SSSAJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Stolt, M.H.
Right arrow Articles by Groover, V.A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Stolt, M.H.
Right arrow Articles by Groover, V.A.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Stolt, M.H.
Right arrow Articles by Groover, V.A.
Related Collections
Right arrow Soil Methods/Instrumentation
Right arrow Wetland Soils
Right arrow Spatial Distribution
Soil Science Society of America Journal 65:527-535 (2001)
© 2001 Soil Science Society of America

DIVISION S-10-WETLAND SOILS

Spatial Variability in Palustrine Wetlands

M.H. Stolta, M.H. Genthnerb, W.L. Danielsc and V.A. Grooverc

a Dep. of Natural Resources Science, Univ. of Rhode Island, Kingston, RI 02881
b P.O. Box 258, Ashfield, MA
c Dep. of Crop and Soil Environmental Sciences, Virginia Polytechnic Institute and State Univ., Blacksburg, VA 24061-0404

Corresponding author (mstolt{at}uri.edu)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY
 REFERENCES
 
Wetlands are complex ecosystems having considerable spatial variability. Understanding soil spatial relationships in wetlands is difficult because of the number of factors that affect soil properties. We established a nested sampling design within five small, forested and scrub-shrub palustrine wetlands in Virginia to examine soil spatial variability within and among sites. Sampling was based on relative elevation intervals within each wetland and soil depth within each sampling unit. Soils were analyzed to determine variability in nutrient status, pH, organic C content, and particle-size distribution (PSD). Elevation contributed the least amount to the total variability (variability among sites) for nearly every parameter. Depth from the soil surface explained the most total variability in regard to PSD, indicating that parent material stratification in these alluvial wetlands strongly influences soil physical properties. Most of the total variability in the soil chemical parameters was explained by site. Within sites, elevation trends were observed for particle-size and chemical parameters in most of the wetlands. Elevation trends were related to water table levels and the depositional environment. Within elevation sampling units, particle-size and chemical parameters were shown to be significantly related to depth from the soil surface (at the 0.05 level). These relationships could be attributed to the stratified nature of alluvial soils and the accumulation of organic matter at or near the soil surface. Pedon sampling locations were spaced {approx}1 m apart and therefore showed less random variability than elevation sampling locations spaced throughout the 0.25- to 0.35-ha study areas. Soils were classified as Endoaquepts, Humaquepts, Dystrudepts, Endoaquents, and Fluvaquents, depending on the wetland site.

Abbreviations: CEC, cation-exchange capacity • TKN, total Kjeldahl N


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY
 REFERENCES
 
WETLANDS are complex ecosystems exhibiting considerable spatial variability (Reddy, 1993). A number of factors control the spatial relationships in these systems, making the separation of systematic from random soil components difficult. Systematic spatial relationships in wetland soils are the result of differences in parent material, elevation, erosional or depositional environment, frequency of flooding, vegetation, pedogenic effects, and hydrology (Johnston et al., 1984; Hayati and Proctor, 1990; Gaston et al., 1990; Farrish, 1991; Reese and Moorhead, 1996). Random effects are attributed to unrecognized differences in these parameters, as well as differences due to sampling and laboratory error (Wilding and Drees, 1983). These random effects often obscure or confound soil–elevation, soil–vegetation, or soil–hydrology relationships. Therefore, to understand spatial relationships in wetland soils, random variability needs to be recognized and separated from systematic variability.

Numerous studies have examined spatial variability in upland soils (Harradine, 1949; Ball and Williams,1968; Drees and Wilding, 1973; Campbell, 1978; Mausbach et al., 1980; Edmonds et al., 1985; Thomas et al., 1989; Mausbach and Wilding, 1991; Stolt et al., 1992). A few studies have made spatial comparisons between soils of uplands and adjacent wetlands. Hammer et al. (1987) compared soil variability within three landscape units: first-order bottomlands, sloping landscapes, and level uplands. Soils on bottomlands were found to have the most spatial variability. Reese and Moorland (1996) examined changes in soils properties from the basin of a Carolina Bay depressional wetland to the upland rim. Significant differences in organic C and clay content were observed between the wetland and surrounding upland rim. Considerable differences in soil parameters were also recognized within the wetland portion, even though the change in elevation was minimal. Both of these studies concluded that using a stratified sampling design within landscape units, such as wetlands, may be very important in quantifying soil variability. However, studies reporting on spatial variability of wetlands using stratified sampling are limited (Reese and Moorland, 1996). Gaston et al. (1990) examined soil morphology of Aquods and Aquults for a 60-ha flatwood site in Florida. Despite a generally uniform terrain across the site, the depth to and presence of spodic and argillic horizons varied considerably. Spatial variability of soil chemical properties was examined in a sloping wet heath in England (Hayati and Proctor, 1990). Trend surface analysis indicated significant relationships among elevation, vegetation, and soil chemical properties across the wetland.

