| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Geography and Geology, University of Wisconsin–Whitewater, Whitewater, WI 53190
b Department of Agronomy, University of Georgia, Athens, GA 30602
c Department of Agronomy and Soils, Auburn University, Auburn, AL 36849
* Corresponding author (jacobsp{at}mail.www.edu)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONNature of the Area...METHODS AND MATERIALSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONNature of the Area...METHODS AND MATERIALSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
2 and Fe-concentration or depletion areas have become commonly used indicators of seasonal saturation in soils. Veneman et al. (1998) summarized these relationships, indicating that the depth to a seasonal high water table during the growing season is often correlated with the presence of redox concentrations, whereas the presence of redox depletions or a reduced matrix often indicates significant duration of saturation. In the southeast USA, studies of Coastal Plain soils have generally found that soil morphology is impacted by seasonal groundwater saturation and furthermore that the duration of seasonal saturation during the growing season can be interpreted from morphology. In North Carolina, Daniels et al. (1971) found that soils with high water tables had thin E horizons, minimal eluviation of fine material, gray colors, and low contrast mottling. High water tables occurred in soils located further from the dissected edge of upland divides. Soils located along the dissected edge of a divide have thick E horizons, fine-textured B horizons and high chroma colors, features attributed to a deep water table.
Genthner et al. (1998) observed good correlation between redox features and a seasonal high water table on the upper Coastal Plain in Virginia. They found that depth to Fe depletions or a reduced matrix underestimated the water-table depth in well-drained and moderately well drained soils. They suggested underestimation occurred because the horizons with these features are deep in the soil and lack abundant organic C, and also because the seasonally high water table occurs in winter months when soil temperatures are cool, two critical factors that can limit biological reduction of Fe. Genther et al. (1998) suggested redox concentrations are better indicators of the seasonal high water table in these soils. In more poorly drained soils, Fe-reduction features tend to occur higher in the profile than the average seasonal high water table, perhaps because of a more abundant C source to drive reduction early in the growing season and possibly because reduction may continue through the cooler winter months in these soils (Genther et al., 1998).
West et al. (1998) studied southwest Georgia Ultisols and found that redox concentrations are associated with a mean cumulative duration of saturation of 20%, with a range of from 4 to 47%. Redox depletions were associated with mean duration of saturation of 40%, ranging between 10 and 62%. Horizons with a low chroma matrix were saturated with a mean of 50% of their 805-d study period, ranging from 12 to 67%. There were no consistent trends in their data to explain the overlap in range of time saturated for each category of redox feature. Therefore, West et al. (1998) suggested some of the features may be relict, having developed when the soils were more poorly drained either because of a wetter climate or less relief.
Additional morphologic properties that relate to or control the occurrence of redox features in southeastern soils are clayey textures and hydraulically restrictive horizons. Megonigal et al. (1993) proposed that clayey horizons, especially deeper in a soil, may reduce O diffusion rates to the point of maintaining redox features in soils only periodically saturated. Hydraulically restrictive horizons at depth are common features on the Coastal Plain and have been implicated in perching water, inducing lateral flow across the landscape, and in the formation of plinthite (Blume et al., 1987; Daniels et al., 1978; Shaw et al., 1997).
The generally accepted relationship between redoximorphic features and seasonal saturation is being utilized for the classification, mapping, and interpretations of soils in the rapidly suburbanizing areas of the Georgia Coastal Plain, but little data is available to define or refine these relationships in the region around Valdosta, Lowndes County, GA. Verification of the relationship between redoximorphic features and the occurrence and duration of seasonal saturation will aid land managers and environmental health officials with proper classification of the soils and in protecting surface and groundwater quality in a major recharge area of the Floridian aquifer (Huddlestun, 1997). The objective of this study was to relate the presence and duration of seasonal saturation to redoximorphic features in some sandy and clayey soils of the Coastal Plain in south central Georgia.
