| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a U.S. Army Corps of Engineers, Environmental Assessment Branch, 26 Federal Plaza, New York, NY 10278-0090 USA
b Dep. of Biology, George Mason Univ., Fairfax, VA 22030-4444 USA
josephine.r.axt{at}usace.army.mil
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMaterials and methodsResultsDiscussionConclusionsREFERENCES |
|---|
to remove dissolved inorganic P from solution, and we compared the P sorption capacities of wetlands with those of streambanks (within wetlands) and adjacent uplands. We hypothesized that wetland soils would have higher P sorption capacities than streambank and upland soils due to the higher concentration of noncrystalline (oxalate-extractable) Al and Fe (Alo and Feo) favored by periodic flooding. We found that P sorption capacities varied both as a function of landscape position and soil depth. Wetlands had higher P sorption capacities than uplands in surface soils (0–15 cm), while below 50 cm the relationship was reversed. Streambank areas within wetlands generally had the lowest P sorption capacities. As hypothesized, Alo was correlated with P sorption capacity in wetland soils
, but so was soil organic matter (as estimated by mass loss on ignition [LOI])
; in fact, Alo and organic matter were positively correlated in wetland soils
. In contrast, clay and silt content were the two soil parameters most highly positively correlated with P sorption capacity in upland soils
. Overall, these results suggest that differences in soil chemistry exist among landscape positions (wetland, streambank, upland) that have important implications with regard to P sorption capacity. Since wetlands and uplands may remove P from different hydrologic sources (i.e., surface runoff in wetlands and groundwater in uplands), hydrology may be a key factor in determining water quality functioning.
Abbreviations: Alo, oxalate-extractable Al • BC, Bernard's Cabin site • BP, Berger Preserve site • CF, Catherine's Furnace site • CP, Coastal Plain • dwe, dry weight equivalent • Feo, oxalate-extractable Fe • HG, Hazel Grove site • LD, Lee Drive site • LOI, loss on ignition • PD, Piedmont • PFO, palustrine forested wetland • PSI, P sorption index • PVC, polyvinyl chloride • SC, Spotsylvania Courthouse site
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and methodsResultsDiscussionConclusionsREFERENCES |
|---|
Approximately one-third of the Chesapeake Bay's watershed lies in Virginia (USEPA, 1988), the majority in the PD and CP physiographic provinces (Markewich et al., 1990). Palustrine forested wetlands (bottomland hardwood forests) are a common type of non-tidal wetland found in this region. For example, Walbridge and Struthers (1993) found that small PFOs accounted for 66.0% of the non-tidal wetlands in Caroline County, in the Virginia CP. The majority were PFO1As and PFO1Cs — bottomland hardwood forests with relatively short hydroperiods (i.e., temporarily or seasonally flooded; Cowardin et al., 1979) along lower-order streams and drainageways. Because of their short hydroperiods, these wetlands have, in the past, been in danger of losing their status as "jurisdictionally protected" wetlands (National Academy of Sciences, 1995). The rapid development expected in the near future in Virginia's "fertile crescent" (the area between the major urban centers of Washington, DC and Richmond and Norfolk, VA) (Year 2020 Panel, 1988) suggests additional pressure favoring wetland loss, and makes understanding the P retention functions of these wetlands particularly timely and important. While non-tidal PFOs represent a significant component of the wetland resources in Virginia's PD and CP, the processes controlling P removal and retention in these wetlands and their surrounding uplands are poorly understood (Walbridge and Struthers, 1993).
For forested wetlands in the southern USA, two mechanisms appear primarily responsible for P retention (Walbridge and Lockaby, 1994): (i) the deposition of sediment and particulate organic P (primarily a physical process), and (ii) the adsorption and/or precipitation of dissolved phosphate (a chemical process involving Al and Fe minerals in acid soils and Ca minerals in alkaline soils) (Wild, 1950; Larsen, 1967; Goldberg and Sposito, 1985). The processes of adsorption and/or precipitation are believed to represent the most important long-term sink for P for a wide range of wetland soils (Richardson et al., 1988; Cooke, 1992; Walbridge and Struthers, 1993). Inputs of sediment and particulate organic P are most commonly associated with overbank flooding, and are likely to be of minor importance in wetlands bordering lower-order streams where hydrologic and material inputs are dominated by riparian transport (cf., Brinson, 1993). Thus, we focused on P sorption (sensu Scheidegger and Sparks, 1996) as the mechanism most likely to control long-term P retention in PFOs in eastern Virginia.