Spatial variability is important to consider when assessing the environmental and ecological functions of a wetland. Many of these functions, such as floodwater storage, traps for sediment, and sinks for various non-point source pollutants are difficult to measure directly. In lieu of direct measurements, soil and landscape properties can be recorded and then related to the potential of the wetland to function in one or more of these capacities (Maltby, 1987; Federal Interagency Committee for Wetland Delineation, 1989; Kentula et al., 1992). The nature of the spatial variability must be considered to ensure that the full range of soil, landscape, and associated wetness conditions are described. Natural wetlands that are recognized to serve a number of functions are now used as reference wetlands. Reference wetlands provide natural comparative sites to use as a marker of the success or failure of adjacent newly created wetlands (D'Avanzo, 1989; White et al., 1990; Kentula et al., 1992; Brinson, 1993; Bishel-Machung et al., 1996).

Our long-term goal was to establish a set of reference wetlands that could be used to make comparisons with a number of constructed wetlands. As an initial step, we examined spatial variability of soil physical and chemical properties within forested and scrub-shrub palustrine wetlands in Virginia. Our objectives were (i) to partition the sources of spatial variability for the overall study and estimate the relative contribution of each, (ii) to estimate and describe the amount of variability in soil properties (particle-size distribution, organic C, N, CEC, pH, Ca, and Mg) in both pedon and elevation sampling units within wetlands, and (iii) to examine differences in site characteristics and soil morphology among wetlands.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 SUMMARY
 REFERENCES
 
Palustrine wetlands were examined in a reconnaissance study in the Piedmont and Coastal Plain physiographic provinces of Virginia (Daniels and Gambrell, 1978; USDA-SCS, 1981; Colquhoun et al., 1991) to find representative forested and scrub-shrub wetlands of these areas that could serve as reference wetlands for comparisons with constructed wetlands. The relatively undisturbed wetlands were in close proximity to wetlands that were constructed for mitigation purposes. Five wetlands, considered representative of scrub-shrub and forested wetlands in the Piedmont and Coastal Plains of Virginia, were chosen for detailed study (Fig. 1) . Soils, vegetation, landscape features, and wetland hydrologic indicators, such as high water marks, rafted leaves, and stratified soils, were used to establish the presence and extent of each wetland. Predominant vegetation was noted for each wetland site.



View larger version (12K):
[in this window]
[in a new window]
  		Fig. 1. Location of the five palustrine study wetlands in Virginia

 
Field Methods
At each wetland site relative elevations were determined within a representative 0.25- to 0.35-ha area of the wetland using a rod and level. Elevations were measured and recorded on a grid spaced at 10-m centers. The centers were flagged so that these elevations would be known for sampling purposes. We estimate that the measurement error for our elevation observations was <1 cm. Three equivalent elevation intervals, spanning the range in elevation within areas of each site considered as a wetland, were established for sampling purposes. Within each elevation interval random locations of a known elevation were chosen and the same sampling protocol was followed. Three samples, spaced 1 m apart, were collected at depths of 5 to 15, 40 to 50, and 90 to 100 cm at three locations (A, B, and C) with a bucket auger. For Locations A and B the three samples for each depth were mixed to form a composite sample for that depth. In order to examine the variability within pedons, at Location C the three samples for each depth were collected and each bagged separately. Soils were described at each sampling location using standard procedures (Soil Survey Staff, 1993). Three wells, screened from 30 to 150 cm, were installed at random locations within each wetland site to monitor water table levels. Water table levels were recorded on a monthly basis. Well locations were not coordinated with elevation intervals.

