Comparative Analysis of Metal Concentrations and Sediment Accumulation Rates in Two Virginian Reservoirs, USA: Lakes Moomaw and Pelham
Lacustrine sedimentation and trace metal accumulation are naturally occurring processes that can be altered by anthropogenic activities. Indices of sediment or metal dynamics are important for the management and operational use of man-made reservoirs and their drainage basins. In this study, we compared two reservoirs in Virginia, USA, to quantify the effect of varying watershed characteristics on sediment and metal fluxes. Lake Pelham is a human-impacted reservoir surrounded by agricultural fields and anthropogenic developments, whereas Lake Moomaw is an undeveloped reservoir surrounded by moderate to extremely sloping forested landscapes. Three sediment cores were taken from each reservoir to estimate 210Pb-based sediment accumulation rates, organic matter content, and indices of trace metal enrichment and accumulation. The average 210Pb-based sediment accumulation rates were 0.348 ± 0.053 and 0.246 ± 0.043 g cm−2 year−1 for Lake Pelham and Lake Moomaw, respectively. The sediment trace metal results showed strong correlation with sediment organic content, and both reservoirs had moderate to high enrichment of Cu and little enrichment of Zn and Pb. Overall, Lake Moomaw had relatively low sediment accumulation and metal enrichment. Comparatively, Lake Pelham had significantly greater metal concentrations, which were highest in the upper reaches of the reservoir. Lake Pelham also had higher sediment accumulation rates and higher metal enrichment, reflecting the impact of human development within the greater watershed. Results from this study suggest that urbanization can increase reservoir sediment and metal fluxes, but atmospheric deposition is also important in forested watersheds that have not undergone anthropogenic land-use change.
KeywordsReservoir sedimentation 210Pb sediment accumulation Trace metals Enrichment factors
Anthropogenic alteration of land since the European settlement of the eastern USA has disrupted the natural cycles of watersheds by affecting sedimentation rates, watershed ecology, and the quality of the water in aquatic systems (Middleton 1953; Dean and Gorham 1998; Walling 2006; Conrad et al. 2007; Trimble 2008; Balogh et al. 2010). Water bodies such as lakes and artificial reservoirs are especially subjected to the consequences of human landscape alteration, which accelerates erosion and sediment fluxes to surface water bodies (F’inkenbine et al. 2000; Newman et al. 2006; Colosimo and Wilcock 2007; Walter and Merritts 2008). Sediments have a major impact in lakes and reservoirs because they contribute to a loss of storage capacity (Odhiambo and Boss 2004, 2006; Newman et al. 2006; Odhiambo and Ricker 2012), which affects the longevity of a reservoir and its potential use for municipal water consumption. Sediments are also the main conveyors of sediment-bound nutrients, metals, and other persistent pollutants to surface waters (Walling 2004; Wang and Cui 2005; Houser et al. 2006; Odhiambo and Ricker 2012). Sediment core analyses are important in establishing historical sediment accumulation rates and pollutant loading (e.g., trace metals) history to a water body.
Metals are introduced into a reservoir in either a soluble or particulate form through various natural and anthropogenic sources such as watershed erosion, point source pollution, and wet deposition (Mueller et al. 1989). Wet deposition occurs when precipitation is enriched with anthropogenic trace elements that become concentrated on atmospheric fine particles and form condensation (Conko et al. 2004). Trace metal contamination usually occurs within industrialized areas and the metals can enter a reservoir through waste effluents, mining activities, fossil fuel combustion, sewage discharge, fertilizers, and pesticides (Förstner and Wittmann 1979).
A fundamental problem with trace metals is that they are not biodegradable, and therefore are persistent in the environment. Once a trace metal has been introduced into a lacustrine system the metals are distributed into the water column, settle on the bottom, and accumulate in sediments (Fichet et al. 1998). As metals adsorb to clays and other particles, the sediments in lakes and reservoirs contain a historic signature of trace metal fluxes (Crecelius and Piper 1973; Shirahata et al. 1980; Renberg 1986; Swain et al. 1992; Foster and Charlesworth 1996; Engstrom and Swain 1997; Brännvall et al. 2001; Balogh et al. 2010). Lacustrine sediment cores have shown increases in metal concentrations since the 1800s and even larger increases in the 1900s (Conrad et al. 2007; Balogh et al. 2010). With the onset and expansion of industrialization, metal concentrations dramatically increased but have recently declined because of the implementation of environmental laws and the banning of leaded gasoline (Conrad et al. 2007).