Phosphorus sorption capacity has been widely estimated with a single point P sorption index (PSI), calculated as x/logc, where x is P adsorbed by the soil in mg P per 100 g of soil and c is the resulting equilibrium phosphate concentration in solution (µmol L-1) (Bache and Williams, 1971). The value of the PSI as a comparative index of P sorption capacity has been supported by a number of studies that have found strong positive correlations between the PSI and soil noncrystalline (oxalate-extractable) Al concentrations in a wide variety of both wetland and upland soils (Richardson, 1985; Richardson et al., 1988; Walbridge et al., 1991; Darke, 1997). Both crystalline and noncrystalline Al and Fe oxides sorb P by the same mechanism, but noncrystalline forms tend to dominate soil P sorption reactions when present in significant amounts because of their greater reactive surface area per unit soil volume (Hsu, 1989; Schwertmann and Taylor, 1989; Freese et al., 1992). Although P sorbed by Fe minerals may be released when Fe (III) is reduced to Fe (II) during periods of low redox potential, it has been suggested that this may be counteracted by the flooding-induced transformation of crystalline Fe to noncrystalline forms (Kuo and Mikkelsen, 1979; Willet and Higgins, 1980; Sah and Mikkelsen, 1989; Sah et al., 1989). Flooding also promotes the persistence of organic matter, which can further inhibit the crystallization of Al and Fe minerals (Schwertmann, 1966; Kodama and Schnitzer, 1980), as organic anions are adsorbed by the noncrystalline Al and Fe oxides; the stronger (i.e., more stable) the organic anion–metal complex, the greater the organic anion's ability to obstruct Al and Fe precipitation (Harter, 1985; Fox et al., 1990).
Spatial variability within PFOs (e.g., streambanks) may affect P sorption capacity, since even infrequent overbank flooding can cause streambank zones to accumulate higher percentages of sand than the wetland as a whole (Leopold et al., 1964). Soils high in clay and silt generally have higher P sorption potentials than sandy soils (Sample et al., 1980), even when their P sorption potentials are increased by coatings of metal oxides, hydroxides, or secondary clay minerals (Harris et al., 1996). In addition, the magnitude and intensity of P loading to wetland areas will depend in part on the degree to which adjacent upland soils can reduce dissolved phosphate concentrations. In combination with the direct effects of flooding on soil chemistry, differences in (i) throughflows of both water and nutrients, and (ii) sediment depositional regimes, could result in differences in P sorption capacity between wetlands and adjacent uplands.
Virginia's PD and CP have different soil parent materials. The PD is characterized by igneous and metamorphic crystalline rocks and relatively shallow soils (
1 m deep), while in the CP, sedimentary rocks are overlain with typically thick soils (2–8 m deep) (Markewich et al., 1990). Fine-grained saprolitic parent material in the PD produces soils that contain more clay, and therefore potentially more Al and Fe oxides, than CP soils.
We measured the physical and chemical properties (e.g., soil texture, P sorption capacity, noncrystalline Al and Fe oxide concentration) of surface (0–15 cm) and subsurface (15–30, 30–50, 50–100 cm) soils from three landscape positions (upland, wetland, and streambank) in six non-tidal PFOs in Virginia (three each in the PD and CP) to (i) examine variations in P sorption capacity among these soils, and (ii) determine the soil physical and chemical properties important in predicting P sorption capacity. We hypothesized that: (i) wetland soils would have higher P sorption capacities than both upland and streambank soils, and (ii) concentrations of noncrystalline Al and Fe would be the most important predictors of soil P sorption capacity.
