Four decades of continuously applied tillage or no-tillage on soil properties and soil morphology

Abstract

As increasing amounts of cropland are managed using no-tillage (NT), information is needed to assess long-term impacts of this practice on soil profile properties. A well-drained Wooster fine-loamy (mixed, active, mesic, Oxyaquic Fragiudalf) soil and a poorly drained, Hoytville silty clay loam (fine, illitic, mesic Mollic Epiaqualf) were sampled. For comparison, adjacent undisturbed forested and grassed areas were sampled. Bulk samples were characterized using physicochemical, mineralogical, and micromorphological methods. At both sites stronger structure and more bioturbation was evident in the topsoil of NT than in PT leading to lower bulk density values and increased, highly connective macroporosity. Evidence for the formation of incipient E horizons was noted in the lower A horizons of soil in the NT plots. The cation exchange capacity (CEC) was increased at both sites with NT and is associated with the increase in soil organic C. The C/N ratio of the NT pedon was closer to that of the A horizon of the forest pedon than to that of the PT pedon. Comparisons of C levels with those estimated in the same soils prior to establishment of the grassed areas or tillage plots suggest that at the Wooster site the grass, NT, and PT pedons all sequestered C. Only the grass pedon sequestered C at the Hoytville site. For both sites, and especially for the well-drained Wooster silt loam soil, continuous, long-term NT management can sustain or even enhance soil functions as compared with long-term PT management.

Abbreviations

BD

bulk density

CC

continuous corn

CEC

cation exchange capacity

NT

no tillage

OC

organic carbon

PT

plow tillage

XRD

X-ray diffraction

1 INTRODUCTION

Soil structure plays a major role in determining the physical characteristics of a soil (Lin et al., 1999a, 1999b). It influences the ability of topsoil to withstand natural and anthropogenic erosion processes (Zucca et al., 2006) and controls porosity, particularly macroporosity (Stoops, 2003), which in turn influences water flow and the movement of soluble and particulate constituents through the soil. When tillage is introduced into an agroecosystem, soil structure and associated properties are dynamically modified, and the soil processes controlled by these physical characteristics are altered (Bronick & Lal, 2005; Lin et al., 1999a, 1999b; Valette et al., 2006). The alterations from tillage not only affect soil physical properties but negatively affect chemical and biochemical processes, which have been shown to reduce the overall soil health of an agroecosystem (Nunes et al., 2018).

Living soil organisms facilitate the creation of new aggregates and participate in the reconfiguration of old peds in a continuous feedback process (De Gryze et al., 2006; Qin et al., 2006). Old root channels and faunal tubes such as earthworm burrows create macroporosity, stimulate macroaggregation (Shipitalo et al., 2002; Velasquez et al., 2007), and aid in the development of granular and crumb structure in A horizons (Stoops, 2003).

Soil disturbance from tillage can weaken aggregation and reduce soil microbial community structure and function (Alhameid et al., 2017; Kumar et al., 2012). Tillage can lead to deterioration of structure due to disturbance, and eventually the creation of a plow layer, resulting in constraints for root growth related to the presence of a dense pan (Singer, 2006). No-tillage (NT) also can initially lead to unfavorable conditions for rooting of some crops, due primarily to compaction (Qin et al., 2006; Shipitalo & Protz, 1987; Singer, 2006) and differential nutrient distribution with depth. However, compaction in NT systems can be partially counteracted by the formation of a surface soil layer rich in organic matter and nutrients (Dick, 1983; Guzman et al., 2006; Lal & Jarecki, 2005; Six et al., 2000).

The effects of tillage do not only influence soil structure, but also the dynamics of chemical characteristics such as concentrations of organic C (OC), N, and P. This is due to the intensity by which tillage breaks up aggregates and mixes soil, fertilizers and residues (Alvarez et al., 1998; Dick, 1983; Valette et al., 2006). Organic C accumulation under NT occurs, especially in the surface soil layer, because of deposition of plant residues on the soil surface and less mixing into the soil matrix. This leads to reduced aeration and soil temperatures (Qin et al., 2006; Shipitalo et al., 2002; Shukla et al., 2003; Six et al., 2000). In NT, roots and exudates decompose in situ, thus affecting soil nutrient cycling (Barré et al., 2018).

Although impacts of continuous NT cultivation on soil have been previously reported, few studies exist where NT management practices have been continuously maintained for 40 or more years. With renewed interest in how reducing tillage may affect soil properties such as soil OC, questions naturally arise as to the long-term environmental and economic impacts of continuous NT application. The Triplett–Van Doren plots, located across Ohio, provide a unique opportunity to investigate these impacts under a continuous corn (CC, Zea mays L.) cropping regime. These plots were established in the early 1960s at various geographical sites in Ohio with soils of contrasting drainage classes and characteristics (Dick et al., 1986a, 1986b, 1991). Depending on the site in Ohio, the tillage and crop rotation treatments on these plots have been continuously maintained for 43 or 44 yr. In the current study, adjacent forest and grassed areas were included as a reference.

The objectives of this study were to document the different effects of long-term NT or plow till (PT) cultivation on two contrasting Ohio soils related to (a) chemical and physical soil properties and (b) soil aggregate characteristics using macro- and micromorphological observation methods.

Core Ideas

• Continuous no-tillage resulted in greater organic C in surface soil compared with plow tillage.
• Accumulation of soil organic C under no-tillage started at the surface and increased downward.
• No-tillage did not (Hoytville) or did (Wooster) increase organic C compared to initial levels.
• No-tillage, compared with plow-tillage soils, exhibited strong structure and lower bulk density.
• Biological mixing in no-tillage soils led to highly connective porosity compared with plow- tillage.

2 MATERIALS AND METHODS

2.1 Site and plot descriptions

Soil pits were excavated at two long-term experimental sites in Ohio where a tillage and rotation experiment had been maintained for 43 or 44 yr. One site is located in northwest Ohio (41°00′ N, 84°00′ W), on a Hoytville silty clay loam (fine, illitic, mesic, Mollic Epiaqualf) that developed in glacial-lacustrine deposits (glacial till reworked by wave action on a nearly level lake plain) (USDA-SCS, 1973). Hereafter it is referred to as the Hoytville site. The mean annual air temperature at Hoytville is 9.9 °C, and moderate amounts of precipitation occur during the year with mean annual precipitation (MAP) being 845 mm, primarily in the warm season (Dick et al., 1986a, 1991). The other site is located in northeast Ohio (40°48′ N, 82°00′ W) and is co-dominated by soils of the Wooster (fine-loamy, mixed, active, mesic, Oxyaquic Fragiudalf) and Riddles (fine-loamy, mixed, active, mesic Typic Hapludalf) series, with the Wooster soil predominating in the area of the plots and the pits (USDA-NCSS, 2011). Hereafter it is referred to as the Wooster site. The mean annual temperature at Wooster is 9.1 °C with an average of 173 frost-free days annually and a MAP of 905 mm (Dick et al., 1986a, 1986b, 1991). The amount of precipitation falling during the growing season at these Ohio sites is generally inadequate for maximum crop yields due to frequent drought stress during the months of June, July, and August (Dick et al., 1986b, 1991). Rainfall events that occur can be erosive due to the high intensity of the storms that lead to excessive water and soil runoff, especially at the Wooster site (Dick et al., 1991).

At the Hoytville site, tile drainage systems or open ditches were constructed to facilitate the removal of excess water from this poorly drained soil, particularly during winter and spring (Dick et al., 1986a, 1991). The original vegetation of the region was a deciduous swamp forest, which was drained for agriculture over 140 yr ago (USDA-SCS, 1973). Wooster site soils, by contrast, are well-drained upland soils with a fragipan at 50–90 cm (USDA-SCS, 1984). The parent material is low-lime glacial till with a thin, discontinuous loess mantle of up to 51-cm thickness. The native vegetation was a continuous cover of hardwood forest (red, white, and black oak). Relic forest remnants occur along creeks and in small lots nearby (USDA-SCS, 1984).

The plots sampled at both the Hoytville and Wooster sites have the same experimental design. The design is a factorial experiment arranged in randomized complete blocks where tillage treatments (moldboard PT and NT) are combined with three crop rotations. Records of management practices of these plots can be found in Dick et al. (1986a, 1986b) and in electronic form (Dick, 1997a, 1997b, 2013). These sites represent more than 40 yr of replication of treatments.