|
Nature of the Area and Site Location |
|---|
|
TOPABSTRACTINTRODUCTIONNature of the Area...METHODS AND MATERIALSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
Series mapped in the region in Plinthic, Arenic, and Grossarenic subgroups of Paleudults and Kandiudults (Tifton, Dothan, Lucy, Wagram, Albany) are recognized to have Fe depletions and concentrations in lower Bt horizons. These series occur on all hillslope positions higher than footslopes. Moderately well and somewhat poorly drained soils that occur on lower backslope positions are often included in these well-drained map units because they occur in narrow bands too small to delineate at the scale of mapping. Paleaquults and Alaquods commonly occur at the base of slopes in the region and on fluvial terraces.
Within this soil geomorphic setting, two transects, a clayey hillslope transect and a sandy stream terrace transect, were selected to represent common landscape and soil sequences mapped in the area. The slopes, mapped soil series, and landscape setting of the transects are typical, and we believe representative, for the region. Seven pedons were studied along the two transects. Both transects are located on managed forest land with dense stands of loblolly (Pinus taeda L.), slash (Pinus ellioti Engelm.), and longleaf (Pinus palustris P.) pine. None of the pedons had evidence of deep soil disturbance during plantation establishment, although Ap horizons with subsoil inclusions indicate past disturbances at most sites. Harvest of pole-size trees at the clayey transect sites occurred at the end of the study period.
Soils of the clayey transect are located at 30°47'32''N lat., 83°23'W long, and were formed in the late Pliocene Miccosukee Formation, a sandy-clay marine deposit (Huddlestun, 1997). The soils are located along a low gradient hillslope and include Plinthic Kandiudults in the upper (4% slope) and lower (2% slope) backslope, and a Plinthaquic Paleudult in the nearly level (1% slope) footslope position. The Kandiudult profiles contain ironstone, plinthite, and reticulate mottling.
Four soils were included in the sandy terrace transect. They are located at 30°47'31''N lat., 83°2'W long. The terraces are associated with the Withlacoochee River, a major stream in the region. The topographically highest soil is a Typic Quartzipsamment formed in either barrier island sands preserved on a marine terrace (Huddlestun, 1997) or an unrecognized stream terrace remnant. The next soil on the transect was a Grossarenic Kandiudult that formed on the backslope separating the upland from a high terrace. A Grossarenic Kandiaquult and an Ultic Alaquod formed in alluvium on a high terrace of the Withlacoochee River (Huddlestun, 1997) were the topographically lowest members of the transect (Table 1).
|
|
METHODS AND MATERIALS |
|---|
|
TOPABSTRACTINTRODUCTIONNature of the Area...METHODS AND MATERIALSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Piezometers were monitored between September 1995 and September 1997, typically at least twice monthly. In some cases, an additional measurement was taken within 12 to 24 h after heavy rains to check for short-term perching of water. The percentage time that a horizon experienced saturation was estimated from graphs of water-table elevations. A horizon was considered saturated if >75% of the horizon was under the free water table. The percentage time was calculated as the number of days saturated divided by the 740-d duration of the monitoring period. Differences in the percentage time of saturation associated with each group of redox feature (concentrations, depletions, low chroma matrix) were evaluated statistically by a two-sample t-test of means assuming unequal variance.
Precipitation received during the study was within the range of typical for the region. We obtained National Climatic Data Center records for the cities of Valdosta and Quitman, located 15 km, nearly equidistant, east and west from the study area. At Valdosta, reported precipitation throughout the study period was less than the 1961 through 1990 mean in most months. At Quitman, reported precipitation values were nearly at the mean, although spring 1996 precipitation was higher than mean values (Fig. 2) . Cumulative annual precipitation totals for each year are within 25% of the mean, although individual monthly values ranged between 14 to 925% of mean. Consequently, we believe that levels of seasonal saturation in the soils studied were typical for the region. The growing season for Lowndes County is 24 February through 30 November.