There are multiple sediment core methods appropriate to determine trends in trace metal fluxes and sediment accumulation. Concentrations of fallout 137Cs and 210Pb are often used to estimate the rates of sedimentation in coastal waters, lakes, and wetlands on scales of several decades to 100 years (Odhiambo et al. 1996; Mizugaki et al. 2006; Conrad et al. 2007; Odhiambo and Ricker 2012). 210Pb is a naturally occurring radionuclide which comes from the decay series of 238U and has a half-life of 22.3 years. 210Pb can attach to the sediment from 222Rn decay (naturally occurring gas found in soil or the atmosphere from 238U decay) and also from 226Ra in the soil. Unsupported or excess 210Pb comes from wet precipitation or atmospheric fallout, and supported 210Pb comes from naturally occurring 222Rn and 226Rn in the soil (Appleby and Oldfield 1992; Jaeger et al. 2009). The total 210Pb in the sediment contains both unsupported and supported 210Pb. 210Pb was used in this study to provide a timeline for studying the historical loading of metals into the study reservoirs.
The objective of this work was to examine sediment deposition in two reservoirs in Virginia, USA, and analyze the historical trends and concentrations of trace metals in the reservoirs. The study watersheds have varying levels of human development, local geology, soil types/erodibility, and topography. One reservoir, Lake Moomaw, is located in the forested, protected Blue Ridge Mountains, but may be subjected to atmospheric influxes of metals from coal production and paper mills in the region. The contrasting reservoir, Lake Pelham, is surrounded by industrial complexes, suburban development, and agricultural fields. This study seeks to establish historical trends of metal fluxes and compare the concentrations to background levels in order to quantify the extent of metal enrichment in modern sediments.
1.1 Background and Regional Setting
Lake Moomaw (Fig. 1) is a 10.24-km2 reservoir located in the Shenandoah Valley region of western Virginia and impounds the Jackson River of the greater James River system. The lake was constructed in 1978 for downstream water quality augmentation, flood control, and recreation. The 857-km2 watershed is largely forested because it lies within the Gathright Wildlife Management Area and the George Washington National Forest (VDGIF 2009). The regional geology is primarily limestone and dolomite and the major soil complexes include the Dekalb-Lily Complex and Weikert-Berks-Rough Complex (Soil Survey Staff 2011). Although the watershed and area surrounding the reservoir are largely undeveloped, there is a large pulp and paper mill approximately 30.6 km downstream of the reservoir. The locale produces “activated carbons” used for chemical production, food purification, and other materials (MWV Specialty Chemicals 2012). The emissions from the paper mill include sulfuric acid, lead compounds, zinc compounds, ammonium, manganese compounds, and many others (US EPA 2010).
2 Sampling and Methods
2.1 Field Sampling
A gravity coring device with plastic liners was used to collect three sediment cores from Lake Moomaw and a manual coring device with plastic liners was used to collect three sediment cores from Lake Pelham (Fig. 1). The sampling site selections were based on water depth and reservoir geomorphology, which influences sediment deposition. After retrieval, the cores were transported in ice to the Earth and Environmental Sciences Laboratory at the University of Mary Washington (Fredericksburg, Virginia, USA), where each core was extruded and subsampled at increments of 2 cm down the core. A fraction of each subsample was used in 210Pb analysis, and the remaining materials were used for trace metal and organic matter analysis.
2.2 Laboratory Analysis
2.2.1 210Pb Analysis and Sediment Accumulation Model
Samples for 210Pb analysis were sent to CORE International Research Labs, Winnipeg, Canada. The 210Pb analysis is based on the method of Eakins and Morrison (1978) in which 210Po is distilled out of sediments at a high temperature, digested by acid, and then plated onto silver disks for alpha spectrometry analysis. The 210Pb is assumed to be in equilibrium with the 210Po, which occurs when sediments are greater than 2 years old. The minimum detection limit for a 0.5-g dry sample was approximately 0.2 dpm 210Po/g dry sample at a confidence level of 95 % for a counting time of 30,000 s. A duplicate was run every 10th sample, a blank run every 20th sample, and detector blanks run every 90 days for quality assurance.