Laboratory Methods
Sand fractions were determined after mechanical sieving. Percentage silt and clay were determined by pipette (Gee and Bauder, 1986). Organic C was determined by dry combustion (Rabenhorst, 1988). Total Kjeldahl N (TKN) was determined after digestion using the colorimetric method described by Mulvaney (1996). Soil pH was determined at a 1:1 soil/water ratio. Samples were analyzed for extractable bases and exchangeable acidity (Thomas, 1982; Rhoades, 1982). Cation-exchange capacity (CEC) was determined by summing the bases and the acidity.

Statistical Methods
Percentage of the variability explained by each component was estimated using a nested design (Webster, 1977; Webster and Oliver, 1990). The value for an individual variable (Yijkl) can be explained by the ideal statistical model

(1)
where u represents the population mean; Si the site effects; Eij the elevation effects; Dijk the depth from the soil surface effects; and Rijkl is the random error. The total variance percentage attributed to each component was estimated by dividing component variance by total variance (SAS Institute, 1985).

Analysis of variance (F statistics) was used to test whether sample means were significantly different at the 0.05 level among elevations within sites and depths within elevations (Zar, 1984). The significance of the difference among means was evaluated using a least significant different (LSD) test after analysis of variance testing. Means and coefficients of variation (CV) were calculated following Zar (1984).


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
SUMMARY
 REFERENCES
 
Wetland Study Site Descriptions
Soils within study wetlands have formed in alluvial materials and are classified as Dystrudepts, Fluvaquents, Humaquepts, Endoaquents, and Endoaquepts (Tables 1 and 2). At the Duntons Mill and Elmwood Baylor wetlands, woody shrubs are very common and herbaceous plants are predominant. The woody shrubs include hazel alder (Alnus serrulata Ait.), buttonbush (Cephalanthus occidentalis L.), and silky willow (Salix sericea Marsh.). Herbaceous plants consist primarily of sensitive fern (Onoclea sensibilis L.), arrow arum [Peltandra virginica (L.) Schott & Endl.], smartweed (Poly- gonum hydropiper L.), and soft rush (Juncus effusus L.). Smartweed and soft rush are more common in Duntons Mill wetland. Both wetlands are adjacent to agricultural fields. Duntons Mill is also adjacent to an old mill pond. The age of the pond is unknown. A small stream near Elmwood Baylor occasionally floods this wetland. Water table levels are at or near the surface for most of the year and show minimal seasonal fluctuation (Fig. 2) . The wetlands are located low on the overall landscape and directly adjacent to a stream or pond, therefore we assumed that these were discharge wetlands. Water table levels are relatively static, which suggests that these wetlands are strongly influenced by groundwater (Hunt et al., 1999). Total precipitation measured at a weather station within 5 km of Duntons Mill was 5 cm less than the 30-yr normal for the 12-mo water table monitoring period. Only in August (10 cm above normal) was there more than a 5 cm difference between monthly precipitation totals and the 30-yr average for a given month. Similar precipitation patterns were observed at the other sites, suggesting that the water tables observed are representative of a typical year.


View this table:
[in this window]
[in a new window]
  		Table 1. Characteristics, soil classification, and topographic relief of the study wetlands{ddagger}

 

View this table:
[in this window]
[in a new window]
  		Table 2. Profile descriptions at the highest relative elevation in the five palustrine wetlands

 


View larger version (24K):
[in this window]
[in a new window]
  		Fig. 2. Water table levels in the five palustrine study wetlands. Levels represent the means of three wells randomly located within each of the wetlands

 
Western Freeway and Fort Lee are mature forested wetlands that show little evidence of anthropogenic disturbance. These wetlands are adjacent to second-order streams (Strahler, 1952), and surface water is occasionally added during stream flooding. Western Freeway is adjacent to an interstate highway and agricultural field. Water table levels at Western Freeway are within 30 cm of the soil surface throughout the year (Fig. 2). Vegetation in the wetland is dominated by red maple (Acer rubrum L.), sweet gum (Liquidambar styraciflua L.), willow oak (Quercus phellos L.), sweetbay magnolia (Magnolia virginiana L.), sweet pepperbush (Clethra alnifolia L.), lizards tail (Saururus cernuus L.), and false nettle [Boehmeria cylindrica (L.) Swartz].

