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 ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fiedler, S.
Right arrow Articles by Sommer, M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Fiedler, S.
Right arrow Articles by Sommer, M.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Fiedler, S.
Right arrow Articles by Sommer, M.
Related Collections
Right arrow Redox Processes
Right arrow Wetland Soils
Right arrow Watershed and Landscape Processes
Published in Soil Sci. Soc. Am. J. 68:326-335 (2004).
© 2004 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA

DIVISION S-10—WETLAND SOILS

Water and Redox Conditions in Wetland Soils—Their Influence on Pedogenic Oxides and Morphology

Sabine Fiedler*,a and Michael Sommerb

a Univ. of Hohenheim, Institute of Soil Science and Land Evaluation, Emil-Wolff-Str. 27, D-70593 Stuttgart, Germany
b GSF—National Research Center of Environment and Health, Institute of Biomathematics and Biometrics, Ingolstädter Landstr. 1, D-85764 Neuherberg, Germany

* Corresponding author (fiedler{at}uni-hohenheim.de).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
For wetland soil studies, it would be desirable to estimate soil redox status and its effect on element translocations based on diagnostic redoximorphic features. However, their identification and interpretation is often complicated and require a regional calibration. This study was conducted (i) to determine the relationship between water table and redox potential (EH), (ii) to investigate their influence on mass balances of pedogenic Mn- and Fe-oxides and soil morphology, and (iii) to evaluate the use of soil color to identify redox status in hydromorphic soils. Three representative geomorphic units were chosen from low relief positions in the cool-humid Allgäu (Germany): an alluvial plain and two pond margins. Within each unit two paired study sites were established along a wetness gradient. Thus in six wetland soils (Aquepts) groundwater level (GWL) and EH were measured over 1 yr. Monitoring results were presented as (i) percentage of time of water saturation, (ii) percentage of time the EH is <170 mV (onset of Fe-oxide reduction), and (iii) related to static soil properties (mass balance of pedogenic oxides). Both, GWL and EH were linearly correlated (r2 = 0.88) and indicated increased Fe mobilization with increasing duration of saturation. Thus, higher Fe losses occurred with increasing duration of EH < 170 mV, whereas no correlation existed between Mn and the duration of reducing conditions. Calculation of Mn and Fe mass balances indicated losses of both elements within the pedons when reducing conditions were located near the surface (<10 cm). However, no losses were detected when the reductive conditions occurred at depths >50 cm. The element redistribution induced by soil redox conditions was reflected by the soil color index of subsoil horizons (C2h), within a sensitive range between 6 and 12. It was shown, that this index is an adequate tool to delineate wetland soils according to the duration of water saturation and Fe-reducing conditions. It could be concluded that C2h may be used as a proxy to estimate the intensity of water/redox conditions.

Abbreviations: AE, Aeric Endoaquept • AH, Aeric Humaquept • C2h, soil color index • Corg, organic carbon • Ct, total contents of carbon • EH, reduction-oxidation (redox) potential • Fed, dithionite citrate bicarbonate extractable Fe • Feo, ammonium oxalate extractable Fe • Fep, pyrophosphate extractable Fe • Fet, total contents of Fe • FH, Fluvaquentic Humaquept • GWL, groundwater level • MAP, mean annual precipitation • ME, Mollic Endoaquept • Mnd, dithionite citrate bicarbonate extractable Mn • Mno, ammonium oxalate extractable Mn • Mnt, total contents of Mn • PDI, profile darkening index • PVC, polyvinylchloride, TH1, Typic Humaquept Pond Margin A • TH2, Typic Humaquept Pond Margin B


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
REDOXIMORPHIC FEATURES have being used for more than three decades for identifying hydrologic and soil wetness conditions (Blume, 1968, 1973, 1988; Schlichting, 1973; Veneman et al., 1998; He et al., 2002). Nevertheless, studies that link time-dependent field conditions (e.g., water saturation), pedogenetic processes (reduction of Fe-oxides, element transport, mass balance, soil color) and landscape conditions are scarce, even though these connections are logical keys to the understanding of soil processes and their ecological consequences such as element cycling.