Plow tillage involved a moldboard plow treatment in the spring at Wooster and in the autumn at Hoytville to a depth of 20–25 cm, and generally two or more secondary tillage operations (disk) applied to a depth of about 10 cm for seedbed preparation prior to planting. At least one disking operation was used each year. No-tillage involved planting directly into the previous year's residues with a coulter-type NT planter, without any other kind of topsoil disturbance. Fertilizer, lime, and pesticide application amounts to the plots were the same for both NT and PT at each location, but the application method differed. These materials were broadcast on the surface for NT and incorporated into the soil with PT. For a more detailed description of the agronomic practices consult Dick et al. (1986a, 1986b, 2013) and (Dick, 1997a, 1997b).

2.2 Soil sampling and characterization

For the purposes of this study, only the CC rotation was sampled. The Hoytville site was sampled in July and the Wooster site in August by excavating four pits with a backhoe at plot locations (near edges of plots) where crop yields are not taken. The soil profiles were cleaned by hand (shovel and knife). At both sites, two pits were placed in the long-term plots under CC with one pit each in a PT plot and NT plot. Pits were also excavated in a grassed alley adjacent to the plot areas, and in a nearby forest area. The soil in the grassed alley areas had not been disturbed for at least 50 yr and the soil in the forest areas had never been tilled, although some harvesting of trees had occurred at least 50 yr previously, primarily using hand tools and no heavy equipment traffic. Only one pit was excavated to generate the minimum possible disturbance to the long-term plots.

Soil profile descriptions were recorded by using the horizon designations and description nomenclature provided by the Soil Survey Manual (SSS, 1993). These descriptions included structure, field texture, consistency, color (on wet and dry specimens), roots, and the abundance, distribution and color of mottles, clay films and carbonates, when present, following the NRCS Field Book (Schoeneberger et al., 2012).

A horizon-wise scheme was followed for soil sampling. Undisturbed soil samples were taken from the A horizon at each pit using tin cans to obtain bulk density (BD) measurements (cylinder method; USDA-NRCS, 2004). In order to describe changes in microstructure and microporosity, one undisturbed soil sample of each Ap horizon under NT and PT at each site was impregnated with Scotchcast epoxy resin #3, cured, and cut vertically and horizontally to expose faces for thin section preparation (USDA-NRCS, 2004). The final cutting and polishing were done by Spectrum Petrographics. A Nikon UFX-II microscope with a FX-35A microphotography device was used to take pictures of the vertically oriented soil thin sections, and they were described using the terminology and concepts provided by Stoops (2003). To estimate porosity, 10 random transects were established in each sample at 2× magnification, and the porous space and aggregate distribution was described and porosity was measured along each transect.

Disturbed bulk soil samples of each horizon were also collected, air-dried, and analyzed by standard characterization procedures according to the manual of the National Soil Survey Laboratory (USDA-NRCS, 2004). Analyses included particle size distribution, total C (dry combustion with adsorption in ascarite), and total N (dry combustion and thermal detection) (USDA-NRCS, 2004). The C/N ratio was calculated as the quotient of total C and N in a given sample. The C pool (volumetric C content to a depth of 31 cm) was calculated from the weighted soil total C concentration and BD.

To compare tillage effects on C pool sizes in the soil profiles, a common depth was used for the tillage treatments at both research sites to ensure that we account for all A horizon OC. The surface profile layer thickness used in this study for C pool calculations was 31 cm, as this represented the depth of the thickest Ap horizon (i.e., the PT plot at the Wooster research site). Below this depth, we did not observe consistent tillage effects. Extractable bases in 1 M ammonium acetate, extractable acidity in triethanolamine-BaCl₂ solution, and pH (potentiometric method, 1:1 soil/water and 1:2 soil/CaCl₂) were also determined according to USDA-NRCS (2004). Cation exchange capacity (CEC) was calculated by summation (extractable acidity + extractable bases,) and base saturation was calculated by determining the sum of the extractable bases (Ca, Mg, Na, and K) as a percentage of the CEC. 

A complete mineralogical characterization was performed for selected horizons of the grassed pedons of both sites, because no changes in the mineralogical composition due to tillage were expected, just the main influence of the parent material and the processes of soil formation on the mineralogical suite. These analyses included total digestion with aqua regia (HF – H₃BO₃) in Parr bombs (Bernas, 1968), followed by total K determination by flame emission and total Fe determination by atomic absorption using a Varian spectrophotometer (SpectrAA-5). The extraction of Fe by the citrate–dithionate–bicarbonate procedure allowed for the determination of the Fe content in Fe oxides (Mehra & Jackson, 1960). Total surface area of Mg-saturated clays was determined when constant weight was reached in an atmosphere depleted in water vapor and previously exposed to ethylene glycol monoethyl ether (Cihacek & Bremner, 1979). The external surface area was obtained using oven-dried Mg-saturated clays and measuring the adsorption of a nonpolar gas (N₂) on the external surface (Brunauer–Emmett–Teller [BET] isotherms; Brunauer et al., 1938). Semiquantitative identification of clay minerals in dried films of Mg- and K-saturated clays was determined by the XRD (X ray diffraction) scanning technique (Whittig & Allardice, 1986) after the following treatments: Mg-saturated clays were exposed to 25 °C and to ethylene glycol, and K-saturated clays were exposed to 25, 350, and 550 °C. The source of radiation was a Cu tube operating at 35 kV and 20 mA, the scanning step was set at 4 s per 0.05° 2θ and the equipment used was a Philips diffractometer (Philips Electronic Instruments). The Mg-25 °C slides were scanned from 2 to 32° 2θ, whereas the other slides were scanned from 2 to 15° 2θ. Pedons were classified according to Soil Taxonomy (SSS, 2014) using field description and laboratory characterizations.

All of the data collected in this research are available in Mestelan (2008).

3 RESULTS AND DISCUSSION

3.1 Soil C and N dynamics

Total OC and total N concentrations decrease with depth at both sites (Tables 1 and 2), and contents near the surface can be attributed primarily to land use (Balesdent et al., 2000; Dick & Gregorich, 2004) and soil type (Dick & Gregorich, 2004; Lal & Jarecki, 2005). The primary variables in the four profiles studied at each site (i.e., forest, grass, and NT and PT corn plots) are tillage and vegetative cover. These variables result in differences in the amount, quality, seasonality and mode of incorporation of residues, which are factors responsible for sequestering OC in the soil (Balesdent et al., 2000; Dick & Gregorich, 2004; Zucca et al., 2006). The total C contents at the soil surface of the forest pedons at both experimental sites are higher than for the other pedons (Tables  1 and 2). The OC content in the Ap1 of the grassed Hoytville pedon and the Ap1 horizon of the NT Wooster pedon are, however, only somewhat less than in the A horizons of the forest pedons. The lowest OC concentrations in the uppermost A horizon occur in the PT plots at the Hoytville and Wooster sites.

