SSSAJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 21 June 2006
Published in Soil Sci Soc Am J 70:1407-1414 (2006)
DOI: 10.2136/sssaj2005.0355
© 2006 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Blanco-Canqui, H.
Right arrow Articles by Shipitalo, M. J.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Blanco-Canqui, H.
Right arrow Articles by Shipitalo, M. J.
Agricola
Right arrow Articles by Blanco-Canqui, H.
Right arrow Articles by Shipitalo, M. J.
Related Collections
Right arrow Structure and Properties
Right arrow Tillage
Right arrow Soil Organic Matter

Soil & Water Management & Conservation

Organic Carbon Influences on Soil Particle Density and Rheological Properties

Humberto Blanco-Canquia,*, R. Lala, W. M. Postb, R. C. Izaurraldec and M. J. Shipitalod

a Carbon Management and Sequestration Center, FAES/OARDC, School of Natural Resources, The Ohio State Univ., 210 Kottman Hall, 2021 Coffey Rd., Columbus, OH 43210-1085
b Environmental Science Div., Oak Ridge National Lab., Oak Ridge, TN 37831
c Joint Global Change Research Inst., 8400 Baltimore Ave., Suite 201, College Park, MD 20740-2496
d USDA-ARS, North Appalachian Experimental Watershed, P.O. Box 488, Coshocton, OH 43812-0488

* Corresponding author (blanco.16{at}osu.edu)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Soil particle density ({rho}s) is not routinely measured and is assumed to range between 2.60 and 2.70 Mg m–3 or to be a constant (2.65 Mg m–3) when estimating essential properties such as porosity, and volumetric water and air relations. Values of {rho}s for the same soil may, however, differ significantly from the standard range due to management-induced changes in soil organic carbon (SOC) concentrations. We quantified the {rho}s and Atterberg limits of a Rayne silt loam for five long-term (>22 yr) moldboard-plowed continuous corn (Zea mays L.; MP), no-till continuous corn (NT), no-till continuous corn with beef cattle manure (NTm), pasture, and forest systems. We also assessed the relationships of SOC concentration with {rho}s and the Atterberg limits and the impact of {rho}s on soil porosity. Mean {rho}s across NT, NTm, and pasture (2.35 Mg m–3) was ~7% lower than that for MP in the 0- to 10-cm soil depth (2.52 Mg m–3, P < 0.01). Forest had the lowest {rho}s of all soils (1.79 Mg m–3). The NTm caused a greater reduction in {rho}s and a greater increase in SOC concentration, liquid limit (LL), plastic limit (PL), and plasticity index (PI) than NT. Surface soils under MP had the highest {rho}s and {rho}b and the lowest SOC concentration, LL, PL, and PI. The SOC concentration was correlated negatively with {rho}s (r2 = 0.75) and positively with Atterberg limits (r2 > 0.64) at >20-cm depth. Estimates of soil porosity for NT, NTm, and pasture using the constant {rho}s overestimated the "true" porosity by 12% relative to that using the measured {rho}s.

Abbreviations: LL, liquid limit • MP, moldboard plow • NT, no-till with no beef cattle manure • NTm, no-till with beef cattle manure • OM, organic matter • PI, plasticity index • PL, plastic limit • {rho}b, bulk density • {rho}s, particle density • SOC, soil organic carbon


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
ARELIABLE MEASURE of {rho}s, the mass per unit volume of the inorganic and organic soil solids, is required to calculate total porosity, rates of particle sedimentation, heat capacity, thermal conductivity, and water and air relations on a volumetric basis (Hillel, 1998). These {rho}s-based parameters are essential to understanding and modeling several processes including water, air, and heat flow as well as chemical transport through the soil. Such numerous and important applications of {rho}s necessitate an accurate measurement of this soil physical property. The {rho}s is not, however, routinely measured during soil characterization because it is often assumed to range between 2.60 and 2.70 Mg m–3 or equal a constant value of 2.65 Mg m–3 for most quartz-dominated soils (Hillel, 1998). Thus, derivation of soil properties requiring {rho}s as input is normally based on the assumed constant value.