2.3 Trace Metal Analysis
The concentrations of Fe, Al, Ba, Cd, Cr, Cu, Pb, Mn, and Zn were analyzed in all core sections. Based on the method of Zwolsman et al. (1993), subsamples from each 2 cm portion were dried at 100 °C in an oven overnight. The dry sediment was then disaggregated with a mortar and pestle and 1 g of <63 μm size sediments was digested with 20 mL of aqua regia acid solution of three parts hydrocholoric acid: one part nitric acid: one part nanopure water (3HCl/1NO3/3H2O) in a 32.2 °C water bath for 2 h, left in a shaker overnight, and centrifuged for 3 h. The leachate was then filtered into acid resistant bottles and refrigerated to preserve the samples. Each sample was diluted to 3 % concentration solution using nanopure ultra-deionized water. Analysis for trace metals was done using a Thermoscientific iCAP 6000Series ICP-OES. The following wavelengths (nm) were used for analysis to avoid interferences: 396.1 (Al), 233.5 (Ba), 226.5 (Cd), 267.7 (Cr), 224.7 (Cu), 239.5 (Fe), 259.3 (Mn), 220.3 (Pb), and 213.8 (Zn). Recovery analysis was performed using the 0.1 μg/g Cd, 0.1 μg/g Pb, 0.024 μg/g Cd, and 0.024 μg/g Pb standards. The results indicated that the intensity data is substantially consistent with only 0.64 % error for 0.100 μg/g Cd, 0.10 % error for 0.100 μg/g Pb, 1.40 % error for 0.024 μg/g Cd, and slightly greater error of 18.74 % for 0.024 μg/g Pb. Though Cd was analyzed and used for recovery analysis, the analyzed samples were below detection limit and therefore Cd is not discussed in the results section.
Average metal concentrations in the Lake Pelham, Lake Moomaw, natural sediments, and regional areas (micrograms per gram for all elements except Al and Fe)
2.4 Organic Content Analysis
Where LOI 550 is the percent organic matter, DW105 is the weight of the initial dry sample, and DW550 is the sample weight after combustion.
2.5 Statistical Analysis
A (2 × 3) two-way analysis of variance (ANOVA) test was conducted in Sigma Plot 11.2 (Systat Software, Inc., San Jose, CA) in order to examine the relationship between the two reservoirs (Lake Pelham and Lake Moomaw) and the average metal concentrations (Cu, Pb, and Zn) by core location (cores located at the upper, middle, and dam sections). Metal concentrations were aggregated into mean values using the individual 2-cm subsamples (N = 60) from each study core. Interaction and main effects were analyzed for all cores. If interaction effects were not significant, the main effects were analyzed with Tukey’s honestly significant difference tests. Pearson product–moment correlation coefficients (r) were also quantified for each 2-cm subsample to elucidate the relationships among lacustrine sediment characteristics (sample depth, sedimentation rate, organic matter) and trace metal concentrations. All statistical tests were considered significant at α = 0.05.
3 Results and Discussion
3.1 Sediment Accumulation Rates
Comparison of the models-Pb-210 Regression (grams per square centimeter per year) and Pb-210 CRS Model (grams per square centimeter per year)
Pb-210 regression model
Pb-210 CRS model average
Pb-210 regression model
Pb-210 CRS model average
The regression model results from Lake Moomaw showed an accumulation rate of 0.601 g cm−2 year−1 at LM1 near the dam in the deepest part of the reservoir (Fig. 2; Table 2). The mid-section of the reservoir at LM2 had an accumulation rate of 0.335 g cm−2 year−1, whereas the upper reaches at LM3 had an accumulation rate of 0.402 g cm−2 year−1 (Fig. 2; Table 2). The CRS average accumulation rates for the three sites, LM1 (0.295 g cm−2 year−1), LM2 (0.231 g cm−2 year−1) and LM3 (0.212 g cm−2 year−1) are slightly lower than the regression model results, but show similar spatial patterns. The temporal variation depicted in the CRS models shows that accumulation rates have not changed significantly in this basin, with a small modern increase of about 0.06 g cm−2 year−1 estimated for the upper reaches of the reservoir (LM3). The modern accumulation increase relative to the base of lacustrine sediments is significantly higher, 0.12 g cm−2 year−1, in the lower part of the reservoir near the dam (LM1). The spatial pattern in accumulation rates in this reservoir also suggests the dominance of suspended sediment influx, with higher accumulation in the deeper area near the dam. In addition to external fluvial sources, the higher rate of accumulation in the lower reaches may also be attributed to extensive, visible shoreline erosion from the reservoir’s steeply sloping shores. Sedimentation from shoreline erosion has been commonly observed in other regional mountain reservoirs. For example, Fayyad (2010) reported sedimentation from shoreline erosion concentrated along the entire reservoir margins in both Smith Mountain and Leesville lakes in southwestern Virginia. The majority of eroded shoreline materials tend to be concentrated as bench deposits in the littoral areas of mountain reservoirs, but significant amounts of sediment may be redistributed beyond the shallow littoral zones depending on shore slope steepness, slumping, and the intensity of wave action. In addition, shoreline erosion may be prevalent in the early years after reservoir construction, but typically decreases as more resistant underlying rocks are encountered (Fayyad 2010).