Water dynamics have profound influence on the genesis of soils, especially for wetland soils. As a result of water saturation O2, diffusion into soil is drastically curtailed (Callebaut et al., 1982). The O2 trapped in the soil or present in the water is consumed within few hours by microbes. The resultant waterlogged soil is practically devoid of molecular O2 (Ponnamperuma, 1985). As long as molecular O2 is available it acts as the preferred electron acceptor, followed by NO3, Mn oxide, Fe (hydr)oxides, SO4, and finally CO2 (Stumm and Baccini, 1978). Reduction of the different acceptors is accompanied by typical ranges of EH. Its measurements in soils can be used to quantify the tendency of soils to oxidize or reduce substances (Faulkner and Patrick, 1992). For example, EH values that are <170 mV (pH = 7) indicate that reduction of Fe (hydr)oxide takes place (Brookins, 1988). Ferrous iron is far more soluble than ferric iron, thus creating an increase in Fe mobility. The Fe2+ can be transported within soils and landscapes via soil solution along redox gradients (Reuter and Bell, 2001). If the Fe2+–specific threshold of EH value is exceeded, then ferrous Fe becomes immobilized and accumulates (Fortescue, 1980). Conversely, areas with frequent reducing conditions lose Fe2+ with outflowing water. This redistribution is visible in the field by the change from brown to gray colored soil horizons (Franzmeier et al., 1983). Consequently, beside pedochemical indicators like Mnd/Fed (McDaniel et al., 1992) and crystallinity of Fe oxides (Richardson and Hole, 1979) soil color has been used as a general indicator of redox processes (Evans and Franzmeier, 1988; Singleton, 1991; Jacobs et al., 2002; Jenkinson et al. 2002). However, it is known that these pedogenic and morphological indicators can lead to false conclusions: (i) Inherited soil color from parent material (Elless et al., 1996; Uzunoglu, 1973), or organic matter can cover or coat redoximorphic features (Veneman et al., 1998; Vepraskas, 2000); (ii) redoximorphic features may be relicts (Schlichting, 1973; Greenberg and Wilding, 1998); or (iii) similar morphological patterns might be induced through element translocations under acid conditions (Spodosols, Mokma and Sprecher, 1994). Therefore, GWL and redox conditions should always be verified. But, measurements of these soil parameters are time-consuming and expensive, and are rarely available in most soil reports.

A systematic procedure that can provide linkage between intensive long-term, expensive in situ measurements (e.g., water table and redox potential) and both soil morphology as well as landscape parameters (e.g., slope position) will allow extrapolations from pedon scale to greater areas (Fiedler et al., 2002). However, the formation of redoximorphic features, including their intensity of expression, is site-specific and not always directly related to the period of saturation. Redoximorphic feature identification and interpretation is often complicated because they are connected through complex, interacting soil-forming processes dominated by redox chemistry (Faulkner and Patrick, 1992). This requires a regional calibration of any proxies to use.

Our study looks for the relationship between reducing conditions and translocation processes of Fe and Mn and the resultant consequences on soil color morphology. The objectives of our study were (i) to determine the relationship between water table and redox potential, (ii) to investigate their influence on mass balances of pedogenic Mn and Fe oxides and soil morphology, and (iii) to evaluate the use of soil color to identify redox status in hydromorphic soils. We selected six soils with redoximorphic features in the humid region of southwest Germany and tested the following hypothesis: increasing wetness corresponds to decreasing redox potentials, which leads in turn to negative mass balances of pedogenic Fe and Mn oxides and should be reflected by distinct soil color morphology in each soil (horizon). The confirmation of this hypothesis would justify the proposition that pedogenetic processes reflect environmental conditions and vice versa (e.g., soil color is a proxy of water and redox regime).