While a {rho}s of 2.65 Mg m–3 can be accurate for some soils, variations in the composition of soil solids, such as an increase in SOC concentration, can significantly lower the {rho}s value since the organic fraction is an important component of soil solids. Tillage and cropping systems are not expected to considerably affect the {rho}s, but changes in SOC concentration resulting from these management systems may significantly modify {rho}s. For example, conversion of traditional tillage to reduced and no-till systems increases SOC pools (Allmaras et al., 2004; Lal, 2004; Hooker et al., 2005). Yet, the effects of such SOC increases on {rho}s have not been well documented, and the general assumption is that any change in {rho}s would be negligible. Singh et al. (1994) reported that the {rho}s for tilled soils was slightly higher than that for NT soils due to lower SOC concentration in a plowed clay loam soil. Studies comparing {rho}s values as a result of changes in SOC concentration under long-term tillage and cropping management systems are not widely documented.

Total soil porosity, an important property estimated directly from {rho}s, is widely reported when assessing soil physical quality under different tillage and cropping systems, but it is often computed with the assumed standard value of {rho}s (2.65 Mg m–3), thus overlooking possible management-induced changes (e.g., Radcliffe et al., 1988; Eynard et al., 2004; Shukla et al., 2004). The reliability of data thus obtained is questionable if {rho}s within the same soil varies significantly as a function of soil management. Thus, characterization of {rho}s across contrasting management systems with differential C input for the same soil is a research priority. Implications of any changes in {rho}s due to soil management for the calculation of {rho}s-based soil properties have not been quantified.

Another important soil physical parameter that can be affected by changes in SOC concentration is the Atterberg limits, which describes the limits of soil consistency such as LL, PL, and PI. Measurement of these limits during routine analyses of agricultural soils is, however, very uncommon (Mapfumo and Chanasyk, 1998). Knowledge of soil water contents at LL and plastic limit PL is fundamental to predicting the tillage and wheel traffic effects on soil consistency. For example, plastic limit is the highest water content at which a soil can be tilled without negatively affecting its structure (Dexter and Bird, 2001; Arvidsson et al., 2004). Data on Atterberg limits are needed to assess long-term land use and tillage impacts on soil mechanical and rheological behavior (Terzaghi et al., 1988). The LL and PL of the soil depend primarily on clay and SOC concentrations (De Jong et al., 1990). The magnitude of dependence of these properties on SOC can be variable and influenced by soil texture, organic matter (OM) source, and soil–crop management. The LL and PL can be strongly, weakly, or not correlated with SOC concentration, depending on other properties (De Jong et al., 1990; Zhang, 1994; Mbagwu and Abeh, 1998; Joosse and McBride, 2003). In a laboratory study, Malkawi et al. (1999) reported that the PI of clay soils increased with the addition of peat, but additions >100 g kg–1 of SOC decreased the PI. In contrast, Zhang (1994) observed a linear decrease in the PI with additions of peat from 0 to 80 g kg–1 of SOC. These contradictory findings warrant further research to clarify the Atterberg limits vs. SOC relationships. The {rho}s and Atterberg limits are two fundamental soil physical properties that have not been comprehensively assessed under long-term tillage management practices with contrasting C input. Moreover, functional relationships of {rho}s and Atterberg limits with SOC concentration for long-term NT systems have not been well documented. These correlations are important to understanding the magnitude of SOC concentration effects on {rho}s and Atterberg limits.