Total sediment accumulation was calculated using Lake Pelham and Moomaw surface areas (LP = 8.96 × 105 m2; LM = 1.02 × 107 m2), average lacustrine sediment densities, and the average sedimentation rate at the three locations in each reservoir. Annual accumulation rates of 6,837 and 70,178 m3/year were estimated for Lakes Pelham and Moomaw, respectively. The total annual accumulation rates translate to basin sediment yields of 1.01 and 0.82 mm m−2 year−1 for Lake Pelham (67.8 km2 watershed) and Moomaw (857 km2 watershed), respectively. The impact of the estimated annual accumulation rates is about 1 % annual loss of storage capacity in both reservoirs, not taking sediment compaction into account. The two reservoirs show comparable sediment accumulation rates and the associated loss of capacity using 210Pb regression model estimates. This is unusual because although the two watersheds have a similar climatic environment and similar sandy or silt loam top soils (soil erodibility Kf = 0.41 − 0.45), they have vastly different geomorphic and land-use characteristics. The accumulation rates in Lake Moomaw are likely attributed to highly erodible soils on the steep (>20°) slopes that dominate the watershed as well as significant lake shoreline erosion. However, the sediment flux rates are likely to be lessened by the dense vegetated land cover and the thin soils that characterize the basin. By contrast, the Lake Pelham watershed is relatively flat topographically (0–5° slopes), and the lacustrine sediment fluxes are most likely influenced by land use rather than geologic or geomorphologic characteristics. The absence of a continuous vegetated riparian zone around the reservoir, which currently consists of a major roadway, golf course, agricultural fields, and a residential development, is probably a source of sediment into the reservoir. The sediment fluxes into Lake Pelham are also likely curtailed by a well developed deltaic-wetland zone in the mouth of the main channel where significant amounts of sediments may be trapped before reaching the reservoir; unlike in Lake Moomaw where steep sloping energetic tributaries empty sediment loads into the reservoir without impedance. In Lake Moomaw, stream incision will only be a factor in sediment yield until the system is bedrock controlled. Considering the steep, valley-constricted nature of the first- and second-order tributaries to the reservoir, debris flows are likely the source of sediment to the valley floor rather than vertical incision (Taylor and Kite 2006). Intrinsic stream sediment production is likely important in both basins, although for different reasons. In Lake Moomaw, the basin contains steep high velocity streams and thus higher rates of stream bed incision (until bedrock is reached), sediment transport, and bank erosion (i.e., valley-stream widening) can occur. Whereas in Lake Pelham impervious surfaces associated with human establishments likely magnify runoff and bankfull flooding in reservoir tributaries after storm events, leading to stream incision and internal stream sediment production as well.
The sediment accumulation estimates for both reservoirs in this study are comparable to regional values for reservoirs located in the Piedmont province of Virginia (Lake Anna, Ni Reservoir) that have a sediment accumulation range of 0.023–1.02 g cm−2 year−1 (Odhiambo and Ricker 2012; Pope and Odhiambo 2013). Other regional reservoirs with steeper slopes and anthropogenic alterations, such as Smith Mountain and Leesville, have undergone >5 % storage capacity reduction due to sedimentation (Fayyad 2010). The spatial and temporal patterns in sedimentation in most regional surface water bodies seem to be correlated with anthropogenic stresses and associated land-use and cover changes within the drainage basins (Fayyad 2010; Odhiambo and Ricker 2012; Ortt et al. 2000; Pope and Odhiambo 2013; Saenger et al. 2010).