|
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONNature of the Area...METHODS AND MATERIALSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
Seasonal Saturation and Redox Features in the Clayey Soils
A clear relationship exists between seasonal saturation and redoximorphic features for the footslope pedon, but not for the backslope pedons. Pedon 92-1, a Plinthic Kandiudult in the upper backslope position, did not have a free water table at <2-m depth during the study period, even within 24 h following rain events as large as 140 mm (5.5 in.). Morphologically, this pedon has 10% platy and nodular plinthite in the lower B horizons and many 10YR 8/1 (white) Fe-depletion areas as part of a reticulately mottled C horizon. Thick clay films were observed in a thin transition zone (B/C horizon) just above the C horizon, suggesting water percolates to that depth. The profile was never saturated during the period of monitoring, and when the pit was excavated (June 1995), the C horizon was so dry that a large backhoe had difficulty breaking through the material, even though the previous year (1994) received 139% of mean precipitation. Clearly the C horizon is hydraulically restrictive, as other studies of similar soils have found (Blume et al., 1987; Shaw et al., 1997), but lateral flow downslope must be rapid enough to remove all percolating water.
While conclusions from a 2-yr study can be tenuous, we speculate that much of the observed redox features in Pedon 92-1 are relict and not reflective of the contemporary hydrological conditions at this site. Given the location on an upper backslope, it is conceivable that the reticulate mottling has been preserved since incision of the local drainage way (Daniels and Gamble, 1967; Nettleton et al., 1989). An edge effect explanation is complicated, however, by the occurrence of the morphologically similar Pedon 92-2 in the lower backslope position. Similar morphology indicates pedon development following or during incision of the landscape. A prehistoric wetter climate, that produced a higher water table, is a plausible explanation for the apparently relict morphology, but is not investigated further here.
In Pedon 92-2, a Plinthic Kandiudult on the lower backslope, the lower solum contains Fe-concentration and depletion areas in a reticulately mottled Btcv2 and C horizon. The Btcv2 horizon has common 10YR 8/2 Fe-depletion areas and as much as 15% platy and nodular plinthite. This horizon was saturated for 4% (30 d total over 2 yr) of the monitoring interval (Fig. 3). The C horizon, with many 10YR 7/2 Fe-depletion areas, experienced saturation for 9% (70 d total, as many as 35 consecutive days) of the monitoring period. The C horizon was moist and easy to dig at the time of excavation, unlike the C horizon in Pedon 92-1. The water table rose from beneath 2-m depth. Water table levels in adjacent piezometers were identical; no water was observed perched above either the reticulately mottled or plinthite-rich horizons, a finding unlike previous studies of similar soils (Carlan et al., 1985; Daniels et al., 1978; Guthrie and Hajek, 1979). The relatively short duration of saturation associated with distinct redoximorphic features is similar to findings in southwest Georgia by West et al. (1998), who suggested the features may form during extremely wet periods, or are relict, having formed under a previously wetter climate, or prior to landscape incision. Clearly, the Kandiudults of the rolling landscapes of southern Georgia are complex, and more study is needed to understand the origin of subsoil redoximorphic features and their implications for classification, use, and management.
Pedon 92-3, a Plinthaquic Paleudult in the footslope, showed a strong relationship between seasonal saturation and redoximorphic features. Iron-depletion features occur in horizons that were saturated 5 to 66% of the study period, with loamy B horizons saturated >20% of the time. Iron masses occur in horizons saturated for 1 to 78% of the study period. Coarse nodular plinthite occurs only in horizons with free water >30% of the time, in contrast to the Kandiudults up slope that were saturated only 1 to 4% of the study period. Matrix color with chroma 2 is associated with saturation 78% of the study period, a duration greater than the maximum range of 67% West et al. (1998) found in southwest Georgia soils.