Thus, the objectives of this study were to: (i) compare the soil {rho}s and consistency of a Rayne silt loam under five long-term (>22 yr) land use and management practices in the North Appalachian region; (ii) compare differences, if any, between the total porosity estimated from the measured {rho}s and that estimated from the assumed constant value; and (iii) determine correlations among SOC, {rho}s, and soil consistency parameters. We hypothesized that: (i) {rho}s for the same soil varies significantly with agricultural management due to differences in C input and mineralization dynamics, and its value differs from the assumed standard range, (ii) soil properties such as porosity computed using measured {rho}s values differ significantly from those computed using the assumed constant value of {rho}s within agricultural soils, and (iii) soil consistency is highly sensitive to differences in C input and can thus be used as a key indicator of management effects on the soil physical state.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Description of the Study Site and Experimental Watersheds
This study was conducted at the USDA North Appalachian Experimental Watershed (NAEW; 40°16'19'' N, 81°51'35'' W) in Coshocton County, OH, in early May 2005. Five adjacent watersheds under long-term (>22 yr) management practices were selected for this study and included: MP, NT, NTm, pasture, and forest (Table 1). The cultivated watersheds were small (~1 ha), while the two watersheds under forest and pasture were relatively large (>1 ha). All watersheds have undulating slopes (>10%). The MP, NTm, and NT watersheds were planted in continuous corn. Tillage operations were practiced on the contour in all cropped watersheds, portraying the typical farming practices in the region. The dominant species is orchardgrass (Dactylis glomerata L.) in the watershed under pasture, and white oak (Querus alba L.) and red oak (Querus rubra L.) in the forested watershed. Additional management details for each watershed are given in Table 1. The soil across the five watersheds is Rayne silt loam (fine-loamy, mixed, active, mesic Typic Hapludult). These soils are unglaciated and have four well-defined horizons (A, B, C, and R) developed from sedimentary rocks including coarse-grained sandstone, shale, and some limestone as dominant bedrock materials (Kelley et al., 1975).


View this table:
[in this window]
[in a new window]
  		Table 1. Management history of the small watersheds.

 
Soil Sampling and Analyses
An area of 0.5 by 0.5 m was selected at six locations about 3 m apart within the summit position of each watershed for soil sampling. Because the watersheds at the NAEW are not replicated, the six sampling locations within the watersheds were used as pseudoreplicates for the study. Intact soil cores (7.6 cm long by 7.6 cm in diameter) were collected with a hammer-driven sampler. Bulk soil samples of ~1000 g were obtained from each sampling location at depth increments of 0 to 10, 10 to 20, and 20 to 30 cm for the determination of {rho}b, {rho}s, LL, PL, and SOC concentration before spring tillage. The {rho}b was determined by the core method and the {rho}s by the pycnometer method (Grossman and Reinsch, 2002).

The LL was determined using the Casagrande device (Campbell, 2001). The PL was determined using the "thread method" (Campbell, 2001). The PI was computed as the difference in water content between the LL and the PL. The total SOC concentration was determined by the dry combustion method (900°C) using a CN analyzer (Vario Max, Elementar Analysensysteme, Hanau, Germany; Nelson and Sommers, 1996). Soil samples for the determination of SOC concentration were air dried at 20°C for 72 h, gently ground, and passed through a 0.25-mm sieve. Total soil porosities were computed using the standard {rho}s (2.65 Mg m–3) and measured {rho}s to test whether variations in {rho}s, if present, among management systems had significant implications for the derivation of soil porosity.

Statistical Analysis
A one-way ANOVA was used to test differences in {rho}s, {rho}b, LL, PL, PI, and SOC concentration among the five contrasting scenarios of land use and management for each soil depth. The data were analyzed assuming a randomized experiment using the six sampling locations within watersheds as pseudoreplicates. The pseudoreplication approach has been used for data analyses in numerous previous studies conducted in these watersheds (e.g., Shipitalo et al., 1997; Butt et al., 1999; Owens et al., 2002; Shukla et al., 2003; Blanco-Canqui et al., 2005). It is assumed that management treatments are mostly responsible for the differences in {rho}s, {rho}b, LL, PL, PI, and SOC concentration among watersheds because these watersheds are adjacent to each other and are within the same soil map unit. The five watersheds are arranged in a natural field block confining the five watersheds on a very similar landscape position, slope, and soil, and they have been continuously managed under the same long-term systems (>22 yr; Table 1), providing data on tillage and cropping systems in the northern Appalachians. Shukla et al. (2003) reported that differences in soil texture among these watersheds were not significant. Simple correlations and regression fits were performed to assess interrelationships among the measured soil properties. All analyses were conducted using SAS statistical software (SAS Institute, 1999).