TABLE 1. Physicochemical characterization of pedons at Hoytville

Soil source	Horizon	Thickness	Total OC^a	Total N^b	C/N ratio^c	BD^d	OC pool^e	N pool^f	Sand	Silt	Clay
		cm	g kg^−1			g cm^–3	kg m^–2		%
Forest	A1	0–7	71.1 (4.2)	5.9 (0.1)	12.0 (0.5)	0.77 (0.01)	11.9	1.10	11.3 (0.2)	46.4 (1.0)	42.3 (0.8)
	A2	7–13	58.7(3.3)	5.1 (0.1)	11.5 (0.2)	0.87 (0.05)			11.4 (0.0)	40.3 (0.7)	48.3 (0.7)
	AB	13–21	39.2 (0.4)	3.7 (0.1)	10.6 (0.3)	1.06 (0.05)			11.0 (1.0)	38.6 (0.8)	50.4 (1.8)
	Btg1	21–33	10.5 (0.8)	1.3 (0.1)	7.9 (<0.1)	1.54 (0.14)			11.4 (0.5)	39.6 (0.9)	49.1 (1.4)
	Btg2	33–45	7.2 (0.4)	1.1 (0.1)	6.8 (0.2)				10.1 (0.1)	39.7 (1.1)	50.2 (1.0)
	Btg3	45–58	3.3 (0.1)	0.7 (0.1)	4.4 (0.1)				10.0 (0.5)	43.1 (1.5)	47.0 (1.0)
Grass	Ap1	0–7.5	63.3 (3.7)	5.8 (0.1)	11.0 (0.5)	0.72 (0.02)	8.36	1.03	10.2 (0.2)	53.5 (1.2)	36.3 (1.0)
	Ap2	7.5–14	27.6 (1.0)	3.0 (0.1)	9.0 (<0.1)	1.17 (0.02)			13.6 (0.4)	49.0 (1.7)	37.4 (1.3)
	Ap3	14–26	20.5 (0.8)	2.2 (0.1)	9.5 (0.1)	1.47 (0.16)			12.7 (0.4)	50.6 (0.7)	36.7 (0.3)
	BAg	26–34	9.9 (0.2)	1.4 (0.1)	7.0 (0.5)	1.50 (0.20)			14.4 (0.1)	48.1 (1.8)	37.5 (1.9)
	Btg1	34–50	5.6 (0.2)	1.0 (0.1)	5.8 (0.3)				14.1 (0.5)	45.6 (1.7)	40.3 (1.2)
No till	Ap1	0–5	38.9 (2.7)	3.3 (0.2)	11.6 (0.3)	1.01 (0.06)	7.35	0.87	15.9 (0.5)	52.6 (1.2)	31.5 (0.7)
	Ap2	5–11	19.9 (1.8)	2.3 (0.1)	8.5 (0.2)	1.25 (0.11)			14.6 (0.3)	49.7 (1.2)	35.7 (0.9)
	Ap3	11–21	14.9 (0.8)	2.0 (0.1)	7.5 (0.2)	1.47 (0.11)			14.1 (0.2)	49.0 (1.0)	36.9 (0.8)
	BAg	21–36	8.9 (0.5)	1.3 (0.1)	6.8 (0.2)	1.55 (0.14)			12.5 (0.4)	47.4 (0.9)	40.1 (0.5)
	Btg1	36–50	4.7 (0.1)	0.8 (0.1)	5.6 (0.2)				12.3 (0.3)	45.0 (0.8)	42.7 (1.1)
Plow till	Ap1	0–5	19.7 (0.6)	2.2 (0.2)	8.91 (0.2)	1.20 (0.13)	6.34	0.96	12.4 (0.3)	49.5 (1.0)	38.1 (0.7)
	Ap2	5–14	18.6 (0.4)	2.3 (0.2)	8.23 (0.2)	1.37 (0.29)			13.4 (0.4)	51.8 (0.9)	34.8 (0.5)
	Ap3	14–27	15.3 (0.5)	2.2 (0.2)	7.04 (0.2)	1.67 (0.18)			15.1 (0.2)	47.3 (0.7)	37.6 (0.5)
	BAg	27–36	9.3 (0.3)	1.2 (0.1)	7.74 (0.3)	1.53 (0.17)			12.9 (0.4)	45.9 (.8)	41.2 (0.4)
	Btg1	36–47	5.1 (0.3)	0.9 (0.1)	5.92 (0.2)				12.9(0.2)	43.9 (1.0)	43.2 (0.8)
	Btg2	47–66	3.2 (0.2)	0.7 (0.1)	4.63 (0.2)				13.6 (0.3)	43.5 (0.8)	42.9 (0.5)

• Note. Standard deviation is provided in parentheses.
• ^a Total OC, total soil organic C (dry combustion). 
• ^b Total N, total soil N (dry combustion).
• ^c Quotient of the OC soil content to the total N in the sample.
• ^d BD, bulk density.
• ^e OC pool, organic C pool or volumetric C contents as the result of combining total OC, BD, and thickness of a given horizon, summarized up to a 31-cm depth.
• ^f N pool, N pool calculations identical to the C pool calculations but using total N values instead.

TABLE 2. Physicochemical characterization of pedons at Wooster

Soil source	Horizon	Thickness	Total OC^a	Total N^b	C/N ratio^c	BD^d	OC pool^e	N pool^f	Sand	Silt	Clay
		cm	g kg^–1			g cm^–3	kg m^–2		%
Forest	A1	0–4	37.9 (2.3)	2.7 (0.3)	14.1 (1.0)	1.15 (0.05)	6.72	0.57	14.1 (0.5)	71.6 (1.0)	14.3 (0.5)
	A2	4–13	28.7 (1.6)	2.2 (0.3)	12.9 (0.9)	1.28 (0.07)			13.2 (0.5)	73.3 (1.0)	13.5 (0.4)
	E	13–28	8.0 (0.6)	0.9 (0.1)	9.2 (0.4)	1.25 (0.05)			17.0 (0.3)	72.7 (0.8)	10.3 (0.5)
	BE	28–41	3.9 (0.3)	0.7 (0.1)	6.0 (0.4)	1.64 (0.12)			16.0 (0.4)	69.3 (0.7)	14.7 (0.3)
	Bt1	41–62	3.3 (0.4)	0.6 (0.1)	5.3 (0.3)				14.5 (0.2)	62.7 (0.5)	22.8 (0.3)
Grass	Ap1	0–4	30.8 (0.8)	3.3 (0.2)	9.3 (0.3)	1.32 (0.04)	7.45	0.84	11.8 (1.0)	72.0 (1.2)	16.2 (0.2)
	Ap2	4–10	27.2 (0.6)	2.8 (0.2)	9.8 (0.5)	1.42 (0.03)			13.6 (1.3)	71.1 (1.5)	15.3 (0.2)
	Ap3	10–20	16.7 (0.5)	1.9 (0.3)	8.7 (1.2)	1.54 (0.03)			13.8 (1.4)	70.1 (1.5)	16.1 (0.1)
	Bt1	20–33	5.3 (0.4)	0.8 (0.2)	6.8 (1.2)	1.63 (0.14)			17.5 (0.8)	65.7 (1.3)	16.8 (0.5)
	Bt2	33–51	2.0 (0.3)	0.5 (0.1)	3.9 (0.2)				21.1 (0.3)	56.9 (0.7)	22.0 (0.4)
No till	Ap1	0–3	34.7 (0.6)	3.2 (0.1)	10.9 (0.2)	1.49 (0.03)	6.75	0.71	13.8 (0.7)	72.5 (1.1)	13.7 (0.4)
	Ap2	3–10	23.0 (0.5)	2.3 (0.2)	9.8 (0.7)	1.57 (0.06)			15.5 (0.6)	69.4 (1.2)	15.1 (0.6)
	Ap3	10–22	10.8 (0.7)	1.2 (0.2)	8.8 (0.9)	1.53 (0.07)			15.8 (0.2)	69.2 (0.6)	15.0 (0.4)
	BA	22–33	4.6 (0.4)	0.6 (0.1)	7.6 (0.6)	1.70 (0.10)			21.6 (0.3)	61.0 (0.8)	17.4 (0.5)
	Bt1	33–48	2.3 (0.5)	0.4 (0.1)	5.4 (0,2)				25.5 (0.3)	52.2 (0.7)	22.3 (0.4)
	Bt2	48–77	1.5 (0.2)	0.5 (0.1)	3.0 (0,2)				29.0 (0.4)	43.5 (1.0)	27.5 (0.6)
Plow till	Ap1	0–3	12.3 (0.6)	1.2 (0.1)	10.4 (1.2)	1.58 (0.04)	5.56	0.62	22.4 (0.8)	64.1 (1.2)	13.5 (0.4)
	Ap2	3–17	11.2 (0.6)	1.2 (0.2)	9.2 (0.9)	1.67 (0.05)			20.2 (0.7)	65.0 (0.9)	14.8 (0.2)
	Ap3	17–31	10.2 (0.5)	1.2 (0.2)	8.4 (1.0)	1.65 (0.08)			23.7 (0.6)	63.9 (0.8)	12.4 (0.2)
	Bt1	31–44	1.9 (0.3)	0.5 (0.1)	3.7 (0.2)	1.65 (0.07)			26.1 (0.4)	57.7 (0.7)	16.2 (0.3)
	Bt2	44–57	1.3 (0.3)	0.5 (0.1)	2.5 (0.2)				28.2 (0.1)	52.7 (0.5)	19.1 (0.4)

• Note. Standard deviation is provided in parentheses.
• ^a Total OC, total soil organic C (dry combustion). 
• ^b Total N, total soil N (dry combustion).
• ^c Quotient of the OC soil content to the total N in the sample.
• ^d BD, bulk density.
• ^e OC pool, organic C pool or volumetric C contents as the result of combining total OC, BD, and thickness of a given horizon, summarized up to a 31-cm depth
• ^f N pool, N pool calculations identical to the C pool calculations but using total N values instead.