Seasonal Saturation and Redox Features in the Sandy Soils
A clear relationship exists between seasonal saturation and redoximorphic features for the terrace pedons, but not for the upland or backslope pedons. Pedon 92-4, a Typic Quartzipsamment on a flat upland surface, has low chroma colors (2) in deep horizons seasonally saturated for >18% of the study period. The sandy texture and siliceous mineralogy of this soil likely enhances the low chroma colors. Iron masses were observed in the C2 horizon, between 53 to 79 cm, that was not saturated during the study period. The Fe concentrations in this horizon may reflect either extreme precipitation years not represented during this study or is the result of capillary rise. Iron depletion features in the C3 horizon with free water only 1% of the study period also suggests the saturation may be occasionally higher for longer periods. Alternatively, the redox features in both the C2 and C3 horizons may be relict.
Pedon 92-5, a Grossarenic Kandiudult on the backslope between the upland surface and a high terrace, has many Fe-depletion features in a reticulately mottled 2C horizon in what appears to be the Miccosukee Formation. Within the solum, the E2 horizon has Fe-concentration features but it was never saturated during the study period. The sandy Bt also has Fe-concentration features only, mostly as Fe masses, but <5% nodular plinthite as well. The 10YR 6/6 Bt was saturated 4% of the study period, during short intervals no longer than 2 wk, indicating that short periods of saturation are not sufficient to induce formation of Fe depletions in this soil. This supports the earlier suggestion that the formation of Fe-depletion features in the Kandiudults may require more extreme precipitation years or are relict features.
Pedon 92-6, a Grossarenic Kandiaquult on a flat terrace surface, showed a good relationship between the duration of seasonal saturation and redoximorphic features. Matrix colors 2 occur in saturated horizons for 28 to 53% of the study period. Iron-depletion areas are present in an E horizon saturated for 15% of the time. The sandy texture, with a low surface area, may contribute to the formation of depletions in a relatively short period of time. Iron-concentration features were recognized in all horizons except the Ap horizon. Iron masses occur in sandy E horizons saturated during 1 to 15% of the study period, while plinthite and soft masses are present in finer-textured horizons that were saturated for 28 to 53% of the study period.
Pedon 92-7, an Aquod on a flat terrace surface, was the wettest soil in the study. The duration of saturation ranged from 91% of the study period in the C horizon to 12% in the E horizon at 15 cm, although the entire soil was saturated to the surface for several days following large rain events. The 10YR 8/1 E horizon contains no redox concentrations even though it was saturated for 12% of the study, an observation common in Florida Aquods and attributed to a lack of Fe in the system (Kuehl et al., 1997). The Bh horizon was saturated for 20 to 30% of the study period, much less than the >60% of time saturated reported by Tan et al. (1999) in some Florida Aquods. Iron-concentration features only occur beneath the Bh, in low chroma horizons that were saturated from 45 to >90% of the study period. With greater depth in the profile, Fe-concentration features changed from nodules and Fe masses to almost exclusively pore linings (rhizospheres) along live roots.
Table 2 summarizes the mean and range of the duration of saturation and the occurrence of redoximorphic features in these south central Georgia soils. The mean duration of saturation associated with Fe-concentration features is 25%, compared with 20% for southwest Georgia soils studied by West et al. (1998). The mean duration of saturation associated with Fe-depletion features is 18%, compared with 41% for West et al. (1998) and 50% by Daniels et al. (1971). Notice that the range of saturation data for Fe-concentration and depletion features includes values of 0%. Excluding the horizons with these features that did not experience saturation changes the mean a few percent, but the data set still includes several horizons that were saturated only briefly, often 1% of the study period. Matrix colors dominated by 2 chroma colors occur in horizons that were saturated between 18 to 91% of the study, with a mean of 57%. We excluded horizons with significant organic matter (A, AE, and Bh) and the E horizon of the Aquod. The value compares well with the mean of 51% reported by West et al. (1998) and the >50% reported by Daniels et al. (1971). The t-tests indicate that the duration of saturation associated with matrix colors dominated by a 2 chroma color is significantly different (
< 0.05) than that associated with Fe-concentration or depletion features. Duration of saturation associated with concentration and depletions features is statistically similar. Excluding zero values did not change the statistical relationships between groups.
