| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Dep. of Geography, 314 Natural Science Building, Michigan State Univ., East Lansing, MI 48824-1115
* Corresponding author (weisenbo{at}msu.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: AAO, acid ammonium-oxalate • CD, sodium citrate-dithionite • RDP, related distribution pattern • SEM, scanning electron microscopy
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
In the Great Lakes region, fragipans are found in soils on both acidic and calcareous parent materials. In Michigan, however, fragipan horizons formed in acidic drift are far more common and spatially extensive than those formed in calcareous drift (Soil Survey Division, 2004). Because fragipans on acidic drift are more common and extensive in Michigan and the Great Lakes region, it seems reasonable that they may also be better developed than their calcareous counterparts. Such a question has not been addressed in the fragipan literature. Furthermore, only a few studies have explored the formation of soils with fragipans in Michigan (e.g., Yassoglou and Whiteside, 1960; Bockheim, 2003) despite the wide occurrence of these soils, and no studies have investigated the highly variable fragipan expression here. The purpose of this study was to evaluate the varying degree of fragipan expression in Michigan's soils and assess differences in development among fragipans formed in acidic and calcareous drift. The degree of fragipan expression in selected soils was characterized through morphological, physical, chemical, and microscopic observations and analyses, to elucidate their possible pedogenic pathways.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
Field and Laboratory Methods
After sampling sites were identified, soil pits were dug by a backhoe. Physical setting and profile descriptions were conducted according to the Soil Survey Division Staff (1993) and Schoeneberger et al. (1998). Soil pit faces were first sketched and photographed and then genetic horizons were sampled. Samples were air-dried and sieved to remove coarse fragments (>2 mm). The remaining fine earth fraction was analyzed in the laboratory. Particle size distribution was determined by pipette with prior H2O2 pretreatment to remove organic matter for A, Bhs, and Bs horizons (Sheldrick and Wang, 1993). Sand separates were determined by dry sieving the sand fraction. The clay-free sand content was calculated and used to illustrate changes in the immobile particle fraction by horizon, which has been recommended for detecting lithologic discontinuities (Schaetzl, 1998). A 2:1 water/soil mixture was used to measure pH. Extractions for Fe, Al, and SiO2 were performed on the <2 mm fraction of all genetic horizons using sodium citrate-dithionite [CD; Fed, Ald, (SiO2)d] and acid ammonium-oxalate [AAO; Feo, Alo, (SiO2)o] (Ross and Wang, 1993; Loeppert and Inskeep, 1996). Extracts were analyzed by flame atomic absorption spectrophotometry. Thick eluvial and illuvial parts of transitional horizons were extracted and analyzed separately. We interpreted the extraction data in the following way. Fed, Ald, and (SiO2)d represent noncrystalline and crystalline free oxide forms of Fe (Jackson et al., 1986; Dahlgren, 1994), and both Al and Si to a lesser extent (McKeague et al., 1971). Feo, Alo, and (SiO2)o represent noncrystalline (organically complexed and inorganic) forms of Fe (Farmer et al., 1983; Parfitt and Childs, 1988), and both Al and Si to a lesser extent (McKeague et al., 1971). Core samples were also collected from the soil pit face for bulk density analysis (Blake and Hartage, 1986). Genetic horizons thinner than the core sampler used were sampled in combination with the subjacent horizon. Triplicate bulk density values were calculated for each horizon or horizon pair, on an oven-dry, coarse-fragment-free basis and then averaged.
Microscopic Methods
Undisturbed bulk samples were collected from all protofragipan and fragipan horizons. Thin section samples for soil micromorphology were cut from samples collected from the pit face in 10 x 6.5 x 5 cm metal tins. These samples were impregnated with 3M Scotchcast (epoxy) resin under vacuum, and oven-cured at 40 to 50°C for 3 d. Oversized (38 x 75 mm) thin sections were cut and polished to a thickness of 
30 µm (National Petrographic Service, Inc., Houston, TX). Thin sections were examined using a petrographic microscope in plane- and cross-polarized light, using temporary cover slips in immersion oil to aid viewing. Micromorphological descriptions were made according to Bullock et al. (1985). Digital images (micrographs) of selected micromorphological features (in plane-polarized light) were collected using a digital imaging system (Pixera Professional).
Intact clod samples, used for SEM imaging, were also collected. These were first divided by hand into subsidiary peds and then subdivided along planes of weakness under a stereographic microscope using a dissecting needle. Of the single, primary peds obtained, only a few, minimally disturbed representative samples were chosen for imaging. Each selected ped or subped sample was mounted onto an aluminum SEM stub using epoxy (Flegler et al., 1993). Carbon paste was used as the conducting medium between the top of the sample and the stub. Samples were then carbon and gold coated to reduce charging. The SEM observations and imaging were conducted on a JOEL JSM-6400V SEM (Joel Inc., Boston) equipped with a Soft Imaging System AnalySIS for image capture and analysis.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The upper sequum of the Feldhauser profile has albic and cambic horizons, both with weakly developed color and structure (Fig. 2 ; Table 1). The lower sequum contains a 2E/B and 3E'&Bt horizon whose Bt parts contain illuvial clay, particularly as intergrain bridges or lamellae. When dry (as at the time of sampling), the 2E/B horizon is notably brittle. This horizon exhibits fragic soil properties, but does not meet all of the additional fragipan criteria. Specifically, this horizon fails to meet these fragipan criteria: bodies of oriented clay and clay films as evidence of pedogenesis, 
10 cm between structural units, and virtually no roots in 60% of horizon (Soil Survey Staff, 1999). Thus, we consider the 2E/B horizon a protofragipan due to its weakly expressed fragipan character. Soil structure is weakly developed and most of the horizons have loose to very friable consistence. Although we found no evidence of reduced soil matrices within the Feldhauser solum, the 2E/B horizon exhibited common, faint, finely disseminated iron depletions. We found evidence of faunal- and floralturbation in the form of surficial worm casts and treethrow activity. Rooting appeared to be restricted in the 2E/B horizon.
|
|
The Glennie upper sequum has albic and cambic horizons that have weakly developed color and structure (Fig. 2; Table 1). Both eluvial and illuvial horizons of the lower sequum exhibit strongly expressed fragipan characteristics. The B parts of the (E/B)x horizon and the Btx horizon have argillans and bridges of illuvial clay. The 2Cd horizon is massive, dense, and strongly effervescent. Overlying horizons have moderately well-developed structure. Upper sequum horizons are loose or very friable in moist consistence, whereas lower sequum horizons are firm or very firm. The lower sequum is dense and appeared to present an obstacle for both root growth and vertical water movement. Iron depletions and concentrations were observed in the lower sequum. Redox depletions occur in the form of both pigment and clay losses in the Ex horizon and the E part of the (E/B)x horizon, with colors that distinctly contrast with the parent material matrix. Redox concentrations in the Bt part of the (E/B)x horizon and the Btx horizon appeared in the form of ferriargillans on ped faces and vertical root channels. We found little evidence of pedoturbation. Very few root traces were found in or below the Ex horizon, but a few roots extended to the (E/B)x horizon along planes of weakness.
Based on field observations, key morphological properties of the protofragipan and fragipans tend to be distinctive to either the eluvial or illuvial horizons; transitional fragipan horizons have varying degrees of both. Predominantly eluvial fragipan horizons tend to lack clay coats/films and contain variable amounts of albic materials that eventually form glossic horizons. The eluvial protofragipan and fragipan horizons tend to have a high value ( 5) and low chroma (
3) moist colors, which indicate moderate to strong eluviation. Moist consistence is firm. Tonguing of the eluvial fragipans' albic materials into the underlying (transitional or illuvial) fragipan horizon suggests eluvial fragipan horizons may contain zones of degradation within these bisequal sola. Predominantly illuvial fragipan horizons exhibit clay coats/films and flows on ped faces and vertical root channels. Clay bridging was observable between sand grains. Moist colors range from brown to reddish brown, and moist consistence is very firm. All protofragipan and fragipan horizons, however, exhibit brittle failure, fine vesicular pores, reduced root presence or root restriction, variable structure ranging from weak, thick platy to strong, coarse subangular blocky, and evidence of oxyaquic conditions, suggestive of periodic episaturation. For reasons we do not fully understand, all of the fragipans lacked the prismatic structure that is commonly used, but not required, as a distinguishing criterion. While coarse, bleached prism faces generally do not occur in eluvial fragipans, they are commonly associated with illuvial fragipans (Bryant, 1989). It is possible that the relatively coarse texture of our fragipans (Fig. 3)
as also observed by Yassoglou and Whiteside (1960) or the periodicity and rapidity of profile wetting and drying may have influenced structural development in our soils (Bryant, 1989; Chadwick and Graham, 2000). We are aware, however, of fragipans near the Munising profile that exhibit strong, coarse prisms with bleached faces.
|
Surface textures are loamy (i.e., loamy sand to sandy loam), and textures for deepest horizons are sand or loam (Table 1). In general, most fragipan parent materials are either loamy (Hallmark and Smeck, 1979; Soil Survey Staff, 1999) or texturally heterogeneous (Olson and Hole, 1967). The protofragipan and fragipans studied here are loamy sand, sandy loam, sandy clay loam, or loam in texture (Table 1; Fig. 3). Predominantly illuvial fragipans are finer textured than are their eluvial counterparts. While the protofragipan and fragipans studied tend to have sandier textures (Fig. 3) than fragipans in Pennsylvania (Peterson et al., 1970), other authors have reported similar textures (loamy sand, sandy loam, and loam) for fragipans formed in glacial drift (Yassoglou and Whiteside, 1960; Hallmark and Smeck, 1979; Veneman and Bodine, 1982; Habecker et al., 1990; Miller et al., 1993). Illuvial fragipans with even finer textures have also been reported (DeKimpe et al., 1976; Hallmark and Smeck, 1979; Habecker et al., 1990).
Clay-free sand data were calculated to identify lithologic discontinuities (Table 2; Schaetzl, 1998), which are thought to influence fragipan evolution (Habecker et al., 1990; Van Vliet and Langohr, 1981). Smeck et al. (1989) observed that fragipans in Ohio form in association with weathering discontinuities, which can form near or at lithologic discontinuities. Each profile has at least one lithologic discontinuity close to its protofragipan or fragipan based on trends in clay-free sand and coarse-fragment contents (Table 1).
|
Chemical Properties
In the three profiles, pH values are relatively low in the eluvial portion of each sequum and increase with depth (Table 2). Beneath the fragipan or protofragipan in the lower sequum, pH increases and reaches a maximum in the deepest horizon. Miller et al. (1993) found that the fragipan horizons, on average, had lower pH values than overlying and underlying horizons. Similarly, most protofragipan and fragipan horizons in this study have pH values lower than their overlying and underlying horizons, ranging from 4.8 to 6.8, despite the acidic or calcareous nature of the parent material. These values are similar to other fragipans formed in glacial drift: 4.6 to 6.5 (Yassoglou and Whiteside, 1960; Ransom et al., 1987; Miller et al., 1993).
According to the Soil Survey Staff (1999), pH values for fragipans beneath spodic horizons are often high relative to other fragipans. This trend, however, was not observed. The protofragipan and fragipan in the Feldhauser and Glennie profiles, both Alfisols, have higher pH values than the fragipan in the Munising (Spodosol) profile. The explanation for this trend lies in the nature of the parent materials. The Munising pedon has formed in acidic glacial drift, whereas the Feldhauser and Glennie soils formed in calcareous parent materials.
Soil extraction data have been used to interpret dominant pedogenic pathways and processes, and to evaluate the nature of fragipans. Some fragic properties, such as brittleness and the ability to slake (possibly due to the solubility of particle-binding/bridging agents) have been attributed to the presence of free- or combined-oxide forms of Fe, Al, and Si (Veneman and Lindbo, 1986; Norfleet and Karathanasis, 1996; Duncan and Franzmeier, 1999). Thus, increased extractable values of Fe, Al, or Si in fragipan horizons can be used as evidence of a possible causative agent(s) (e.g., Harlan et al., 1977; Hallmark and Smeck, 1979; Karathanasis, 1989; Norfleet and Karathanasis, 1996). Failure of extractable values to increase in the fragipan may imply the lack of such agent(s) (e.g., Wang et al., 1974; DeKimpe et al., 1983; Miller et al., 1993).
In all three profiles, Fed > Feo, indicating that most of the Fe in these profiles exists in free-oxide (noncrystalline or crystalline) forms. Values of Alo > Ald in the profiles suggest that active, noncrystalline (organically complexed and inorganic) forms of Al also are dominant. Examination of the extraction data for Fe and Al reveals a few notable trends that are consistent with the findings of others (i.e., Blume and Schwertmann, 1969). In nearly all cases, maximum extractable values for Fe and Al are in the uppermost B horizon (e.g., Bw1 or Bhs). The position of these maxima suggests that podzolization (and associated subprocesses, such as chelation and cheluviation) is active and is responsible for the translocation of Al accompanied by Fe or organic matter or both and the relative enrichment of SiO2 in the overlying eluvial zone. While the upper sequa are the loci of (advanced) podzolization within these sola, the lower sequa are associated with lessivage. Thus, there are no obvious associations among CD- and AAO-extractable oxides and noncrystalline forms of Fe, Al, and Si and the protofragipan/fragipan character of the horizons. These data, however, do not preclude the presence of a fragic-property (brittleness) agent that contains these elements in combination.
Micromorphological Properties
A number of micromorphological features have been consistently observed in thin sections from fragipans: (i) close-packing or interlocking of skeleton grains, (ii) sepic fabric (similar to a porphyric-related distribution pattern [RDP]), (iii) intergrain bridging by clay or amorphous silica, (iv) oriented clay in the form of grain, ped, or channel argillans (grain, aggregate, or void coatings) or pedotubules (infillings), and (v) degraded argillans (coatings) in close proximity to, or within the fragipan. In SEM samples, commonly described micromorphological properties of fragipans are closely packed fabrics, intergrain bridging, and intact and degraded void coatings. All protofragipan and fragipan horizons observed in thin section and under the SEM showed evidence of these properties.
Protofragipan and fragipans described in thin section exhibit a closely packed groundmass that is largely uninterrupted by voids, has a porphyric-RDP and massive microstructure (Fig. 4A) . Payton (1993a) also reported closely packed, interlocked grains with a porphyric-RDP and massive microstructure in fragipan horizons. Micromorphological evidence of close-packing or interlocking of grains has been observed by others working with fragipans in glacial (Yassoglou and Whiteside, 1960; DeKimpe and McKeague, 1974; Lindbo and Veneman, 1993; Miller et al., 1993) and other parent materials (Thompson, 1980; James et al., 1995). Moreover, these authors attributed fragipan hardness or consistence to close-packing or interlocking of mineral grains. We observed closely packed fabrics in all SEM samples of the protofragipan and fragipans (Fig. 4B), as did Thompson (1980) and Payton (1983).
|
Void coatings (Fig. 4D) and infillings (Fig. 4E) were documented in thin section in each of the protofragipan and fragipan horizons. Void coatings are either typic or crescentic and composed of limpid, dusty, or silty clay, silt, or unsorted separates of various sizes. They range in thickness from 5 to 2000 µm with internal fabrics that are nonlaminated, microlaminated, layered, or compound. Void infillings are dense and either completely or incompletely filled with soil separates having similar compositions as those associated with void coatings. The most prominent void infillings, however, are composed of silty clay with a gray, grainy appearance. This material was observed filling voids up to 2 mm wide, and exhibited a variety of internal fabrics. Both clay-sized particles and silt (sometimes with sand) have been documented as infilling fragipan voids, as homogeneous fabrics (Yassoglou and Whiteside, 1960; DeKimpe and McKeague, 1974; Langohr and Pajares, 1983; Miller et al., 1993) or heterogeneous fabrics (Collins and O'Dubhain, 1980; Payton, 1983; Thompson and Smeck, 1983).
Some of the void coatings and infillings associated with the predominantly eluvial protofragipan and fragipan horizons exhibit a gray, grainy appearance (Fig. 4E). Collins and O'Dubhain (1980) reported that silt concentrations, best developed in fragipan horizons, exhibit a similar appearance in some Irish Spodosols. These silt concentrations were determined to be illuvial, oriented, very densely packed with no pore space, and sometimes microsorted void coatings and infillings. Ransom et al. (1987) observed the alteration of argillans to grainy cutans in a fragipan horizon in Ohio, which, along with extensive albic neoskeletan formation, were indicative of argillic-horizon degradation. Gray, grainy clay coatings in the upper portion of a fragipan, studied by Payton (1993a)(1993b), were associated with the deposition of Fe-depleted clay and very fine silt and are sometimes manifested as compound, illuvial coatings, as they may incorporate silt coatings and Fe coatings. According to Payton (1993b), clay illuviation under oxidizing conditions was followed by the mobilization of Fe oxides. Further changes in fragipan redox potential were associated with void-wall and clay-coating destabilization and degradation. Bleached clay and silt particles along these surfaces then became available for translocation via water (i.e., lessivage and pervection [Frenot et al., 1995], respectively), resulting in void surfaces exhibiting skeletal residues. Our observations suggest this may be occurring in eluvial zones of the protofragipan and fragipans.
Void-coatings and void-coating degradation were also observed in the protofragipan and fragipans using SEM (Fig. 4F, 4G). According to Payton (1993b), remnant coating patches become smaller and more isolated as the coating degrades. The few, isolated remnants of the channel coating (composed predominantly of silt with some clay-sized particles) in Fig. 4G suggest that it has experienced advanced degradation, which may eventually lead to channel-wall destabilization and the remobilization of the adjacent fabric. Although degradational features in fragipans have been reported in the literature, they have been evidenced mostly in thin sections or macromorphological investigations (Langohr and Pajares, 1983; Payton, 1983; Lindbo and Veneman, 1993; Miller et al., 1993; Lindbo et al., 2000; McDaniel et al., 2001).
Lessivage, pervection, and eluviation of Fe may be involved in the degradation of the void coating and fine- and coarse-material migration and microerosion (Payton, 1993b). Additionally, the association of silt with degraded coatings (Fig. 4G) and the silt accumulations discussed earlier suggest that water flow has influenced the migration and organization of the protofragipan or fragipan horizons' mobile components (i.e., plasma) (Langohr and Pajares, 1983; Payton, 1993a). Other mechanisms for the destabilization and translocation of silt within horizons such as fragipans are rapid wetting of dry soil, rapid dewatering of saturated soil, thawing of frozen soil, or access to a silt-rich source (Nettleton et al., 1994). Deposition of these grains is favored by pore-size discontinuities, low Ca and Mg content, high silt content, low organic carbon content, and low aggregate stability (Nettleton et al., 1994), many of which are common in the fragipans.
Void coatings composed of clay-sized material that appears to be amorphous based on its surface morphology were observed in protofragipan and fragipan horizons (Fig. 4H). This clay material is often in the form of micrometer-sized beads that is sometimes coalesced into larger aggregates. Although similar material has been reported as being associated with intergrain bridges (Payton, 1983), it has not (to our knowledge) been documented in electron micrographs of fragipan void coatings. Compositional data for this amorphous material was not collected due to its thin, discontinuous nature.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Three soils with varying degrees of fragipan expression from across northern Michigan were chosen for characterization. The Feldhauser soil, formed in sandy loam glacial till with a finer-textured cap, contains a protofragipan: a 2E/B horizon with fragic soil properties. The Munising soil, formed in sandy loam glacial drift, contains a moderately developed fragipan: (B/E)x, (E/B)x, 3(E/B)x'. The Glennie soil, formed in dense loamy glacial till, contains a strongly developed fragipan: Ex, (E/B)x, and Btx. All four soils have bisequal profiles and have similar seasonal hydrologic characteristics.
A few inferences concerning possible pedogenic pathways for fragipan evolution can be based on the data presented. In the three soils studied, the sola have been sufficiently acidified for the initiation of subsequent pedogenic processes. The upper sequum appears to be the locus of (advanced) podzolization and associated subprocesses (e.g., chelation and cheluviation) and translocation of silts and clays, which have guided the differentiation of the upper eluvial–illuvial horizon pair. The lower sequum, containing the protofragipan or fragipan horizons, appears to be the locus of lessivage (and pervection, to an extent), which have contributed to differentiation of the eluvial and illuvial portions of the protofragipan or fragipan. The lower sequum appears to also be affected by seasonal episaturation, at or above the protofragipan or fragipan, and associated saturation and reduction processes, which enhance the eluvial portion of the lower sequum and the degradation of the illuvial fragipan.
Closely packed fabrics, intergrain bridges, void coatings and infillings, and degraded void pedofeatures were observed in the protofragipan and fragipans studied. The presence of a fragic-property (brittleness) agent was not precluded by CD- and AAO-extraction data.
Data presented illustrate that the degree of fragipan expression in Michigan soils is not entirely controlled by the amount of carbonates in the parent material. Furthermore, our findings may have wider applicability within the Great Lakes region: an area in which studies of fragipan genesis are few.
|
ACKNOWLEDGMENTS |
|---|
Received for publication December 4, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 | |||