|
Materials and methods |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and methodsResultsDiscussionConclusionsREFERENCES |
|---|
, within 17 km of the fall line dividing the two physiographic provinces (Fig. 1)
. Physiographic provinces were verified in the field by R.J. Diecchio, Dep. of Geography & Earth Systems Science, George Mason University. Average annual precipitation in this region is 1156 mm; mean annual temperature is 12.9°C (Elder, 1985). Five sites were located on U.S. National Park Service lands in Spotsylvania County, Virginia (38°N, 77°W): Hazel Grove (HG) and Catherine's Furnace (CF) in the Chancellorsville National Battlefield Park, Spotsylvania Courthouse (SC) in the Spotsylvania National Battlefield Park, and Lee Drive (LD) and Bernard's Cabin (BC) in the Fredericksburg National Battlefield Park. A sixth site, Berger Preserve (BP), was located in adjoining Caroline County, Virginia, on Nature Conservancy land. Sites were chosen as representative of non-tidal palustrine forested wetlands in this region. Wetlands are small (<5 ha), relatively undisturbed, and drained by either a central stream (HG, CF, SC, BP) or ephemeral drainageway (LD, BC). Forest canopies and subcanopies are dominated by broad-leaved deciduous [e.g., red maple (Acer rubrum L.), sweetgum (Liquidambar styraciflua L.), river birch (Betula nigra L.)] and/or evergreen [sweetbay magnolia (Magnolia virginiana L.)] trees, with composition varying somewhat across sites. Wetland–upland boundaries were delineated using the 1987 United States Army Corps of Engineers' Wetland Delineation Manual (United States Army Corps of Engineers, 1987) (E. Sherwin, 1992, personal communication). Soils of upland sites are mapped as clayey, thermic Typic Hapludults. The PD wetland soils are mapped as either fine-silty, thermic Typic Ochraquults (HG) or a mixture of Fluvaquents and Udifluvents (CF, SC); CP wetland soils are mapped as either Aquic Hapludults (BP) or Aquults with a gravelly substratum (LD and BC) (Elder, 1985; S. Martin, 1994, personal communication). Soil identifications based on mapped units were verified by L. Heidel, NRCS, Shenandoah Co., Virginia.
|
of streambank, wetland, and upland plots were established on the east side of the stream only; a road and 19th century iron furnace precluded plot establishment west of the stream channel. At HG, it was possible to establish four streambank plots on the north side of the stream, but basin configuration and the presence of a road allowed the establishment of only two upland plots (a full complement,
, of streambank, upland, and wetland plots were established south of the stream). Transects were used solely to randomly establish plot locations, and were not referenced again during either soil sampling or statistical analysis. In each plot, three soil cores were collected to a depth of 15 cm, excluding surface litter, and composited to form one surface soil sample per plot (n = 8 per landscape position per site). Subsurface soil samples (15–30, 30–50, and 50–100 cm) were collected similarly, but soils from two adjacent plots of the same landscape position were composited into one sample (n = 4 per landscape position per site). Soils were stored in polyethylene bags in the field, and kept cool until returning to the lab, where they were stored field moist at 4°C. Prior to analysis, soils were homogenized by hand, and roots and other recognizable plant material were removed. Surface soil samples were collected during summer 1993; subsurface soil samples were collected during summer 1994.
Soil bulk density was estimated at each site during fall 1994, using the same sampling strategy as described above for soil collections. In both wetland and upland soils, surface and 15- to 30-cm-horizon samples were collected by hammering a 5.2-cm-diameter polyvinyl chloride (PVC) pipe with a sharpened rim into the ground. Both very wet and very dry soil conditions in wetlands and uplands, respectively, often made it impossible to use a PVC pipe for bulk density soil collections below 30 cm; in those cases, samples were collected with either a 2.0-cm-diameter punch probe or a 7.1-cm-diameter auger.
Physical and Chemical Analyses
Soil pH was determined by pH electrode in a 1:2 slurry of soil/deionized water, following a 10- to 20-min equilibration. Soil organic matter was estimated by mass LOI in a muffle furnace for 14 h at 550°C (Lim and Jackson, 1982). Soil texture (sand, silt, and clay) was estimated by the Bouyoucos (1962) method.