A comparison of the OC pools (Table 1) for the various profiles confirms that the forest pedon contained the most C followed by grass, NT, and PT at the Hoytville site in accordance with patterns reported in other research works (Balesdent et al., 2000; Hopmans et al., 2005; Priano et al., 2017). The greater OC amounts stored in the forest pedons is due to the accumulation of litter on the soil surface (Balesdent et al., 2000; Hopmans et al., 2005; McColl & Gressel, 1995), and in situ incorporation of dead roots and faunal transport of residues (Balesdent et al., 2000; McColl & Gressel, 1995). At the Wooster site (Table 2), the grassed pedon had the greatest C pool, followed by the forest and NT pedons (no difference between these two pedons), and the PT pedon had the smallest C pool. At the Wooster site, the sampled forest area is at the edge of a narrow, forested area and may have lost some litter due to removal by wind. The grassed areas at both sites may also have received some lime and fertilizer inputs that could have improved plant biomass productivity.

Overall, the values of the C pools are higher in the pedons of the Hoytville series than in the pedons of the Wooster series. This is, in part, explained by the native C pool sizes at the two sites. The Hoytville pedons had higher native OC contents because they formed in a poorly drained landscape and in fine textured parent materials (Collins et al., 1999; Dick, 1983; Dick & Gregorich, 2004). Collins et al. (1999) reported a pool size value of 5.98 kg m^–2 to a depth of 20 cm at the Hoytville site prior to any application of tillage treatments in 1964. This value reflects the effect of previous cultivation (Collins et al., 1999).

We also used estimations of BD for the 20-to-31-cm layer, the pedotransfer function developed by Calhoun et al. (2001) for Ohio soils, and data for the C content of the 20-to-31-cm depth of samples collected in the soil survey of this soil before the experiment was established to calculate the original total OC pool. The result was that we estimated a pool value of 7.50 kg m^–2 as the total OC pool present to a depth of 31 cm in the Hoytville soil immediately prior to establishment of the PT and NT plots. To the same depth, the soil C pool in the Hoytville pedons 43 yr later was 11.9, 8.36, 7.35, and 6.34 kg m^–2 for forest, grassed, NT, and PT pedons, respectively (Table 1). That indicates a loss of OC at a rate of 3.33 g m^–2 yr^–1 for NT and 27.3 g m^–2 yr^–1 for PT. The loss of C under PT is almost eight times greater than under NT. At the beginning of the fourth experimental decade, only 1.87% of the OC content in the cultivated area just prior to the establishment of the plots was lost under NT, but 15.4% was lost under PT. It must be noted that pedon sampling does not imply statistical replication that accounts for spatial OC\ variability.

 At the Wooster site, a C pool value of 4.40 kg m^–2 was calculated for the cultivated soils at the beginning of the experiment. Calculations for OC pool to a depth of 0–31 cm were performed as for the Hoytville site using soil survey information available for the Wooster experimental site. To the 0-to-31-cm depth, the soil C pool in the Wooster forest, grassed, NT, and PT pedons after more than four decades was 6.72, 7.45, 6.75, and 5.56 kg m^–2, respectively (Table 2). Since the establishment of the plots, C was sequestered at a rate of 57.6 g m^–2 yr^–1 for NT and 28.7 g m^–2 yr^–1 for PT.

Because BD values are needed in C pools calculations, and BD values increased due to cultivation in the 0-to-31-cm depth, it is possible that the C pool value increases in both soils and for both tillage treatments may be a function of changes in bulk densities rather than increases in OC. Guzman et al. (2006) found no significant change in the amount of C (mass basis) in the upper 15 cm of soil in Kansas under continuous PT, although continuous application of NT resulted in an increase of the amount of OC. Finally, modifications in the soil climate such as reduced O₂ diffusion and reduced number of wet–dry cycles may lead to reduced OC decomposition (Shipitalo et al., 2002; Shukla et al., 2003). The development of more stable aggregates may also contribute to increased soil C amounts in the forest, grass, and NT plots compared with the PT plots (Balesdent et al., 2000; Six et al., 2000; Six, Callewaert, et al., 2002; Six, Conant, et al., 2002). Values of OC loss with PT are similar to the range reported in Dick and Gregorich (2004), and somewhat low in comparison with values reported by Collins et al. (1999) for different long-term sites in the U.S. Corn Belt. However, even longer-term studies than this one would be needed to determine when, or if, a new equilibrium OC level has been reached in these soils.

Anthropogenic activity is believed to be responsible for the high total soil N values of the Ap horizons of the grass, NT, and PT pedons relative to the forest pedons at both sites (Tables 1 and 2). All of the cultivated pedons had narrower C/N ratios in the Ap horizons than those of the A horizons of the forest pedons (Hopmans et al., 2005). Nitrogen inputs via fertilizer (Guzman et al., 2006) and retention of stabilized organic N compounds in the soil may explain the differences in N contents between managed agroecosystems and the native forest areas. The C/N ratios of the Ap horizons under NT tend to stabilize at higher values than those under grass or PT pedons, suggesting that greater N losses (gaseous processes or leaching via percolation) occur with NT than with grass cover or PT (Brye et al., 2001; Jacinthe et al., 2002; Martens & Dick, 2003). Alternatively, the added C in the plant residues is not respired as quickly or as completely under NT as compared with PT. However, Alvarez et al. (1998) reported that NT generated more light density and intermediate density OC fractions that were enriched with N but showed high mineralization rates in in vitro tests, suggesting that N in these fractions may have high turnover rates. Alternatively, Soon et al. (2001) reported greater organic N mineralization under NT compared with PT in a coarse-loamy, Typic Cryoboralf in Alberta. Nitrogen retention by clay silicates prevalent in B horizons may contribute to narrower C/N ratios with depth (Nommik & Vahtras, 1982; Young & Aldag, 1982). Hoytville pedons with large amounts of illites and vermiculites (Table 3) may fix or retain significant amounts of N compared with the Wooster pedons that are dominated by interstratified minerals and kaolinites (Table 3) (Nommik & Vahtras, 1982; Young & Aldag, 1982).

TABLE 3. Mineralogical characterization of soils under grass at Hoytville and Wooster

Site	Horizon^a	Thickness	Clay fraction composition^b	Clay-CEC^c	Total K^d	Iron in clays^e	Surface area^f
							Total	Internal
		cm		cmol ^+ kg^–1 clay	% K2O	% Fe2O3	m^2 g^–1
Hoytville	Ap1	0–7.5	Micas/Illites, Verm; Kao, Qtz	29.3	4.37	2.06	161		111
	Btg2	50–68		29.4	4.17	6.47	180		115
	Cx	+160		27.6	4.21	4.24	108		58.1
Wooster	Ap1	0–4	Interstratified, Kao, Qtz	36.7	3.24	4.38	90.2		38.2
	Bt1	20–33		25.9	2.74	6.07	147		94.6
	Btx	76–109		21.0	3.62	7.20	210		136
	C	+146		30.9	3.21	9.05	204		132

• ^a Only selected horizons of the pedon under grass at both sites (Hoytville and Wooster) were analyzed.
• ^b Determined by X-ray diffraction (XRD) analysis of the clay fraction. Verm, Vermiculite; Kao, kaolinite; Qtz, quartz. Minerals are mentioned in decreasing order of abundance.
• ^c CEC, cation exchange capacity of Ca-saturated clays.
• ^d Wet digestion in aqua regia–HF–H₃BO₃ with flame emission quantification.
• ^e Extractable Fe in CBD (citrate–bicarbonate–dithionate extracts).
• ^f Total surface area of clays by the ethylene glycol monoethyl ether (EGME) technique and internal area as determined by N₂ adsorption.

Overall, the Hoytville site had greater total N amounts, and the N pool in the forest area at Hoytville was particularly high as compared with the same vegetation at Wooster, and both values were in the range observed by Marty et al. (2015). These observations are attributed to the mineralogical richness of the Hoytville site. The forest area at Wooster seems to be rather limited in its OC accretion capacity due to its N status, according to Marty et al. (2015). At both places the grass had a higher N pool than the other agricultural uses (NT and PT), suggesting diminished nitrate leaching and potential denitrification (Brye et al., 2001; Jacinthe et al., 2002; Martens & Dick, 2003).