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Particle Density
Soil management had a significant effect on {rho}s at all depths (P < 0.01; Fig. 1). The NT, NTm, and pasture treatments had significantly lower {rho}s than the MP treatment in the 0- to 10-cm depth. The {rho}s averaged across these three treatments (2.35 Mg m–3) was ~7% lower than the {rho}s (2.52 Mg m–3) for the MP treatment. The {rho}s was the lowest (2.33 Mg m–3) under NTm for agricultural soils but did not differ significantly from that under NT and pasture treatments in the upper depth. The MP treatment had significantly higher {rho}s than the NTm treatment at all depths. The {rho}s for the MP treatment was 8.5% higher in the 0- to 10-cm depth, 7% higher in the 10- to 20-cm depth, and 4% higher in the 20- to 30-cm soil depth compared with the NTm treatment (P < 0.01). The {rho}s under pasture was also significantly higher by ~5% compared with NTm at the 10- to 30-cm depth. The forest soils had the lowest {rho}s of all management systems and increased from 1.79 Mg m–3 for the 0- to 10-cm depth to 2.40 Mg m–3 for the 20- to 30-cm depth (P < 0.01). These results are in accord with the hypothesis that {rho}s differs among agricultural systems, particularly within cultivated soils. As expected, the {rho}s increased with increasing soil depth and the magnitude of differences among treatments in the >20-cm depth was relatively small.


Figure 1
View larger version (16K):
[in this window]
[in a new window]
  		Fig. 1. Mean soil particle density at three depths for moldboard plow (MP), no-till with no beef cattle manure (NT), no-till with beef cattle manure (NTm), pasture, and forest long-term management practices at the Northern Appalachian Experimental Watershed in Coshocton County, OH. The LSD bars at a given soil depth indicate the significance of particle density among the five management practices.

 
The higher surface {rho}s for MP than NT and pasture is supported by the results of Singh et al. (1994), who observed that plowed soils had higher {rho}s (2.49 Mg m–3) than NT (2.45 Mg m–3) on a clay loam, and with those of Sparling et al. (2000), who reported that plowed soils (2.60 Mg m–3) had a slightly higher {rho}s than pasture (2.57 Mg m–3) for a silt loam in New Zealand. Differences in {rho}s among management systems, in this study, were about four times greater than those reported by Singh et al. (1994) and Sparling et al. (2000), possibly due to differences in the duration of management. Singh et al. (1994) and Sparling et al. (2000) evaluated {rho}s in soils under <9 yr of management, whereas the soils in our study were under continuous management for >22 yr (Table 1). Long-term (>9 yr) studies comparing {rho}s between conventionally tilled and NT soils have not been widely reported. The lower {rho}s in forest vs. agricultural soils was expected due to the abundance of organic materials on forest floors. Effects of forest management on {rho}s cited in the literature are highly variable. Redding et al. (2005) reported that {rho}s for various species ranged between 1.52 and 1.60 Mg m–3, which is lower than the forest {rho}s measured in this study. Other researchers (van Donk and Tollner, 2000) have also reported lower values of {rho}s (<1.52 Mg m–3). These inconsistencies of {rho}s values among forest soils may be due to differences in SOC concentration, forest species, methodology, and climatic conditions.

The {rho}s range within agricultural soils (2.33–2.52 Mg m–3) measured in the 0- to 10-cm depth for this study was consistently below the range of assumed values of 2.60 to 2.70 Mg m–3. It is also lower than those reported by Singh et al. (1994) and Sparling et al. (2000). While the decrease in {rho}s from 2.52 to 2.33 Mg m–3 may be attributed to high input of organic residues, differences in {rho}s between MP (2.52 Mg m–3) and the standard range (2.60–2.70 Mg m–3) may be due to variations in parent material, chemical composition and structure of the inorganic solids, and topography. Soils of this study have steep slopes (>10%) and low sand content (<209 g kg–1). Ball et al. (2000) reported that the {rho}s in sloping soils was lower than in flat terrains in Scotland, and their {rho}s ranged from 2.36 to 2.87 Mg m–3. While the {rho}s for MP and pasture treatments in our study increased to ~2.60 Mg m–3 for the 0- to 30-cm depth, the {rho}s for NT, NTm, and forest increased to levels significantly (<2.52 Mg m–3) below the assumed standard range.