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Environmental Setting
The research area is located in the Allgäu region of southwest Germany. The landscape formed during the last glaciation period. The moraine landscape has a distinct topography of rolling hills with numerous closed depressions that has only a few streams with their affiliated alluvial plains. An alluvial plain and two pond margins (A, B in Table 1) were chosen as representative geomorphic units of lower relief positions. These pond margins occur in an area equivalent to 7% (approximately 42 km2) of the total Allgäu area (approximately 600 km2). Within each unit two paired study sites were established along a soil wetness gradient (Table 1). The soils of the two pond margins are influenced by past colluviation due to the former use of the catchment as arable land until the mid 1960s (Sommer et al., 2004).


View this table:
[in this window]
[in a new window]
  		Table 1. Description of geomorphic units and study sites.

 
The temperate-humid climate of the site (6.5°C mean annual temperature [MAT], 1300 mm, mean annual precipitation [MAP]) has annual evapotranspiration rates that range between 500 and 800 mm. The excess of precipitation over evapotranspiration results in an annual amount of water percolating through the solum of 500 to 1000 mm (Kleber, 1997).

The water table levels vary from year to year in response to precipitation. In addition, typically the climate and variable water tables should be accompanied by distinct EH conditions. However, the available redox and water table data were collected within two studies (Fiedler, 1997; Fiedler and Sommer, 2000) with different measurement periods (April 1992 to October 1993 and July 1996 to July 1998). Therefore, it was necessary to find out time windows with minimal deviation from MAP. These time windows were 21 July 1992 to 20 July 20 1993 for Pond Margin A, and 21 July 1996 to 20 July 1997 for both, the Pond Margin B and alluvial plain. Precipitation was 3% (35 mm) lower than MAP during 1992 and 1993; and 10 to 12% (120–150 mm) lower than MAP during 1996 and 1997.

Field Methods
Water table depths were determined with 6-cm i.d. polyvinyl chloride (PVC) wells (Fa. J. Stockmann, Warendorf/Einem, Germany). The entire belowground portion of the well was slotted on two sides in 1-cm intervals by horizontal slits (1 mm wide and 35 mm long), wrapped in filter fabric and installed in holes made with a 10-cm diameter bucket auger. The ends of the wells were covered with PVC caps. Boreholes were back-filled with a coarse quartz sand. Additionally, the soil/bore hold contact was sealed at the surface with bentonite to prevent by-pass flow of surface water into the wells. The wells were installed to a depth of 100 cm for weekly GWL monitoring. The depth of the water in the well was measured using a flashlight attached to the end of a retractable tape. All investigated soils, except for Aeric Endoaquept (AE) (alluvial plain) were saturated in one or more horizons within 100 cm for the first 4 wk of the study. Therefore, at the AE soil a well of 200 cm length was installed to ensure continuous water level monitoring.

The EH was measured using permanently installed electrodes and have been previously described (Fiedler and Fischer, 1994; Fiedler, 1997; Fiedler and Sommer, 2000). Briefly, to determine the distinct ranges of oxidation-reduction conditions in each horizon, 2 to 4 Pt electrodes were installed per horizon and one Ag/AgCl reference electrode was installed per profile. During the study period July 1992 to July 1993 a personal computer supported unit was used for automatic hourly readings (Fiedler and Fischer, 1994). During the second period (July 1996 to July 1997), a conventional portable pH/EH meter (WTW, Weinheim, Germany) was used for manual weekly readings (Fiedler and Sommer, 2000). To exclude differences in sampling frequency (hourly vs. weekly), we used from the data pool of hourly readings only one value (Wednesday 1300 h) per week for the present study.

Soil Characterization
The morphology of each profile was described in detail. Special emphasis was placed on the redoximophic features description as potential predictors of redox conditions in soils. Basic soil data given in Table 2 were determined according to methods of Schlichting et al. (1995). Bulk density was measured using undisturbed soil cores. Stone contents (vol. %) were estimated in the field according to Schlichting et al. (1995). Standard soil analyses were performed on the fine earth fractions (<2mm): particle-size distribution, pH (CaCl2), and total C (Ct, dry combustion, Leco CN-2000, Leco Instruments, GmbH, Krefeld, Germany). In all investigated soils, organic C (Corg) is equivalent to Ct, because soils are carbonate-free (except C horizons). Pedogenic Fe and Mn oxides were extracted by dithionite citrate bicarbonate (Fed, Mnd) (Mehra and Jackson, 1960), NH+4 oxalate (Feo, Mno) (Schwertmann, 1964), and analyzed with atomic absorption spectroscopy (AAS, Varian SpectrAA-200, Varian, Mulgrave Victoria, Australia). The total elemental analysis (Mnt, Fet) was conducted using X-ray fluorescence (Siemens SRS-200, Bruker AXS, Karlsruhe, Germany).


