decreases the rate at which oxygen (O2) can diffuse 1West Virginia University, Division of Plant and Soil Sciences, PO Box 6108, Morgantown, WV 26506. *Corresponding author - jskousen@wvu.edu. Southeastern Naturalist K. Stephens, J. Sencindiver, and J. Skousen 2015 Vol. 14, Special Issue 7 41 through the soil (Mitsch and Gosselink 2000). In submerged soils, the rate of O2 diffusion through the water-filled pores is four to five orders of magnitude slower than the rate of O2 diffusion in a well-drained soil (Howeler and Bouldin 1971). Under these conditions, the remaining dissolved O2 in the subsurface is rapidly consumed through microbial respiration, and the redox potential (Eh) decreases as alternative electron acceptors such as iron oxides are used during anaerobic incubation (Faulkner and Patrick 1992, Gambrell and Patrick 1978, Ponnamperuma, 1972). When all of the O2 is removed from the system, the soil is considered reduced. After the O2 is removed, manganese (Mn), iron (Fe) and other elements may undergo reduction, allowing for the development of gray colors, or gleying, in the subsurface horizons (Environmental Laboratory 1987). If the soil is periodically saturated, mottles (rust- or gray-blue-colored spots, depletions and or concentrations, respectively) may develop. These features are indicators of alternating periods of oxidation and reduction (Faulkner and Patrick 1992), during which Fe and Mn are mobilized by reduction and then immobilized in the form of oxides during the subsequent drying cycle by oxidation (Pickering and Veneman 1984). The presence of reduced conditions may be determined by measuring soil Eh with platinum-tipped redox electrodes. The critical Eh value indicating that the soil has been reduced is related to soil pH. Thus, critical Eh varies among soils with different pHs. For analysis, this variability is nullified by adjusting Eh values to a pH of 7. A soil meets the anaerobic conditions of the recommended hydric soil technical standards if Eh measures ≤175 mV at pH 7 (National Technical Committee for Hydric Soils 2004). In submerged hydric soils, decomposition rates are drastically reduced compared to aerobic non-hydric soils, thus allowing for the accumulation of organic matter. McLatchey and Reddy (1998) reported that about three times more C is decomposed in aerobic soils compared to anaerobic soils. The reduced decomposition rates are a result of anaerobic respiration, which is slowed by microbial use of electron acceptors other than O2 (Furhmann 1998). In incompletely submerged hydric soils, decomposition rates are reduced, but the rate will be a function of the hydrologic regime. Soil horizons are considered organic if they contain 12–20% organic C, with the variation related to percent clay (Soil Survey Staff 1993). Three types of organic horizons are identified in soils: 1) Fibric (Oi) = least decomposed organic materials, 2) Hemic (Oe) = partially decomposed organic materials, and 3) Sapric (Oa) = most decomposed organic materials. If >40 cm of organic material accumulates in a soil, the soil is considered to be an organic soil, also known as a Histosol. If <40 cm of organic material accumulates, the soil is considered a mineral soil. Mineral soil horizons are designated as either A, B or C. A horizons show signs of soil development and usually contain more carbon and roots than B and C horizons. Like A horizons, B horizons exhibit some degree of development. C horizons are undeveloped, lack structure, and are found beneath A and B horizons. Southeastern Naturalist K. Stephens, J. Sencindiver, and J. Skousen 2015 Vol. 14, Special Issue 7 42 Northern peatlands are dominated by Histosols that have an accumulation of peat over mineral materials. The wetland soils of Canaan Valley (hereafter, the Valley), WV, and the adjacent Beaver Creek watershed situated north of the Valley, are similar to northern peatland soils. Peatlands are characterized by a perched water table, low pH, high cation exchange capacity soils, and an accumulation of peat. Depending on a site’s vegetation and hydrology, bogs, marshes, fens, and swamps can develop in peatlands. Fens and marshes prevail in the Valley and the Beaver Creek watersheds. Histosols and mineral hydric soils are common in these areas. Wetland soils in both areas have developed in similar parent materials, on similar topographic positions, under similar climates, and in the same amount of time, resulting in similar soil types. The differences between the Valley’s wetland soils and those in the Beaver Creek watershed are the presence of multiple, partially reclaimed Upper Freeport and Bakerstown mine spoils and active mining operations in the Beaver Creek watershed. Acid mine drainage (AMD) and sediments from past mining activities flow into many wetlands of the Beaver Creek watershed, resulting in wetland degradation. AMD is acidic, sulfate-rich runoff that forms upon exposure of pyrite (FeS2) or other sulfidic materials to O2 and water. Though the chemistry is variable, AMD typically has high metal concentrations, with Fe and Mn in the range of 20–2000 and 4–126 mg L-1, respectively, and pH of 2–4.5 (Appalachian Regional Commission 1969). Furthermore, elevated aluminum (Al) and sulfate (SO4 2-) concentrations are