3.2 Other soil chemical properties

Soil reaction and CEC are also impacted by management at the Hoytville and Wooster research sites. The content of extractable Ca and extractable Mg on the exchange sites and on pH values in the upper 51–57 cm of the grassed, NT, and PT pedons relative to those in the forested pedon at the Wooster site reflect the fact that the cultivated pedons received periodic liming applications (Table 4). The chemistry of the forest pedons is characteristic of native conditions at both sites (Tables 4 and 5). In the NT pedons at both research sites, some acidification is observed in the Ap2 and Ap3 horizons relative to the Ap1 with pH values decreasing to close to those of the forest pedon. These pH values are accompanied by decreases in exchangeable bases and base saturation and increased extractable acidity (Tables 4 and 5). This suggests that even though NT may benefit from surface applications of lime (Mengel & Ruiz, 2013), the benefits of liming seem primarily restricted to the Ap1 horizon. Below the Ap1 horizon in NT pedons, the impact of lime applications is limited due to broadcast surface application and the lack of physical incorporation of the lime into underlying horizons by tillage. Additions of fertilizers containing ammonium also increases soil acidity, especially at the surface of the NT soil, as a result of nitrification (Guzman et al., 2006). Similarly, after four decades of NT land use, Lal and Jarecki (2005) reported a pH decrease for the 0-to-15-cm layer in a well-drained silt loam Ohio soil.

TABLE 4. Chemical characterization of pedons at Wooster

						NH4Oac-extractable bases^c
Soil source	Horizon	Thickness	pH^a water	pH CaCl2	BaCl2–EA^b acidity	Ca	Mg	K	Na	CEC^d	BS^e
		cm	1:1	1:2	cmol+ kg^–1						%
Forest	A1	0–4	5.7 (<0.1)	5.5 (<0.1)	11.2 (0.2)	10.3 (0.9)	2.0 (<0.1)	0.42 (<0.1)	0.02 (0.00)	23.9 (0.2)	53.2
	A2	4–13	5.4 (<0.1)	5.1 (<0.1)	11.3 (0.2)	7.4 (0.1)	1.4 (0.1)	0.31 (<0.1)	0.02 (0.00)	20.5 (0.2)	44.9
	E	13–28	4.4 (0.1)	4.0 (0.1)	10.0 (0.1)	0.9 (0.1)	0.3 (<0.1)	0.10 (0.0)	0.02 (0.00)	11.3 (0.1)	11.5
	BE	28–41	4.2 (<0.1)	3.8 (<0.1)	10.9 (0.2)	0.5(0.0)	0.2 (<0.1)	0.11 (0.0)	0.02 (0.00)	11.8 (0.1)	7.5
	Bt1	41–62	4.1 (0.1)	3.7 (0.1)	13.3 (0.1)	0.9 (0.1)	0.5 (<0.1)	0.10 (0.0)	0.03(<0.01)	14.9 (0.1)	10.7
Grass	Ap1	0–4	5.2 (0.1)	5.1 (0.1)	8.9 (0.4)	7.3 (1.0)	2.3 (0.1)	0.52 (<0.1)	0.02 (0.00)	19.1 (0.1)	53.3
	Ap2	4–10	5.2 (0.1)	5.0 (0.1	8.1 (0.1)	7.1 (0.5)	2.2 (0.1)	0.25 (<0.1)	0.02 (0.00)	17.7 (0.1)	54.2
	Ap3	10–20	5.6 (0.1)	5.4 (0.1)	6.2 (<0.1)	6.1 (0.2)	2.0 (0.0)	0.24 (<0.1)	0.02 (0.00)	14.6 (0.1)	57.5
	Bt1	20–33	6.2 (0.1)	5.8 (0.1)	4.5 (<0.1)	4.8 (0.1)	2.2 (<0.1)	0.14 (<0.1)	0.03 (<0.01)	11.7(0.1)	61.4
	Bt2	33–51	6.1 (0.1)	5.7 (0.1)	4.1 (<0.1)	4.6 (0.1)	2.7 (0.0)	0.13 (0.0)	0.04 (<0.01)	11.6 (0.1)	64.3
No till	Ap1	0–3	5.5 (0.1)	5.5 (0.1)	10.9 (0.1)	7.4 (0.2)	2.2 (<0.1)	1.21 (0.2)	0.02 (0.00)	21.8 (0.2)	49.9
	Ap2	3–10	5.7 (0.1)	5.4 (0.1)	8.6(0.1)	6.2 (0.2)	1.9 (0.0)	0.53 (0.1)	0.02 (0.00)	17.3 (0.1)	50.2
	Ap3	10–22	5.9 (0.1)	5.8 (0.1)	6.5 (0.1)	4.8 (0.1)	1.8 (<0.1)	0.37 (<0.1)	0.02 (0.00)	13.5 (0.1)	51.8
	BA	22–33	6.5 (0.2)	6.3 (0.1)	5.1 (0.1	3.5 (0.0)	1.6 (0.0)	0.22 (0.0)	0.03 (<0.01)	10.4 (0.1)	51.1
	Bt1	33–48	6.3 (0.1)	6.0 (0.1)	7.3 (0.1	4.8 (0.1)	2.5 (<0.1)	0.20 (<0.1)	0.03 (0.00)	14.8 (0.1)	50.5
	Bt2	48–77	5.5 (0.1)	5.1 (0.1)	6.9 (0.1)	5.6 (0.1)	3.2 (0.1)	0.20 v	0.04(<0.01)	16.0 (0.1)	56.9
Plow till	Ap1	0–3	6.0 (0.1)	5.8 (0.1)	5.9 (0.1)	4.8 (0.1)	1.6 (0.0)	0.61 (<0.1)	0.01 (0.00)	13.0 (0.1)	54.5
	Ap2	3–17	6.1 (0.1)	6.0 (0.2)	5.8 (0.1)	5.5 (0.2)	1.8 (0.1)	0.39 (<0.1)	0.02 (0.00)	13.5 (0.1)	56.9
	Ap3	17–31	6.7 (0.1)	6.5 (0.1)	5.1 (0.1)	5.8 (0.1)	2.0 (0.1)	0.46 (<0.1)	0.02 (0.00)	13.3 (0.1)	62.1
	Bt1	31–44	6.8 (0.1)	6.4 (0.1)	5.2 (0.1)	3.5 (<0.1)	1.8 (0.0)	0.26 (<0.1)	0.02 (0.00)	10.8 (<0.1)	52.2
	Bt2	44–57	6.5 (0.1)	6.1 (0.1)	5.5 (0.1)	3.7 (<0.1)	2.1 (0.0)	0.21 (0.0)	0.03 (0.00)	11.6 (<0.1)	52.5

• Note. Standard deviation is provided in parentheses.
• ^a Potentiometric method; 1:1 solid/extractant (water) or 1:2 (CaCl₂) relationship.
• ^b Extractable acidity in BaCl₂–triethanolamine
• ^c Cations extracted in NH₄–acetate.
• ^d Summation method.
• ^e BS, base saturation; the ratio of the sum of basic cations to the cation exchange capacity, expressed as a percentage.

TABLE 5. Chemical characterization of pedons at Hoytville

						NH4Oac-extractable bases^c
Soil source	Horizon	Thickness	pH^a water	pH^a CaCl2	BaCl2–TEA^b acidity	Ca	Mg		K	Na	CEC^d	BS^e
		cm	1:1	1:2	cmol+ kg^–1							%
Forest	A1	0–7	6.1 (0.1)	6.1 (0.1)	12.6 (0.4)	35.6 (1.8)	3.5 (0.2)		0.4 (<0.1)	0.06 (<0.01)	52.2 (1.5)	75.8
	A2	7–13	6.1 (0.1)	6.1 (0.1)	12.5 (0.3)	31.1 (1.5)	2.8 (0.1)		0.2 (<0.1)	0.05 (<0.01)	46.7 (1.0)	73.2
	AB	13–21	6.0 (0.1)	5.7 (0.1)	11.8 (0.2)	24.9 (1.0)	2.3 (0.1)		0.2 (<0.1)	0.04 (<0.01)	39.3 (0.7)	70.0
	Btg1	21–33	5.7 (<0.1)	5.3 (<0.1)	8.8 (0.1)	16.2 (0.3)	2.1 (0.1)		0.2(<0.1)	0.06(<0.01)	27.5 (0.2)	67.9
	Btg2	33–45	6.4 (<0.1)	6.1(<0.1)	7.5 (0.1)	16.4 (0.3)	2.0 (0.0)		0.2 (0.0)	0.07 (<0.01)	26.2 (0.2)	71.4
	Btg3	45–58	6.9 (<0.1)	6.6 (<0.1)	6.6 (0.1)	16.0 (0.1)	2.0 (0.0)		0.2 (0.0)	0.09 (<0.01)	25.0 (0.1)	73.5
	Btg4	58–78	6.9 (<0.1)	6.6 (<0.1)	5.41 (0.2)	16.5 (0.1)	1.9 (0.0)		0.3 (0.0)	0.08 (<0.01)	24.2 (0.1)	77.7
Grass	Ap1	0–7.5	6.0 (0.1)	6.0 (0.1)	13.3 (0.2)	18.5(0.3)	5.9 (0.2)		1.0 (0.1)	0.03 (0.00)	38.7 (0.2)	65.7
	Ap2	7.5–14	5.9 (0.2)	5.7 (0.1)	9.2 (0.1)	17.6 (0.1)	3.9 (0.1)		0.5 (<0.1)	0.03 (0.00)	30.9 (0.1)	71.5
	Ap3	14–26	6.1 (0.1)	5.9 (0.1)	8.8 (0.2)	18.8 (0.3)	3.5 (0.1)		0.3 (0.0)	0.04 (0.00)	29.2 (0.2)	77.8
	Bag	26–34	6.2 (<0.1)	5.9 (<0.1)	6.5 (0.1)	17.6 (0.1)	2.9 (<0.1)		0.2 (0.0)	0.06 (0.00)	26.5 (0.1)	78.6
	Btg1	34–50	6.6 (<0.1)	6.2 (<0.1)	5.7 (<0.1)	16.2 (0.1)	2.5 (<0.1)		0.2 (0.0)	0.07 (0.01)	23.8 (0.1)	79.3
No till	Ap1	0–5	6.0 (0.1)	5.9 (0.1)	8.4 (0.1)	13.2 (0.1)	5.2 (0.1)		0.9 (0.1)	0.02(0.0)	27.6 (0.1)	69.7
	Ap2	5–11	5.7 (0.1)	5.6 (0.1)	10.3 (0.3)	13.2 (0.2)	5.2 (0.3)		0.9 (0.1)	0.02 (0.0)	29.6 (0.2)	65.1
	Ap3	11–21	5.5 (0.1)	5.2 (0.1)	10.7 (0.2)	11.7 (0.1)	4.6 (0.1)		0.5 (<0.1)	0.02 (0.0)	27.5 (0.1)	61.1
	BAg	21–36	6.3 (<0.1)	5.9 (0.1)	7.0 (0.1)	11.7 (0.1)	4.0 (0.0)		0.2 (0.0)	0.03 (<0.01)	22.9 (0.1)	69.5
	Btg1	36–50	6.7 (<0.1)	6.6 (0.1)	5.9 (<0.1)	15.7 (0.2)	3.7 (0.0)		0.2 (0.0)	0.04 (<0.01)	25.5 (0.1)	76.7
Plow till	Ap1	0–5	6.0 (0.1)	5.8 (0.2)	8.0 (0.1)	14.7 (0.1)	2.6 (0.2)		0.3 (0.1)	0.03(<0.01)	25.7 (0.1)	68.9
	Ap2	5–14	6.1 (0.1)	5.6 (0.1)	8.7 (0.1)	14.8 (0.1)	2.5 (0.1)		0.3 (0.0)	0.03 (0.00)	26.3 (0.1)	66.9
	Ap3	14–27	6.5 (<0.1)	6.0 (<0.1)	6.1 (<0.1)	12.2 (0.1)	2.3 (0.1)		0.3 (0.0)	0.04 (0.00)	20.9 (0.1)	70.8
	BAg	27–36	6.9 (<0.1)	6.4 (<0.1)	7.6 (<0.1)	16.1 (0.2)	2.8 (0.1)		0.2 (0.0)	0.04 (<0.01)	26.7 (0.1)	71.5
	Btg1	36–47	7.3 (<0.1)	6.8 (<0.1)	6.4 (<0.1)	15.5 (0.2)	3.1 (0.0)		0.2 (0.0)	0.07 (0.01)	25.2 (0.1)	74.7
	Btg2	47–66	7.5 (<0.1)	7.0 (<0.1)	6.4 (<0.1)	14.7 (0.2)	2.6 (0.0)		0.2 (0.0)	0.07 (0.01)	24.0 (0.1)	73.1

• Note. Standard deviation is provided in parentheses.
• ^a Potentiometric method; 1:1 solid/extractant (water) or 1:2 (CaCl₂) relationship.
• ^b Extractable acidity in BaCl₂–triethanolamine
• ^c Cations extracted in NH₄–acetate.
• ^d Summation method.
• ^e BS, base saturation; the ratio of the sum of basic cations to the cation exchange capacity, expressed as a percentage.

The impact of lime application with depth in NT soils relative to PT soils is illustrated by the distribution of exchangeable Ca and Mg (Tables 4 and 5). Exchangeable Ca and Mg contents in the NT Wooster pedon (Table 4) decrease from the Ap1 through the Ap3 horizon, whereas exchangeable Ca and Mg contents in the PT pedon are essentially constant throughout the Ap horizons due to mixing of these horizons by tillage. The return of cations via plant residues may also be confined to the soil surface in NT. Variation in nutrient content with depth was reported at both research sites (Dick, 1983; Dick et al., 1986a, 1986b, 1991; Lal et al., 1990, 1994) and was confirmed at the Hoytville site by Lal and Jarecki (2005). Lal and Jarecki (2005) also found that nutrient stratification in Hoytville soils is less pronounced than in the more loamy Crosby silt loam (Aeric Epiaqualf).

Although Lal and Jarecki (2005) did not find an increase in CEC due to long-term NT, our study found greater CECs in Ap horizons of NT pedons than PT pedons (Tables 4 and 5). The increase in CEC and exchangeable bases in our study is attributed primarily to the increase in OC pools in the NT pedons. Extractable K contents also show stratification (decreasing contents with depth) in the Ap horizons of the NT pedons at both research sites whereas extractable K contents are relatively uniform throughout the Ap horizons of the PT pedons at both sites (Tables 4 and 5). Ismail et al. (1994) reported that exchangeable Ca, Mg, and K, were significantly higher in the surface 5 cm of NT soils when compared with PT soils after 20 yr of CC, but differences were not evident below 5 cm. Similar results were observed in Kansas on a Smolan silty clay loam soil (fine, smectitic, mesic Pachic Argiustoll) by Guzman et al. (2006) after 23 yr of continuous cultivation with sorghum [Sorghum bicolor (L.) Moench]. Franzluebbers and Hons (1996) found no change in the content of Ca or Mg in the surface of a calcareous fine Fluventic Ustochrept of Texas after 8.5 yr with NT, but the pH decreased in the A horizon.

3.3 Bulk density, macro- and microstructure, and porosity

Bulk density of the A horizons was strongly affected by land use at both experimental sites, but there was no impact of land use on BD in the uppermost B horizons (Tables 1 and 2). The lowest values occur in the A horizons of the forest pedons and reflect the effects of high OC contents and the enhanced conditions for aggregation provided by fauna, tree roots, and absence of repeated disturbance. Bulk densities are lower in the A horizons of grassed pedons relative to NT or PT, suggesting that grass cover provides better condition for biological activity than continuous cropping, and the lack of tillage since the establishment of the plots mitigated the compaction effects inherited from previous cultivation. Bulk density values in this study were compared with those measured in the plots in 1971 and 1980 (Dick et al., 1986a, 1986b). In the early years of these plots, soil BD was greater under NT, but after two to four decades the densest A horizons are observed with PT (Tables 1 and 2), indicating deterioration of structure and compaction due to OC loss and repeated soil disruption (Balesdent et al., 2000; Lal et al., 1994; Mestelan et al., 2006; Singer, 2006). The high BD present in the Ap3 horizon of the PT pedon at Hoytville suggests the development of a plow pan. (Shipitalo & Protz, 1987). Lal et al. (1994) also reported relatively low BD values under NT and CC at the Wooster site after 28 yr of continuous cropping, and Shipitalo et al. (2002) and Shukla et al. (2003) found decreased BD in the 0-to-5-cm layer in long-term NT experiments established in northeast and central Ohio, respectively. Organic C incorporation and enhanced biological activity, which result in greater porosity, are the possible explanations for low BD with long-term NT (Calhoun et al., 2001; Shipitalo et al., 2002; Mestelan et al., 2006; Shukla et al., 2003).