Phosphate adsorption isotherms were developed by equilibrating 2.0 g dry weight equivalent (dwe) of fresh soil (n = 2 per sample) in 50-mL centrifuge tubes with 25 mL of 0.01 M CaCl2 solutions containing 16, 33, 130 or 260 mg P L-1, added as KH2PO4. Samples were shaken mechanically for 24 h, and then centrifuged for 20 min at 0.314 g. Orthophosphate concentrations in the clear supernatants were analyzed colorimetrically by the method of Murphy and Riley (1962), using a Technicon II Autoanalyzer and method number 696-82W (Bran and Luebbe Inc., 1989). Sorption data from the 130 mg P L-1 addition level were used to develop a PSI for each site, at each landscape position and profile depth (Bache and Williams, 1971). The index is calculated as x/logc, where x = P adsorbed by the soil (mg P per 100 g of soil) and c = the equilibrium solution phosphate concentration after 24 h of shaking (µmol L-1).
Oxalate-extractable (noncrystalline) Al and Fe were estimated by shaking replicate 0.4-g-dwe subsamples for 4 h in darkness with 40 mL of 0.2 M acid ammonium oxalate adjusted to pH 3.0 (USDA, 1972; Walbridge et al., 1991). Solutions were clarified by centrifugation (10 min, 0.314 g). Extractable Ca was estimated by shaking 4.0-g-dwe subsamples with 25 mL of 1.0 M ammonium acetate for 30 min, and then repeating the process with fresh extractant (Thomas, 1982). Samples were centrifuged to clarity (10 min, 0.185 g) after both extractions, and the supernatants combined. The Al, Fe, and Ca concentrations were analyzed using a Perkin Elmer Model 2380 Atomic Absorption Spectrophotometer (Perkin Elmer, 1982).
Statistical Analyses
Data were analyzed for normality using the Shapiro Wilk test and normal probability plots. Homogeneity of variance was determined using Bartlett's test, and by visually examining plots of observed vs. expected residuals. Comparisons of soil characteristics between and among different physiographic provinces, study sites, landscape positions, and depths in the soil profile were made using an unbalanced partial hierarchal Mixed Model GLM with missing cells (SAS Institute, 1990); non-normal data were ranked using Wilcoxon rankings and then analyzed with the same GLM procedure (SAS Institute, 1990). When the GLM indicated statistical significance, differences among means were analyzed with the Student-Newman Keuls multiple comparison test (Sokal and Rohlf, 1981).
|
Results |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and methodsResultsDiscussionConclusionsREFERENCES |
|---|
|
|
Phosphorus Sorption
Phosphorus sorption capacities varied both as a function of landscape position and soil depth. On average, P sorption capacities were highest in wetlands soils above 50 cm (particularly above 15 cm), and highest in upland soils below 50 cm; streambank soils had consistently low P sorption capacities (Fig. 2)
. The PD upland and streambank soils had higher P sorption capacities than CP upland and streambank
soils in surface horizons (0–15 cm), but the reverse was true below 50 cm (Fig. 3)
. The CP and PD wetland soils had similar P sorption capacities in surface horizons, but CP soils exhibited much higher P sorption capacities below 30 cm (Fig. 3).
|
|
, respectively) (Fig. 4)
. Stepwise multiple regression identified a two-term model based on Alo and LOI that explained 83% of the variation in wetland
. A similar model for streambank soils also accounted for 83% of the variation in PSI, but LOI
, rather than Alo, was the first term in the model. In contrast, clay percentage was the best predictor of the PSI in upland soils (Fig. 5)
, while stepwise multiple regression identified a two-term model based on clay and silt that explained 87% of the variation in upland
. Oxalate-extractable Al and LOI were highly positively correlated
in wetland soils (Fig. 6)
. The correlation decreased when streambank soils were included
, and Alo and LOI were not correlated in streambank soils alone
.