Particle Density vs. Soil Organic Carbon
There were large differences in SOC concentration (P < 0.01; Fig. 2) among management systems at all depths. Because trends of concentration and mass of SOC were not significant, results are discussed based on SOC concentration only. The SOC concentration in the 0- to 10-cm depth increased in the order: forest > NTm > pasture > NT > MP, and the order was almost the opposite for {rho}s (Fig. 1). Thus, these significant differences in SOC concentration explain much of the difference in {rho}s. In fact, the linear regression in Fig. 3A shows that variations in SOC concentration explained ~75% of the variability in {rho}s (P < 0.001) in the 0- to 10-cm depth. For example, the MP management had the lowest SOC concentration and the highest {rho}s while the forest management had the highest SOC concentration and the lowest {rho}s in the 0- to 10-cm depth. These results show that changes in SOC concentration due to management can indeed have large and significant effects on {rho}s as hypothesized. Surprisingly, Fig. 3A shows that the effect of SOC concentration on {rho}s can be as large as that on {rho}b in the 0- to 10-cm depth. The r2 values and slopes of the linear regression fits of SOC concentration vs. {rho}s and {rho}b are similar, suggesting that {rho}s is as sensitive as {rho}b to changes in SOC concentration in these soils. The decrease in {rho}s with increasing SOC concentration is attributed to the dilution effect of the mineral particles (Hillel, 1998). In this long-term study, the highly significant {rho}s vs. SOC concentration correlation (r = –0.87; Fig. 3A) in the 0- to 10-cm depth was much higher than that (r < 0.50) reported in similar studies (Ball et al., 2000). These inconsistencies may be attributed to the differences in duration and type of management practices. The {rho}s vs. SOC concentration correlation was weaker between 10 and 30 cm than the 0- to 10-cm soil depth, which is probably due to the smaller differences in SOC concentrations among management systems. The SOC concentration explained 54% of the variability in {rho}s for the 10- to 20-cm depth, and 45% for the 20- to 30-cm depth, and these correlations were significant (P < 0.01).


Figure 2
View larger version (14K):
[in this window]
[in a new window]
  		Fig. 2. Mean soil organic C concentration at three depths for moldboard plow (MP), no-till with no beef cattle manure (NT), no-till with beef cattle manure (NTm), pasture, and forest long-term management practices at the Northern Appalachian Experimental Watershed in Coshocton County, OH. The LSD bars at a given soil depth indicate the significance of soil organic C concentration among the five management practices.

 

Figure 3
View larger version (21K):
[in this window]
[in a new window]
  		Fig. 3. (A) Relationships of bulk and particle density of soil with SOC (soil organic carbon) concentration for the 0- to 10-cm depth and (B) depth distribution of bulk density under moldboard plow (MP), no-till with no beef cattle manure (NT), no-till with beef cattle manure (NTm), pasture, and forest long-term management practices at the Northern Appalachian Experimental Watershed in Coshocton County, OH. The LSD bars at a given soil depth indicate the significance of bulk density among the five management practices.