Significant differences in soil morphological properties are evident between the native and cultivated pedons of the Hoytville and Wooster research sites (Table 6). The A horizon under forest at the Hoytville experimental site is 13 cm thick and has two distinguishable subhorizons (A1 and A2) with differences in structure and root abundance and distribution. The structure of the A1 horizon is strong medium granular, whereas the structure of the A2 is moderate very fine subangular blocky (Table 6). Water table at or near the surface precluded the formation of E horizons in Hoytville pedons. A transitional AB horizon occurs below the A2 and has the same type, but coarser structure, than the A2 horizon (Table 6). The Ap1 horizon of the grassed pedon at Hoytville has the same structure as the A1 horizon of the forest pedon and has a weak to strong fine subangular blocky structure in the Ap2 and a firm fine prismatic to moderate fine subangular blocky structure in the Ap3, indicating former structure degradation under agricultural use (Table 6). Under NT, a 5-cm Ap1 horizon is found enriched with organic residues and with fine granular and fine platy structure. The Ap2 horizon shows signs of compaction (massive structure) and weak subangular blocky structure (Table 6). The Ap1 and Ap2 horizons of the PT pedon show signs of structural degradation. Structureless soil samples or platy structures are commonly attributed to compaction (Table 6; Shipitalo and Protz, 1987; Stoops, 2003).

TABLE 6. Macro- and microstructure and porosity of Hoytville and Wooster A horizons

Site	Soil source	Horizon	Thickness	Macrostructure^a	Microstructure^b	Microporosity^c	Porosity^d
			cm				%
Hoytville	Plow till	Ap1	0–5	1 m pl	1′ abk dominant / 3′ verm	Common chn, scarce cpv + apl	38.7
		Ap2	5–14	4 structureless / 1 f sbk	3′ c sbk	Common cpv + apl, scarce chn, ch	35.6
		Ap3	14–27	5 structureless	3′ verm / 1′ internal abk	cpv, napl, chn + ch	33.4
	No till	Ap1	0–5	1 f pl / 3 f gr	3′ wv pl + lent / gr + cr	apl + cpv / chn	40.8
		Ap2	5–11	5 structureless / 1 c sbk	sbk to abk / wv pl + lent	cpv + apl	37.3
		Ap3	11–21	4 to 3 sbk	3′ sbk + verm /wv pl	cpv + napl + apl	30.2
	Forest	A1	0–7	3 m gr	—Micromorphology not performed—
		A2	7–13	2 vf sbk
		AB	13–21	2 m sbk
	Grass	Ap1	0–7.5	1 to 3 m gr
		Ap2	7.5–14	1 to 3 f sbk
		Ap3	14–26	4 f pr / 2 f sbk
Wooster	Plow till	Ap1	0–3	1 f pl / 2 vf gr	3′ sbk / pl	ch + chn / cpv	36.7
		Ap2	3–17	1 f pl / 1 f sbk	3′ sbk + pl	Abundant ves / chn	33.2
		Ap3	17–31	1 f sbk	1′ sbk +structureless	Common ves / chn	30.5
	No till	Ap1	0–3	3 f gr	verm / gr + sbk	cpv, apl, chn, ch	42.1
		Ap2	3–10	2 m gr / 1 pl	gr + sbk /structureless	chn + cpv	38.4
		Ap3	10–22	2 m to c gr	gr + sbk / structureless	chn + cpv	34.7
	Forest	A1	0–4	3 m gr	—Micromorphology not performed—
		A2	4–13	2 m gr
		E	13–28	1 pl/ 1 f sbk
	Grass	Ap1	0–4	f to m gr
		Ap2	4–10	1pl/2 m gr
		Ap3	10–20	1 pl/1 m gr

• ^a Macrostructure: 1 = weak; 2 = moderate; 3 = strong; 4 = firm; 5 = very firm; sbk = subangular blocky; gr = granular; pl = platy; pr = prismatic; structureless; vf = very fine; f = fine; m = medium, c = coarse. Two structure forms separated by a slash (/) indicate compound structure.
• ^b Microstructure: pedality definition: 1′ = microaggregates faintly expressed; 2′ = microaggregates of intermediate definition; 3′ = strongly defined aggregates; abk = angular blocky; cr = crumbs; verm = vermicular; lent = lenticular; wv lent = wavy lenticular; pl = plates.
• ^c Microporosity: chn = channels; ch = chambers; cpv = compound packing voids; apl = accommodating planes; napl = nonaccommodating planes; ves = vesicles.
• ^d Estimated using image analysis of thin sections.

The Wooster pedon under native vegetation (deciduous forest) has a 13-cm A horizon and is underlain by a 15-cm eluvial horizon (E) followed by a transitional BE (Table 6). In the grassed pedon at Wooster, the Ap1 has fine to medium granular structure, followed by an Ap2 and Ap3 that combine granular and platy structure (Table 6). The NT soil has a 3-cm dark Ap1 with strong fine granular structure, probably as a result of biological activity and OC accretion. The structure of the Ap2 (moderate granular and weak platy) shows structure degradation after agricultural land use. Some amelioration is found in the Ap3 that has medium to coarse granular structure (Table 6) accompanied by a great abundance of roots (Table 6). Comparing the soil structural features in NT and grassed pedons, it seems that NT allowed more structure development than grass at Wooster after four decades since establishment of the tillage treatments. The PT soil exhibits a thin surface crust (Ap1) with weak platy and moderate very fine granular structure overlying a dense Ap2 with weak thin platy and subangular blocky structure formed by secondary tillage, overlying a dense Ap3 (fine subangular blocky structure, Table 6).

Finally, whereas the structure is strong or moderate in the forested pedons at both research sites and the NT pedon at Wooster, the structure in all other pedons is weak which is indicative of degradation and compaction (presence of platy structure) that is thought to have mostly occurred prior to the establishment of the research plots. The combined thickness of Ap horizons in the grass, NT and PT soils are 20, 22, and 31 cm, respectively, at Wooster (Table 2). Erosion losses and compaction due to tillage both prior to and after establishment of the Wooster plots explain the lesser thickness of the Ap horizon in the cultivated and grass pedons as compared with the combined thickness of the A, E, and BE horizons (41 cm) in the native forest pedon. The Ap horizon of the PT soil is thicker than either the NT or grassed pedons because of deeper plowing due to the increase in the size of machinery that has occurred with time. This annually incorporates more of the Bt horizon into the Ap horizon. Although there is no evidence of increased OC below the Ap horizon in this study, Balesdent et al. (2000) suggests that the burial of residues due to PT facilitates the translocation of OC below the plow layer, giving some characteristics of A horizon to rather deep layers. Evidence for erosion of the A horizon under PT can be found in the shallower depth to the Btx horizon (57 cm vs. 76 and 77 cm in the PT, grass, and NT pedons, respectively) (Table 2).

Detailed analysis of thin sections, collected from the A horizons of the NT and PT pedons sites, revealed certain agreement between morphological field descriptions of structure and the microstructure for the Wooster pedons (Table 6). Faunal pedoturbation is most strongly expressed in the Ap horizons of the NT soils (as an example, see Figure 1a). At 10× microscopic resolution, nematodes (probably fungi or bacteria predators) are common in A horizons of NT pedons (both sites) in macropores containing fresh organic residues where bacteria and fungi dwell. Increased faunal activity has been consistently reported by Shipitalo and Protz (1987), Edwards et al. (1990), Granovsky et al. (1994) and Shipitalo et al. (2002) in continuous NT soils of Ohio. Similar findings were reported by Drees et al. (1994) for a soil in Kentucky after 20 yr of NT.

FIGURE 1

Microstructure and porosity of Ap1 horizons under different tillage intensities. (a) Hoytville no-till (NT) showing compound packing voids and biogenic aggregates, (b) Hoytville plow till (PT) showing microaggregates, organic matter (OM) browning, and OM punctuation, (c) Wooster NT showing fractures in lenses, and (d) Wooster PT showing vesicular and nonconnective porosity. All pictures were taken with normal (plane) light at 2× magnification. The pictures are all vertically oriented with the top being the upper boundary

Earthworm burrows create channels for preferential flow and generate macroaggregates by reconfiguration of former peds (Shipitalo and Protz, 1987; Edwards et al., 1990; Granovsky et al., 1994; Shipitalo et al., 2002). Burrows and aggregates created or reshaped by biological activity are rather permanent features in NT soils and have been previously reported for the Wooster site (Mestelan et al., 2006). Inspection of thin sections corroborated that interpedal (macro)porosity was enhanced with NT (Table 6, Figure 1a, c). Although medium to high values of porosity also occur with PT, the pores have low connectivity due to occurrence of vesicles (see Figure 1c and compared with Figure 1d). Shipitalo and Protz (1987), however, indicated that half of the macropores were reduced in size due to compaction with NT, a situation that was not corroborated in our observations. At both sites under either PT or NT, where the horizons had platy structure (Table 6), there were more elongated pores (i.e., channels with accommodating or nonaccommodating planes). Shipitalo and Protz (1987) reported similar observations under NT in glacial till soils of Canada. Fresh roots, and roots in different degree of decomposition (i.e., roots of previous crops), occur primarily in Ap horizons of PT pedons at both sites (Figure 1b). Regardless of the degree of compaction in PT or NT pedons, NT nutrient stratification tends to create greatest root concentration in the uppermost soil horizons (Qin et al., 2006). In our study, the largest concentrations of root remnants occur in the Ap3 horizons of the PT soils.

3.4 Edaphic control on organic C sequestration

The sorption of organic compounds by clays results in physically and chemically protected OC in the soil (Jastrow & Miller, 1998; Six, Conant, et al, 2002; Indraratne et al., 2007). In addition, clays (smectites and other 2:1 minerals) with high specific surface areas and CECs tend to generate stable aggregates (Bronick & Lal, 2005). According to XRD analysis and the total K content of the clay fraction, the clays in the pedons at the Hoytville site are dominated by illites or vermiculites (Table 3), and the pedons classify in the illitic family of Soil Taxonomy (SSS, 2014). Total surface and specific surface areas suggest the presence of smectites (Table 3), but their presence was not confirmed by XRD. Nevertheless, the high surface and internal specific area of the clay fraction of the Ap1 and the Btg2 horizons (Table 3) indicate a high potential for interaction with particulate OC and possibly even more decomposed (and more chemically stabilized) organic residues.

Kaolinite, quartz, and Fe oxides are present in low amounts in the Hoytville soil (Table 3). The high OC values of native soil at the Hoytville site is undoubtedly a function of the mineralogical characteristics of the clay fraction (Table 3), the high clay content (Table 1), and the naturally poor drainage with a deciduous swamp-forest native vegetation. Subsequent drainage and tillage enhanced OC decomposition, due to increased gaseous exchange with the atmosphere and increased microbial accessibility to OC, resulted in a decline of OC in Hoytville soils, and a reduced ability of NT to prohibit the OC loss. Bronick and Lal (2005) reported that illites and smectites lose some effectiveness in protecting C compounds due to wet–dry cycles that have been more frequent since the Hoytville experimental research facility was drained. Also, OC levels in Hoytville soils developed after drainage and once agricultural practices have been adopted reflect an equilibrium condition that can be changed only little by application of NT (Dick & Gregorich, 2004; Six, Feller, et al., 2002).

In the Wooster soils, Al–interlayer vermiculites, kaolinite, and quartz dominate the clay fraction (Table 3). Some of these minerals may be inherited from the parent material but most form as a result of weathering. A significant content of Fe can be found in the clay fraction in the A horizon of the grassed pedon (Table 3). In tropical and subtropical soils, Six, Feller, et al. (2002) and Zinn et al. (2007) report that kaolinite and iron oxides play an important role in OC stabilization. In temperate soils, Six et al. (2000) found that the presence of kaolinite and Fe and Al oxides in soils with dominance of 2:1 clays exerts some control on OC accumulation and aggregate stabilization, suggesting that eventually there is some mineralogical control of the OC accretion in soils such as Wooster. However, due to the rather low clay content in Wooster soil, it seems that the main control of OC stabilization is structural, via microaggregation (Table 2).

With a conventional optical petrographic microscope, the finest particles that can be distinguished are fine silts. With this constraint in mind, we attempted to describe different levels of residue–OC organization and their interaction with the mineral fraction as “domains.” At least three domains, further described below, can be defined in the soils under study and correspond to different levels of distribution of the organic residues, OC chemical–biochemical alteration, and physical allocation.

Some organic residues that have been incorporated into the soil system are not tightly associated with soil minerals but instead are allocated in macropores (Figure 1c). In fact, the highest density of microorganisms and nematodes occur close to organic remnants in NT soils. This form of OC represents the first domain and is called particulate OC (Chenu, 2006; Jastrow & Miller, 1998; Olk & Gregorich, 2006). The second domain is based on the evidence of decomposition of organic residues due to browning or bleaching of organic substances (Stoops, 2003; Figure 1b). There is also evidence of mixing with mineral particles. Such degraded residues occur in proximity to fresh residues in macropores and may eventually generate microaggregates that may then form macroaggregates (Figure 1b; Jastrow& Miller, 1998). The third domain is represented by highly decomposed organic residues that exhibit chemical–biochemical stabilization, and generally show as dark brown or black masses on micrographs (i.e., spots of 5 μm in diameter). The smallest OC constituents that we were able to recognize by the petrographic microscope, and that show as brownish or brownish yellowish stains, appear to be intimately associated with the mineral matrix of aggregates (i.e., stained lenses; Figure 1c). These can be included in the last domain and can be considered physically protected. These observations are valid for both soils and essentially both tillage treatments. The differences are due to the distribution and abundance of these domains.

Fresh organic matter in macropores is ubiquitous in the Ap1 with both tillage intensities at both sites, and in the Ap3 with PT at both sites, whereas punctuations and altered residues are common in Ap2 and Ap3, particularly at the Wooster site. Microaggregates close to macropores and fresh organic residues and/or altered remnants can be seen in soil from both sites with both tillage intensities but are more numerous with NT at Wooster. Stains are visible in all the A horizons analyzed from both sites but are conspicuous and darker in the Hoytville pedons. Its seems that in Wooster pedons, organic matter has a better environment for stabilization, especially for fresh organic residues, resulting in enhanced possibilities of OC retention due to protection given by OC interaction with clay minerals and by retention in microaggregates.

4 SUMMARY

At the Hoytville (clayey, poorly drained soil) and Wooster (loamy, well-drained soil) sites we found the following:

1. Soil profile development occurred under continuous NT and grassed pedons, as indicated by better defined and strong structure. This was accompanied by increased OC accumulation that could explain lower BD values for NT compared with PT pedons.

2. Biological mixing in the NT pedons due primarily to deep burrowing earthworms, root channels, and biogenic aggregates, as reported by Mestelan et al. (2006), contributed to greater and highly connective porosity that yields lower bulk densities than in PT pedons. 

3. The lack of tillage in NT pedons was considered to have limited aggregate disruption and gaseous exchange, and moderated temperature and moisture contents. This resulted in a greater concentration of OC in the surface layers of the NT soil pedon as compared with the PT soil pedon. The depth of OC accumulation after 44 yr extended downward about 50 cm and, based upon historical records, this depth of accumulation started from the surface and increased over time from the moment of imposing NT.

4. At least three levels of organization of OC and the mineral matrix (domains) were observed including fresh, loose organic residues in macropores, OC associated with microaggregates but still only partially decomposed, and OC as discrete dark punctuations on the mineral matrix or stains of diffuse boundaries. Microaggregates are commonly associated with fresh organic residues in macropores. This suggests the formation of macroaggregates through the clustering of microaggregates.

5. At the Hoytville site, we found that, although OC concentrations were higher with NT than PT and especially in the A horizons, the total amount of C could not reach the values that existed at the beginning of the experiment. At the Wooster site, liming effects were observed in the cultivated soil to depths of 50–57 cm. In addition, at Wooster, the amount of C was greater under NT than under PT, and also greater than the amount of soil C at the beginning of the experiment.

6. At the Wooster site especially, NT often generated greater crop yields and a higher residue return than PT (Dick et al., 1991; Triplett and Dick, 2008) and there was less decomposition under NT due to restrained gaseous exchange and protection given to the new OC by minerals and the physical allocation in microaggregates. These factors tended to result in greater changes in the amounts of OC in the Wooster soil than in the Hoytville soil. For both sites, and especially for the well-drained Wooster silt loam soil, it is clear that continuous, long-term NT management can improve soil properties to sustain or even enhance crop yields and soil quality as compared with long-term PT management.

ACKNOWLEDGMENTS

The authors acknowledge the pioneering work of Dr. Glover Triplett and Dr. Dave VanDoren who established the tillage and rotation plots in Ohio in the early 1960s. Silvia Mestelan thanks the Department of State of the United States for being granted with a Fulbright Scholarship that allowed her to pursue Ph.D. studies.

AUTHOR CONTRIBUTIONS

Silvia Mestelan: Conceptualization; Formal analysis; Investigation; Methodology; Writing-original draft. Neil Smeck: Conceptualization; Investigation; Methodology. Christine Sprunger: Writing-review & editing. Ashly Dyck: Writing-review & editing. Warren Dick: Conceptualization; Data curation; Formal analysis; Funding acquisition; Methodology; Project administration; Writing-review & editing.

CONFLICT OF INTEREST

The authors declare no conflict of interest.