|
|
|
(Fig. 7)
. The Pmax data were combined with soil bulk densities to estimate total P sorption potentials (kg ha-1) for each profile (0–100 cm) (Fig. 8)
. Upland soils had the highest total P sorption potentials at four of the six sites, including all PD sites, while wetlands had the highest total P sorption potentials in two of the three CP sites (Fig. 8). Streambank soils had the lowest P sorption potentials at all sites (Fig. 8).
|
|
|
|
Discussion |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and methodsResultsDiscussionConclusionsREFERENCES |
|---|
We also observed higher Feo (but not Alo) concentrations in wetland (including streambank) vs. upland soils, significantly higher below 15 cm (Table 2). Fuad (1995) also found significantly higher Feo (but not Alo) concentrations in wetland vs. upland soils at eight non-tidal PFO sites in Caroline County, part of Virginia's CP. Periodic flooding in wetlands may promote the transformation of crystalline Fe to noncrystalline forms (Sah and Mikkelsen, 1986). We had expected CP wetland soils to be sandier and thus lower in P sorption potential than PD soils, but found instead that CP sites had relatively high average clay contents (24–30%) across all depths, and were similar in average clay content to PD sites (20–32%). Weathering and flooding processes may have deposited clay and associated Al and Fe in these CP wetland soils (Schlesinger, 1997). Since all sites are within 17 km of the fall line (i.e., upper CP and lower PD), similarities in soil textures of PD and CP soils were not unreasonable (Table 1).
Phosphorus Sorption
In surface soils (0–15 cm), the average ability of non-tidal PFOs to remove and retain dissolved phosphate was greater than the ability of adjacent uplands (Fig. 2). In the Ogeechee River floodplain in Georgia, the P sorption capacities of wetland surface soils were also generally greater than upland surface soils (Darke, 1997). Walbridge (1993) observed a greater contrast in the P sorption capacities of wetland and upland surface soils in the Texas CP, primarily due to very low P sorption capacities in the upland soils. Unlike in this study, where upland plots were adjacent to wetland plots, the Texas upland soils were collected farther upslope, and included sandy ridgetop soils.
When averaged across physiographic province
, wetland P sorption capacities exceeded those of upland and streambank soils above 50 cm, with the greatest differences observed in surface soils (Fig. 2). In eight Virginia CP PFOs, Fuad (1995) also found that average P sorption capacities were greater in wetlands vs. uplands, although this relationship was not always consistent within individual sites. Differences observed in this study suggest that wetlands and uplands might provide different water quality benefits with respect to P retention — high P sorption capacities in wetland surface soils would be most important in removing phosphate from surface runoff; high P sorption capacities in upland subsurface soils would be more effective in removing phosphate from waters that percolate through upland profiles before entering groundwater pools. These observations are consistent with the ideas of Darke (1997), that wetlands may act as a second line of defense in protecting downstream waters from non-point source P inputs.
Theoretical P sorption capacities reflect P isotherm data, but weighting soil horizons according to length provides an overall estimate of a soil profile's total P sorption capacity. For example, the high P sorption capacities of CP wetlands below 30 cm and CP uplands below 50 cm (Fig. 3) translated into CP wetland and upland profile maxima (0–100 cm) that were greater than or equal to PD wetland and upland profile maxima, respectively (Fig. 8). These maxima measure potential P sorption capacity; in practice, each site's unique hydrology would affect which areas of the soil profile were the most important with regard to P sorption (i.e., do the majority of P inputs enter through riparian transport, overbank flooding, etc.). The magnitude of P sources in the watershed, as well as the ability of adjoining upland soils to retain P, would also affect the amount of P a wetland receives (Walbridge and Struthers, 1993). Although these results require confirmation across a greater number of CP and PD sites, they are potentially significant since CP watersheds can be especially prone to eutrophication via P originating from agricultural areas (cf., Mozaffari and Sims, 1994; Harris et al., 1996).
Dividing riparian PFOs into wetland and streambank zones uncovered meaningful within-wetland variation. Streambank soils had consistently low P sorption capacities (Fig. 4–6), presumably due to their lower concentrations of organic matter and clay than wetland and upland soils (Table 1). Richardson et al. (1988) also found that within-wetland soil heterogeneity affected P sorption capacity in three North Carolina CP swamps. A detailed examination of a Georgia floodplain forest indicated that although wetland soil microtopography (i.e., lower elevation swales and higher elevation ridges) had a limited effect on P sorption capacity, soil Al and Fe content were affected. On an annual basis, lower elevation swale microsites tended to accumulate Feo and lose Alo (Darke, 1997). Spatial and temporal variability in hydrologic and sediment inputs can affect the distribution of soil constituents that are important in P sorption (Johnston, 1991). Our data agree with these studies, and suggest that it is misleading to consider riparian wetlands as homogenous units in terms of their ability to remove dissolved inorganic phosphate from incident surface and subsurface waters.
A strong relationship was observed between the PSI, a single point P sorption index based on the 130 mg PL-1 level of addition, and Pmax, a maximum P sorption capacity based on four levels of P addition and the Langmuir equation (Fig. 7). Mozaffari and Sims (1994) found a similar correlation between the PSI and Pmax
in Delaware CP soils (0–100 cm), as did Simard et al. (1994) in A, B, and C horizons of Canadian Appalachian soils (r2 
0.94 in all horizons). These results support the use of the single point index as a reliable gauge of a soils's P sorption potential. Estimating the PSI is not only less time-consuming than developing P sorption isotherms, it facilitates comparisons with related soil physical and chemical parameters.
We found a strong positive correlation between the PSI and both Alo and LOI in wetlands (Fig. 4); a two-term model of Alo and LOI was slightly better than Alo alone
, and Alo and LOI were themselves highly correlated
. Numerous other studies have observed similar correlations between Alo and the PSI in both wetland and upland soils (Richardson, 1985; Richardson et al., 1988; Walbridge et al., 1991; Yuan and Lavkulich, 1994; Fuad, 1995), while the relationship between Alo and LOI is consistent with the findings of Darke (1997).
Using a dry combustion technique like LOI to estimate organic matter content in soils with >25% clay can artificially inflate organic matter estimates because of weight changes in thermally reactive crystalline and noncrystalline clay minerals (Grewal et al., 1991). Part of our observed relationship between LOI and both Alo and the PSI in high clay wetland soils could be due to the contribution of clay minerals unintentionally included in the LOI estimate of organic matter. When we reanalyzed those wetland soils with 25% or less clay content (14 of 24 site–depth combinations; Table 1), we found that LOI was still highly correlated with Alo
but less so with the PSI
. In future research we would recommend using a more robust method of determining organic matter content, particularly when examining relationships between organic matter and other soil characteristics is a primary objective.
Despite the uncertainties surrounding LOI as an appropriate measure of organic matter in high clay soils, the strong correlative relationship observed between LOI and Alo suggests that Al–organic matter complexes could play an important role in controlling P sorption potential in these wetland soils. Haynes and Swift (1989) suggested that not only did Al–organic matter complexes maintain Al in a highly sorptive noncrystalline state, but that steric hindrance effects of organic matter on noncrystalline Al were reduced after soil drainage. In their proposed mechanism, bonds with the noncrystalline Al were broken as organic molecules dried and condensed, thereby creating fresh P adsorption sites. Appelt et al. (1975) suggested that P sorption was dependent on the OH/Al ratio, such that ligand exchange occurred between P and Al-associated hydroxyl groups as organic matter-Al polymers grew. Organic matter appeared to both foster the hydrolysis of Al and obstruct hydroxy-Al crystallization, both of which increase P sorption (Appelt et al., 1975; Huang and Violante, 1986). These mechanisms could provide an explanation for the high P sorption potentials observed in the surface wetland soils in this study (Fig. 2) and in many other periodically flooded mineral wetland soils (Kuo and Mikkelsen, 1979; Richardson, 1985; Richardson et al., 1988).