Vegetation in the Fort Lee wetland is predominantly red maple, sweet gum, swamp chestnut oak (Q. michauxii Nutt.), sweetbay magnolia, sweet pepperbush, greenbriar (Smilax rotundifolia L.), lizards tail, and cardinal flower (Lobelia cardinalis L.). Water table levels range from near the surface in the spring, above the surface in the winter, and more than 1 m below the surface in the late fall (Fig. 2). The average difference between water table levels among the three randomly located wells was <20 cm for the winter and spring water table readings. However, differences between the wells in some of the summer readings were more than 45 cm. In some cases wells will give erroneous water table levels because of stratified soil conditions. This could result in potential errors in recorded water levels for the stratified soils at Fort Lee (Table 2) during the summer and late fall.

Cub Creek is a forested wetland located adjacent to a third-order stream that commonly floods. This wetland shows minimal evidence of anthropogenic disturbance. Vegetation is at the mature stage and the predominant species include red maple, sycamore (Platanus occidentalis L.), sweet gum, sweetpepper bush, poison ivy (Toxicodendron radicans L.), and greenbriar. During the summer and late fall the water table was consistently below 90 cm, and water tables during the rest of the year rarely reach into the upper 30 cm of the soil (Fig. 2). Hydric soil indicators are absent from the upper part of the soil (Table 2), suggesting that this wetland does not meet the criteria of a hydric soil (Environmental Laboratory, 1987). However, the site commonly floods, and soils show stratified layers and thick accumulations of recent overwash (Table 2). The recent sediment, stratification, and common flooding indicate that this riparian area functions in at least two wetland capacities, sediment entrapment and floodwater storage. Such palustrine wetlands are commonly found in the Piedmont and should be investigated as reference wetlands.

Total Soil Variability within Study
Maximum differentials in elevations in the wetlands range from 60 to 120 cm (Table 1). Considerable microrelief occurs within the wetlands (Fig. 3) . However, when all of the wetlands were compared, elevation contributed the least amount to the total variability for nearly every parameter (Table 3). For these palustrine wetlands, differences in site and depth from the soil surface are apparently more important than elevation in explaining soil characteristics among the wetlands. Differences among wetlands can be attributed to factors such as hydrology, chemistry of the inundating water and groundwater, vegetation, alluvial parent materials, and the relative potential energy of the adjacent stream or water body (by relative potential energy we mean the energy each stream has when at flood stage relative to the other wetland sites).



View larger version (32K):
[in this window]
[in a new window]
  		Fig. 3. Contour map of the Western Freeway wetland

 

View this table:
[in this window]
[in a new window]
  		Table 3. Total soil property variability explained by site, elevation, depth, and error

 
Most of the total variability for the TKN, pH, CEC, exchangeable Ca and Mg, and organic C parameters was explained by site (Table 3). The Cub Creek wetland floods after substantial rainfall, but water tables are often 1 m below the soil surface (Fig. 2). At Fort Lee, water tables range from >1 m below the soil surface in the fall to extended periods in the winter and spring above the soil surface. Water tables at the Western Freeway, Duntons Mill, and Elmwood Baylor wetlands are within 15 cm of the soil surface for most of the year. Such differences in duration of saturation may explain the variability in CEC, TKN, pH, and organic C among wetland sites. Higher water tables would reduce the amount of organic matter decomposition. Therefore organic C contents, CEC, and TKN (especially organic N) would be expected to be higher in the wetter sites. Some differences in exchangeable cations, TKN, and pH may be attributable to influences from the surrounding uplands. The three wetland sites on the lower Coastal Plain are located adjacent to agricultural fields. Runoff of agricultural amendments such as fertilizer and lime could increase these values. The Western Freeway wetland also collects runoff from a nearby interstate highway.