The Lake Moomaw accumulation rates, even the lower CRS results, seem relatively high compared with other mountainous lakes where natural geomorphic conditions and climate variability are the dominant factors controlling short and long term sediment yields. Appleby and Piliposian (2006) estimated sedimentation in the Tatra Mountain lakes and reported mean rates ranging from 0.0055 to 0.046 g cm−2 year−1 and Guevara et al. (2010) also estimated sedimentation of about 0.013 g cm−2 year−1 in the Patagonia lakes. These other mountain lakes, however, are located in younger geologic landscapes compared with Lake Moomaw and should be expected to yield more sediments. These data suggest that although the valley and ridge province of Virginia is dominated by highly erodible colluviums (Mills 1988), the three cores used here might be overestimating sediment accumulation rates in Lake Moomaw. Possible explanations for the highest sediment accumulation recorded (closest to the dam) may be that the LM1 core captured a relict fluvial channel or an area containing slumping shoreline materials. If this were the case, the LM1 core may not be a true reflection of the entire lower reaches of the basin. Future close core sampling intervals and our ongoing high resolution geophysical sediment surveys in these reservoirs will likely provide more clarity regarding spatial variation in sediment thicknesses.
3.2 Metal Concentrations
3.3 Organic Matter
The lacustrine organic matter results are presented in Fig. 4. Lake Moomaw had lower organic matter content compared with Lake Pelham, ranging from approximately 3 to 9 % at all core depths. All three Lake Moomaw cores showed >2 % increase in organic content with depth, peaking at the 8- to 12-cm interval, corresponding to the pre-/post-impoundment boundary (Figs. 2 and 4). Increased organic matter towards the bottom of the three cores may represent buried soil surface horizons (O or A horizons) and relict vegetation left from reservoir construction that are overlain by clastic mineral lacustrine sediments.
Sediment organic content in Lake Pelham ranged from 2 to 13 %, with LP1 location recording the highest values (Figs. 1 and 4). The temporal trends also showed >2 % increases in organic content at the pre/post-inundation boundary in all the cores (Fig. 4). These data indicate that organic matter flux has increased after the creation of Lake Pelham. The progressive increase in organic matter may be associated with greater allochthonous C fluxes, derived from watershed sources (including soil erosion), or autochthonous C in the shallow areas of the reservoir from the phytoplankton, periphyton, and other organisms residing in the reservoir itself (Kraus et al. 2011). Studies have shown that reservoirs damming fluvial waterways typically receive large allochthonous organic matter loads (Wetzel 1975; Groeger and Kimmel 2009); therefore organic matter inputs from terrestrial sources are likely the main source of the increase in organic matter within Lake Pelham. For both reservoirs, the apparent shift in lacustrine sediment organic matter concentration correlates to the pre-/post-impoundment boundary conditions, with the less disturbed Lake Moomaw receiving less external nutrient fluxes that support aquatic vegetation production.
3.4 Statistical Analysis of Metal Distributions
Summary of (2 × 3) two-way ANOVA analysis for major trace metal concentrations (micrograms per gram) in Lakes Pelham and Moomaw
Source of variation
Lake × position
Lake × position
Lake × position
The general trends of metal concentrations are similar to those of many reservoirs situated on fluvial waterways, where sediments drop from suspension before reaching the lower sections of the lake (Shotbolt et al. 2005). Metal concentrations in Lake Moomaw tended to strongly reflect this relationship, with progressively less metals from the upper sections down towards the dam core position. In Lake Pelham there was a more variable relationship from the upper to dam core positions. There was a significant drop in Pb and Zn concentrations from the upper to middle cores (Fig. 5a) but an increase in these metals in the dam core. The upper core of Lake Pelham receives fluvial influx from adjacent roadways and agricultural fields, while the middle core is disconnected from fluvial inputs and surrounded by patches of forested land (Fig. 1). The forested buffer and distance to upper fluvial waterways has likely contributed to lower metal concentrations in the center of the lake. By contrast, the dam core is subjected to metal influx from a small 2nd order tributary creek (Vaughn Branch) that enters the lower lake from the north (Fig. 1). Small tributary creeks can be significant sources of sediment to large reservoirs in the region (Odhiambo and Ricker 2012), and this creek drains extensive low- to medium intensity developments (Homer et al. 2007) with many road crossings that may have contributed to the higher Pb and Zn levels observed at the dam core position (Yesilonis et al. 2008).
Correlation matrices of trace metal and lacustrine sediment characteristics
Sediment rate (g cm−2 year−1)
Organic matter (%)
Lake Moomaw (N = 24)
Lake Pelham (N = 36)
3.5 Enrichment Factors
The Lake Moomaw EF estimates, relative to average crustal abundances, showed moderate enrichment of Cu with the exception of some spikes at the 10- to 12-cm depth in LM3 and 6- to 8-cm in LM2 (Fig. 6). The cores were significantly enriched in Pb and Zn, ranging from 4.95 to 10.44 for Pb and 5.47 to 7.59 for Zn in LM2 and LM3, respectively. The EFs, relative to regional geochemistry, showed very different results from the average crustal abundances. Pb was deficient in all three cores with EF values of 1.05–2.05. Zn EF values ranged from 0.58 to 4.59 and were deficient to slightly enriched in all three cores. Significant to very high enrichment was observed in the EF values of Cu in LM2 and LM3.
The concentrations of Pb and Zn in Lake Moomaw may be attributed to the second largest pulp and paper mill in Virginia emitting approximately 1,361 metric tonnes of Pb and Zn compounds into the air, causing wet deposition of metals in Lake Moomaw 30.6 km upstream. Lake Moomaw is located north of the pulp and paper mill, and therefore, northerly winds could subject the lake to atmospheric deposition of Pb and Zn. The atmospheric trace metals in Lake Moomaw may also be associated with its proximity to West Virginia, where coal mining processes and coal-fired power generation plants are common (US EPA 2010). The enrichment of Cu in Lake Moomaw may be attributed to any number of sources. Statistical analyses showed that surface concentrations of metals were not uniform spatially or temporally throughout the reservoir. These results may indicate that atmospheric deposition of metals is not the dominant source, because the distribution of metals is fairly heterogeneous. As there are no major industrial operations within the watershed, the Cu enrichment may be attributed to natural parent material sources. Natural sources such as local mineralized lithologic zones, forest fires, biogenic processes, and windborne soil particles may contribute to excess Cu concentrations (Weant 1985) and result in heterogeneous metal accumulation within lacustrine sediments.
3.6 Metal Accumulation Rates
Metal accumulation rates (micrograms per square centimeter per year) for all cores
The sources of trace metals tend to differ between relatively pristine and human-impacted watersheds. In forested watersheds with minimal human developments and lack of point source pollution, metal accumulation is associated with local geology, forest fires, or atmospheric deposition from outside sources (Weant 1985; Baron et al. 1986; Scudlark et al. 2005; Odhiambo et al. 2013). Our results suggest that enrichment of Pb and Zn in Lake Moomaw is likely from atmospheric deposition and Cu is likely derived from local mineralized parent materials. Similar observations have been made in other forested lakes of the eastern USA. Abernathy et al. (1984) showed the significance of atmospheric input to Fontana Lake (a forested mountain lake in North Carolina, USA) where the surface sediment concentrations of Cu, Zn, and Mn were found to be similar to areas receiving industrial pollution. High Cu values in their study were also attributed to local mineralized schists, which is a probable source of Cu to many forested lakes of the region. Wet and dry atmospheric deposition has also been observed in other locations of the eastern USA; for example, Dabous (2002) studied metal enrichment in a pristine lake in Florida, USA, and attributed metal enrichment to wet deposition in recharge areas coupled with an increase in sediment input to the lake from bedrock, soils, and surficial sand erosion. The dynamics of atmospheric deposition in remote mountainous landscapes was also detailed by Lovett and Kinsman (1990) who showed that atmospheric inputs are a major source of pollutants in high elevation ecosystems. Their analysis suggested that increases in atmospheric fluxes of Cd, Cu, Pb, and Zn were greatest at high elevations as a function of more precipitation, relative to adjacent lowlands.