 
The lowest {rho}s, under NTm (<2.49 Mg m–3), within the agricultural soils at the three depths indicates that long-term manuring (>42 yr) can greatly affect the profile distribution of {rho}s by lowering the {rho}s values in contrast with unmanured treatments. Manuring in interaction with lack of soil disturbance increased the SOC concentration and reduced the {rho}s (2.33 Mg m–3) in the 0- to 30-cm depth even more than NT without manure (<2.53 Mg m–3) in which {rho}s increased rapidly between 10 and 20 cm depths. These results corroborate the significant implications that manuring has for increasing the SOC concentration and altering the {rho}s with soil depth. Compared with soils under MP, those under NTm had 3.7 times higher SOC concentration for the 0- to 10-cm depth, 2.3 times higher for the 10- to 20-cm depth, and 1.5 times higher for the 20- to 30-cm depth, which probably explains the lower {rho}s under NTm. These results show that manuring increased SOC concentrations compared with MP even below the 10-cm soil depth. Similar studies assessing the combined effects of manuring vs. a NT system on {rho}s vs. SOC relationships and the depth distribution of SOC concentrations are few. Schjønning et al. (1994) reported that long-term cattle manuring (>90 yr) decreased {rho}b slightly from 1.64 to 1.57 Mg m–3 and {rho}s from 2.64 to 2.63 Mg m–3 in a plowed sandy loam in Denmark. The magnitude of the decrease in {rho}s in that study was not as large as in our study (2.52 vs. 2.33 Mg m–3), probably due to differences in soil texture and mineralogy.

Particle Density vs. Soil Porosity
The {rho}b, an essential input for computing soil porosity, was also significantly affected by management, particularly in the 0- to 20-cm depth (Fig. 3B). It was higher under MP than under other treatments in the 0- to 10-cm depth (P < 0.01). Total soil porosity computed using the standard {rho}s and measured values of {rho}s and {rho}b by management for each soil depth is shown in Fig. 4. Porosity computed using the measured {rho}s values differed significantly from that computed using the constant value of {rho}s at all depths. The variable fmeas was used to signify porosity computed using measured {rho}s, and fcalc to signify porosity computed using the standard {rho}s (2.65 Mg m–3). The fcalc was consistently larger than fmeas for all management systems (P < 0.01). Even the fcalc (0.46 ± 0.005) for the MP management was slightly, but significantly, higher than the fmeas (0.43 ± 0.003) due to the fact that the {rho}s for the MP treatment (2.52 Mg m–3) was lower than the assumed standard value of 2.65 Mg m–3 in the 0- to 10-cm depth. Discrepancies between fcalc and fmeas were the smallest for the MP treatment. The fmeas for the MP treatment was 7% lower than the fcalc at the 0- to 10-cm depth, while the fmeas for the NT, NTm, and pasture treatments were ~12% lower than the fcalc across the three management systems at the upper depth (Fig. 4). The fmeas for the forest management was 27% lower than fcalc in the uppermost depth. Averaged across depths, the difference between fcalc and fmeas for NT and NTm was two times as high as that for MP, which is due to the management-induced changes in {rho}s. These results show that the use of the standard {rho}s value for estimating total porosity overestimates the "true" porosity in these soils, and can thus lead to erroneous conclusions concerning management effects on total porosity. The constant value of 2.65 Mg m–3 may apply only to soils with both a low SOC concentration and a high presence of quartz sand fractions (Redding et al., 2005).


Figure 4
View larger version (19K):
[in this window]
[in a new window]
  		Fig. 4. Total soil porosity using assumed (2.65 Mg m–3) and measured particle density for moldboard plow (MP), no-till with no beef cattle manure (NT), no-till with beef cattle manure (NTm), pasture, and forest long-term management practices at three depth intervals. Different letters within the same treatment indicate significant differences between the two porosities at the 0.05 probability level.

 
The data presented suggest that the results of some long-term studies of agricultural systems in which there were large and significant changes in SOC concentrations may need to be reevaluated if constant {rho}s values were used to compute total porosity. These results also emphasize the critical importance of including {rho}s in any routine characterization of soil properties while assessing the impacts of different agricultural systems on soil physical quality, given the sensitivity of {rho}s to changes in SOC concentration. While {rho}s may not be affected by soil disturbance or structural deterioration like {rho}b and other dynamic soil structural properties, it can be significantly altered by changes in SOC concentration. Additional comparison of {rho}s effects on soil porosity across diverse management practices and soils are needed to further understand the magnitude of changes.