View this table:
[in this window]
[in a new window]
  		Table 2. Selected soil properties of the study sites.{dagger}

 
Calculations
The seasonal trends of water table fluctuations and reducing conditions published in previous papers (e.g., Fiedler, 1997, 2000) were used to integrate dynamic parameters (EH, GWL) and relate them to static soil properties (e.g., mass balances of pedogenic oxides).

Within the measurement period, the water table depths were transformed into percentage of time with water saturation. As decision criterion for water saturation at horizon scale, we used the following procedure: recorded water table levels fluctuating mostly within one horizon were used to separate the horizon into saturated and unsaturated zones. We arbitrarily regarded the horizon as saturated when ≥60% of the horizon thickness was situated below the water level. For example, the Bg1 (AH) was located at the depths between 52 and 90 cm below surface. During reading dates with water table depth of 80 cm below surface, 26% of the horizon (thickness) was influenced by water table. In this case, we referred the horizon as unsaturated. On the other hand, at a water table depth of 65 cm below surface, 66% of the horizon (thickness) was influenced by water table. In this case, we referred the horizon as saturated.

All redox data were corrected for the standard H electrode by adding 215 mV at 10°C to the field readings. Since reduction of Fe oxides are a function of both EH and pH, a correction to a given pH is needed to interpret EH data. Therefore, a laboratory experiments was conducted to calibrate EH to pH (Fiedler, 1997). Based on this, the investigated soils responded to reducing conditions as follows: The pH increased in the acid topsoils by 1.2 units, and in the nearly neutral subsoils decreased by 1.5 units. The maximum pH value was 7.0, therefore the onset of pedogenic Fe and Mn oxides reduction was calculated for this pH (Brookins, 1988; Brümmer, 1974). Analogous to water saturation, the percentage of time with <170 mV (Fe reduction) and <450 mV (Mn reduction) was calculated for each horizon. At the profile scale, we marked the depth at which ≥60% of the time EH values <170 mV were measured.

We calculated mass balances for Fe and Mn using the ratio-method of Sommer and Stahr (1996). Relative gains or losses of pedogenic oxides (Fed, Mnd) can be achieved by comparing the extracted element: clay ratio in a pedon (P ratio) with the element: clay ratio of the (mixture of) parent material (C ratio).

The Fed and Mnd masses as well as the clay masses per total soil volume were calculated as follows:

Mx, Mass of Fed, Mnd, or clay in the pedon fine earth (kg m–2 profile depth–1); xi, Fe, Mn, or clay (<2 mm) content in horizon i (g kg–1 fine earth); n, number of horizons to profile depth; {rho}B, bulk density (Mg m–3); yi, thickness of the horizon i (cm); cfi, coarse fraction (>2mm) of the horizon i (vol. %).

Using the P and C ratios we were able to quantify relative Fe losses (original element mass > actual element mass) and gains (original element mass < actual element mass) by:

Results from the ratio method have been highly correlated to mass balance calculations on Fet using index elements (Sommer and Stahr, 1996). Nevertheless, only differences greater than ±25% should be interpreted in terms of gains and losses (Sommer and Stahr, 1996). The ratio method can also be used for assessment of gains and losses at the horizon scale (dividing Fed, Mnd by clay content). Soil color for mineral horizons was characterized using the color index (C2h) developed by Evans and Franzmeier (1988). A and O horizons were characterized by profile darkening index (PDI) of Thompson and Bell (1996).