Comparatively, the sources of metals in human-impacted reservoirs are diverse and can include waste effluents, fossil fuel combustion, mining activities, and fertilizers (Förstner and Wittmann 1979). There are many human-impacted reservoirs with metal concentrations comparable to those of our human-impacted watershed draining to Lake Pelham. Lake Anna in central Virginia serves as a cooling water supply for a nuclear power plant and has documented enrichment of Pb, Cu, Zn, and especially Cd in the lacustrine sediment attributed to old mine tailings in the basin and waste materials from the power plant (Odhiambo et al. 2013). The sediment of Lake Anna closest to the nuclear power plant has concentrations similar to the developed Lake Pelham basin with concentrations of Cu, Pb, and Zn at 42.6, 118.7 and 253.7 μg/g, respectively (see Table 1; Odhiambo et al. 2013). Elsewhere in the region, the major sources of metals in the Chesapeake Bay have been identified as point sources, storm-water runoff, and atmospheric deposition (Conko et al. 2004; Conrad et al. 2007; Hartwell and Hameedi 2007). According to Hartwell and Hameedi (2007), the trends of metal input to the Chesapeake Bay have been static or decreasing since 1986 following a peak in the 1970s. The modern sediment in the Bay is enriched by 1.5 to 3.5 times the background (natural) concentrations from pre-European settlement. Lake Pelham is likely subjected to metal fluxes from similar sources; however, contrary to regional trends our data indicate that the metal input has been increasing since the 1980s. In the Lake Pelham watershed, development has only recently occurred as the historically agricultural area has developed into a commuter-rich suburb of the greater Washington, DC region. Increased population growth may be contributing metals to the reservoir through fossil fuel combustion, abrasion of road surfaces, and degraded vehicle components (Conko et al. 2004) as commuter traffic has increased in the area. Comparatively, the sources of metals in the Lake Moomaw watershed are less obvious and are likely atmospheric or geochemical, reflecting the undeveloped, forested nature of the watershed.
The six sediment cores and 210Pb chronologies provided a framework for spatial and temporal analysis of variations in both sediment and trace metal fluxes in Lakes Pelham and Moomaw. The results imply that although Lake Moomaw is still largely undeveloped, the basin’s steep slopes coupled with highly erodible colluvial soils and the prevalence of shoreline erosion has resulted in lacustrine sediment accumulation rates comparable to the anthropogenically altered Lake Pelham. Most of the trace metals present in the sediments of Lake Moomaw may be attributed to watershed erosion on steep slopes, bank erosion, and local atmospheric deposition from nearby paper mill industries in the region as well as regional atmospheric sources such as West Virginia coal processing plants. The geomorphic configuration and characteristic valley-confined streams in the Lake Moomaw watershed makes the reservoir vulnerable to significant increases in sediment accumulation and metal concentrations, if road construction and urbanization were to occur in the basin. By contrast, Lake Pelham sediments show the impact of recent watershed urbanization, as reflected by the progressive increase in metal concentrations, sediment organic content, and sediment accumulation rates during modern time periods.
The authors would like to thank the University of Mary Washington for their financial support during the course of this project and Dr. L. Giancarlo, Sunnan Yoon, and Laura Pilati for their assistance in field sampling and laboratory analysis. The authors would also like to thank the two anonymous reviewers for their insightful suggestions on the earlier versions of this manuscript.
- Appleby, P. G., & Oldfield, F. (1992). Application of lead-210 to sedimentation studies. In M. Ivanovich & R. S. Harmon (Eds.), Uranium-series disequalibrium: applications to earth, marine, and environmental sciences (pp. 731–778). Oxford: Clarendon Press.Google Scholar
- Chillrud, S. N., Bopp, R. F., Simpson, H. J., Ross, J. M., Shuster, E. L., Chaky, D. A., Walsh, D. C., Choy, C. C., Tolley, L. R., & Yarme, A. (1999). Twentieth century atmospheric metal fluxes into Central Park Lake, New York City. Environmental Science and Technology, 33, 657–662.CrossRefGoogle Scholar