Atterberg Limits
Soil management had a significant impact on the Atterberg limits also, particularly in the 0- to 20-cm soil depth (Fig. 5). Differences in water content at the LL among the five treatments were large for the 0- and 20-cm depth and increased in the order: NTm > forest > pasture > NT > MP. The soil water content at the LL for NTm soils was 6% higher than for forest, 24% higher than for pasture, 33% higher than for NT, and 85% higher than for MP soils. Water content at the PL in the 0- and 10-cm depth increased in the order: NTm = forest > pasture = NT > MP. At the PL, NTm and forest treatments retained 30% more water than pasture and NT and 100% more than MP, while pasture and NT retained 55% more water than MP. At the 10- to 20-cm depth interval, NTm and forest retained ~36% more water at the PL than the rest of treatments. Differences in the PI among management systems were smaller than those in the LL and the PL. The NTm management had a significantly higher PI than NT and MP at the three depths. No significant differences in the PI were observed among NTm, forest, and pasture treatments, but the value of the PI averaged across these three management systems was 23% higher than that for NT and 84% higher than that for the MP treatment in the 0- to 10-cm depth. At the 10- to 20-cm depth, the PI for NTm was 16% higher than that for pasture, 42% higher than that for NT, and 92% higher than that for MP. While differences in the PI between NTm and forest were not significant in the 0- to 20-cm depth, the PI for NTm was 50% higher than that for the forest management in the 20- to 30-cm depth.


Figure 5
View larger version (18K):
[in this window]
[in a new window]
  		Fig. 5. Soil water contents at the liquid limit (LL), plastic limit (PL), and plasticity index (PI) for three soil depths for moldboard plow (MP), no-till with no beef cattle manure (NT), no-till with beef cattle manure (NTm), pasture, and forest long-term management practices at the Northern Appalachian Experimental Watershed in Coshocton County, OH. The LSD bars at a given soil depth indicate the significance of LL, PL, and PI among the five management practices.

 
These results show that management significantly affected the soil consistency at all depths. No-till in combination with manuring had a strong effect on increasing the LL, PL, and PI values. The water retention values for the PI, LL, and PL constants under NTm were ~30% higher than those under NT alone, indicating that manured NT soils had better soil structure or consistency. Manure supplies OM and improves the water retention capacity of the soil (Arriaga and Lowery, 2003). The higher LL and PI for NTm than for forest in the 20- to 30-cm depth interval suggest that manure additions may stabilize the soil better than forest-derived organic materials at lower depths. Because the Atterberg constants were highly sensitive to management, they could be used as indicators of soil physical quality for long-term land use and management systems for these and similar soils. The consistency limits are important parameters related to soil structural stability (Marinissen, 1994). Indeed, Watts and Dexter (1998) stated that the PL in particular is a useful index of soil physical quality. For instance, the higher PL value for NTm in our study indicates that soils under this management have better physical structure than under the other agricultural systems in this study.

The study results also show the merit of measuring the Atterberg limits for determining the optimum water content for tillage. The large differences in LL, PL, and PI suggest that the optimum water content for tillage varies significantly among the selected cultivated watersheds. When the soil is too wet, tillage operations can have detrimental effects on soil structure (Mueller et al., 2003). The knowledge of differences in soil consistency limits can allow a better scheduling of tillage operations and traffic. Mueller et al. (2003) identified the PL as one of the most sensitive parameters to estimate the highest water content for optimum tillage across a broad range of soils. The higher PL value indicates that these soils are trafficable even at higher water contents under NTm and NT than under MP land use. In other words, MP soils are prone to vehicle-induced compaction at water contents lower than NTm and NT soils. The PL under NTm soils was two times as high as under MP. Soil macro- and microstructural properties can be significantly altered by untimely tillage ignoring the dynamics of soil consistency (Adam and Erbach, 1992; Barzegar et al., 2004).