|
Finally, while not an initial objective of this study, our saturation data can be used to assess whether any of these pedons meet the criteria for classification as hydric soils (USDA-NRCS, 1998). Pedons 92-7 and 92-3, with sandy loam or coarser textures within 50 cm, were saturated to 15-cm depth for at least 14 d during the February 24 through November 30 growing season. Based on our 2 yr of data, only Pedon 92-7, the Aquod, appears saturated above the critical depth as part of the seasonal cycle of saturation. This pedon has no clear hydric soil indicators. The color value (4) of the A horizon is too high and the proportion of coated to stripped soil particles is less than the required 70% for indicator S7 (dark surface). In addition, the stark white E horizon lacks Fe-concentration features or humus staining, precluding application of indicators S5 (sandy redox) or S6 (stripped matrix). All organic matter accumulation above the A horizon is fibric or hemic organic materials with recognizable pine needles, cones, and other woody detritus. Pedon 92-3, on the other hand, was saturated only when the level of saturation rose in response to large precipitation events and remained high for at most about 14 d, causing it to marginally classify as a hydric soil. This pedon also lacks any well expressed hydric soil indicators. Redox concentrations are present beneath the A horizon, but the value and chroma of these features are too high for indicator F12 (Fe/Mn masses).
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONNature of the Area...METHODS AND MATERIALSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
2. A gray matrix (i.e., 2 chroma color) is indicative of seasonal saturation, on average for >50% of the time. Based on occurrence frequency, sandy horizons are more likely to develop a gray matrix, probably because of their lower surface area and siliceous mineralogy.
3. Low chroma Fe-depletion features are associated with a mean duration of saturation of 18% of the study period, less than is commonly reported.
4. High chroma Fe-concentration features are associated with a mean duration of saturation of 25%, similar to other regional reports, but these features are not reliable indicators of the top of the zone of seasonal saturation, as measured by piezometers. Iron-concentration features may reflect capillary action up from the zone of saturation or possibly reflect high water table levels in periods of extreme precipitation.
5. Kandiudults with plinthite and reticulate mottling do not show a consistent relationship between saturation and redoximorphic features.
In general, only soils in the lower landscape positions appear to have relationships between saturation and redox features similar to other reports in the literature. Pedons in upland and backslope positions do not exhibit such relationships, even though they contain redoximorphic features. Reasons why are not clear from this 2-yr study, but we suspect many of the redox features in these south central Georgia soils are relict, having formed prior to incision of the landscape. More monitoring in a variety of landscape settings is needed to determine landscape, climate, and local hydrologic controls on the occurrence of these features.
|
ACKNOWLEDGMENTS |
|---|
Received for publication July 10, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONNature of the Area...METHODS AND MATERIALSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. P. Morgan and M. H. Stolt Soil Morphology-Water Table Cumulative Duration Relationships in Southern New England Soil Sci. Soc. Am. J., March 29, 2006; 70(3): 816 - 824. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. P. Morgan and M. H. Stolt A COMPARISON OF SEVERAL APPROACHES TO MONITOR WATER-TABLE FLUCTUATIONS Soil Sci. Soc. Am. J., March 1, 2004; 68(2): 562 - 566. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Fiedler and M. Sommer Water and Redox Conditions in Wetland Soils--Their Influence on Pedogenic Oxides and Morphology Soil Sci. Soc. Am. J., January 1, 2004; 68(1): 326 - 335. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. He, M. J. Vepraskas, D. L. Lindbo, and R. W. Skaggs A Method to Predict Soil Saturation Frequency and Duration from Soil Color Soil Sci. Soc. Am. J., May 1, 2003; 67(3): 961 - 969. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 | |||