- Colosimo, M. F., & Wilcock, P. R. (2007). Alluvial sedimentation and erosion in an urbanizing watershed, Gwynns Falls, Maryland. JAWRA, 43, 499–521.Google Scholar
- Dean, W. E. (1974). Determination of carbonate and organic matter in calcareous sediments and sedimentary rocks by loss on ignition: comparison with other methods. Journal of Sedimentary Petrology, 44, 242–248.Google Scholar
- Demographic Analysis. (2010). Culpeper County Comprehensive Plan. http://web.culpepercounty.gov/Portals/0/Departments/Planning_and_Zoning/Documents/Chapter%202%20-Fina.1.pdf
- Environmental Protection Agency. (2010). Releases: Facility Report. <http://www.epa.gov/cgi-bin/broker?zipcode=24426>. Accessed 07 June 2011
- Fayyad, M.J. (2010). Appalachian power company Smith Mountain Project No. 2201: Sedimentation monitoring plan. A report for the USA Federal Energy Regulatory Commission, Project No. 2210–197, 23p.Google Scholar
- Hartwell, S.I., & Hameedi, J. (2007). Magnitude and Extent of Contaminated Sediment and Toxicity in Chesapeake Bay. NOAA Technical Memorandum NOS NCCOS 47, Silver Spring, MD. Available http://ccma.nos.noaa.gov/publications/NCCOSTM47.pdf
- Homer, C., Dewitz, J., Fry, J., Coan, M., Hossain, N., Larson, C., Herold, N., McKerrow, A., VanDriel, J. N., & Wickham, J. (2007). Completion of the 2001 National Land Cover Database for the Conterminous United States. Photogrammetric Engineering and Remote Sensing, 73, 337–341.Google Scholar
- Middleton, A. P. (1953). Tobacco coast: Richmond. Whittet and Shepperson: Virginia. 482 p.Google Scholar
- Mills, H. H. (1988). Surficial geology and geomorphology of the Mountain Lake area, Giles County, Virginia, including sedimentological studies of colluviums and boulder streams. US Geological Survey Professional Paper No., 1469, 62 p.Google Scholar
- Mmolawa, K. B., Likuku, A. S., & Gaboutloeloe, G. K. (2011). Assessment of heavy metal pollution in soils along major roadside areas in Botswana. African Journal of Environmental Science and Technology, 5(3), 186–196.Google Scholar
- Mueller, C. S., Ramelow, G. J., & Beck, J. N. (1989). Spatial and temporal variation of heavy metals in sediment cores from the Calcasien River/Lake complex. Water, Air, & Soil Pollution, 43, 213–230.Google Scholar
- Odhiambo, B. K., & Boss, S. K. (2006). Watershed physiography, land use and sediment yield: case study from NW Arkansas, USA. Journal of Spatial Hydrology, 6(1), 29–51.Google Scholar
- Ortt, R.A., Kerhin, R.T., Wells, D., & Cornwell, J., (2000). Bathymetric survey and sedimentation analysis of Loch Raven and Prettyboy Reservoirs: Maryland Geological Survey, Coastal and Estuarine Geology File Report No.99–4.Google Scholar
- Regional Water Supply Plan. (2011). Town of Culpeper, Virginia and County of Culpeper, Virginia. 83 p. http://web.culpepercounty.gov/LinkClick.aspx?fileticket=YM5k0GdU6Ek%3D&
- Smith, SM (2006) Geochemistry of Delaware, Maryland, Virginia, and West Virginia national uranium resources evaluation hydrochemical and stream sediments reconnaissance program. USGS Report OFR 97-292: NURE HSSR Data. http://pubs.usgs.gov/of/1997/ofr-97-0492/state/nure_dmv.htm. Accessed March 2012
- Soil Survey Staff. (2011). Natural Resources Conservation Service, United States Department of Agriculture. Web Soil Survey. Available http://websoilsurvey.nrcs.usda.gov/. Accessed 25 May 2011
- Trimble, S. W. (2008). Man-induced soil erosion on the southern Piedmont (2nd ed.). Ankeny, IA: Soil and Water Conservation Society.Google Scholar
- Virginia Department of Game and Inland Fisheries (VDGIF). (2009). Lake Moomaw. <http://www.dgif.virginia.gov/fishing/waterbodies/display.asp?id=88>. Accessed 8 April 2011.
- Virginia Department of Mines, Minerals and Energy (VDMME). Generalized Geologic Map of Virginia. Map. <http://www.dmme.virginia.gov/Dgmr/pdf/VA%20geol%20map.pdf>. Accessed 18 July. 2012.
- Walling, D.E. (2004). Using environmental radionuclides to trace sediment mobilization and delivery in river basins as an aid to catchment management. Proceedings of the Ninth International Symposium on River Sedimentation. Yichang, ChinaGoogle Scholar
- Wang, X., & Cui, P. (2005). Support soil conservation practices by identifying critical erosion areas within an American watershed using the GIS-AGNPS model. Journal of Spatial Hydrology, 5, 31–44.Google Scholar
- Weant, G.E. (1985). Sources of Copper Air Emissions. United States Environmental Protection Agency Air and Energy Engineering Project SummaryGoogle Scholar
- Wetzel, R. G. (1975). Limnology. Philadelphia: W.B. Saunders.Google Scholar