Atterberg Limits vs. Soil Organic Carbon
The significant differences in soil consistency were attributed to variations in SOC concentration with management systems. The water contents at LL and PL including the PI followed a quadratic relationship with SOC concentration in the 0- to 10-cm depth (Fig. 6A). Variations in the SOC concentration explained 98% of the variability in the LL, 93% in the PL, and 68% in the PI. These strong correlations show that management-induced changes in the SOC concentration have a strong effect on soil consistency. Unlike studies by De Jong et al. (1990) and Zhang (1994), who reported variable correlations, in our study soil consistency was highly correlated with the SOC concentration. The positive correlations in this study, however, support the conclusion of Odell et al. (1960), who reported that LL, PL, and PI were all positively correlated with the SOC concentration. The significant PI vs. SOC correlations reported by Odell et al. (1960) and by our study contrast with the study by Ball et al. (2000), who reported no significant dependence of PI on SOC concentration. These inconsistencies may be due to differences in soil inherent characteristics such as management duration, soil parent material, clay content and mineralogy, and the type and nature of the OM (De Jong et al., 1990).


Figure 6
View larger version (22K):
[in this window]
[in a new window]
  		Fig. 6. Liquid limit, plastic limit, and plasticity index as a function of soil organic carbon (SOC) concentration for the (A) 0- to-10-cm and (B) 10- to-20-cm soil depths across five long-term management practices at the Northern Appalachian Experimental Watershed in Coshocton County, OH. The r2 were all significant at the 0.001 probability level.

 
The quadratic regression curves in Fig. 6A for the LL and PL increased rapidly up to ~48 g kg–1 of SOC and then decreased slightly with a further increase in the SOC concentration. The decline of water contents after ~48 g kg–1 of SOC may be due to the qualitative differences in the OM between the agricultural and forest soils. We hypothesize that the OM in the agricultural soils may be strongly associated with the mineral fraction, which would account for its effect on soil consistency, whereas the OM in the forest soil was probably less closely associated with the mineral fraction. The increase in the LL and PL with increasing SOC concentration is probably due to the high surface area of the organic materials. The LL, PL, and PI increased linearly with increasing SOC at low concentrations (<30 g kg–1) for the 10- to 20-cm depth (P < 0.01; Fig. 6B). At this depth, the SOC concentration explained ~76% of the variability in the LL, 83% in the PL, and 44% in the PI. At the 20- to 30-cm depth, the SOC concentration was not significantly correlated with Atterberg limits except the PI, where changes in the SOC concentration explained about 70% of the variability in the PI. As with {rho}s, the lower correlation coefficients at lower depths could be attributed to the decrease in SOC concentrations. At lower depths, increases in clay content may play a dominant role in explaining changes in the consistency limits.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Soil particle density and consistency (Atterberg limits) were significantly affected by land use and management practices across five watersheds on a Rayne silt loam in the northern Appalachians. No-till management, particularly in combination with manure, decreases the {rho}s and increases soil consistency significantly compared with MP due to differential C input. Total porosity computed using the constant value (2.65 Mg m–3) of {rho}s significantly overestimates the "true" porosity, thereby suggesting the need to measure {rho}s values to reliably compute related soil properties. The Atterberg limits can be sensitive parameters to evaluate the physical condition of soils because of large differences among management systems. Results also suggest that measurement of {rho}s and soil consistency must be included in any routine soil tests because these soil properties can differ significantly among management practices. Both {rho}s and Atterberg limits were significantly influenced by management-induced changes in SOC concentrations. Since improved soil conservation practices such as reduced tillage and no-till in combination with manuring for promoting SOC sequestration are being increasingly adopted, additional studies are needed to further understand the effects of contrasting and enhanced management systems on {rho}s and related soil parameters.


   ACKNOWLEDGMENTS
 
Research conducted within the CSiTE (Consortium for Research on Enhanced Carbon Sequestration in Terrestrial Ecosystems) program. CSiTE is supported by the U.S. Dep. of Energy, Office of Science, as part of the Terrestrial Carbon Sequestration program in the Office of Biological and Environmental Research.

Received for publication October 23, 2005.


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





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Blanco-Canqui, H.
Right arrow Articles by Shipitalo, M. J.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Blanco-Canqui, H.
Right arrow Articles by Shipitalo, M. J.
Agricola
Right arrow Articles by Blanco-Canqui, H.
Right arrow Articles by Shipitalo, M. J.
Related Collections
Right arrow Structure and Properties
Right arrow Tillage
Right arrow Soil Organic Matter

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome