HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Web of Science (6) Citing Articles via Google Scholar Google Scholar Articles by Lapen, D. R. Articles by Wang, C. Search for Related Content PubMed Articles by Lapen, D. R. Articles by Wang, C. GeoRef GeoRef Citation Agricola Articles by Lapen, D. R. Articles by Wang, C.

DIVISION S-10-WETLAND SOILS

Placic and Ortstein Horizon Genesis and Peatland Development, Southeastern Newfoundland

^a Eastern Cereal and Oilseed Research Centre, Agriculture Canada, Ottawa, ON, Canada K1A 0C6

lapend{at}em.agr.ca

	ABSTRACT

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion Summary and conclusions REFERENCES

 
There is considerable speculation about relationships between soil development and local-scale blanket bog formation in southeastern Newfoundland. This study examined podzolization along a catena in southeastern Newfoundland. The specific objective was to investigate pedogenic processes, in particular placic and ortstein horizon development, as well as potential links between soil and peatland formation. Chemical, hydrological, and physical processes potentially responsible for pedogenesis were examined. Placic genesis appears to be linked to the precipitation of Fe at oxidizing and higher soil pH contacts at depth in the soil. Placic formation in the well-drained heathlands is apparently precluded by constraints on Fe mobilization at the surface, in particular, Fe-oxidizing conditions and the presence of sufficient supplies of metals neutralizing mobile organic substances. Organic-complexed Al is the dominant cementing agent in ortstein horizons. Cementation is augmented by minimal bioturbation by roots and high coarse-fragment content, which concentrates Fe and Al to a limited soil matrix. Weathering, podzolization, elevated water tables, and peatland processes are promoted by groundwater discharge from adjacent bogs as well as by sesquioxide pans that impede vertical drainage.

Abbreviations: Al_d, dithionite aluminum • Al_o, oxalate aluminum • Al_p, pyrophosphate aluminum • D_b, bulk density • Fe_d, dithionite iron • Fe_o, oxalate iron • Fe_p, pyrophosphate iron • K_sat, bulk hydraulic conductivity • K_x, horizontal hydraulic conductivity • K_z, vertical hydraulic conductivity • Mn_d, dithionite manganese • LB, lowland bog • LPF, lowland poor fen • OC, organic carbon • P1–8, pedon numbers • S, relative saturation • Si_o, oxalate silicon • TDR, time domain reflectometry • UB, upland bog • UPF, upland poor fen

	INTRODUCTION

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion Summary and conclusions REFERENCES

 
PROLONGED SATURATION, or near saturation, of the ground surface is required for peat bog development (Ivanov, 1978). Impermeable and semi-impermeable sesquioxide pans, such as placic and ortstein horizons, have been linked to peatland formation because they can significantly impede vertical drainage (Karavayeva, 1968; Ugolini and Mann, 1979). Placic horizons are thin (commonly <1 cm thick), indurated sesquioxide pans composed primarily of Fe that are generally impermeable to water flow (Agriculture Canada Expert Committee on Soil Survey, 1987). Ortstein horizons are variably permeable (Wang et al., 1978), cemented sesquioxide (primarily Al and Fe) pans, which are at least 3 cm thick (Agriculture Canada Expert Committee on Soil Survey, 1987). By virtue of their ability to impede drainage, pans can encourage peat formation on otherwise well drained soils. On the southeastern tip of Newfoundland, sesquioxide pans commonly occur in the coarse-textured Spodosols that underlie blanket bogs (peatlands covering vast expanses of irregular terrain) and in fens (peat-forming areas fed by groundwater) (Heringa, 1981; Crum, 1988). Spodosols in adjacent heathlands often lack pans, and their surface soils are typically well drained.

Placic and ortstein horizon genesis is not precisely understood (Soil Survey Staff, 1975; Miles et al., 1979), and causal links between these horizons and blanket bog formation in the region are speculative (Irwin, 1994). One of the first steps toward a more complete understanding of these relationships is examination of the mechanisms that potentially promote or constrain local-scale pan formation. In this paper, two general hypotheses on sesquioxide pan development are put forth: Hypothesis 1 states that pans will form when sesquioxides, mobilized via reducing conditions in upper soil horizons, translocate to a point at which oxidation and precipitation of sesquioxides results from abrupt increases in soil pH and/or soil redox potential (Crampton, 1963; Damman, 1965; McKeague et al., 1968). Hypothesis 2 states that sesquioxide pans will form when sesquioxides complex with soluble organic matter and translocate to points in the soil profile where they can precipitate (McKeague et al., 1967; Schwertmann and Fischer, 1973; De Coninck, 1980). Although studies by Damman (1965), McKeague et al. (1967, 1968), Wang et al. (1978), and Miles et al. (1979) provided useful data on the composition and nature of some placic and ortstein soils in Maritime and Atlantic Canada, they could not isolate genetic mechanisms.

This study characterized soil pedons and examined processes potentially responsible for the development of placic, ortstein, and nonpan Spodosols in peatland and heathland environments in southeastern Newfoundland. The primary purpose of this study is to provide insights into sesquioxide pan genesis and potential relationships between local-scale peatland development and podzolization.

	Materials and methods

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion Summary and conclusions REFERENCES

 
Soils, Vegetation, and Hydrology
The study area is located at Cape Race, Newfoundland, Canada (Fig. 1) . Average total yearly precipitation at Cape Race is 1379 mm (Environment Canada, 1982). Most slope gradients are <4°, and a sandy to loamy (Day, 1983) lodgement till, generally <2 m thick, is the parent material. The till originates from local bedrock formations of greenish-gray sandstone and siltstone, slate, and argillite (Heringa, 1981). Spodosols are the major soils in the area and a cobble pavement commonly occurs over their surfaces. Blanket bogs and heathlands predominate in the region.

View larger version (46K): [in this window] [in a new window]   Fig. 1 Study site location, Pedons 1 through 8 (P1 to P8), and locations of 14C dates of organic soils (yr BP). Pedon 2* is located in the upland poor fen in a better drained topographic position SW of P2. See Table 1 and 3 for pedon descriptions

Blanket bogs in the region are covered by Sphagnum spp. and contain living to well decomposed organic material (Wells and Pollett, 1983). The upland (UB) and lowland bogs (LB) are generally <2.0 m deep. Blanket bog water tables in the region are typically between 0.05 to 0.20 m below the surface (Northlands Assoc., 1989) and groundwater discharge from the bogs occurs primarily in the direction of surface slope aspect (Lapen, 1998). Placic horizons underlying the UB and LB perch the groundwater (Lapen et al., 1996).

Vegetation in the UPF includes Sphagnum spp., and relative to the bogs, larger proportions of Cyperaceae spp. (sedges) and heath (Ericaceae spp.). Surface peat and muck are approximately 0.1 to 0.3 m thick. The water table is usually within 0.2 m below the surface (0.05–0.10 m depth) and placic horizons effectively perch the groundwater (Lapen et al., 1996). The fen receives, on a continuous basis, discharge from the UB, and groundwater in the fen drains downslope into the heathlands (Lapen et al., 1996; Lapen, 1998).

Heathland vegetation is composed mainly of heath, sweet gale (Myrica gale L.), Schreber's moss [Pleurozium schreberi (Brid.) Mitt.], and lichens (Cladonia and Cladina spp.). A porous humus mat loosely drapes the mineral soil surface and a cobble pavement. The soils are predominantly excessively to well drained.

Sedges, sweet gale, and Sphagnum spp. are the main vegetation types in the LPF. Peat and muck are 0.05 to 0.2 m thick and ortstein horizons effectively perch the groundwater. The LPF receives groundwater discharge from the adjacent LB, and groundwater in the fen, in turn, drains into the adjacent heathlands. The water table in the LPF usually fluctuates above 0.2 m below the surface (Lapen et al., 1996; Lapen, 1998).

Field and Laboratory Measurements
Volumetric water content () was estimated gravimetrically and by time domain reflectometry (TDR) using a Tektronix 1502B cable tester (Wilsonville, OR) and balanced twin-lead wave guides (0.15 m). The wave guides were inserted horizontally into selected horizons (2–3 probes per horizon). Volumetric water contents of mineral and organic soils were estimated according to the calibration methods described in Topp et al. (1980) and Topp et al. (1994), respectively. Relative saturation (S) was determined by

(1)

where n is total porosity. Porosity (assumed to equal at saturation) of the mineral soils at most TDR probe locations was estimated when the soils became saturated during a very heavy rain storm during the summer of 1993. Wells made of slotted PVC pipe were used to monitor water table elevations at selected locations along the catena. Well lengths were equivalent to peat thicknesses in bog, to the thicknesses of the soil above the placic and ortstein horizons in the fens, and to the thickness of the soil above bedrock in the heathlands.

Soil redox potentials were measured in selected horizons using a Jensen (Tacoma, WA) Model P5E hand-held redox meter, an Ag–AgCl reference electrode, and Pt-coated microelectrodes. For each measurement, electrodes were inserted horizontally into horizons near the TDR probe locations and allowed to equilibrate (usually <1 h). There were typically three replicate measurements per horizon. Groundwater pH was determined using a Fisher Scientific Limited (Nepean, ON, Canada) model 119 pH and temperature meter; at least 10 replicate measurements per measurement site discussed in this study were made during the summers of 1993 and 1994. Daily average soil temperatures were measured from thermocouples installed in a soil pit near Pedon no. 5 (P5). A tipping-bucket rain gauge was used to measure total daily rain amounts. The redox, rain, temperature, and S data reported in this study were collected from 5 July (Day 186) to 26 July (Day 206) 1994.

Soil corers (5.50 and 7.62 cm diam.) were used to acquire bulk density (D_b) samples for uncemented horizons, while the clod D_b method was employed if the soil was cemented (Blake and Hartge, 1986). Particle density and total porosity (n) were determined for each core and clod sample (Carter and Ball, 1993). The saturated horizontal (K_x) and vertical (K_z) hydraulic conductivity (K_sat) of soil cores and clods were estimated using a constant-head permeameter (Freeze and Cherry, 1979); for clods (pan soils), sample margins were sealed in wax to help prevent peripheral leakage.

Loose soil samples were collected from each soil horizon. The samples were air dried and passed through a 2.0-mm mesh. Soil pH was estimated in a 1:1 soil to water ratio, organic carbon (OC) was determined by dry combustion, and Fe and Al were extracted by dithionite–citrate, oxalate, and pyrophosphate methods (Sheldrick, 1984). Manganese and silicon were extracted by dithionite and oxalate methods, respectively. Clay and silt fractions were determined using a Micromeritics (Norcross, GA) 5100 Sedigraph particle-size analyzer. The samples were pretreated with dithionite–citrate and H₂O₂. Large bulk samples (0.03 m³) of soil were used to determine coarse-fragment contents (>2.0 mm) of soil sections encompassing several soil horizons.

Basal and mid-profile peat samples from the UB and LB were collected with a minimized Macauley-style peat sampler (Day et al., 1979). Calibrated radiocarbon (¹⁴C) dates (Stuiver and Pearson, 1993; Pearson and Stuiver, 1993) of the peat samples were determined by the Radiocarbon Laboratory, Brock University, St. Catherine's, ON (Fig. 1).

	Results and discussion

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion Summary and conclusions REFERENCES

 
The largest rain event (>30 mm) occurred on Day 197 (Fig. 2) . Average daily precipitation during the study period was 2.9 mm d^-1. The 30-yr average daily precipitation for July at Cape Race is 2.7 mm (Environment Canada, 1982). Soil temperatures recorded at the site are consistent with July normals at similar soil depths for areas on the south coast of Newfoundland (Banfield, 1983).

View larger version (31K): [in this window] [in a new window]   Fig. 2 Average daily soil temperatures (point graph), total daily rain depths (bar graph), and water table depths at selected locations in the peatland complexes (point graph) during the study period. Symbol legends for soil temperature and water table depth indicate measurement depths below surface and well locations, respectively. Note: water tables were recorded in soils above the sesquioxide pans

View larger version (35K): [in this window] [in a new window]   Fig. 3 Physical properties of pedon mineral soil horizons. Pedon number is given in small box in upper left hand corner of graph sets. Grain size (GS) on bar graph: clay = black, silt = white, sand = diagonal stripes; horizontal hydraulic conductivity Kz = , vertical hydraulic conductivity Kx = ; bulk density Db = • (averages of at least 5 samples). Saturated hydraulic conductivity Ksat values are geometric averages of at least 3 samples

View larger version (36K): [in this window] [in a new window]   Fig. 4 Average daily relative saturation S () and redox potential (•) profiles for selected pedons on Days 193, 194, 196, and 204. Dotted S and redox lines are for Pedon 2*. Dashed lines indicate standing water level in the soil pits. Redox and measurements were taken after the soil pits were pumped free of standing water. Redox potentials are averages of three measurements

In the UPF, reducing conditions were observed above the placic horizons even though the soils were unsaturated. During unsaturated conditions, the UPF organic soils maintained high water contents; more so at P2 than at P2*. In fact, reducing conditions above the placic horizon at P2 were observed on all days prior to Day 197. Generally, redox potentials above the UPF placic horizons increased during and immediately following rain, after which, potentials gradually decreased over time. For instance, at P2* after the rain on Day 197, redox potentials in the middle of the A horizon dipped from values >600 mV to values <400 mV on Day 204 (some replicate measurements were <200 mV). The redox and S data in this study suggest that oxidizing conditions can occur beneath placic horizons in a bog and fen environment and that reducing conditions can be maintained in unsaturated soils during typically wet, as well as uncharacteristically dry situations.

Assuming Fe_d represents total Fe, inorganic forms of Fe dominate in placic horizons at P1 and P8 (Table 2) . In fact, the Fe_p:Fe_d ratio in the placic horizons at P1 and P8 are only 0.25 and 0.22, respectively. The Al_o:Fe_d (0.19 and 0.13) and Mn_d:Fe_d (0.004 and 0.002) ratios in the placic horizons at P1 and P8 are relatively small. Another chemical cementing agent, Si (Wilding et al., 1977), is not a significant cementing material in these soils, because Si_o:Fe_d ratios in the placic horizons at P1 (0.04) and P8 (0.02) are also small. The average pH of soils above the placic horizons is 5.0, while the average pH of the soils immediately below the placic horizons is 5.7 (Table 2).

Pedogenesis
The placic horizons at P1 and P8 are composed primarily of inorganic Fe and redox–pH processes may have been important mechanisms in their formation. Soil redox and episaturation were strongly influenced by the placic horizons' impedance of drainage; therefore, it was impossible to determine whether the observed redox regimes existed prior to or during the early stages of placic development. However, at P3, a heathland pedon currently lacking sesquioxide pans, saturated and reducing conditions in the Bw horizon were observed immediately above a permeable, unsaturated and oxidized BC horizon (Fig. 4). For example, redox potentials in the Bw horizon at P3 abruptly decreased from 500 mV on Day 201 to <-90 mV on Days 203 to 206. On a daily basis, the soils immediately above and below the Bw horizon at P3 supported oxidizing conditions. This redox contact occurred at 0.3 m below the mineral soil surface, a depth corresponding with upper boundary of the BC and Bw horizons in the placic horizon pedons.

Although redox measurements were made only during a 3-wk period, these observations demonstrate that abrupt reducing-to-oxidizing contacts can occur in permeable soils that lack cemented horizons at the site. Such conditions, when combined with an increase in soil pH, can facilitate the abrupt oxidation of Fe at depth in the placic pedons (Hypothesis 1).

In Newfoundland, placic horizons have been observed at or near fine to coarse soil-texture contacts (McKeague et al., 1967, 1968). The textural–structural contact at the placic locales in this study are primarily coarse to fine and/or structureless to weakly platy. These physical properties and associated decreases in BC and Bw horizon K_z, tend to slow soil solution infiltration, thus prolonging solution exposure to oxidizing conditions in the underlying soil. Nevertheless, the idea that placic genesis is due predominantly to the precipitation of organically-complexed Fe cannot be entirely refuted. Subsequent decomposition of organic matter by microorganisms might explain the abundance of inorganic Fe in the placic horizons at P1 and P8. Differences between Al and Fe contents in illuvial horizons, which were also observed in other studies (Adams et al., 1980; McKeague et al., 1967, 1968), may have been due to fulvic acid having a greater affinity for Fe(III) than for Al(III) (Mokma and Buurman, 1982), or for the Al that is transported out of the system via acidity (Adams et al., 1980).

While the Bhsm and Bsm horizons at P2 are composed mainly of organically-complexed Fe, redox–pH processes may have been important to pan genesis. Hypothesis 1 is supported by the locations of observed reducing-to-oxidizing and lower-to-higher pH soil contacts, and by the fact that the most indurated portion of the pan, the Bhsm,* is dominated by inorganic Fe. However, abrupt soil-solution contact with more oxygenated soil horizons can precipitate both inorganic- and organic-complexed Fe (Schwertmann and Fischer, 1973). The proportions of Fe_p in the Bhsm and Bsm horizons may have also been the result of Fe oxides adsorbing organic substances (Adams et al., 1980; McKeague et al., 1986) at the placic locale and/or less decomposition of pan organic material.

Causal links between placic genesis and bog initiation were not examined specifically; however, the onset of cool, wet climate conditions 2800 yr BP (Wells and Pollett, 1983) suggests bog formation was climatically induced, while the location of initiation (bottom of depressions) implies bog inception was spatially constrained by topographic and soil factors. In the UPF, placic genesis probably preceded basal peat accrual (basal peats 135 yr BP). Wet–acidic conditions in the UPF currently encourage peat-forming processes and podzolization. These conditions are promoted by placic horizons that impede vertical drainage, as well as by lateral inputs of acidic–low redox groundwater from the UB (Lapen, 1998). In fact, the range of groundwater pH at the UB vs. UPF boundary and the UPF vs. heath boundary were measured to be 4.4 to 4.6 and 4.8 to 5.9, respectively. Groundwater originating in a bog environment can lower pH and redox potentials in adjacent environments and can provide significant inputs of dissolved organic substances (Wetzel, 1983) that complex with Fe and Al (De Coninck, 1980).

Heathland Soils (Pedons 3, 4, and 5)
Soil Characteristics
A loose consistency, root activity, and the occurrence of pellets and/or aggregates all suggest that surface Bhs horizons are bioturbated by roots (Table 3) . There is also a marked increase in the clay plus silt contents below the Bw horizons (Fig. 3). Coarse-fragment content (>2.0-mm fraction) in mineral soil horizons above the Bw horizons at P3 and P5 are 500 g kg^-1; at P4 it is 900 g kg^-1.

On a daily basis, oxidizing and unsaturated conditions prevailed in the pedogenic horizons at P4 and P5; however, during the storm on Day 197, very thin, spatially variable zones (0.01 m thick) of temporary episaturation were visually observed in the Bw horizon at P5 and the upper portion of the Bsm horizon at P4. Redox potentials measured at these sites during this time were >550 mV. Daily S values at P4 and P5, which were similar in value and profile (P4 shown in Fig. 4), were notably less than those at P3. These differences were attributed in large part to lack of direct throughflow inputs from peatlands, and to the slopes' enhancement of lateral drainage. In general, S values of all heath organic horizons were smaller than those in the bogs and fens.

The heath Bhs horizons have an abundance of organic-complexed Fe and Al and occur immediately below O horizons (Table 4) . In the case of P3, high percentages of Fe_p + Al_p also occur in the Bs horizon (22.6 mg g^-1). Of all the heath Bhs horizons, the one at P4 has both the highest content of, and the most organically-complexed Fe and Al . Although more conclusive chemical data are required, the relatively high (Fe_d + Al_o):OC ratios (0.79) and crystalline Fe content (11.5 mg g^-1) in the Bhs horizon at P4, suggest that high metal content immobilizes organic substances (De Coninck, 1980). While Fe_p abruptly decreases in the Bsm horizon at P4 (4.7 mg g^-1), Al_p remains fairly high (13.2 mg g^-1). The (Fe_d + Al_o):OC ratios in the Bhs horizons at P3 (Bhs1 = 0.29 and Bhs2 = 0.39) and P5 (0.25) are smaller; however, the amount of Fe + Al in the profile reaches a maximum in these horizons. Although the vertical distributions of Fe + Al and OC at P5 are similar to those at P4, the concentrations are smaller. The Si_o contents in the Bhs horizons at P4 and P5 are also small, indicating that the occurrence of pedogenic aluminosilicates is minimal.

On the other hand, even if saturated heathland soils could become highly reduced, groundwater residence time in surface Bhs horizons is typically short. For instance, during the summer of 1993, surface Bhs horizons were observed to be variably saturated for <2 d after a 4.1 mm hr^-1 storm that produced 83 mm of rain. Thin zones of episaturation (usually <0.02 m thick) can occur at P4 and P5, as previously described; however, they are generally short-lived (lasting a few hours to <2 d), spatially variable, and were not observed to support abrupt shifts in redox potentials at depth in the soil profile. Water table response during very heavy precipitation events, soil color (7.5YR 3/2), general lack of redoximorphic features, and the S and redox data all suggest eluviation of Fe(II) is probably not a significant process in the heath pedogenic horizons.

Mobilization of sesquioxides from surface horizons may also be inhibited by significant quantities of metals that immobilize the soluble organic substances produced at the surface. This immobilization mechanism is supported by the fact that soil pH in all the heath pedon horizons above the BC horizon is poorly correlated with Al_o (r = 0.06). Moreover, groundwater pH in heathland B horizons ranged between 4.8 and 5.9 (pH 5.0 at P4 and P5). The high Fe + Al content in the Bhs horizons was likely augmented by (i) a high percentage of coarse-fragments that concentrated Fe and Al to a limited soil matrix, (ii) an abundance of weatherable Fe + Al bearing materials at and near the surface, (iii) oxidizing conditions, and (iv) organic matter binding. Surface soils with an abundant supply of minerals that release Fe and Al at a moderate rate often lack well developed albic horizons (McKeague et al., 1978).

The smallest percentage of Al_o in the surface Bhs horizon occurs at P3 . The more direct hydrological link between the UPF and marginal heathlands has made P3 slightly more acidic relative to the other heath pedons. The acidic, often low redox potential (<350 mV) UPF groundwater can drain as throughflow into nonplacic heath soils. At P3, acidity (soil pH = 4.1) may have promoted greater eluviation of Al, and to a lesser extent Fe, from the surface Bhs horizon; however, more significant depletion of Fe + Al in the Bhs and Bs horizons is likely checked by the processes previously described. Although the Bhs2 and Bs horizons have relatively high amounts of organic-complexed Al, a dominant cementing agent in ortstein horizons (Miles et al., 1979), cementation processes were possibly inhibited by root activity (Table 3), and/or cryoturbation (Henderson, 1968).

Ortstein formation is evident at P4. Soil pH (5.2) and higher metal concentrations promote illuviation of organically-complexed Al. Cementation in the Bsm horizon, however, is reinforced by minimal root activity, high concentrations of organically-complexed Al, and high coarse-fragment content. The latter factor may be critical since it decreases the volume of fine-textured soil in which illuviated cementing material can accumulate; such a factor can also augment precipitation of organic-complexed Fe and Al by increasing metal concentrations in the soil. McKeague and Wang (1980) noted that strong cementation requires the cementing agent to occur as a continuous phase. The reduction in soil matrix as a result of high coarse-fragment content can help achieve this because less cementing agent is necessary to link available soil grains. Ortstein formation in the heathlands was observed only where coarse-fragment contents in illuvial horizons were high. Nevertheless, slope position and the permeability of the Bsm horizon (K_z > 10^-3 cm s^-1) reduce the probability of surface saturation and, subsequently, peat formation. The K_sat of ortstein soils in eastern Canada can be quite high; for example, 2 x 10^-2 cm s^-1 (Wang et al., 1978).

Pedon 5 is located on a nearly level footslope, where bog formation processes in the region are generally expected to occur. However, permeable soils help preclude surface saturation. Roots, which extend well into a permeable C horizon, and lower percentages of coarse-fragments in the illuvial Bs horizons have likely curtailed significant cementation and, potentially, the development of lower K_sat pans. Because regional climate has supported sesquioxide pan development in other heathlands on the southern Avalon Peninsula of Newfoundland (Heringa, 1981), field data suggests that local-scale factors are largely responsible for the lack of strong and continuous pan development in the heathland soils in this study.

Lowland Poor Fen Soils (Pedons 6 and 7)
Soil Characteristics
Vertical drainage in the LPF is impeded by the low K_z of the ortstein horizons (Fig. 3). The low K_z resulted largely from the illuviation of sesquioxides that reduces the volume of water-conducting pores. Microscopic investigation of the ortstein soil and BC horizons revealed horizontal–platy sesquioxide accumulations (ortstein horizons) and abandoned rootlet holes lined with precipitated sesquioxides (BC and ortstein horizons) (Table 1). High D_b in the surface horizons is due to high percentages of coarse-fragments. The coarse-fragment content in the mineral soils above the BC horizons vs. in and below the BC horizons is 650 g kg^-1 vs. 250 g kg^-1, respectively.

Immediately after Day 197, the LPF water table rose to slightly below the surface (Fig. 2). The soils in the lower portions of the Bhs2 horizon at P6 and Bs1 horizon at P7 were saturated and supported redox potentials <350 mV on all days prior to Day 194 (Fig. 4). Unlike redox potentials in the unsaturated soils at P2, redox potentials abruptly increased when the measurement sites became unsaturated, even though S values were relatively high. After Day 197, redox potentials in the materials above the ortstein horizons gradually decreased through time. The weakly cemented ortstein horizons and the horizons immediately below them remained unsaturated during the entire study period. Redox measurements in the ortstein horizons at P7 were also >500 mV. The LPF redox observations and soil color suggest that the ortstein horizons and soils directly below them support oxidizing conditions throughout longer periods of time; moreover, reducing conditions can occur in the soils immediately above the semipermeable and oxidizing ortstein horizons.

Unlike the heath pedons, Fe + Al maxima in the LPF pedons do not occur in the surface mineral horizons (Table 2). The (Fe_d + Al_o):OC ratios in the Bhs1 and Bh horizons at P6 and P7 are 0.15 and 0.18, respectively. The Al and Fe in the LPF soils are predominantly organically-complexed, and like the Bsm horizon at P4, the primary cementing agents in the ortstein horizons are organic-complexed Al [average Al_p:(Fe_d + Al_o) ratios in ortstein horizons at P6 and P7 are 0.66 and 0.60, respectively], and to a lesser extent, organic-complexed Fe [average Fe_p:(Fe_d + Al_o) ratios in the ortstein horizons at P6 and P7 are 0.26 and 0.34, respectively]. The average (Fe_d + Al_o):OC ratios for the ortstein horizons are 0.52 at P6 and 0.57 at P7. Horizons from the Bhs2 to Bs2 at P6 have essentially similar Fe_d + Al_o concentrations. The Fe_d + Al_o concentrations in the ortstein horizons at P7 are slightly greater than those at P6. The average pH of soils above, within, and below the ortstein horizons in the LPF are 4.4, 5.1, and 5.0, respectively (Table 2).

Pedogenesis
Due to a combination of the low K_z ortstein horizons and gentle slope gradients, drainage conditions in the LPF are fairly poor. These conditions have helped to (i) elevate the water table, (ii) decrease rooting depths, (iii) support peat formation, (iv) enhance weathering of mineral material, and (v) increase Fe and Al mobilization.

Like P4, coarse-fragment content and lack of root activity has reinforced cementation processes in the ortstein horizons. Illuviation of sesquioxides is linked to a combination of oxidizing conditions, higher soil pH (5.1), and greater concentrations of metals in the illuvial horizons. While coarse-fragment content likely contributed to cementation processes in all ortstein pedons, as described above, it is perhaps owing to greater Fe and Al eluviation–illuviation that the ortstein horizons in the LPF are more strongly developed than those at P4. On a total material basis, assuming the coarse-fragment contents given previously, the ortstein horizon Fe_d + Al_o contents at P4, P6, and P7 are 2, 4 to 5, and 6 to 10 mg g^-1 of total material, respectively. Wang et al. (1978) observed that ortstein horizons are more strongly expressed in poorly drained environments, rather than in well drained environments. Field evidence suggests that the influence of the LB on podzolization and weathering is most pronounced at P7. Albic horizon characteristics are more strongly expressed at P7 than they are at P6, and coarse material in and above the surface soils becomes progressively more bleached toward the LB. In fact, ranges of groundwater pH above the ortstein horizons at the LPF vs. LB and heath vs. LPF margins are 4.7 to 5.2 and 5.0 to 5.2, respectively. These pH observations, combined with the location of P7 being immediately adjacent to the LB, where it receives more direct contributions of acidic–low redox bog groundwater discharge, may explain greater soil development at P7 relative to P6.

The fact that soil pH is positively correlated with Al_o (r = 0.68) in horizons above LPF BC horizons suggests that pH may have been critical to the stability of organic-complexed Al and/or the solubility of hydrous oxides (Carbera and Talibudeen, 1977). For similar horizons, Fe_d has a significantly weaker correlation with pH (r = 0.28). Relative to the heath pedons, it seems likely that the wetter and more acidic conditions in the fen helped to mobilize Al in the upper horizons.

Weakly cemented ortstein, rather than placic or strongly cemented ortstein, may have been influenced in part by the relatively recent role that lateral drainage inputs from the LB played in fen podzolization. For instance, assuming an LB water table at a depth of 0.15 m below the surface and a vertical peat accrual rate of 0.034 cm yr^-1, the water table at the 1820 yr BP basal date location in the LB likely became elevated above the LPF BC horizon 1300 yr BP and above the LPF mineral soil surface 400 yr BP. For typical bog–water table elevations, these dates roughly represent times at which the LB could have supported groundwater discharge directed immediately into LPF pedogenic horizons. However, most significant discharge inputs probably occurred within the past several hundred years when more favorable lateral hydraulic gradients (i.e., increased slope of water table from LB to LPF due to LB peat accrual) in the LB became more firmly established. Nevertheless, relatively recent inputs of LB groundwater could have been instrumental in accelerating pedogenic processes in the LPF because significant podzolization can occur in very short time periods (Stoner and Ugolini, 1988). Progressive cementation, promoted by increasingly beneficial surface moisture conditions and minimal root activity, would positively reinforce these processes by retarding vertical drainage. The subsequent growth of Sphagnum spp. would increase both the LPF acidity and the supply of soluble organic matter at the surface (Karavayeva, 1968; Wetzel, 1983).

It is yet unclear whether the LPF will develop strongly cemented ortstein or indurated placic horizons; notwithstanding these speculations, the ortstein horizons currently serve to impede vertical drainage to a point at which moisture conditions favor peat-forming processes and the establishment of bog-forming vegetation.

	Summary and conclusions

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion Summary and conclusions REFERENCES

 
This study examined pedogenic processes along a hillslope catena and potential relationships between soil and peatland development in the blanket bog region in southeastern Newfoundland. General findings are the following.

(i) Placic horizon genesis appears to be linked to the precipitation of mobile Fe at higher pH and oxidizing soil contacts 0.3 m below the mineral soil surface. Placic development in adjacent heathlands is likely constrained by oxidizing conditions and high concentrations of metal that neutralize mobile organic compounds in surface soils; thus, heath soils remain well drained and unsuitable for blanket bog development.

(ii) Organic-complexed Al appears to be the dominant chemical cementing agent in ortstein horizons. Cementation is likely encouraged by minimal root activity and high coarse-fragment content, which decreases the volume of soil in which illuviated cementing material can accumulate. The latter condition appears especially critical to ortstein formation at well drained sites. Relative to heathland soils, more strongly expressed ortstein horizons in the poorly drained LPF may be linked to greater weathering and mobilization of sesquioxides via reducing, wetter, and more acidic soil conditions. Illuviation in the LPF is augmented by oxidizing conditions at depth in the soil profile, higher soil pH, and greater concentrations of metals.

(iii) Podzolization, peat formation, and Sphagnum spp. growth are positively reinforced by wet surface soils. Because groundwater discharge from adjacent bogs and impermeable to semipermeable sesquioxide pans encourage such hydrological conditions, these factors are important to fen maintenance and development, and potentially, to the lateral spread of blanket bogs at these sites.Northlands Associates. 1989

	ACKNOWLEDGMENTS

 
Funding was provided by Horton Research Grant, awarded to D.R. Lapen by the American Geophysical Union, and Natural Sciences and Engineering Research Council Grants to J.S. Price, Geography, University of Waterloo, Ontario, Canada, and to R. Gilbert, Geography, Queen's University, Kingston, Ontario. The authors appreciate the laboratory analysis by W.D. Reynolds and B. Sheldrick. Helpful comments were provided by G.C. Topp.

Received for publication February 24, 1997.

	REFERENCES

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion Summary and conclusions REFERENCES

 

• Agriculture Canada Expert Committee on Soil Survey. The Canadian system of soil classification, 2nd ed Ottawa: Agric. Can. Pub. 1646, 1987.
• Adams W.A., Raza M.A., Evans L.J. Relationships between net redistribution of Al and Fe and extractable levels in Podzolic soils derived from lower Paleozoic sedimentary rocks. J. Soil Sci. 1980;31:533-545.
• Banfield C.E. Climate. In: South R., ed. Biogeography and ecology of the island of Newfoundland. The Hague, the Netherlands: Dr. W. Junk Publ, 1983:37-106.
• Blake G.R., Hartge K.H. Bulk density. In: Klute A., ed. Methods of soil analysis. Madison, WI: Part 1. SSSA Book Ser. 5. ASA and SSSA, 1986:363-375.
• Carbera F., Talibudeen O. Effect of soil pH and organic matter on labile aluminum in soils under permanent grass. J. Soil Sci. 1977;28:259-270.
• Carter M.R., Ball B.C. Soil porosity. In: Carter M.R., ed. Soil sampling and methods of analysis. Boca Raton, FL: Lewis Publ, 1993:581-588.
• Crampton C.B. The development and morphology of iron pan Podzols in mid and south Wales. J. Soil Sci. 1963;14:282-302.
• Crum H. A focus on peatlands and peat mosses. Ann Arbor, MI: Univ. Michigan Press, 1988.
• Damman, A.W.H. 1965. Thin iron placic horizons: Their occurrence and the conditions leading to their development. Information Rep. N-X-2. Can. Dep. Forestry, St. John's, NF.
• Day J.H., Rennie P.J., Stanek W., Raymond G.P. Peat testing manual. Technical Memorandum 125. Ottawa: Nat. Res. Council of Canada, Associate Committee on Geotechnical Research, 1979.
• Day J.H. Manual for describing soils in the field. Agriculture Canada, Ottawa: Land Resour. Res. Inst. Contrib. 82-52. Research Branch, 1983.
• De Coninck F. Major mechanisms in formation of spodic horizons. Geoderma 1980;24:101-128.
• Environment Canada Canadian climatic normals 1951–1980: Temperature and precipitation, Atlantic Provinces. Downsview, ON: Atmospheric Environment Service, 1982.
• Freeze R.A., Cherry J.A. Groundwater. Englewood Cliffs, NJ: Prentice Hall, 1979.
• Gray D.M. Handbook on the principles of hydrology. Ottawa: Canadian Nat. Committee for the Internat. Hydrological Decade, 1970.
• Henderson E.P. Patterned ground in southeastern Newfoundland. Can. J. Earth Sci. 1968;5:1443-1453.
• Heringa P.K. Soils of the Avalon Peninsula, Newfoundland. Land Resources Res. Inst. Pub. 113. Rep. 3. Newfoundland Soil Survey. Ottawa: Agriculture Canada, 1981.
• Irwin, T.E. 1994. Development of a blanket bog at St. Shotts, Avalon Peninsula, Newfoundland. Ph.D. diss. Univ. of Toronto, Canada.
• Ivanov K.E. Water movement in mirelands. London, UK: Academic Press, 1978.
• Karavayeva N.A. Formation and evolution of cemented ortstein horizons in the Taiga zone. 9th Int. Cong. Soil Sci. 1968;4:451-458.
• Lapen, D.R. 1998. Potential hydrologic and pedologic factors that govern blanket bog development and maintenance in southeastern Newfoundland. Ph. D. diss. Queen's Univ., Kingston, ON, Canada.
• Lapen D.R., Moorman B.J., Price J.S. Using ground-penetrating radar to delineate subsurface features along a wetland catena. Soil Sci. Soc. Am. J. 1996;60:923-931.[Abstract/Free Full Text]
• McBride M.B. Environmental soil chemistry. New York: Oxford Univ. Press, 1994.
• McKeague J.A., Cheshire M.V., Andreux F., Berthelin J. Organo-mineral complexes in relation to pedogenesis. In: Huang P.M., Schnitzer M., eds. Interactions of soil minerals with natural organics and microbes. SSSA Spec. Pub. 17. Madison, WI: SSSA, 1986:549-592.
• McKeague J.A., Damman A.W.H., Heringa P.K. Iron–manganese and other placic horizons in some soils of Newfoundland. Can. J. Soil Sci. 1968;48:243-253.
• McKeague J.A., Ross G.J., Gamble D.S. Properties, criteria of classification, and concepts of genesis of Podzolic soils in Canada. In: Mahaney W.C., ed. Quaternary soils. Norwich, UK: Geological Abstracts, 1978:27-60.
• McKeague J.A., Schnitzer M., Heringa P.K. Properties of an Ironpan Humic Podzol from Newfoundland. Can. J. Soil Sci. 1967;47:23-32.
• McKeague J.A., Wang C. Micromorphology and energy dispersive analysis of ortstein horizons of podzolic soils from New Brunswick and Nova Scotia, Canada. Can. J. Soil Sci. 1980;60:9-21.
• Miles N.M., Wang C., McKeague J.A. Chemical and clay mineralogical properties of ortstein soils from the Maritime Provinces. Can. J. Soil Sci. 1979;59:287-299.
• Mokma, D.L., and P. Buurman. 1982. Podzols and podzolization in temperate regions. ISM Monograph 1. Int. Soil Museum, Wageningen, the Netherlands.
• Northlands Associates Research on bog preparation and peat sod drying techniques in Newfoundland (1986–1988). St. John's, NF, Canada: Northlands Assoc. Ltd, 1989.
• Pearson G.W., Stuiver M. High-precision bidecadal calibration of the radiocarbon time scale 500–2500 BC. Radiocarbon 1993;35:25-34.[Web of Science]
• Schwertmann U., Fischer W.R. Natural "amorphous" ferric hydroxide. Geoderma 1973;10:237-247.
• Sheldrick, B.H. 1984. Analytical methods manual. Land Resour. Res. Inst. Contrib. 84–30. Agriculture Canada, Ottawa.
• Soil Survey Staff. Keys to soil taxonomy, 6th ed Washington, DC: U.S. Gov. Print. Office, 1994.
• Soil Survey Staff. Soil taxonomy. USDA-SCS Agric. Handb. 436. Washington, DC: U.S. Gov. Print. Office, 1975.
• Sparling J.H. Studies on the relationship between water movement and water chemistry in mires. Can. J. Bot. 1966;25:2227-2233.
• Stoner M.G., Ugolini F.C. Arctic pedogenesis: 2. Threshold-controlled subsurface leaching episodes. Soil Sci. 1988;145:46-51.
• Stuiver M., Pearson G.W. High-precision bidecadal calibration of the radiocarbon time scale, AD 1950–500 BC and 2500–6000BC. Radiocarbon 1993;35:1-24.[Web of Science]
• Topp G.C., Davis J.L., Annan A.P. Electromagnetic determination of soil water content: Measurements in coaxial transmission lines. Water Resour. Res. 1980;16:574-582.
• Topp, G.C., S.J. Zegelin, and I. White. 1994. Monitoring soil water content using TDR: An overview of progress. p. 67–80. In Symp. and Worksh. on Time Domain Reflectometry in Environmental, Infrastructure, and Mining Applications. Evanston, IL. 7–9 Sept. 1994. U.S. Dep. Interior Bureau of Mines, Spec. Pub. Sp 19–94.
• Ugolini F.C., Mann D.H. Biopedological origin of peatlands in southeast Alaska. Nature (London) 1979;281:366-368.
• Wang C., Beke G.J., McKeague J.A. Site characteristics, morphology and physical properties of selected Ortstein soils from the Maritime Provinces. Can. J. Soil Sci. 1978;58:405-420.
• Wells E.D., Pollett F.C. Peatlands. In: South R., ed. Biogeography and ecology of the island of Newfoundland. The Hague: Dr. W. Junk Publ, 1983:207-265.
• Wetzel R.G. Limnology. New York: Saunders College Publ, 1983.
• Wilding L.P., Smeck N.E., Drees L.R. Silica in soils: Quartz, cristobalite, tridymite, and opal. In: Dixon J.B., Weed S.B., eds. Minerals in soil environments. 1st ed. SSSA Book Ser. 1. Madison, WI: SSSA, 1977:471-552.

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Web of Science (6) Citing Articles via Google Scholar Google Scholar Articles by Lapen, D. R. Articles by Wang, C. Search for Related Content PubMed Articles by Lapen, D. R. Articles by Wang, C. GeoRef GeoRef Citation Agricola Articles by Lapen, D. R. Articles by Wang, C.