Soil depth explains most of the total variability for the soil particle-size fractions (Table 3). Soils in the palustrine wetlands are minimally developed Entisols and Inceptisols, showing little evidence of pedogenesis (Table 2). Therefore, differences in particle size with depth are not likely related to pedogenic processes, such as eluviation and illuviation. These soils have multiple C horizons, over-thickened A horizons, and buried horizons, suggesting that stratification during deposition of the alluvial materials is the most important factor in explaining particle-size differences. A sizeable component of the total variability within the sand fraction (as much as 39% for the coarse sand fraction) was explained by site. Sites are located adjacent to a pond, and second- and third-order streams. In relative terms, these environments represent at least three different potential depositional energy environments. The wetland adjacent to the pond (Duntons Mill) would represent a depositional environment of minimal potential energy. The larger third-order stream (Cub Creek) that moves through the dissected Piedmont has more relative potential energy than the second-order streams running through the low gradient Coastal Plain. Streams that have more potential energy can carry a coarser sediment load to deposit in the adjacent wetland during flooding. The sand data, and the location of the sites, suggest that considerable differences exist among sites and that possibly some of these differences can be related to the potential of the streams to carry and deposit sediment into the adjacent wetlands.

Variability among Elevations within Wetland Sites
In order to describe and explain differences within sites, PSD, organic C, TKN, CEC, and exchangeable bases were examined for the samples collected at the 5- to 15-cm depth for the three elevation intervals (Tables 4 and 5). Although the majority of the soil property means are not significantly different (at the 0.05 level) among the three elevations intervals, several trends in the data were observed. For all of the wetland sites, the highest relative elevations had the highest amount of total sand. This trend is strongest for the Western Freeway, Cub Creek, and Duntons Mill wetlands where the clay content is greatest for the lower relative elevation interval. This would be expected in an alluvial environment where the energy of the depositional waters determines the size of the material deposited. Within a given wetland, as the elevation increases the PSD becomes coarser. Waters of high energy are necessary to reach the highest elevations and thereby deposit the coarsest material. Floodwaters do not stay at the high elevations long enough for much of the silt and clay size particles to settle out. Cub Creek is the largest and has the highest gradient of the four streams adjacent to the study wetlands. Cub Creek dissects the Piedmont, has the highest relative potential energy of the wetland sites, and has the most sand at the highest elevation interval. The Duntons Mill wetland in the present form is a low-energy environment. The wetland is presently adjacent to a mill pond and is not subjected to periodic high-energy flooding. However, particle-size distributions are similar to the Western Freeway wetland, which is subject to more dynamic surface water flows. These data suggest that the 5- to 15-cm materials at the higher elevation intervals (35–70 and 70–105 cm) in Duntons Mill were deposited under a different energy environment than presently exists (i.e., before the mill pond was constructed). How much of an effect the mill pond construction has had on the wetland soil characteristics is unknown. The predominant vegetation found at Duntons Mill is typical of scrub-shrub wetlands, and the same plants are found at the other scrub-shrub wetland, Elmwood Baylor. With the exception of TKN levels, the Duntons Mill chemistry data appear to be similar to the Elmwood Baylor data (Table 5), suggesting that these wetlands were possibly in similar settings in the past.


View this table:
[in this window]
[in a new window]
  		Table 4. Sample means of selected particle-size fractions at the 5 to 15-cm depth.{dagger}

 

View this table:
[in this window]
[in a new window]
  		Table 5. Sample means of exchangeable Ca and Mg, cation-exchange capacity (CEC), organic C (OC), and total Kjeldahl N (TKN) at 5- to 15-cm depths.{dagger}

 
For the Western Freeway, Cub Creek, and Duntons Mill sites, fine and medium sand fractions follow the trends observed for total sand, and medium silt follows the trend observed in the clay fraction. These particle-size data support the elevation trends discussed above. Fort Lee has the least change in elevation across the wetland (60 cm), and this may explain why trends in particle size across elevation intervals are not strongly evident.

Cation-exchange capacity decreases as elevation increases in the Western Freeway, Cub Creek, and Duntons Mill wetlands (Table 5). The differences are only significant in the Duntons Mill data. Relationships between CEC and elevation for these three wetlands are directly related to the soil clay and organic C contents. The highest elevations in the Western Freeway, Cub Creek, and Duntons Mill wetlands have the lowest quantity of clay (Table 4) and organic C (Table 5). Intermediate elevation intervals at Fort Lee (20–40 cm) and Elmwood Baylor (25–50 cm) have the highest CEC values, as well as the highest amounts of both clay and organic C, of the three elevation ranges.