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Saturated/Reducing Conditions
Based both on the descriptive statistics of groundwater data given in Table 1 and the time with water saturation given in Fig. 1 increasing soil wetness was observed as follows: AE < Aeric Humaquept (AH) < Mollic Endoaquept (ME) < Typic Humaquept 1 (TH1) < Fluvaquentic Humaquept (FH) < Typic Humaquept 2 (TH2).



View larger version (51K):
[in this window]
[in a new window]
  		Fig. 1. Depth functions of percentage with water saturation (left column) and depth functions of EH (right column) for six soils (Aeric Endoaquept [AE], Aeric Humaquept [AH], Mollic Endoaquept [ME], Typic Humaquept Pond Margin A [TH1], Fluvaquentic Humaquept [FH], Typic Humaquept Pond Margin B [TH2]) in the Allgäu region of southwest Germany. Box plots contain 10, 25, 50(median), 75, and 90%; thick line are depth function of medians, filled scores equals means; (a) 21 July 1992 to 20 July 1993, (b) 21 July 1996 to 20 July 1997.

 
The water tables of the last two soils (FH, TH2) were nearly stationary. In both soils permanent water saturation was found at depths below 30 cm. During half of the field study the horizons below 11 cm were saturated, suggesting small seasonal variability. During snowmelt or flooding events, all sites were flooded with the exception of AE. Snowmelt ponding or flooding from channel overflow occurred for 2% of all sampling dates at ME, 7% at AH, 10% at TH1, 15% at FH, and 19% at TH2. The AE soil had the largest range (13–161 cm below surface) and lowest median water level (100 cm below surface) in the study, which indicates it was the driest and most variable soil in terms of water fluctuations (Table 1). From all data we concluded that an upward gradient in water potentials existed. Consequently, all soils can be regarded as soils with a permanent groundwater table (gleyzation, endosaturation) showing a dominant upward water movement at pedon scale. As in other studies (Faulkner and Patrick, 1992), a relationship between wetness and redox status was observed. High mean water tables coincided with small oxidative zones above them (Fig. 1). For example, in the two wettest soils (FH, TH2) at a depth >20 cm below surface 50% of all measured EH values were <170 mV.

The ranges and means of EH generally decreased with depth (Fig. 1). The standard deviation in the upper horizons (0–20 cm) ranged from 177 (horizon A1 of AE) to 394 (horizon A1 of TH1) and can be explained by seasonally changes in aeration (Fiedler, 2000). The horizons with permanent submerged conditions showed smaller standard deviations (e.g., horizon 2AB of FH, standard deviation = 145) and symmetric distributions of EH values that are demonstrated by the coincidence of the median and mean. The EH data confirm the postulated gleyzation in that oxic conditions were located above anoxic conditions (steep upward EH gradient at the pedon scale).

The linear correlation between the percentage of time with water saturation and the percentage of time with EH < 170 mV indicates that an increasing duration of saturation favors the mobilization of Fe (Fig. 2) . Furthermore, it can be concluded from these data that the Fe(III)/Fe(II) system was likely the dominant redox buffer in the investigated soils.



View larger version (27K):
[in this window]
[in a new window]
  		Fig. 2. Relationship between the percentage of time with EH < 170 mV (mean values of all electrodes from each horizon) and the percentage of time with water saturation of the investigated horizons (i.e., percentage of time with ≥60% of the horizon thickness situated below the water level).

 
Distribution of Pedogenic Manganese and Iron Oxides
Soil Profile Depth Functions
The total Fe and Mn contents of the soils ranged from 36 to 50 g Fet kg–1 and 0.9 to 3.2 g Mnt kg–1 in the AH soil; values of TH2 ranged from 24 to 40 g Fet kg–1 and 0.2 to 0.6 g Mnt kg–1 (Table 2). Total Fe was lowest in the most poorly drained soils (ME, FH, and TH2), indicating loss of Mn and Fe to local groundwater flow. Similar trends were found in other studies (Schwertmann and Fanning, 1976). Generally, a parallel trend of Fet and Fed (crystalline oxides) occurred. According to Blume and Schwertmann (1969) Fed preferentially accumulates in well-aerated horizons (see Bg in AE with 17.9 g kg–1) and in horizons closest to the mean GWL (see Bg1 in AH with 18.7 g kg–1) (Table 2).