Oxalate-extractable Al was not a good predictor of P sorption in Virginia uplands, where clay and silt explained 87% of the variation in P sorption potential. The significance of clay in predicting P sorption potential is widely recognized (Agbenin and Tiessen, 1994; Sanyal et al., 1993; Juo and Fox, 1977). Because clays contain "broken edges", ligand exchange sites exist between open hydroxyl groups and P (Goldberg and Sposito, 1985). In a range of Brazilian soils, maximum P sorption potential and clay content varied concurrently with depth (Agbenin and Tiessen, 1994), while in upland soils of the Delaware Coastal Plain, the PSI was strongly correlated with percentage of clay (Mozaffari and Sims, 1994). The unique set of primary predictors of the PSI between wetlands and uplands point out meaningful differences in wetland vs. upland soil chemistry with regard to P sorption.
|
Conclusions |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and methodsResultsDiscussionConclusionsREFERENCES |
|---|
between Alo and LOI suggests the potential role of Al–organic matter complexes in controlling P sorption capacity. Because flooding and saturated soil conditions can cause the accumulation of both organic matter and noncrystalline Al, hydrologic regime may help explain high P sorption potentials in wetland surface soils. Percentage of clay and silt explained 87% of PSI variation in upland soils. The role of Virginia non-tidal PFOs in improving water quality may vary as a function of the type of hydrologic input. Wetlands had superior P sorption capacities in surface soils (0–15 cm), and thus appear particularly suited for handling surface runoff. The high clay content and correspondingly high P sorption potentials of upland subsurface soils (50–100 cm) suggest their potential for improving groundwater quality. Wetland profile maxima (i.e., the soil profile's cumulative P sorption potential) exceeded those of adjacent upland profile maxima at two of the three CP sites, whereas no PD wetland profile maxima was greater than its adjacent upland maxima. Streambank soils consistently had lower P sorption potentials than wetland soils, suggesting that riparian areas should not be considered homogenous units when estimating their P sorption potentials. Depending on surface and groundwater flow patterns, wetlands and adjacent uplands may act in conjunction, particularly in the CP, to serve as important buffers between agricultural lands and aquatic ecosystems.Bran and Luebbe 1989; Perkin 1982
|
ACKNOWLEDGMENTS |
|---|
Received for publication October 23, 1997.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and methodsResultsDiscussionConclusionsREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. W. Leader, E. J. Dunne, and K. R. Reddy Phosphorus Sorbing Materials: Sorption Dynamics and Physicochemical Characteristics J. Environ. Qual., January 4, 2008; 37(1): 174 - 181. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Novak and D. W. Watts Phosphorus Sorption by Sediments in a Southeastern Coastal Plain In-Stream Wetland J. Environ. Qual., October 27, 2006; 35(6): 1975 - 1982. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. C. Murray and D. Hesterberg Iron and Phosphate Dissolution during Abiotic Reduction of Ferrihydrite-Boehmite Mixtures Soil Sci. Soc. Am. J., June 21, 2006; 70(4): 1318 - 1327. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. L. Bruland and C. J. Richardson Hydrologic Gradients and Topsoil Additions Affect Soil Properties of Virginia Created Wetlands Soil Sci. Soc. Am. J., November 1, 2004; 68(6): 2069 - 2077. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. L. Bruland and C. J. Richardson A Spatially Explicit Investigation of Phosphorus Sorption and Related Soil Properties in Two Riparian Wetlands J. Environ. Qual., March 1, 2004; 33(2): 785 - 794. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. B. Craft and C. Chiang Forms and Amounts of Soil Nitrogen and Phosphorus Across a Longleaf Pine-Depressional Wetland Landscape Soil Sci. Soc. Am. J., September 1, 2002; 66(5): 1713 - 1721. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. D. Bridgham, C. A. Johnston, J. P. Schubauer-Berigan, and P. Weishampel Phosphorus Sorption Dynamics in Soils and Coupling with Surface and Pore Water in Riverine Wetlands Soil Sci. Soc. Am. J., March 1, 2001; 65(2): 577 - 588. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||
= uplands, 
= wetlands, • = streambanks
= PD, 
= 50–100 cm