Higher elevations would be expected to have the lowest water tables relative to the surface, and therefore these locations should have more losses of N and C due to organic matter decomposition. In general, levels of TKN and organic C decrease from the lowest elevation to the highest at the Western Freeway, Cub Creek, and Duntons Mill wetlands (Table 5). Spatial relationships based on elevation were not apparent for soil C and N data at Fort Lee and Elmwood Baylor. Levels of exchangeable Ca and Mg are noticeably higher at Western Freeway, especially in the lowest elevations. Runoff from the adjacent agricultural fields is probably contributing to elevated levels of Ca and Mg from inundating waters at the lowest elevation. Duntons Mill and Elmwood Baylor wetlands are also adjacent to agricultural fields, but relationships between elevation and exchangeable cations are not apparent.

The CV values for the elevation interval means from the 5- to 15-cm depth range from 1 to 120% (Tables 4 and 5). These values suggest considerable random variability within elevation sampling units even though the elevation interval within a given sampling unit is <40 cm. Fort Lee is unusual in having low CV values for particle-size data within elevation sampling units (Table 4). Ten of the 15 means for Fort Lee have CV values of 15% or less. Only 13% of the particle-size means for the other four wetlands have CV values <15%.

Variability among Depths within Elevations
Most of the particle-size fractions showed significant differences with depth from the soil surface (Table 6). These data are indicative of the stratified alluvial sediments that are the parent material for these soils. Stratification occurs because each stream will carry a variety of sediment loads depending on flow rates and flooding conditions. In addition, streams tend to meander over time, especially those in the lower Coastal Plain. Therefore, the distance from the source of the sediment changes with time affecting the size of the particles being deposited.


View this table:
[in this window]
[in a new window]
  		Table 6. Sample means of selected particle-size fractions, organic C (OC), and total Kjeldahl N (TKN) at the highest relative elevation interval.{dagger}

 
At almost every site, the TKN and organic C content decreased with depth from the soil surface (Table 6). The only exception was TKN in Western Freeway. In most cases the differences in the amounts of both TKN and organic C between the 5- to 15-cm depth and the 40- to 50-cm depth were significant ({alpha} = 0.05). The higher TKN values are associated with the accumulation of organic matter as the surface and near surface horizons receive additions of organic matter from plant roots and leaf litter.

Variability within pedons was examined by collecting and analyzing three samples from each of the three depths at sampling location C within each elevation interval. Pedon sampling units have average CV values between 20 and 29% for the medium sand, fine sand, total sand, and clay particle-size fractions, TKN, and organic C means (Table 7). The average CV values for the same parameters calculated from three locations within elevation intervals averaged from 43 to 54% (Table 7). These data show that considerably less random variability occurred within pedon sampling units than the elevation sampling units. Pedon sampling units were spaced within 1 m of each other. In contrast, elevation sampling locations were spread across the 0.25- to 0.35-ha sampling area.


View this table:
[in this window]
[in a new window]
  		Table 7. Pedon and elevation average CV values calculated across all of the sites (n = 45) for selected particle-size fractions, organic C (OC), and total Kjeldahl N (TKN).{dagger}

 
Soil Morphology
Wetland soil morphology is dependent upon the vegetation, fauna pedoturbation (i.e., crayfish [Astacus astacus] and other burrowers), depositional environment, and the height and duration of the water table. The wetlands we studied had differences in each of these factors. Thus, the soils differed in both morphology and classification (Tables 1 and 2). Cub Creek and Western Freeway are both forested wetlands, but their soil morphology is drastically different (Table 2). Cub Creek has a sandy overwash layer at the surface, common redoximorphic features in the B horizons, and few roots below 40 cm. Western Freeway has a much longer duration and shallower depth of saturation than Cub Creek (Fig. 2). Soils in the Western Freeway wetland are dominated by low chroma A horizons with many and common roots to depths exceeding 1 m (Table 2). Organic matter contents in the upper meter are above 50 g kg-1 (Table 6). Soils at Fort Lee appear to be an intermediate between the other two forested sites. The water tables in the winter and spring are similar to those at Western Freeway. During the summer and fall Fort Lee water tables are similar to Cub Creek.