Manganese is mobilized under less reducing conditions than Fe (Gilkes and McKenzie, 1988; McDaniel and Buol, 1991). Thus, higher amounts of Mn in the well-drained soils and progressively less in the wetter soils were found. Continuously, high Mnd contents were detected in the aerobic soils (maximum in the horizon Bg1 of AH, 3.2 g Mn kg–1). Horizons with high accumulation of Mn and Fe were characterized by highly variable redox conditions.

Prevailing reducing conditions in the TH1, FH, and TH2 soils were reflected by minimal Mnd values, which ranged in the mineral horizons between 0.11 and 0.48 g kg–1 (Table 2).

The AE and the AH soils had nearly constant Fed/clay ratios (approximately 0.032) above the median water level. In these landscapes, all soils are Inceptisols and clay translocation is minimal. Comparatively lower Fed/clay (approximately 0.018) ratios were observed in the strongly reduced FH and TH2.

The presence of organic matter retards the crystallization of mineral surfaces (Schwertmann, 1966). Accordingly, the Feo increased with increasing contents of organic matter and depth (Table 2). Well-drained conditions in the AE and the AH soils favored the crystalline Fe oxides which coincided with low Feo/Fed ratios (Table 2). These ranged in the AE and AH soils from 0.49 to 0.52 in topsoil and from 0.05 to 0.4 of subsoil horizons. The trend for greater ratios suggests that the noncrystalline forms of Fe oxides predominate in water-saturated soils. Excluding the humus-rich horizons (Corg > 120 g kg–1) the highest Feo/Fed ratios were observed in the FH (approximately 0.7) followed by TH2 (approximately 0.5). Additionally, the high contents of C prolonged reducing conditions, which we interpreted as a lesser degree of crystallinity that results from less aging of Fe oxides compared to well-aerated conditions.

In general, Fe and Mn were accumulated in the A and O horizons, most probably due to cycling of these elements by vegetation (bioaccumulation) and upward movement, as discussed below. Pyrophosphate extracted a maximum of 70% Feo (horizon AB of FH), indicating that most of the Fe was organically bound (data not shown).

Distribution of pedogenic oxides is typical for gleyzation processes (Schlichting, 1973; Blume, 1973, 1988) and in agreement with the observed gradients in water potential and EH. Under anaerobic conditions both Mn and Fe compounds become reduced. Their solubility after reduction is enhanced by orders of magnitude, which is a factor that increases their mobility within and among soil horizons (Vepraskas, 2000). Upward movement of water from the water table into the more aerobic horizons above leads to the enrichment of Fe and Mn, partially as concentrations. Horizons with a permanent high water table show depletions of Mn and Fe in contrast to the overlying horizons. The intensity of Mn and Fe redistribution was different between the investigated soils. The most useful and suitable pedochemical indicators for identification of gleyzation processes in our study were the Feo and Fed distributions and their ratios.

Mass Balances
The relationship between reducing conditions and Mn and Fe losses per horizon is illustrated in Fig. 3 . The reducing condition used for Mn oxides was % of time with an EH values <450 mV (Fig. 3a), and for Fe was percentage of time with an EH value < 170 mV (Fig. 3c). The figure includes only horizons with losses, because gains are related to catchment conditions and ground water discharge.



View larger version (31K):
[in this window]
[in a new window]
  		Fig. 3. Relationship between relative element losses and redox conditions (only horizons ≤0% of original mass) on horizon scale: (a) relative Mn losses (percentage of original mass) and percentage of time with EH < 450 mV, (b) relative Mn losses (percentage of original mass) and percentage of time with EH < 450 mV, (c) relative Fe losses and percentage time with EH < 170 mV.

 
No relationship between reducing conditions (EH < 450 mV, <170 mV) and Mn was observed in this study (Fig. 3a,b). In contrast, Fe losses increased with the increasing duration of EH < 170 mV as can be seen in Fig. 3c. Although the relationship is rather weak (r2 = 0.62, p < 0.0001, n = 31), one can clearly differentiate between horizons showing reducing conditions <40% of time (smallest losses) and those with >90% (greatest losses). Topsoil horizons showed a greater variation than subsoil horizons. We speculate that the higher variation is due to the differences in the formation of organometal complexes, which have subsequently altered the Fe solubility.

At pedon scale the depth with reducing conditions decreased from 85 to 10 cm along our wetness gradient (Fig. 4a) . Instead, Mn does not show a trend (Fig. 4b). The Fe losses in Fig. 4c are high in three of the wettest soils (ME, FH, TH2). The trend for Fe, however, is not continuous. Instead, thresholds seem to control the losses. No Mn and Fe losses can be detected when prolonged reducing conditions occur at soil depths >50 cm. Whereas losses of both elements occur as measured and calculated under prolonged reducing conditions at soil depths <10 cm. The existing fluctuation of redox conditions in the intermediate profiles ME and TH1 are not likely to be in equilibrium with the solid phases of Mn and Fe. Although time scales are very different (mass balances 104 yr, EH measurements 1 yr), recent conditions in soil solution correspond to long-term solid phase at least for both extremes of redox conditions.



View larger version (57K):
[in this window]
[in a new window]
  		Fig. 4. (a) Depth at which EH < 170 mV continues for >60% of a year (cumulative), (b) relative Mn mass balances, and (c) relative Fe mass balances. (Aeric Endoaquept [AE], Aeric Humaquept [AH], Mollic Endoaquept [ME], Typic Humaquept Pond Margin A [TH1], Fluvaquentic Humaquept [FH], Typic Humaquept Pond Margin [TH2], hatched areas equal ±25% of original mass, only values above or below that range are interpretable as gains or losses).

 
Redoximorphic Features
In general, the results of poor drainage are anaerobic conditions that turn soils from brown to gray (Rowell, 1981). Ferric iron is a strong coloring agent but while ferrous iron is not (Thompson and Bell, 1998). Consequently, soil color should be a suitable indicator of Fe redistribution and redox conditions. To verify these linkages, EH measurements in combination with pedogenic parameters and morphological features are necessary. Extrapolation of pedon results to local or even regional areas appear possible. For instance, the state of the EH regime could be predicted on the basis of soil morphology.

In the following discussion of our study, matrix colors, water levels and redox regimes of all horizons were linked. The median GWL separated the profiles into aerobic and anaerobic zones. Below the median GWL, a clear color change in favor of gray was observed. For instance, the color of FH soil changed from 2.5YR 4/2 to 5GY 3/1 (Table 2). These gray horizons were continually reduced. In agreement with other studies (Franzmeier et al., 1983; Pickering and Veneman, 1984) permanently saturated horizons (redox depletion) are dominated by chroma of 1. Matrix color of chroma ≥5 was found only in the well-drained AE soil. According to Dobos et al. (1990) the most notable gray colors were detected under thick organic horizons (e.g., in FH, TH2 soils). The dark index values (PDI, Thompson and Bell, 1996) of the investigated O and A horizons did not show a regular pattern (data not shown). The relationship between water/redox condition and soil color has been described using different models (Simonson and Boersma, 1972; Megonigal et al., 1993; Blavet et al., 2000). In our study, the soil color was summarized by the soil color index value (C2h), exclusively defined for B horizons (Evans and Franzmeier, 1988). It showed a non-linear relationship to time of water saturation (Fig. 5a) . To verify the hypothesis that soil color is a useful indicator for EH regime and Fe redistribution, we plotted the percentage of time with EH < 170 mV against the color index value (Fig. 5b). The C2h values proved to have a sigmoidal relationship with saturated as well as reducing conditions. These relationships correspond to the Boltzmann fit model:

where x is the C2h value, y is the percentage of time with water saturation, and A1 and A2 are the asymptotic values of the sigmoidal curve to fit. The color index changed between 6 and 12 with time of reducing conditions and 7 to 12 with time of water saturation. Within this sensitive range the soil color can be used for the deduction of water/redox regimes. Below and above these values the time of saturated/reducing conditions remains nearly constant despite changes in C2h. Soil horizons with C2h below 6 were continually reduced (≥80% water saturated), and horizons with C2h above 12 were continually oxidized (≤10% water saturated). Thus, the soil color index is an adequate and practicable tool to gradually differentiate between a wide variation of wetland soils.



View larger version (44K):
[in this window]
[in a new window]
  		Fig. 5. Relationship between soil color index (C2h) of B horizons and (a) the percentage of time with water saturation, and (b) the percentage of time with EH < 170 mV. Shaded areas indicated the sensitive range, within the C2h is an adequate tool to draw gradual distinctions of wetland soils.

 
Another feature of wetland soils is the reduced oxidation of organic matter and the related accumulation of organic matter (Rowell, 1981; Reuter and Bell, 2003). Thus, soil C stock increased with increasing reducing conditions. Carbon stock was highest in the FH2 soil (24.6 kg m–2 1 m–1 depth) and lowest in the AH soil (12.4 kg m–2 1 m–1 depth).


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
It would be desirable for wetland soil studies to estimate redox status and its effect on element translocations based on diagnostic morphological features or simple characterization analysis. To do this, however, different time scales have to be bridged. In our investigation, we regarded static (e.g., Fe distribution, color) and dynamic (e.g., change of water saturation or EH) soil parameters, representing time scales between 1 and 104 yr. We adjusted these scales by integrating the hourly or weekly readings of the dynamic parameters (EH, GWL) over 1 yr and related them to static properties. Previous studies (Fiedler and Sommer, 2000; Sommer and Fiedler, 2002) indicated that the application of the time below certain EH thresholds is more suitable for estimation of redox dynamics in redoximorphic soils than seasonal EH trends. Therefore, we used the theoretical onset of Fe-oxide reduction as a proxy of EH inducted element mobilization. Thus, we calculated the percentage of time with EH < 170mV.

Although the scales still differ by many orders of magnitudes, the recent conditions of the soil solutions fit very well to long-term soil solid phase evolution. Our findings show that the percentage of time with EH < 170 mV is linearly related to the percentage of time with water saturation. High percentage of EH values below the threshold favored Fe2+ mobilization and Fe losses. In this context two different distribution patterns were found. The first represents the element redistribution within the pedon (well aerated AE and AH soils) and the other pattern shows the translocation processes within landscapes, reflected by soils with strong reducing conditions and defined by element losses (FH and TH2 soil). In general, our findings are in agreement with those of Chadwick and Chorover (2001) who claimed soil redox status as a key threshold variable affecting the trajectory of pedogenesis.

The hypothesis, that the soil color is a mirror image of water/redox condition resulting from Fe mobilization can be accepted for the soils of this study. Soil morphology summarized by soil color index, indicated that there was a near linear linking to the percentage of time with (i) water saturation and (ii) EH < 170mV. Up to now, the soil color was mainly used to identify the presence of saturated conditions, but not for the quantification of time with water influence, reducing conditions or element mobility. Our findings permitted an exact differentiation of soil horizons according to their duration within the mentioned conditions. However, horizons with C2h values outside of the sensitive color range cannot adequately be discriminated. It is known that either horizons with almost permanent water logging or short episodic water saturation show such values.

We concluded that, C2h may be used as a proxy to estimate the intensity of water/redox conditions, but it is restricted to redoximorphic soils in equilibrium with actual environmental conditions. For a generalization, it is necessary to validate this linkage regionally. Therefore, it is important to extend this investigation to other hydromorphic soils, landscape positions, and humid regions.

Received for publication September 6, 2002.


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





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 ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fiedler, S.
Right arrow Articles by Sommer, M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Fiedler, S.
Right arrow Articles by Sommer, M.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Fiedler, S.
Right arrow Articles by Sommer, M.
Related Collections
Right arrow Redox Processes
Right arrow Wetland Soils
Right arrow Watershed and Landscape Processes

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