often associated with AMD (Skousen et al. 2000). In the Appalachian coal mining region, FeS2, which is commonly found in Upper Freeport coal seams, is the primary mineral that forms AMD. When exposed to O2 and water, FeS2 is oxidized, and Fe+2, acidity (H+), and SO4 2- are generated (Geidel and Caruccio 2000). The West Virginia Division of Highways (WVDOH) Corridor H project, in Tucker County’s Beaver Creek watershed, will affect wetland habitats. The WVDOH intends to minimize any potential impacts during disturbance of these habitats. The objectives of this study were to assist WVDOH in evaluating wetlands along the proposed route of Corridor H, and to establish the baseline status of the Beaver Creek watershed’s wetlands. We also compared physical and chemical properties of wetlands impacted by AMD to unimpacted wetlands. The study was part of a larger ongoing study to evaluate the impacts of AMD on sulfate reduction and other soil properties. Methods Site descriptions and sampling locations In the summer of 2001, nine transects were established in four Beaver Creek watershed wetlands (Fig. 1). We described soils and sampling locations every 50 or 100 m along the transect, depending on the wetland size. In total, 31 description and sampling locations were included in the study, and each soil was described according to standard soil survey procedures (Soil Survey Division Staff 1993). A Dutch auger and peat sampler were used to expose soils for descriptions Southeastern Naturalist K. Stephens, J. Sencindiver, and J. Skousen 2015 Vol. 14, Special Issue 7 43 and to remove soil samples. If we observed soil uniformity along a transect, we did not take samples from all description sites. We sampled 25 of the 31 described soils for laboratory analyses. Sampling was based on soil profile descriptions, with organic horizons sampled by type (i.e., Oi sampled together, Oe sampled together, Oa sampled together). We sampled subsurface mineral horizons at the following depths: M1 = 0–10 cm, M2 = 10–30 cm, M3 = 30–60 cm, and M4 = 60–100 cm, with M1 the uppermost layer and M4 the deepest. Based upon profile descriptions, the average total thickness of the organic layer and the solum were calculated. The solum is the portion of the soil that has undergone some development and includes all O, A, and B horizons. The four wetlands sampled in this study are identified as Narrow Wetland, Iron Pond, Railroad Grade, and Elder Swamp (Fig. 1). Narrow Wetland is a relatively flat wetland located adjacent and southeast of Rt. 93. We described soils at six sampling locations in this wetland and took samples from three of them. The water table in Narrow Wetland is at or near the surface through much of the year, creating anaerobic conditions. Drainage appeared to be impeded by the presence of Rt. 93, which crosses at the wetland terminus. The soils were impacted by adjacent, partially-reclaimed Upper Freeport coal spoils, located on the wetland’s western side. Tyha latifolia L. (Common Cattail), Carex spp. (sedges), Spahgnum spp. (sphagnum mosses), and Polytrichum spp. (hair-cap mosses) were the prevalent species throughout this wetland. Figure. 1. Aerial photograph of wetland sites in Beaver Creek watershed, Tucker County, WV. Wetlands are green shaded areas with names as designated for our study. Southeastern Naturalist K. Stephens, J. Sencindiver, and J. Skousen 2015 Vol. 14, Special Issue 7 44 Railroad Grade, located on the northwestern side of Rt. 93, was less impacted than Narrow Wetland or Iron Pond. Two transects were established in this wetland, and soils were described and sampled at seven locations: three on one transect and four on the other. Sediments and AMD from the Upper Freeport coal spoils located on the western end of Railroad Grade directly across Rt. 93 appeared to have impacted two of the seven soil sampling locations. The remaining five soils did not appear to be impacted. An abandoned railroad grade on the northeastern side of this wetland impeded drainage, thus creating wetter conditions than would normally prevail in the wetland. Mosses dominated Railroad Grade vegetation, and sparse populations of sedges and cattails were scattered throughout. Iron Pond is the youngest, smallest, and by far the most impacted wetland in the study. Steep, partially reclaimed Upper Freeport coal spoils surround this wetland, and sediments from the spoils have eroded into the wetland and are components of the wetland substrate. This wetland is anthropogenic in nature and received AMD from multiple seeps along the base of the surrounding spoils (Fig. 1). A small stream bisects the Iron Pond wetland before it spills into an orange-colored pond that is impounded by Rt. 93 and surrounding mine soils. Route 93 hinders drainage where it crosses at Iron Pond’s terminus. The combination of Rt. 93 and the presence of mine spoils were the main factors affecting soil development in this wetland. Soils were described and sampled at five locations along one transect spanning the length of Iron Pond (Fig. 2). Sampling locations represented the diverse assemblage of cattails, mosses, and sedges which dominated this wetland’s flora. Elder Swamp, the largest wetland in the study area, is classified as an Exceptional Resource Value and as a Wetland Special Area by West Virginia University (1995). These classifications designate wetlands with unique qualities and properties. Elder Swamp is situated on the northwestern side of Rt. 93 and east of Iron Pond. Five transects were established in the following habitats: one each in alder thicket, forested wetland, and scrub-shrub wetland habitats; and two in fen-marsh wetland areas. Soils were described at three locations and sampled at two locations in the alder thicket, an area dominated by Alnus incana ssp. rugosa L. (Speckled Alder). We described and sampled soils at two locations in the forested wetland, which was located in a thin stand of Pinus strobus L. (White Pine) slightly upslope from the alder thicket and the fen-marsh wetland. Soils were also described and sampled at two locations in the scrub-shrub wetland, which was located adjacent to the forested wetland and dominated by mosses, Vaccinium spp. (blueberries), and Hypericum spp. (St. John’s wort). Soils in the forested and scrub-shrub wetlands were similar, so we grouped them together and will refer to them as the soils of the transitional wetlands. Fen-marsh wetland vegetation was dominant in Elder Swamp, where sphagnum and hair-cap mosses covered the ground and a few sedges were present. The fen borders the marsh, which makes up the central portion of Elder Swamp. The marsh was dominated by cattails growing on peat hummocks. A braided stream that feeds nearby beaver ponds Southeastern Naturalist K. Stephens, J. Sencindiver, and J. Skousen 2015 Vol. 14, Special Issue 7 45 flows through the marsh area. Soils were described at six locations and sampled at four locations on the two Elder Swamp transects. Laboratory analyses All soil samples were air-dried and passed through a 2-mm sieve for analyses. Soil samples were analyzed for total C, total N, and total S on a LECO CNS 2000 analyzer. Soil pH was measured in a 1:1 soil water suspension using a silver/ silver chloride double junction electrode. Many of the soil samples with high C values required water additions to the pH water:soil suspensions. All mineral soils were analyzed for relative particle size using the pipette method (Soil Survey Staff 1996). Other analyses, completed as part of a larger study, are reported elsewhere (Stephens 2003). Redox potential In the fall of 2001, platinum-tipped redox electrodes were constructed according to Faulkner et al. (1989). In December 2001, electrodes were placed in the ground and the first measurements were taken 24 hours after initial placement. Redox measurements were made in the field for 15 months using a portable voltmeter and a saturated calomel reference electrode. Measurements were not taken in February 2003 due to a severe snowstorm which restricted access to the study sites. Meter readings were adjusted by adding 200 mV, so redox potential was based on the standard hydrogen reference electrode, not the saturated calomel electrode. At each redox sampling location, six probes were permanently placed in the soil at a depth of 20 cm and six probes were placed at a depth of 10 cm. Measurements from the six electrodes were averaged and reported as a single redox value. Measurements were taken at three locations in each of the following wetlands: Railroad Grade, Iron Pond, and Elder Swamp. In Railroad Grade, one location was AMD-impacted and the other two locations were not impacted. At Iron Pond, measurements were taken from three locations representing three different vegetative communities. All sites in Iron Pond were AMD-impacted. In Elder Swamp, measurements were taken from soils in the fen, marsh and scrubshrub areas. None of these areas was considered to be AMD-impacted. Redox probes at the 10-cm depth were placed in the soil’s A horizon in the scrub-shrub site and in organic horizons at all other sites. At the 20-cm depth, the probes were placed in a B horizon at the scrub-shrub site and in A horizons at Iron Pond stop 5 and Railroad Grade stop 1 (Fig. 2). At all other sites, the 20-cm probes were placed in organic horizons. Soil pH was also measured in the field at 10-cm and 20-cm depths when each of the monthly redox potential readings was taken. To simplify interpretation of Eh readings and comparisons among wetlands, redox potentials were adjusted to represent Eh at pH 7. This adjustment was made by subtracting 59 mV for each pH unit of decrease below pH 7 (Karathanasis et al. 2003), using field pH measurements, which were taken at the same time as the redox potential Southeastern Naturalist K. Stephens, J. Sencindiver, and J. Skousen 2015 Vol. 14, Special Issue 7 46 measurements. In addition to these field readings, we made measurements in the laboratory using the air-dried and sieved samples. Figure. 2. Map of Iron Pond wetland in the Beaver Creek Watershed. Transect, sampling locations (stops), stream, and AMD seeps are identified. Southeastern Naturalist K. Stephens, J. Sencindiver, and J. Skousen 2015 Vol. 14, Special Issue 7 47 Only results for the measurements at 20 cm from the marsh and scrub-shrub areas of Elder Swamp and an unimpacted area of Railroad Grade are presented here. The values for these wetlands represent the range of redox potential found in all wetlands of this study. Additional redox data may be found in Stephens (2003). Results and Discussion Narrow Wetland The total mean thickness of organic horizons in the Narrow Wetland was 16 cm with a range of 3–42 cm (Table 1). The mean value for Oi horizons was 13 cm with a range of 3–42 cm (Table 1). Oe horizons were described at only three locations, and the mean thickness was 10 cm with a range of 6–17 cm (Table 1). Based on these data, no soil described in this wetland would be classified as a Histosol. No Oa horizons were described in Narrow Wetland, but A horizons were described in every soil. Lack of Oa horizons indicates that the organic material was relatively young and not well decomposed. This wetland most likely became saturated following the construction of Rt. 93 in 1977. This recent formation would explain the presence of thick Oi and Oe surface horizons, but no Oa horizons. Furthermore, the presence of well-developed A horizons is an indicator of soils that have developed under non-saturated conditions. Total C for the organic material had a mean value of 24.1% with a range of 16.2–29% (Table 2). Mean C values sharply decreased from the organic layer to the mineral layer, but the C values were relatively high for a mineral soil, ranging from 6.3–6.7%. Total S was highest in the organic layer at 0.9% with a range of 0.3–1.6%. The highest S value (1.6%) was observed in a soil located near an AMD seep. This value is similar to those reported for high S tidal marsh soils (Rabenhorst and Hearing 1989), indicating that this soil received more S than normal fresh water wetlands. Sulfate introduced by inflowing AMD was most likely the source of this additional S. Values for N (0.4%) and pH (5.6) were highest in the organic layer, and decreased in the mineral layers (Table 2). The average texture in the M1 layer was silty clay loam, and the texture of the M2 and M3 layers was silt loam (Table 3). Iron Pond In Iron Pond, five locations were established along one transect, with stop 1 located at the head of Iron Pond and stop 5 located closest to Rt. 93 (Fig. 1). Of the five locations, all except stop 1 received some AMD. Soils at each location had an Oi horizon consisting mainly of moss fibers. The mean thickness of the Oi horizons (19 cm) was the highest for any wetland in this study (Table 1). No Oa horizons were described at any location, and only two soils contained Oe horizons (Table 1). As with the Narrow Wetland, Iron Pond had no well-decomposed Oa horizons. Since Oi horizons were dominant, the mean total thickness of the organic horizon was similar to the mean thickness of the Oi horizon (Table 1). The solum thickness of 31 cm was the thinnest of any wetland (Table 1), indicating that the soil was the youngest in the study. Southeastern Naturalist K. Stephens, J. Sencindiver, and J. Skousen 2015 Vol. 14, Special Issue 7 48 Table 1. Thickness of individual organic horizons (Oi, Oe, and Oa), total thickness of all organic horizons, A horizon thickness, and solum thickness. Values expressed as a mean (in cm) of all soils sampled in each wetland, with n representing the number of soils in each the wetland that contained the horizon type. Solum thickness defined as sum of all organic, A and B horizons. Organic layer Oi Oe Oa thickness A Solum Wetland ID n Mean Range n Mean Range n Mean Range n Mean Range n Mean Range n Mean Range Narrow Wetland 5 13 3–42 3 10 6–17 0 - - 6 16 3–42 6 9 2–15 6 39 10–69 Iron Pond 5 19 18–23 2 12 11–13 0 - - 5 29 20–50 1 6 6 5 31 20–50 Railroad Grade 7 16 7–33 7 8 4–12 4 16 12–24 7 33 17–54 4 8 6–19 7 52 25–66 Elder Swamp: alder thicket 1 3 3 3 6 3–7 3 16 6–31 3 22 13–34 1 6 6 3 101 94–108 transitional soils 0 - - 2 6 4–7 3 6 4–9 4 7 4–11 4 4 2–5 4 106 90–123 fen marsh 5 12 4–23 6 20 7–38 6 55 12–99 6 94 48–120 0 - - 7 102 90–120 Southeastern Naturalist K. Stephens, J. Sencindiver, and J. Skousen 2015 Vol. 14, Special Issue 7 49 The average texture for the mineral layers was loam or silt loam, but the ranges indicate extreme differences in texture among individual soils (Table 3). The sand content of the mineral layers decreased along the transect from the head of Iron Pond to the terminus of the transect. As a result, the proportion of soil clay and silt increased along the transect. The decrease in sand across the wetland probably related to deposition of sands from sediments entering at the head of the wetland via the stream. Table 2. pH and total carbon, sulfer, and nitrogen, for wetland soils of the Beaver Creek watershed. n = the number of soil samples taken from each wetland, Horizon = soil layer from which samples were taken. Values are expressed as mean for each wetland followed by range. BDL = below detectable limit. Wetland ID/ pH % carbon % sulfur % nitrogen Horizon n Mean Range Mean Range Mean Range Mean Range Narrow Wetland O 3 5.6 5.4–5.9 24.1 16.2–29.0 0.9 0.3–1.6 0.4 0.4–0.5 M1 3 4.9 4.5–5.6 6.7 5.1–9.0 0.6 0.1–1.2 0.1 0.1–0.1 M2 3 5.1 4.5–5.7 6.3 5.1–8.0 0.3 0.1–0.3 0.1 BDL–0.2 M3 3 5.2 4.6–6.1 6.5 5.5–8.2 0.2 0.1–0.3 0.1 0.1–0.2 Iron Pond O 5 4.2 3.2–6.0 15.4 9.6–21.4 0.8 0.2–1.6 0.2 0.2–0.3 M1 5 4.9 4.5–5.7 4.0 2.5–8.7 0.3 0.1–0.7 0.1 BDL–0.1 M2 5 4.5 4.0–4.8 1.3 0.8–1.5 0.1 0.1–0.2 BDL BDL M3 5 4.5 4.4–4.9 2.2 1.4–4.3 0.1 BDL–0.1 BDL BDL M4 2 4.9 4.4–4.8 2.8 1.4–4.3 0.1 0.1 BDL BDL Railroad Grade O 7 4.5 4.0–5.2 28.8 18.7–35.0 0.8 0.5–1.6 0.4 0.1–0.8 M1 7 4.2 3.9–4.7 6.9 4.3–11.2 0.2 0.1–0.7 0.1 BDL–0.2 M2 7 4.3 3.9–4.9 9.6 1.6–25.4 0.5 0.1–2.1 0.3 0.1–0.3 M3 4 4.3 4.3–4.4 6.8 1.2–7.0 0.3 0.1–0.5 0.1 BDL–0.1 Elder Swamp: alder thicket O 2 5.4 5.0–5.8 31.9 29.0–34.8 0.6 0.5–5.0 1.2 1.0–1.4 M1 2 4.8 4.7–4.8 9.3 9.2–9.5 0.1 0.1 0.1 0.1 M2 2 4.7 4.7–5.0 9.2 8.1–10.4 0.2 0.1–0.2 0.1 0.1–0.2 M3 2 4.8 4.6–5.0 7.9 4.6–11.3 0.3 0.2–0.4 0.2 0.1–0.3 Elder Swamp: transitional soils O 4 4.0 3.8–4.3 30.5 23.2–33.8 0.2 0.2–0.4 1.1 0.9–1.5 M1 4 3.7 3.6–3.8 6.4 2.4–8.2 0.1 0.1 0.2 0.1-–0.2 M2 4 3.9 3.9–4.0 2.5 1.6–3.6 BDL BDL BDL BDL–0.1 M3 4 4.2 4.1–4.3 2.0 0.7–4.6 BDL BDL BDL BDL M4 4 4.6 4.4–4.8 2.0 0.7–3.9 BDL BDL–0.1 BDL BDL Elder Swamp: fen marsh O 4 4.6 3.8–5.7 36.5 25.2–47.0 0.7 0.5–1.1 1.1 0.9–1.3 M1 2 4.3 3.9–4.7 4.9 2.0–7.8 0.1 BDL–0.1 0.1 BDL–0.1 M2 1 4.8 4.8 4.8 4.8 0.1 0.1 BDL BDL M3 1 4.8 4.8 4.9 4.9 0.1 0.1 0.1 0.1 Southeastern Naturalist K. Stephens, J. Sencindiver, and J. Skousen 2015 Vol. 14, Special Issue 7 50 The mean pH of the organic layer was 4.2, the lowest in the study, but values ranged from 3.2–6.0 (Table 1). From stop 1 to stop 5, the pH of the Oi horizons sharply decreased as follows: 6.0 > 5.0 > 3.4 > 3.6 > 3.2 (Fig. 1). The lower pH reflected the addition of AMD along the length of the wetland from stop 1 to stop 5. The pH of the mineral material was similar to that of the mineral material in the other wetlands. Iron Pond’s organic layer had the lowest total C (15.4%) of any wetland (Table 2). Mean C content of the mineral layers was very low with a range of 1.3% in the M2 layer to 4.0% in the M1 layer (Table 2). Sulfur concentrations were highest in the O layer, which had an average of 0.8% and a range of 0.2– 1.6% (Table 2). The high S values in conjunction with low C values indicated that the wetland retained some S from sulfate imported by AMD. Organic and mineral layer N values were the lowest of any soil in the study (Table 2). Railroad Grade The mean thickness of the organic layer (33 cm) was second only to Elder Swamp’s fen-marsh wetland (Table 1). The organic layer thickness ranged from 17–54 cm (Table 1), with the lower values observed in the two impacted soils. Table 3. Particle size of all mineral layers in each wetland of the Beaver Creek watershed. Values expressed as mean, with n representing the number of mineral layers used to obtain the value. SiL = silt loam, SiCL = silty clay loam, L = loam, LS = loamy sand. % sand % silt % clay Texture Wetland ID Horizon n Mean Range Mean Range Mean Range class Narrow Wetland M1 3 6.8 1.5–7.2 54.6 51.0–58.3 38.6 29.6–47.5 SiCL M2 3 18.7 7.2–31.0 55.3 46.9–58.9 26.0 22.1–28.1 SiL M3 3 29.1 19.5–31.0 50.8 42.41–58.9 20.5 17.8–21.6 SiL Iron Pond M1 5 31.9 4.1–66.3 46.6 20.3–66.4 24.1 12.85–36.9 L M2 5 36.3 6.1–66.0 50.5 20.3–49.4 22.5 14.9–32.1 SiL M3 5 22.9 22.2–23.6 51.0 46.5–55.5 26.1 20.8–31.3 SiL Railroad Grade M1 7 8.7 0.5–13.7 61.6 53.8–86.1 29.2 7.6–41.8 SiL M2 7 8.7 1.4–18.6 56.7 46.4–64.7 29.7 32.8–38.7 SiCL M3 4 23.8 2.6–61.6 49.3 25.6–59.2 27.0 12.8–41.5 SiCL Elder Swamp: M1 2 2.7 1.8–3.6 61.6 29.3–63.9 35.8 32.6–38.9 SiCL alder thicket M2 2 3.6 3.4–3.7 60.5 59.6–61.4 36.0 35.2–36.7 SiCL M3 2 23.9 14.0–33.8 52.6 45.6–59.6 23.5 20.6–26.4 SiL Elder Swamp: M1 4 17.1 3.7–37.6 59.4 45.3–65.6 23.5 17.1–35.1 SiL transitional soils M2 4 25.0 15.3–40.8 48.7 36.0–59.6 26.2 22.8–28.5 SiL M3 4 37.8 33.4–47.1 40.6 32.9–47.3 23.9 20.0–28.0 SiL M4 4 38.4 30.0–53.6 43.0 31.6–47 20.8 14.8–32.4 SiL Elder Swamp: M1 2 18.2 9.3–27.1 53.2 47.1–59.3 28.7 13.7–43.6 SiL fenmarsh M2 1 78.5 78.5 16.9 16.9 4.6 4.6 LS M3 1 50.7 50.7 39.7 39.7 9.7 9.7 L Southeastern Naturalist K. Stephens, J. Sencindiver, and J. Skousen 2015 Vol. 14, Special Issue 7 51 The impacted soils were subjected to an input of sediments and AMD from nearby mine spoils, two factors that probably inhibited organic horizon development. Soils were described and sampled at seven locations in Railroad Grade and all had Oi and Oe horizons (Table 1). Four of the seven soils had Oa horizons and the mean thickness of the Oa was 16 cm, second highest in the study (Table 1). The 52-cm solum thickness at Railroad Grade was greater than the thicknesses at Narrow Wetland and Iron Pond, but only about half the solum thickness at Elder Swamp (Table 1). An abandoned railroad grade, situated along the entire length of the northeastern border of the Railroad Grade wetland, has slowed drainage and changed the wetland hydrology, similar to the effects of Rt. 93 on Narrow Wetland and Iron Pond. Mean values for pH of both the organic layer and mineral layers ranged from 4.2–4.5 (Table 2). Average C concentration in the organic layer was 28.8% with a range of 18.7–35.0%, which was similar to those in the Alder Thicket and Narrow Wetland (Table 2). Average C content in the mineral layers was similar to the values for Narrow Wetland. S content (0.8%) and range (0.5–1.6%) for the organic layer were almost identical to values for Narrow Wetland and Iron Pond (Table 2). As observed at Iron Pond and Narrow Wetland, the higher S concentrations were observed in impacted organic layers, indicating that S is being retained in the AMD-impacted soils. Nitrogen values also were similar to those observed in Iron Pond and Narrow Wetland. The organic layer had a mean total N value of only 0.4%, and the mineral layers had either 0.1% or 0.3% total N (Table 2). Texture of the mineral soils was consistent throughout the wetland, with silty clay loam or silt loam textures observed in all soils. Alder thicket Oa and Oe horizons were described at every location in the Elder Swamp alder thicket habitat. The mean thickness was 16 cm for the Oa and 6 cm for the Oe horizon (Table 1). An Oi horizon was described at only one location. These data indicate that organic material in this wetland decomposed at a higher rate than the decomposition rates that we observed at some of the other wetlands, resulting in the formation of 16-cm-thick, well-decomposed Oa horizons. The presence of only one Oi horizon and thin (6 cm) Oe horizons supported these findings. When soils in this wetland were sampled and described, the water table was 20 cm below the surface. Therefore, this wetland was not saturated at the surface year round. The mean solum thickness was 101 cm, which means that soil in this wetland was more pedologically developed than soils in Iron Pond, Narrow Wetland, and Railroad Grade. The mean thickness of the organic layer was only 22 cm with a range of 13–34 cm (Table 1). Yet in the mineral layers, C contents were almost 10%, indicating this wetland accumulated and incorporated C in the mineral soil. The mean C content of the organic layer was 31.9% with a range of 29.0–34.8% (Table 2). These values were similar to the 26.2% total C that Chambers (1996) reported for a wetland dominated by alder in the Valley. Mean total N values of the organic Southeastern Naturalist K. Stephens, J. Sencindiver, and J. Skousen 2015 Vol. 14, Special Issue 7 52 layer were 1.2% with a range of 1.0–1.4%. These were the highest N values observed in the organic layer of any wetland in the study and they were similar to the 1.3% reported by Chambers (1996) for a wetland dominated by alder in the Valley. At 0.6%, mean S values were lower than those observed in Iron Pond, Narrow Wetland, and Railroad Grade. The alder thicket area is not impacted by AMD, therefore high S values were not expected. The S content of the mineral layers was similar to the mineral layers of other wetlands. The mineral material consisted of silty clay loam textures in the M1 and M2 layers which overlay a silt loam texture in the M3 layer (Table 3). The sand content was never greater than 3.7% in the M1 and M2 layers, but sharply increased to as high as 33.9% in the M3 layer (Table 3). This finding indicated that sediments in the mineral layers (M1, M2, M3) were deposited under different conditions. Transitional wetlands (forested and scrub-shrub wetlands) Forested and scrub-shrub wetlands were saturated at the surface for only short periods of time in most years, as indicated by a thin (7 cm), well-decomposed organic layer (Table 1). All organic horizons in these soils were either Oa or Oe. As we observed in the alder thicket wetland, lack of Oi horizons indicated that organic materials decomposed faster than the organic materials in other wetlands. Soils in these wetlands were the driest in the study. The solum thickness (106 cm) was the greatest in the study, which means that the soils in transitional wetlands were well developed (Table 1). The solum mostly consisted of deep Bt and Bx horizons in the subsurface. The Bt is a horizon of clay accumulation, and the Bx is a fragipan, which is a loamy layer that is dense, brittle, and restricts root and water movement. The fragipan was responsible for perching the water table, which in turn has allowed hydric soil conditions to develop. The mean C content of the organic layer was 30.5% with a range of 23.2–33.8% (Table 2). This value was similar to those observed in Narrow Wetland, Railroad Grade, and the alder thicket portion of Elder Swamp. Unlike those wetlands, the transitional wetlands had much thinner organic layers and were much drier. In the M1 layer, the carbon content was 6.4%, with a range of 2.4–8.2% (Table 2). This layer was primarily made up of A horizons, which are mineral horizons that usually contain the most carbon. In the M2, M3, and M4 layers, the C content dropped to around 2.0% (Table 2). Total N for the organic layer was 1.1% with a range of 0.9– 1.5% (Table 2). N in the mineral layers ranged from below detection limits (BDL) to 0.2% (Table 2). For both the organic and mineral layers, S content was the lowest in the study. Only 0.2% S was observed in the organic layer, and observed mineral layer S was BDL–0.1% (Table 2). All pH values in these soils were near 4.0 (Table 2). Alder thicket soils are characterized by low base saturation. The mineral material consisted of silty clay loam and loam textures (Table 3). Fen-marsh Soils in the Elder Swamp fen-marsh wetland were deep, saturated, and high in organic C. The mean thickness of the organic layer was 94 cm with a range of Southeastern Naturalist K. Stephens, J. Sencindiver, and J. Skousen 2015 Vol. 14, Special Issue 7 53 48–120 cm (Table 1). In most of the described soils, no mineral material was observed to a depth of 120 cm, the maximum depth of sampling. However, mineral material was described at two locations. In one, it was a silty clay loam C horizon and at the other location it consisted of both B and C horizons. The silty clay loam material may be perched and holding the water table in this wetland. The organic layer consisted of Oi, Oe, and Oa horizons. The mean thickness of the Oa horizons was 55 cm, by far the thickest in any wetland (Table 1). Organic C content of the organic layer was 36.5% with a range of 25.2–47.0% (Table 2). This was the highest value observed in the study and is similar to values observed by Chambers (1996) in a moss-lichen wetland in the Valley. Average S concentration in the organic layer was 0.7% with a range of 0.5–1.1% (Table 2), similar to values observed in all other wetlands except the transitional wetlands. Total N in the organic layer was 1.1% with a range of 0.9–1.3% (Table 2). The lowest N values were observed at the two locations sampled in the fen. The average pH of the organic layers was 4.6, but ranged from 3.8–5.7 (Table 2). The range reflected differences in pH between soils in the marsh and the fen areas of the Elder Wetland. The fen is a cation-poor environment that receives no significant inflow and therefore has no significant outflows. Acids generated during the decomposition of organic matter create a low pH environment. In the marsh, a stream provides a constant influx of cations, which buffer the organic acids. The pH in the marsh soils ranged from 4.8–6.1. In stark contrast, the pH in fen soils ranged from 3.7–4.1. Redox potential Eh values in the scrub-shrub wetland (Fig. 3) were variable with the lowest readings occurring in the wetter spring months, indicating a higher water table and the presence of reducing conditions. In the drier summer months and colder winter months, the increased redox potential (>175 mV) indicates that the soil was not reduced. The higher redox potential is the result of a lower water table in summer and decreased microbial activity in the winter months, and helps to explain the presence of a minimal organic layer (7 cm) in the scrub-shrub wetland soils. Organic matter accumulates when persistent reducing conditions are present, but under variable reducing conditions, such as those present in the scrub-shrub wetland, organic materials undergo increased decomposition. Furthermore, these soils contained mottles near the surface, which are expected when a fluctuating water table is present. The soils in Railroad Grade were reduced throughout the entire study period with all Eh values consistently less than 175 mV (Fig. 3). The lower Eh values we observed are not surprising given the 38-cm accumulation of organic material. Yet, the Eh fluctuated as the redox potential increased in August and September of 2002 when the water table is normally the lowest. The most reduced soil in the study was that of the Elder Swamp fen-marsh area (Fig. 3). Except for December 2001 and January 2003, all values for Eh were less than -200 mV, indicating strong reducing conditions that favor organic Southeastern Naturalist K. Stephens, J. Sencindiver, and J. Skousen 2015 Vol. 14, Special Issue 7 54 Figure. 3. Redox potential for fifteen months at 20–cm depth in three wetland soils of the Beaver Creek watershed. Redox potentials expressed as the mean of values from six redox electrodes; values are adjusted to a soil pH reference value of 7. Southeastern Naturalist K. Stephens, J. Sencindiver, and J. Skousen 2015 Vol. 14, Special Issue 7 55 matter accumulation. These findings correspond to the 94 cm of organic material observed in the fen-marsh wetland (Table1). Conclusions Soils have developed under different conditions in each wetland, leading to variation in soil properties among wetlands. Some wetlands contain Histosols that have been developing for extended periods of time, and other wetlands contain young, undeveloped fill material. The fen-marsh wetland in Elder Swamp consisted mainly of Histosols formed by years of organic matter accumulation via reduced decomposition rates. Redox potential data from the fen-marsh site confirm the presence of year-round strong reducing conditions, which favor organic matter accumulation. Upslope from the fen-marsh wetland is the drier transitional wetland with lower total C values. Soils in the transitional wetland are subjected to a fluctuating water table, which has resulted in mottles, thin organic layers, and variable redox conditions. The alder thicket soil is much wetter than the soil in the transitional wetland but it is not as saturated as the fen-marsh soil. However, organic material is accumulating in the alder thicket wetland, as evidenced by the high C content of the mineral layers. Elder Swamp was the only wetland considered to be completely unimpacted by mining activities and AMD. Sediments and AMD from mine spoils impacted the other wetlands: mine spoils border Narrow Wetland on the western edge and an AMD seep was adjacent to one sampling point, which resulted in high S values and low soil pH. Additionally, hydrology in Narrow Wetland was altered by the construction of Rt. 93, which dammed the water creating saturated conditions. As a result, organic matter is beginning to accumulate. Iron Pond is severely impacted by mine spoils. The soil in this wetland consisted of fill material in the subsurface with thick, low pH, Oi horizons present at the surface. Very little soil development was observed in this wetland, yet a diverse vegetative community is growing in Iron Pond. Railroad Grade is partially impacted. The impacted soils had high S values, and contained sediments from nearby mine spoils. The non-impacted soils have thick organic layers, but are not yet Histosols. We believe that the nearby railroad grade is impeding drainage so these soils are in the process of accumulating organic materials. Acknowledgments This research was supported by funds provided by the West Virginia Division of Highways, Charleston, WV and the West Virginia Agricultural and Forestry Experiment Station, Morgantown, WV. We appreciate the assistance of Dr. R. Fortney and Dr. R. Viadero in completing this research. Literature Cited Appalachian Regional Commission. 1969. Acid mine drainage in Appalachia. Congressional House Document 91-180. Vol. I–III. Appendix C. Appalachian Regional Commission, Washington, DC. Southeastern Naturalist K. Stephens, J. Sencindiver, and J. Skousen 2015 Vol. 14, Special Issue 7 56 Chambers, D.B. 1996. Physical, chemical, and biological data for four wetland habitats in Canaan Valley, WV. US Geological Survey. Open-File Report 95-334. 23 pp. Environmental Laboratory. 1987. Corps of Engineers wetlands delineation manual. Technical Report. Y-87-1. US Army Engineer Waterways Experiment Station, Vicksburg, MI. 98 pp. plus appendices Faulkner, S.P., and W.H. Patrick, Jr. 1992. Redox processes and diagnostic wetland indicators in bottomland hardwood forests. Soil Science Society of America Journal 56:856–865. Faulkner, S.P., W.H. Patrick, and R.P. Gambrell. 1989. Field techniques for measuring wetland soil parameters. Soil Science Society of America Journal 53:883–890. Fuhrmann., J.J. 1998. Microbial metabolism. Chapter 10. Pp. 189–217, In D.M. Sylvia, J.J. Fuhrmann, P.G. Hartel, and D.A. Zuberer (Eds.). Principles and Applications of Soil Microbiology. Prentice-Hall, Inc. Upper Saddle River, NJ. 695 pp. Gambrell, R. ., and W.H. Patrick, Jr. 1978. Chemical and microbiological properties of anaerobic soils and sediments. Pp. 375–423, In D.D. Hook and R.M. Crawford (Eds.). Plant Life in Aerobic Environments. Ann Arbor Science Publication, Incorporated, Ann Arbor, MI. 564 pp. Geidel, G., and F.T. Caruccio. 2000. Geochemical factors affecting coal mine drainage quality. Pp. 105–130. In R.I. Barnhisel, R.G. Darmody, and W.L. Daniels (Eds.). Reclamation of Drastically Disturbed Lands. American Society of Agronomy, Madison, WI. 1082 pp. Howeler, R.H., and D.R. Bouldin. 1971. The diffusion and consumption of oxygen in submerged soils. Soil Science Society of America Journal 35:202–208. Kararthanasis, A.D., Y.L. Thompson, and C.D. Barton. 2003. Long-term evaluations of seasonally saturated “wetlands” in western Kentucky. Soil Science Society America Journal 67:662–673. McLatchey., G.P., and K.R. Reddy. 1998. Regulation of organic matter decomposition and nutrient release in a wetland soil. Soil Science Society of America Journal 27:1268–1274. Mitsch W.J., and J.G. Gosselink. 2000. Wetlands. 3rd Edition. John Wiley and Sons, Inc. New York, NY. 936 pp. National Technical Committee for Hydric Soils. 2004. The Hydric Soil Technical Standard. Hydric Soil Technical Note 11. USDA Natural Resources Conservation Service. Washington, DC. 23 pp. Available online at ftp://ftp-fc.sc.egov.usda.gov/NSSC/Hydric_ Soils/HSTS.pdf. Accessed 8 December 2004. Pickering, E.W., and P.L.M. Veneman. 1984. Moisture regimes and morphological characteristics in a hydrosequence in central Massachusetts. Soil Science Society of America Journal 48:113–118. Ponnamperuma, F.N. 1972. The chemistry of submerged soils. Advanced Agronomy 24:29–96. Rabenhorst, M.C., and K.C. Haering. 1989. Soil micromorphology of a Chesapeake Bay tidal marsh: Implications for sulfur accumulations. Soil Science 147:339–347. Skousen, J.G., A. Sexstone, and P.F. Ziemkiewicz. 2000. Acid mine drainage control and treatment. Pp. 131–168. In R.I. Barnhisel, R.G. Darmody, and W.L. Daniels (Eds.). Reclamation of Drastically Disturbed Lands. American Society of Agronomy, Madison, WI. 1082 pp. Soil Survey Division Staff. 1993. Soil Survey Manual. USDA Handbook No. 18. US Government Printing Office. Washington, DC. 315 pp. Southeastern Naturalist K. Stephens, J. Sencindiver, and J. Skousen 2015 Vol. 14, Special Issue 7 57 Soil Survey Staff. 1996. Soil Survey Laboratory Methods Manual. Soil Survey Investigations Report No. 42. Version 3.0 National Soil Survey Center, Lincoln, NE. 716 pp. Stephens, K. 2003. Characterization of wetland soils in the Beaver Creek watershed. M.Sc. Thesis. West Virginia University, Morgantown, WV. 131 pp. Available online at http://kitkat.wvu.edu:8080/files/2864.1.Stephens_Kyle_thesis.pdf. USDA-NRCS. 1995. Changes in hydric soils of the US. Federal Register, Volume 60 (37)/ Friday, 24 February, P. 10,349. US Government Printing Office. Washington, DC. West Virginia University. 1995. Corridor H/ Blackwater River restoration. Unpublished report submitted to the West Virginia High Technology Consortium, WVHTC-FS95- 1011. Report is available from the West Virginia University National Research Center for Coal and Energy, Morgantown, WV. 45 pp.