The Western Freeway wetland has a 10- to 30-cm water table draw-down in the summer and fall that would indicate a slightly drier environment for parts of the year compared with Elmwood Baylor and Duntons Mill (Fig. 2). Sites with longer periods of saturation at or near the soil surface often have higher amounts of organic C because the anaerobic conditions retard decomposition of the plant remains. The upper 5 cm of Elmwood Baylor and Duntons Mill soils are sapric organic materials (Table 2), but below this depth organic C levels drop rapidly in comparison with Western Freeway (Table 6). This may occur because in the scrub-shrub wetlands the root abundance in the lower profile was common or few. Therefore, site conditions such as vegetation may be as important as water table patterns in explaining differences in soil morphology and spatial distributions.


   SUMMARY
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY
REFERENCES
 
In this study we examined five palustrine wetlands to elucidate the sources of spatial variability for the overall study, to determine the variability in soil properties for both pedon and elevation sampling units within each wetland, and to examine differences in site characteristics and soil morphology among wetlands. Site characteristics differed in vegetation, topographic relief, hydrologic pattern, source of surface water, and physiographic region. Soil morphology differed among wetlands and the soils were classified as Endoaquepts, Humaquepts, Dystrudepts, Endoaquents, and Fluvaquents. When the soil properties of all the wetlands were compared, we found that elevation explained only a minor portion (0–16%) of the variability. For these forested and scrub-shrub wetlands, factors related to site conditions or depth from the surface explain a majority of the variability in soil properties for the overall study. Differences among wetlands can be attributed to factors such as hydrology, water chemistry, vegetation, alluvial parent materials, and the relative potential energy of the adjacent stream. Relationships between soil depth and particle-size distribution could be attributed to the stratified nature of the alluvial parent materials. Within each wetland, however, relative elevation was found to be an important factor controlling particle-size distribution and chemical properties. In regard to chemical properties, higher elevations were found to have the lowest levels of N and organic C. Higher elevations would be expected to have the lowest water tables relative to the surface, and therefore these locations have higher rates of organic matter decomposition. Particle size–elevation relationships within most of the wetlands showed that as the elevation increased the particle-size distribution at the 5- to 15-cm depth became coarser. This was expected in these alluvial wetland environments where waters of the highest energy are necessary to reach the highest elevations, and thereby deposit the coarsest material. The observed elevation–soil property relationships within wetlands support the need for stratified sampling in wetlands based on elevation. Random variability within pedons was considerably less than within elevation sampling units. Pedon sampling units were only 1 m apart, whereas elevation sampling locations were spread across the 0.25- to 0.35-ha sampling area. Our long-term goal was to evaluate a set of relatively undisturbed wetlands that could be used for reference purposes when assessing the soil properties of constructed wetlands. This study suggests that spatial relationships attributable to depth from the soil surface and elevation, wetland setting, and wetness regime should be considered when comparing reference and constructed wetlands.


   ACKNOWLEDGMENTS
 
This project was supported by the Transportation Research Council of the Virginia Department of Transportation, Charlottesville, VA, and was coordinated by Mr. Mike Fitch. We would like to thank Ms. Angela Cummings for monitoring the water table levels and recording the vegetation at Fort Lee, and Mr. Steve Nagle for his assistance in the water table monitoring. This manuscript is a contribution from the Rhode Island Agricultural Experiment Station (Contribution no. 3798) and the Department of Crop and Soil Environmental Sciences at Virginia Tech.

Received for publication July 9, 1999.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY
 REFERENCES
 




This article has been cited by other articles:


Home page
 J. Environ. Qual.Home page
R. Corstanje, S. Grunwald, K. R. Reddy, T. Z. Osborne, and S. Newman
Assessment of the spatial distribution of soil properties in a northern everglades marsh.
J. Environ. Qual., May 1, 2006; 35(3): 938 - 949.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Stolt, M.H.
Right arrow Articles by Groover, V.A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Stolt, M.H.
Right arrow Articles by Groover, V.A.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Stolt, M.H.
Right arrow Articles by Groover, V.A.
Related Collections
Right arrow Soil Methods/Instrumentation
Right arrow Wetland Soils
Right arrow Spatial Distribution

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome