Soil Science Society of America Journal 64:1904-1910 (2000)
© 2000 Soil Science Society of America
DIVISION S-10-WETLAND SOILS
Propensity of Soils to Develop Redoximorphic Color Changes
M.C. Rabenhorsta and
S. Parikhb
a Natural Resource Sciences, H.J. Paterson Hall, Univ. of Maryland, College Park, MD 20742-5821 USA
b Dep. of Agronomy, Pennsylvania State Univ., University Park, PA 16802 USA
mr1{at}umail.umd.edu
 |
ABSTRACT
|
|---|
Most soils with seasonally high water tables exhibit redoximorphic features, such as Fe depletions and concentrations. Some soils formed from red parent materials apparently are less typical in their behavior and have in this sense been described as problematic. Regulatory guidelines have mentioned such problematic red soils, and the Field Indicators of Hydric Soils in the United States attempts to address the issue by using test indicator TF2. This study was undertaken in an attempt to develop an approach for quantifying the inherent tendency of soils to form redoximorphic features under reducing conditions. Thirty soils from diverse parent materials in the Mid-Atlantic region was studied, including 13 thought to be problematic soils formed from red parent materials. Samples were treated using a citrate-buffer and Na dithionite at room temperature (25°C) for periods of time ranging from 0.5 to 72 h, and also at 80°C for 4 h. Colors were measured using a digital colorimeter to document changes in Munsell hue, value, and chroma. By comparing initial soil colors with those measured under the treatments of 25°C for 1 h and at 80°C for 4 h, a color change propensity index (CCPI) was developed that effectively discriminated between soils which were and were not thought to be problematic. The utility of the CCPI was tested using 14 additional soil samples, six of which came from outside the Mid-Atlantic region. Recommendations are made to differentiate between problematic and nonproblematic soils based on values of CCPI.
Abbreviations: CCPI, a color change propensity index • CI, chroma index • HI, hue index
|
INTRODUCTION
|
|---|
COLOR is perhaps the most obvious property that a soil possesses and can reveal much about its character. The specific soil color observed is determined by the color of its various components, particularly the color of the silicate (or other primary structural) mineral grains and the quantity and nature of the various natural pigments, which may occur with and coat the mineral grains. The two most common pigments are organic materials and Fe oxides or oxyhydroxides (hereafter referred to as oxides). In horizons without appreciable quantities of pigmenting agents, such as some E horizons or C horizons, soil color is dominated by the color of the mineral grains themselves, which usually are various shades of gray or white. In the upper horizons, which are more strongly affected by biological activity, the pigmenting effects of organic matter are most strongly seen. The impact of the quantity and the nature of the organic matter, and the effects of interactions between soil texture and organic matter on soil color have been demonstrated by Schulze et al. (1993).
In subsoil horizons where the quantity of organic matter is limited, coatings of Fe oxides have a more pronounced effect on soil color. Not only does the quantity of the oxides affect the color, but the mineralogy, the crystal size, and isomorphic substitution of other metals within the oxide crystal structure can also dramatically impact soil color (Schwertmann, 1993). Manganese oxides can be strong soil pigments causing the soil to be dark brown or black, but they are far less abundant than Fe oxides and usually have minimal effects on soil color unless concentrated as coatings, nodules, or concretions.
|
Colors of Redoximorphic Features
|
|---|
It is generally known that when soils become saturated with water, anaerobic conditions can develop if an oxidizable C source is present and the temperatures are warm enough for microorganisms to be active. If anaerobic conditions persist, soils may become sufficiently reduced that ferric (III) Fe in oxide minerals begins to be reduced to the ferrous (II) state. The particular redox potential at which this transformation occurs (or is thermodynamically predicted) is dependent on the soil pH as well as additional factors. While the solubility of ferric Fe oxides is extremely low, ferrous Fe is quite soluble. Thus, the reduction of Fe oxides in soils permits the translocation of Fe in the aqueous phase.
Seasonal variations in rainfall and evapotranspiration lead to fluctuations in water table levels giving rise to alternating conditions of reduction and oxidation with respect to Fe oxides. Mobilization of ferrous Fe during periods of reduction can lead to segregation of the Fe oxides and the formation of some zones that are depleted and others that are enriched in Fe. The processes leading to the segregation of Fe oxides and the factors that determine where depleted and concentrated zones occur within the soil have been described by Vepraskas (1992).
The colors of depleted zones reflect the colors of the uncoated mineral grains and are typically gray or white (low chroma 
2). In more extreme cases, the Fe oxides are not only depleted from small zones but may be largely removed from the soil, leading to what is called a depleted matrix or gleyed matrix. Conversely, the zones where oxides have been concentrated typically appear redder or browner than the matrix soil color, and have been termed masses, coatings or pore linings, depending on where they appear with regard to natural surfaces within the soil (Vepraskas, 1992). These redoximorphic features (formerly called wetness or drainage mottles) constitute distinctive color patterns in the soil that have been used by soil scientists for decades to interpret soil drainage and hydrological conditions (Veneman et al., 1998).
|
Anomalous Soil Hydromorphology
|
|---|
Over the years, soil scientists working in the field have encountered circumstances where soils possessed seasonally high water tables, but did not exhibit the types of redoximorphic features typical of wet soils (Niroomand and Tedrow, 1990; C.E. Robinette, 1992, personal communication; Huddleston et al., 1997). In some cases, these phenomena have been attributed to oxyaquic conditions, where the soils were wet but for some reason did not develop reducing conditions. More recently, instances have been documented that indicate that the soil parent material may play an important role in whether or not a wet soil exhibits redoximorphic features (Elless et al., 1996). Among soils most noted to be problematic in this regard are some of those derived from red parent materials. While not all soils with red parent materials show this difficulty, the phenomenon is sufficiently broad that the difficulty has been addressed in Field Indicators of Hydric Soils in the United States (USDA-NRCS, 1998) and previously by the Federal Interagency Committee for Wetland Delineation (1989).
|
Possible Explanations for Anomalous Soil Hydromorphology
|
|---|
Several possible explanations may be offered to account for the reticence of certain soils with seasonally high water tables to develop low chroma redox depletions. These explanations fall into two general classes: (i) those related to chemical conditions in the soil environment and (ii) those related to inherent properties of the soil itself. Regarding the soil chemical environment, some soils may be saturated but not reducing. The term oxyaquic has been used in the Keys to Soil Taxonomy (Soil Survey Staff, 1998) to describe conditions where the soil is wet but not reducing with respect to Fe. Oxyaquic conditions may occur where (i) soils contain low quantities of decomposable organic matter, (ii) soil temperatures are so low that biologic activity is limited, or (iii) where rapid flow of groundwater may prevent depletion of the O2 levels (Vepraskas and Sprecher, 1997). In these cases, the poor drainage of the soils will not be reflected in the morphology, because soil redox changes are insufficient to induce the formation of redoximorphic features.
There are two possible explanations related to properties inherent in the soils themselves. The first may be that the colors of uncoated mineral grains themselves are brownish in color showing a high chroma, rather than low chroma color. This could prevent soils from appearing gray or exhibiting low chroma colors even when Fe oxide coatings have been removed. High chroma or brown-colored grains are not particularly common, but could be caused by the occlusion of Fe oxides within the interiors of mineral grains such as quartz or small lithic fragments of shales or siltstones, where they are isolated from the reducing conditions in the soil, and thus retain higher chroma colors. A second reason that some soils do not change color under reducing conditions may be that certain Fe oxide minerals can be more resistant to reduction than others. The most common Fe oxide minerals in soils and sediments are hematite (
-Fe2O3), goethite (-FeOOH), ferrihydrite (5 Fe2O3·9H2O), and lepidocrocite (
-FeOOH). Based on thermodynamics, the solubility products of goethite and hematite are nearly the same (pK in the range of 42.2–44.0) and significantly (orders of magnitude) greater than those for lepidocrocite
and ferrihydrite
(Langmuir and Whittmore, 1971; Schwertmann and Taylor, 1989). As one would expect, the most stable phases (hematite and goethite) are the most common Fe oxide minerals in soils. However, it has been shown that Al substitution in the structures of both goethite and hematite can significantly lessen their propensity for being reduced and solubilized (Fey, 1983). Because goethite can accommodate greater amounts of Al (33 mol%) than hematite (16 mol%) (Schwertmann and Taylor, 1989), it has been postulated that naturally occurring goethites should be less susceptible to reduction than hematites. This argument has been used to explain pedogenic yellowing in soils derived from red sediments in Brazil where both hematite and goethite coexist in the parent material (Macedo and Bryant, 1987, 1989; Bryant and Macedo, 1990). In contrast, Fe oxides in soils that show resistance to reduction in the Mid-Atlantic region seem to have dominantly hematitic mineralogy (Elless and Rabenhorst, 1994; Elless et al., 1996; Niroomand and Tedrow, 1990). Elless (1992) has suggested that this may in part be due to Al substitution in the hematite.
|
Interpretational Difficulties
|
|---|
As mentioned above, there are some soils, especially some of those with red parent materials, that are less apt than others to show redoximorphic color patterns associated with saturated soil conditions. In some cases, this phenomenon may be the result of properties inherent to the soil. Possible explanations may be related to the Fe oxide mineralogy and to Al substitution for Fe in the structures, but are complex and thus are not easily predicted. Both soil classification and soil interpretations, such as siting of waste disposal systems or hydric soil delineations, are often tied to the presence and location of redoximorphic color patterns. The test hydric soil indicator TF2 (Red Parent Material) (USDA-NRCS, 1998) was set up to accommodate problematic red parent material soils, but so long as the Munsell hue is 7.5RYR or redder, it does not differentiate between those red soils that are or are not problematic. Therefore, it would be beneficial to know how soils compare with each other regarding their propensity or resistance to develop redoximorphic color changes, and to identify a way to quantify these differences in order to characterize a soil's propensity for or resistence to color change. The objectives of this project were (i) to develop an approach for evaluating the relative inherent propensities of soils to exhibit redox induced color changes and (ii) to use this approach to categorize soils according to their relative inherent propensity to develop redox induced color changes.
|
Materials and methods
|
|---|
Thirty soil samples were used in the initial experiment and were selected from the archived collection at the University of Maryland (Table 1)
. Thirteen samples were selected as problematic soils, which from earlier work were thought to be resistant to color change. These were soils formed from either red Triassic shales of the Piedmont physiographic province (Elless and Rabenhorst, 1994; Elless et al., 1996) or from red Paleozoic shales in the Ridge and Valley province. Seventeen samples were selected as nonproblematic soils that were thought not to be resistant to color change. These soils were collected from across the state, including all physiographic provinces. Samples were air dried and crushed to pass a 2-mm sieve.
In order to document subtle and gradual changes in soil color under reducing conditions, samples were reacted with sodium dithionite in a citrate buffer solution at room temperature (25°C) for varying periods of time, and colors were measured using a digital colorimeter (CR-300, Minolta, Osaka, Japan). Five grams of air-dry soil were mixed with 5 g of sodium dithionite, to which was added 70 mL of citrate buffer solution and then vigorously stirred (Kittrick and Hope, 1963). The time intervals for observation were 0, 0.5, 1, 2, 4, 8, 24, 48, and 72 h (Table 2) . Samples were also extracted at 80°C using a hot water bath. These samples were extracted for 2 h, centrifuged, decanted, and then treated a second time with 5 g of sodium dithionite and 70 mL of citrate buffer for another 2 h at 80°C. All samples were stirred periodically during the extractions to ensure that all sodium dithionite was dissolved (Table 2), and all treatments were run in duplicate. For graphing purposes, the 4-h treatment (two 2-h treatments) at 80°C was treated as equivalent to 400 h at room temperature (following Van Hoff's rule that chemical reactions increase by a rate of two- to threefold for each 10°C increase in temperature).
View this table:
[in this window]
[in a new window]
|
Table 2 Conditions of samples in each of the 10 treatments prior to soil color measurements indicating time interval, stirring frequency, and the temperature
|
|
Extracted soil samples were centrifuged at 1500 rpm for 5 min and decanted. The supernatant was discarded except for samples treated at 80°C, which were saved for later Fe analysis. The soil materials were homogenized and spread on a filter paper until excess liquid was absorbed. Measurements were taken while soil was still moist, 
90 s after application to the filter paper. Measurements were made on four different portions of each sample, and each measurement consisted of three readings averaged by the instrument. Colors were recorded using Munsell notation to the nearest 0.1 unit of hue, value, and chroma. The Munsell notation for hue is typically reported using a combination of numbers and letters. In order to place Munsell hue on a numerical scale, a system was employed based on the Munsell hue circle, which places hues on a scale ranging from
(Post et al., 1993; Chamberlin and Chamberlin, 1980). The supernatant from each sample extracted at 80°C was transferred to a 500-mL volumetric flask and brought to volume with distilled water. The quantity of Fe extracted was determined by atomic adsorption spectrophotometry. Following analysis of the first 30 samples, the data were analyzed and a strategy was developed for differentiating between soils that did or did not show resistance to color change. The color index developed from the developmental phase was then tested using 14 additional soils.
|
Results and discussion
|
|---|
Index Development
The color components (hue, value, and chroma) recorded after various periods of extraction for four samples representative of each of the problematic and nonproblematic groups are presented in Fig. 1, 2, and 3
. The patterns of change for the two groups appeared to be fairly distinct for both hue and chroma, but Munsell values were not distinguishable. The final hue (4 h at 80°C) was generally lower (redder) for the problematic soils, suggesting that colors of the uncoated mineral grains might be a contributing factor to the phenomenon. However, there was no difference in final chroma between the two groups. Differences between the two groups became even more evident when the magnitude of the change in color component from the initial condition was plotted against time (Fig. 4, 5, and 6)
. Nonproblematic soils showed an immediate and progressive change in hue, while the problematic soils showed essentially no change in hue until 8 h. Although the problematic soils showed a small change in chroma during the first 4 h (generally <1 unit), the nonproblematic soils showed a much greater change in chroma during the same period (2–4 units). Changes in value were not different for the two groups and were quite small, which is what one might expect since changes in Munsell value are more likely to occur as a result of extracting organic matter rather than Fe oxides (Schultz et al., 1993).
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 1 Munsell hue of representative problematic (P) and nonproblematic (N) samples measured over time during ongoing dithionite-citrate-bicarbonate treatments
|
|
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 2 Munsell value of representative problematic (P) and nonproblematic (N) samples measured over time during ongoing dithionite-citrate-bicarbonate treatments
|
|
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 3 Munsell chroma of representative problematic (P) and nonproblematic (N) samples measured over time during ongoing dithionite-citrate-bicarbonate treatments
|
|
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 4 Change in Munsell hue (difference from initial hue) of representative problematic (P) and nonproblematic (N) samples measured over time during ongoing dithionite-citrate-bicarbonate treatments
|
|
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 5 Change in Munsell value (difference from initial value) of representative problematic (P) and nonproblematic (N) samples measured over time during ongoing dithionite-citrate-bicarbonate treatments
|
|
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 6 Change in Munsell chroma (difference from initial chroma) of representative problematic (P) and nonproblematic (N) samples measured over time during ongoing dithionite-citrate-bicarbonate treatments
|
|
One might postulate that differences between the two groups are a function of the quantity of Fe oxides in the samples. However, when extractable Fe contents for the two groups are compared, there is no significant difference (Table 3)
. Thus, Fe content per se does not appear to affect the propensity of a soil to change color.
Because the most striking differences between the two groups were in changes in hue and chroma, we attempted to incorporate these data into a color index that would indicate the propensity of a soil to develop redox-induced color changes. Because most nonproblematic soils showed color changes within the first 0.5 to 1 h, while most problematic soils showed little change during this period, we used comparisons in color change after 1 h and after the maximum treatment. Indices using chroma and hue only were derived as shown below. Samples from the two groups could be partially separated using the chroma index (CI) (Eq. 1) and were more effectively separated using the hue index (HI) (Eq. 2).
 | (1) |
 | (2) |
Graphing samples according to both indices simultaneously suggested that even more effective separation could be accomplished by combining the two indices (Fig. 7)
. A CCPI was therefore proposed that incorporates both the HI and the CI (Eq. [3]). Because the magnitude of the HI is approximately four times larger than the CI, the CI is multiplied by four before being added to the HI so that the two components would have roughly the same weighting. By using the CCPI, soils from the two test groups were effectively separated, with the problematic soils having an CCPI of <25 and the nonproblematic soils having an CCPI of >40 (Fig. 8)
.
 | (3) |
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 7 Differentiation of problematic and nonproblematic soils using combination of hue index (HI) and chroma index (CI)
|
|
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 8 Differentiation of problematic and nonproblematic soils using Master Color Index (MCI) equal to hue index (HI) plus four time the chroma index (CI)
|
|
Testing of Color Index
Samples that had not been included in the initial index development were selected to evaluate the utility of the CCPI for differentiating between those soils that did or did not show an inherent resistance to develop redoximorphic color changes. The test samples used are described in Table 4
. Four of these were problem soils from Maryland, four were nonproblematic soils from Maryland, and six samples were provided from other states where the soils were thought to be problem red parent material soils. These soils were treated using the same procedure as the initial experiment except color measurements were taken only at 0 and 1 h at 25°C and after the 4-h treatment at 80°C.
View this table:
[in this window]
[in a new window]
|
Table 4 Samples selected to test the Color Change Propensity Index (CCPI) for use in discriminating between problematic and nonproblematic soils
|
|
The soils from Maryland behaved largely as expected, with nonproblematic soils having CCPIs well above 40 and the problematic soils having CCPIs near or below 25 (Fig. 9)
. Five of the six samples from out of state had CCPIs below 28, with the other sample having an CCPI of 32. Thus, while several of these samples had slightly higher CCPI's than the samples from Maryland, they were still well below the value of 40 suggested as the lower limit of nonproblematic soils. This suggests that the range of CCPIs for problematic red soils should perhaps be extended up to 30, leaving the range from 30 to 40 as the transitional or questionable zone.
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 9 Index values calculated for test samples in assessing the value of the CCPI to differentiate between problematic and nonproblematic soils
|
|
|
Conclusions
|
|---|
By comparing the changes in colors observed for different soil samples under reducing conditions, it was demonstrated that some soils change color relatively easily, while others are inherently more reticent to develop yellower or greener hues and lower chroma colors. Thus some soils with red colors can easily form low chroma redoximorphic features under reducing conditions, while others will not. We believe that application of test hydric soil indicator TF2 Red Parent Material (or any indicator to be developed that is similar to it) should be restricted for use in those soils that inherently demonstrate reticence to develop redoximorphic features. A relatively simple color change index has been developed from color measurements made using a digital colorimeter on soils treated using dithionite-citrate-bicarbonate at room temperature after 1 h and at 80°C after 4 h. The CCPI is derived from a HI based on the percentage of the maximum change in hue that occurs in 1 h and a CI based on the change in measured chroma after 1 h and after 4 h at 80°C. Based primarily on a sample set from the Mid-Atlantic region supplemented by a few samples from Texas, Kansas, and Minnesota, we offer a preliminary recommendation that soils with a CCPI of 30 or less be recognized as problematic soils and that soils with a CCPI of 40 or more be considered to be nonproblematic. Soils with CCPIs ranging between 30 and 40 should receive additional scrutiny as they represent intermediate conditions. This approach should be tested more extensively using samples representing an even greater geographic distribution. Additional investigations are warranted in order to determine the factors that cause particular soils to resist the development of redoximorphic features.Federal Interagency Committee for Wetland Delineation 1989; 1998
|
ACKNOWLEDGMENTS
|
|---|
The authors gratefully acknowledge the assistance of Mickey Ransom at Kansas State University, Larry Wilding and Philip Owens at Texas A&M University, and Marc Diers at the USDA-NRCS in Duluth, MN, for providing soil samples formed from red parent materials to be tested during this study.
Received for publication August 20, 1999.
|
REFERENCES
|
|---|
- Bryant R.B., Macedo J. Differential chemoreductive dissolution of Fe oxides in a Brazilian Oxisol. Soil Sci. Soc. Am. J. 1990;54:819-821.[Abstract/Free Full Text]
- Chamberlin G.J., Chamberlin D.G. Colour—Its measurement, computation and application. Philadelphia, PA: Heyden and Sons, 1980.
- Elless M.P. Morphology, mineralogy, and hydrology, of soils in the Triassic Culpepper basin of Maryland. College Park, MD: Univ. Maryland, 1992 Ph.D. Diss. (Diss. Abstr. 92-34559).
- Elless M.P., Rabenhorst M.C., James B.R. Redoximorphic features in soils of the Triassic Culpepper Basin. Soil Sci. 1996;161:58-69.
- Elless M.P., Rabenhorst M.C. Hematite in the shales of the Triassic Culpeper Basin of Maryland. Soil Sci. 1994;158:150-154.
- Federal Interagency Committee for Wetland Delineation. 1989. Federal manual for identifying and delineating jurisdictional wetlands. Coop. Technical Publ. U.S. Army Corps of Engineers, USEPA, U.S. Fish and Wildlife Service, and USDA Soil Conservation Service, Washington, DC.
- Fey M.V. Hypothesis for the pedogenic yellowing of red soil materials. Tech. Commun. S. Afr. Dep. Agric. Fish. 1983;180:130-136.
- Huddleston, J.H., K. Verble, D. Clausnitzer, and W. Austin. 1997. Field indicators of hydric soils in the Pacific Northwest—How well do they work? p. 74. In Wetlands heritage and stewardship. Abstracts of the Soc. of Wetland Scientists 18th Annual Meeting, Bozeman, MT. 1–6 June 1997.
- Kittrick J.A., Hope E.W. A procedure for the particle size separation of soils for XRD analysis. Soil Sci. 1963;96:319-323.
- Langmuir D., Whittmore D.O. Variations in the stability of precipitated ferric oxyhydroxides. In R.F. Gould (ed.) Nonequilibrium systems in natural water chemistry. Adv. Chem. Ser. 1971;106:209-234.
- Macedo J., Bryant R.B. Morphology, mineralogy, and genesis of a hydrosequence of oxisols in Brazil. Soil Sci. Soc. Am. J. 1987;51:690-698.[Abstract/Free Full Text]
- Macedo J., Bryant R.B. Preferential microbial reduction of hematite over goethite in a Brazilian oxisol. Soil Sci. Soc. Am. J. 1989;53:1114-1118.[Abstract/Free Full Text]
- Niroomand G., Tedrow J.C.F. Importance of the parent material on gley formation. Pochvovedenie 1990;8:39-42.
- Post D.F., Levine S.J., Bryant R.B., Mays M.D., Batchily A.K., Excadafal R., Huete A.R. Correlations between field and laboratory measurements of soil color. In: Bigham J.M., Ciolkosz E.J., eds. Soil color. Madison, WI: SSSA, 1993:35-50 SSSA Spec. Publ. 31..
- Schulze D.G., Nagel J.L., Van Scoyoc G.E., Henderson T.L., Baumgardner M.F., Stott D.E. Significance of organic matter in determining soil colors. In: Bigham J.M., Ciolkosz E.J., eds. Soil color. Madison, WI: SSSA, 1993:71-90 SSSA Spec. Pub. 31..
- Schwertmann U., Taylor R.M. Iron oxides. In: Dixon J.B., Weed S.B., eds. Minerals in soil environments, 2nd ed Madison, WI: SSSA, 1989:379-438.
- Schwertmann U. Relations between iron oxides, soil color, and soil formation. In: Bigham J.M., Ciolkosz E.J., eds. Soil color. Madison, WI: SSSA, 1993:51-69 SSSA Spec. Publ. 31..
- Soil Survey Staff. Keys to soil taxonomy, 8th ed Washington, DC: USDA-Soil Conservation Service, 1998.
- USDA-NRCS. 1998. Field indicators of hydric soils in the United States. Ver. 4.0. G.W. Hurt et al. (ed.) USDA, NRCS, Fort Worth, TX.
- Veneman P.L.M., Lindbo D.L., Spokas L.A. Soil moisture and redoximorphic features: A historical perspective. In: Rabenhorst M.C., et al. , ed. Quantifying soil hydromorphology. Madison, WI: SSSA, 1998:1-23 SSSA Spec. Publ. 54..
- Vepraskas, M.J. 1992. Redoximorphic features for identifying aquic conditions. North Carolina Agric. Res. Ser. Technical Bull. 301. Raleigh, NC.
- Vepraskas M.J., Sprecher S.W. Overview of aquic conditions and hydric soils. In: Vepraskas M.J., Sprecher S.W., eds. Aquic conditions and hydric soil: The problem soils. Madison, WI: SSSA, 1997:1-22 SSSA Spec. Publ. 50..
TOP
ABSTRACT
INTRODUCTION
