JEQ Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 4 January 2008
Published in J Environ Qual 37:219-228 (2008)
DOI: 10.2134/jeq2007.0227
© 2008 American Society of Agronomy, Crop Science Society of America, and 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 PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Udawatta, R. P.
Right arrow Articles by Garrett, H. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Udawatta, R. P.
Right arrow Articles by Garrett, H. E.
Agricola
Right arrow Articles by Udawatta, R. P.
Right arrow Articles by Garrett, H. E.
Related Collections
Right arrow Ecosystem Management
Right arrow Ecosystem Restoration
Right arrow Best Management Practices

TECHNICAL REPORTS

Influence of Prairie Restoration on CT-Measured Soil Pore Characteristics

Ranjith P. Udawattaa,b,*, Stephen H. Andersonb, Clark J. Gantzerb and Harold E. Garretta

a Center for Agroforestry, School of Natural Resources, Univ. of Missouri, Columbia, MO 65211
b Dep. of Soil, Environmental and Atmospheric Sciences, School of Natural Resources, Univ. of Missouri, Columbia, MO 65211

* Corresponding author (UdawattaR{at}Missouri.edu).

Received for publication May 7, 2007.

   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 Materials and Methods
 Results and Discussion
 Conclusions
 REFERENCES
 
Restored prairies are expected to improve soil physical properties, yet little is known about the extent of change to soil properties and how rapidly these changes take place. The objective of this study was to compare effects of prairie restoration on computed tomography (CT)-measured pore parameters. Undisturbed soil cores (76 mm diam. by 76 mm long) from native prairie (NP), restored prairie (RP), conservation reserve program (CRP), and no-till corn (Zea mays L.)-soybean (Glycine max (L.) Merr.; CS) sites were collected with six replicates from the 0- to 40-cm depth in 10-cm increments. Five CT images were acquired from each soil core using a medical CT scanner with 0.2 by 0.2 mm pixel resolution with 0.5 mm slice thickness, and then images were analyzed. Soil bulk density and hydraulic conductivity (Ksat) were also measured. Soils under NP, RP, CRP, and CS areas had 83, 43, 48, and 26 pores on a 2500 mm2 area, respectively, for the 0- to 40-cm depth. The number of pores, number of macropores (>1000 µm diam.), macroporosity, mesoporosity (200-1000 µm diam.), and fractal dimension were significantly higher and pore circularity was lower for NP, RP, and CRP than the CS treatment. The CT-measured mesoporosity and macroporosity of the CS treatment were 20 and 18% of the values for the NP site. CT-measured number of pores and macropores explained 43 and 40% of the variation for Ksat. The study showed that prairie restoration improves CT-measured soil pore parameters and decreases bulk density which are related to soil water infiltration.

Abbreviations: CRP, Conservation Reserve Program • CS, corn–soybean rotation • CT, computed tomography • Ksat, saturated hydraulic conductivity • NP, native prairie • RP, restored prairie


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
Materials and Methods
 Results and Discussion
 Conclusions
 REFERENCES
 
RECENT interest in prairie restoration and its environmental benefits has encouraged the evaluation of restored prairies compared to native prairies and row crop areas in several regions of the US. Replacement of annual crops with perennials alters the quantity and quality of residue added to the soil (Wienhold and Tanaka, 2001). Natural prairie vegetation has been shown to increase hydraulic conductivity and organic matter content and lower soil bulk density values (Brye and Pirani, 2005; Brye and Moreno, 2006). Comparing prairie and agricultural soils in Arkansas for nutrients, Brye and Pirani (2005) found that agricultural activities negatively impacted native soils relative to organic carbon, and total C and N. Prairie soils had significantly greater quantities of soil organic carbon, total C and N and were low in pH, electrical conductivity, Ca, P, and Fe. Kucharik et al. (2006) observed 37% higher below-ground carbon in a remnant prairie than a 65-yr-old restored prairie. Tillage operations have been shown to enhance decomposition of organic matter and wetting and drying of soil, thereby decreasing biogeochemical gradients, microbial diversity, and aggregate stability (Park and Smucker, 2005).

Differences in soil hydraulic conductivity can be attributed to roots, soil organisms, soil properties, and agricultural activities (Fuentes et al., 2004). Meek et al. (1992) and Fuentes et al. (2004) stated that root growth and decay play a major role in determining temporal variation in soil hydraulic conductivity. Tillage operations in crop areas create more pores near the surface and disrupt pore continuity (Bouma, 1991; Beare et al., 1994). Shortly after tillage, saturated hydraulic conductivity in the top soil is larger but decreases with time (Cassel and Nelson, 1985; Messing and Jarvis, 1993). Soils managed under native prairies or non-agricultural practices do not experience changes in pore parameters due to tillage. Furthermore, wheel traffic causes compaction on agricultural soils, destroys larger pores and thereby restricts water movement (Fuentes et al., 2004). It could be speculated that root characteristics, agricultural activities, and an extended growing season in prairie systems contribute to differences in soil hydraulic properties between soils under row crop and prairie management.

Agricultural activities often increase soil bulk density. Seobi et al. (2005) observed higher bulk densities in crop areas as compared to permanent vegetative buffers with grass and trees. Differences were attributed to roots, organic matter, and agricultural activities. Comparing a regularly harvested prairie with a no-harvest prairie in Arkansas, Brye and Moreno (2006) stated that bulk density was higher on a regularly harvested prairie. In Missouri, Kremer and Anderson (2005) observed lower soil bulk density for prairie soils as compared to row crop soil for the 0- to 10-cm soil depth. In the prairie sites, soil organic matter content was higher and may have contributed to reduced soil bulk density.

Changes in soil bulk density and addition of organic matter probably contribute to better soil porosity. Soil pores, especially macropores (diam. > 1000 µm) promote soil water movement through the profile. In addition, better water retention is also important to improve plant growth. Perennial vegetation, such as pasture, grass buffers, and agroforestry buffers, increases soil porosity compared to row crop land (Chan and Mead, 1989; Seobi et al., 2005; Udawatta et al., 2006, 2008). Macropores rapidly channel surplus flow, thereby reducing nutrient leaching through smaller pores (van Noordwijk et al., 1991a) and allowing movement of water and air into the soil. Research shows that macropore characteristics such as shape, size, orientation, and size distribution affect the rate, flow, and retention of water (Rasiah and Alymore, 1998; Udawatta et al., 2006). Therefore, differences in porosity among soils need to be quantified to diagnose changes due to agricultural management practices and to design management guidelines (Pachepsky et al., 1996).

Computed tomography (CT) procedures have been shown to be superior to traditional methods, provide a finer resolution on a mm- to µm-scale (Gantzer and Anderson, 2002; Akin and Kovscek, 2003; Carlson et al., 2003), and closely agree with water retention–derived estimates (Anderson et al., 1990; Rachman et al., 2005). By combining cross-sectional images, information can be available on a three-dimensional scale allowing visualization of pore tortuosity, connectivity, and characteristic pore parameters (Mooney, 2002; Carlson et al., 2003). The best-known advantage of CT is its ability to quickly and nondestructively image the interior of a three-dimensional object while retaining connectivity and spatial variation in pores (Al-Raoush, 2002; Carlson et al., 2003).

The effects of prairie restoration on soil properties and pore parameters are not fully understood. Furthermore, prairie restoration must be examined as only 0.03% of the pre-settlement natural prairies exist in the Midwestern USA. These sites are nearly extinct, and changes in soil properties due to restoration takes a long period of time to be significant (McLachlan and Knispel, 2005). Therefore, information on soil properties is needed to evaluate beneficial effects of prairie restoration. We hypothesize that prairie restoration improves soil physical properties and pore parameters. The specific objectives of this study were to: (i) evaluate differences in CT-measured macropore and coarse mesopore characteristics (number of pores, number of macropores, macroporosity, mesoporosity, pore circularity, and fractal dimension of macroporosity) among native prairie and restored prairie sites as well as a continuous row crop site, and (ii) correlate CT-measured pore parameters and soil bulk density with saturated hydraulic conductivity (Ksat).


   Materials and Methods
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
Results and Discussion
 Conclusions
 REFERENCES
 
Study Area and Management
A cropping system treatment and a conservation reserve program treatment were compared to native prairie and restored prairie ecosystems in central Missouri (Table 1 ). The treatments for the study include the following: Tucker Prairie (NP; native prairie), Prairie Fork (RP; restored prairie), Conservation Reserve Program (CRP), and corn–soybean rotation (CS). The undisturbed Tucker Prairie site has been under native prairie vegetation and consists of big blue stem (Andropogon gerardi Vitman.), little blue stem (Schizachyrium scoparium Nash.), prairie dropseed (Sporobolus heterolepis [A. Gray] A.Gray), and Indian grass (Sorghastrum nutans [L.J. Nash]) (Buyanovsky et al., 1987). The Prairie Fork Conservation Area site was under row crop management for approximately 100 yr and was restored in 1993 with native grasses and legumes. The study area vegetation consisted of little blue stem, side-oats gamma (Bouteloua curtipendula var. curtipendula), and Indian grass. CRP and CS sampling plots are located within the USDA-ARS Agricultural Systems for Environmental Quality site near Centralia, MO which had originally been under cultivation for approximately 100 yr. The CRP sampling sites had been in CRP since 1991 with present vegetation consisting of 95% tall fescue, some orchardgrass, and red clover. The CS sampling sites were managed with mulch tillage since 1991 with 190 kg ha–1 N during corn years and lime, P, and K added based on soil test for a grain yield of 10,100 kg ha–1 for corn and 2500 kg ha–1 for soybean. The sites were in corn in 2005.


View this table:
[in this window]
[in a new window]

  		Table 1. Selected soil physical and chemical properties for the Tucker Prairie, Prairie Fork, and Centralia Research sites.{dagger}

 
Parent Material, Soils, and Climate
The parent materials for the soils in the study area are loess over loamy sediments derived from pre-Illinoian till (Unklesbay and Vineyard, 1992). Soils in the NP, RP, CRP, and CS areas are classified as Mexico silt loam (Fine, smectitic, mesic Vertic Epiaqualfs). The Mexico series consists of very deep and poorly drained soils with an argillic horizon at varying depths on 0 to 4% slopes. The potential for runoff is high to very high and permeability is very slow. The mean annual temperature ranges from 10 to 13°C, and mean annual precipitation ranges from 890 to 1020 mm. Most areas are used to grow corn, soybeans, hay, pasture, and small grains. The native vegetation consists of warm-season grasses and forbs.

Sample Collection and Preparation
Undisturbed soil cores from the surface 0- to 40-cm depth were collected at 10-cm increments with six replicates from 27 May to 3 June 2005 using an Uhland sampler. Sampling Plexiglas rings were 7.62 cm long by 7.62 cm diam., with a 3.2 mm thick wall. Samples for the NP and RP were collected from prairie soils with a uniform vegetative cover. CRP soil cores were collected from areas managed in CRP since 1991. For the CS treatment, soils were collected midway between corn rows in non-trafficked inter-rows. An additional depth (40–50 cm) was sampled for the NP site since the depth to claypan was deeper at this site. The soil inside the cylinders was secured with two plastic caps at each end and with masking tape. The soil cylinders were labeled, placed in plastic bags, sealed, placed in individual cardboard containers, and transferred to the laboratory in a wooden box. Soil samples were stored in a refrigerator at 4°C until analyses were conducted.

The bottom plastic cover was replaced with two layers of fine nylon mesh to secure soil within the cylinder. The top plastic cover was removed and the soil cores were placed in a 15 cm deep plastic tray. The soil cores were slowly saturated from the bottom with a solution containing calcium chloride (CaCl2; 6.24 g L–1) and magnesium chloride (MgCl2; 1.49 g L–1) using a Mariotte system. This concentration has been found to be similar to soil ionic concentrations in claypan soils (Palmer, 1979). After a 24-h saturation period, wet weights were recorded and samples were placed on a –3.5 kPa glass-bead tension table for 24 h for draining. This procedure removed water from macropores and coarse mesopores to allow better image contrast. Samples were weighed again, two plastic end caps were secured with masking tape, and the samples were prepared for transport to the CT scanner. Two phantoms, distilled water in an aluminum tube (outside and inside diam. 2.32 and 1.60 mm), and a solid copper wire (outside diam. 0.55 mm), were attached to the long axis of the Plexiglas cylinder for a standard comparison among scans. Copper has attenuation similar to manganese concretions present in claypan soils. Physical and chemical properties for soils at the sites are presented in Table 1.

Scanning and Image Analysis
Computed tomographic image acquisition was conducted using a Siemens Somatom Plus 4 Volume Zoom X-ray CT scanner. The scan system parameters were set to 125 kV, 400 mAs, and 1.5 s scan time to provide detailed and low noise projections. The field of view, i.e., the cross-sectional dimension, was 100 mm with 512- by 512-mm picture elements (pixels) giving a pixel size of 0.19 by 0.19 mm. The X-ray beam width or "slice" thickness was 0.5 mm producing a volume element (voxel) size of 0.018 mm3. Each soil core was placed horizontally within the scanner so that X-rays intersected the soil core perpendicular to its longitudinal axis. The first scan image was taken from a 15-mm distance from the top of the soil core. Four additional scans were taken at 26, 37, 48, and 59 mm from the top of the soil core. The data were stored on a CD for subsequent image analysis.

Pore characteristics of scanned images were analyzed with public domain software Image-J version 1.27 (Rasband, 2002). A 2500 mm2 region was demarcated using Area Selection Tools as the "Region of Interest" (ROI) to exclude voids near the core walls and to minimize beam hardening interference. The region adjacent to the interior wall may have higher porosity due to discrepancy between the radii of the curvature between the soil particles and the Plexiglas wall (Al-Raoush, 2002). Segmentation or the separation of air-filled pore areas and the other regions within a scan was completed by converting the gray scale image to identify two populations in the image based on their intensity values. The intensity value (relative attenuation value ranged from 0 to 256) from the water phantoms (38–42, mean = 40) was used as threshold values to differentiate air-filled spaces and the other regions within a scan with no overlap. Values lower than the threshold value were identified as air-filled pores and values greater than the threshold value were identified as non-pore (Fig. 1 ). The selected value was in between the values used by Rachman et al. (2005) and Gantzer and Anderson (2002). Number of pores, pore area, pore perimeter, circularity, and porosity statistics were obtained from Analyze Particles Tool. Pore area was used to estimate pore diameter and to classify pores into macro- (diam. > 1000 µm) and mesopore (diam. 200–1000 µm) categories (Scott, 2000). Porosity of a scan was determined by estimating the ratio between total pore area of a scan and the 2500 mm2 area. Pore circularity was estimated by dividing the product of pore area and 4{pi} by the pore perimeter squared (Tuller et al., 1999). The fractal dimensions of images (option in Image-J) were determined with 100 as the threshold value. A different threshold value was used in order (100 vs. 40) to better populate the low porosity samples with pores and allow computation of the fractal dimension (Gantzer and Anderson, 2002).


Figure 1
View larger version (142K):
[in this window]
[in a new window]

  		Fig. 1. Typical 2500 mm2 area scan images of native prairie, restored prairie, Conservation Reserve Program (CRP), and corn–soybean rotational treatments at four scan depths (middle scan is shown) in a selected profile. Air-filled pores are in red, solid areas in gray, and manganese concretions in white.

 
Soil Hydraulic Properties and Bulk Density
After scanning, saturated hydraulic conductivity (Ksat) and dry bulk density were determined on all 102 soil cores. Cores were covered with cheese-cloth at the bottom and saturated in a 15 cm deep plastic tray with water before Ksat was measured. The constant head method was used to determine Ksat (Klute and Dirksen, 1986). Sample cores were air-dried and weighed. A subsample was dried at 105°C for 24 h to determine the air-dried water content. Bulk density was calculated with air-dried core weights corrected to oven-dried conditions with the subsample air-dried water content.

Statistical Analysis
Due to the systematic arrangement of the sampling treatments, homogeneity of variance tests were conducted to evaluate whether the treatments could be compared. Differences in pore characteristics among scans along the soil core were statistically compared to evaluate depth and management influences. Statistical analysis of data was completed assuming a completely randomized design with scans at different depths within a core as a split-plot. The PROC GLM procedure in SAS was used to test differences in depth within and among treatments and to compare differences among treatments (SAS Institute, 1999). Means, standard deviations, and differences among means for the measured parameters were determined with SAS (PROC MEANS). The MIXED procedure was used to determine differences among treatments within the same depth. Statistical differences were declared significant at the {alpha} = 0.05 level.

The Stepwise regression procedure was used to determine relationships between saturated hydraulic conductivity and one parameter, two parameter, or three parameter model combinations of CT-measured pore parameters. The following CT-measured pore parameters were used in the analysis: CT-measured number of pores, number of macropores, macroporosity, mesoporosity, total porosity (macroporosity plus mesoporosity), pore circularity, and fractal dimension of macroporosity as well as bulk density.


   Results and Discussion
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results and Discussion
Conclusions
 REFERENCES
 
Soil Bulk Density
Soil bulk density ranged from 0.98 to 1.45 Mg m–3 among the four treatments (Fig. 2A ). The corn–soybean (CS) treatment had the highest bulk density throughout the profile although bulk density was not significantly different among treatments at some depths. The observed bulk density for the native prairie site was 84% of the cropped sites (Table 2 ); this effect is probably due to the roots of the permanent vegetation and absence of agricultural tillage and traffic. The NP site had the lowest bulk density among the four treatments for the two surface depths. CRP and restored prairie (RP) sites had the lowest bulk densities for the third and the fourth depths, respectively, and these two densities were different from the respective CS density. Similar to our results, Kremer and Anderson (2005) observed the lowest bulk density in a native prairie and the highest for a row crop site. They evaluated only the surface 0- to 10-cm depth and the CRP had slightly lower bulk density than the restored prairie, although not significantly different.


Figure 2
View larger version (22K):
[in this window]
[in a new window]

  		Fig. 2. Mean soil bulk density (A) and saturated hydraulic conductivity (B) for native prairie (NP), restored prairie (RP), Conservation Reserve Program (CRP), and corn–soybean rotational (CS) treatments (n = 6). The bar or number indicate LSD (0.05) values (bulk density and Ksat of the NP treatment for the 40–50 cm depth zone were 1.15 Mg m–3 and 18.6 mm h–1, respectively).

 

View this table:
[in this window]
[in a new window]

  		Table 2. Bulk density, saturated hydraulic conductivity (Ksat), computed tomography (CT)-measured number of pores, number of macropores, macroporosity, mesoporosity, circularity, and fractal dimension as influenced by depth and treatment (n = 6) and the ANOVA.

 
The average bulk density of the 10- to 40-cm depth of the CS treatment was 1.45 Mg m–3; values greater than this are considered to restrict root growth (Udawatta and Henderson, 2003). Establishment of prairie and CRP pasture appeared to significantly reduce (P < 0.05) the bulk density at two depths (0–10, 30–40) throughout the profile. The surface of the restored prairie showed a 14% reduction (P < 0.05) in bulk density as compared to the CS treatment, and this treatment was only 13% greater (P < 0.05) than the native prairie. The lower soil bulk density in the prairies and CRP as compared to CS can be attributed to abundance of roots and biopores (Jiang et al., 2007). The vegetation in the prairies has been identified as grasses and forbs. Deep-rooted forbs and grasses in these areas may have contributed to observed differences in bulk density among the treatments.

Soil Hydraulic Conductivity
Saturated hydraulic conductivity showed effects of management (Fig. 2B; Table 2); however, overall management effects were not significantly different (P = 0.39), and the CS treatment was significantly lower (P < 0.10) than the others. The CS treatment had the lowest Ksat (24 mm h–1) averaged across depths while the two prairie sites had the highest (61 mm h–1). Our results support previous findings that soil hydraulic conductivity for native prairie is greater than crop soils (Fuentes et al., 2004). The Ksat of the CRP (58 mm h–1) treatment was much closer to the prairie than to the CS. Jiang et al. (2007) found a smaller Ksat for the CRP treatment and similar Ksat for the crop treatment in the same study area as the present study.

Differences among the treatments in Ksat for the 0- to 10-cm depth ranged from the lowest 87 mm h–1 for the CS to the highest 200 mm h–1 for the RP treatment. Approximately 60 to 80% of the crop, grass, and tree roots occupy the surface 0 to 10 cm soil and roots and root channels can develop into connected pores (Chaturvedi and Das, 2003; Udawatta and Henderson, 2003). Saturated hydraulic conductivity generally decreased with depth. For the second and third depths, Ksat showed a distinct difference between the CS and the other three treatments. The 10- to 30-cm depth Ksat for the CS and the average for the other three treatments were 3.6 mm and 36 mm h–1, respectively.

The results of this study support previous findings that soil hydraulic conductivity is generally higher under permanent vegetative areas than crop areas (Rachman et al., 2005; Seobi et al., 2005; Udawatta et al., 2006). Differences can be attributed to better soil structure, well-preserved pore networks, and enhanced macropore flow, roots, and organic matter additions (Obi, 1999; Mishra et al., 2003). However, the degree of change in Ksat varies with soil, vegetation, and duration of the permanent vegetation (Mazurak et al., 1960; Schwartz et al., 2003; Fuentes et al., 2004; Jiang et al., 2007). For example, Mazurak et al. (1960) in Nebraska observed infiltration rates under perennial grasses within 16 yr of establishment approaching those of native grassland on a silt loam soil, while Schwartz et al. (2003) on the Southern Great Plains found little impact of CRP management on improving hydraulic conductivity on fine-textured soils after a 10-yr period. In this study, restored prairie vegetation was on the site for 12 yr and CRP management for 14 yr. These treatments have already started to create positive changes in Ksat. It is anticipated that, as plant roots occupy more soil volume, soil hydraulic properties will further improve. However, it may take a considerable amount of time to observe significant improvements in the deeper horizons.

CT-Measured Pores
Two terms, depth zone and scan depth, will be used to distinguish between the four depth zones or soil core depths (0–10, 10–20, 20–30, and 30–40 cm) and the 20 scan depths (five scans per depth zone), respectively, to explain CT-measured pore characteristics. The distribution of CT-measured number of pores varied among the treatments and depth zones studied (Fig. 3A , Table 2). Significant differences in number of pores were found among the treatments (P < 0.01). Native prairie, RP, CRP, and CS had an average of 83, 43, 48, and 26 pores across all scan depths on a 2500 mm2 scan area, respectively. Native prairie had 120, 72, 68, and 72 pores at the 0 to 10, 10 to 20, 20 to 30, and 30 to 40 cm depth zones, respectively, while the CS treatment had the lowest number of pores (62, 24, 17, and 0) for these depth zones (P < 0.05). The number of pores in each soil depth zone was similar between the RP and CRP treatments and they were significantly higher (P < 0.01) than the CS treatment.


Figure 3
View larger version (27K):
[in this window]
[in a new window]

  		Fig. 3. Number of computed tomography (CT)-measured pores (A) and macropores (B) for native prairie (NP), restored prairie (RP), Conservation Reserve Program (CRP), and corn–soybean rotational (CS) treatments by depth (n = 6). Bars indicate LSD (0.05) values (average number of pores and number of macropores of the NP treatment for the 40–50 cm depth zone were 57 and 14, respectively).

 
The NP treatment had more pores than the other three treatments at each scan depth, except for the 1.7-cm depth (Fig. 3). Generally, the number of pores decreased with soil depth for all treatments and CS had no pores below 27 cm (Fig. 1; Table 2). Among the 20 scan depths, NP had two times or more CT-measured pores as compared to the CS treatment, except for three depths. On average, these three scan depths (3.9, 5.0, and 17.2 cm) of NP had 1.65 times more pores than the CS treatment. Restored prairie and CRP treatments had two times or more pores for 15 scan depths as compared to the CS treatment.

The CT-measured number of macropores in the CS treatment was significantly lower (P < 0.01) than the other three treatments (Fig. 3B; Table 2). Native prairie, RP, CRP, and CS treatments had an average of 23, 13, 12, and 4 macropores across all scan depths per 2500 mm2 area and the number of macropores decreased with soil depth for all treatments. The rate of decrease was gradual between two consecutive scans within the first depth zone. The largest decrease occurred between the first and the second depth zones, except for the CS treatment. However, the rate of decrease was smaller for the NP treatment. It had 34 and 18 macropores per 2500 mm2 for 0 to10 cm depth zone and 10 to 40 cm depth zones, respectively, whereas RP had 35 and 6 macropores per 2500 mm2 for those respective depths. The CRP treatment had 40 and 3 pores per 2500 mm2 for those respective depth zones. The CS treatment had 8, 6, and 3 macropores per 2500 mm2 area for 0 to 10, 10 to 20, and 20 to 30 cm depth zones. No macropores were detected below 27 cm for the CS treatment while a sharp reduction in macropores was detected below 27 cm for RP and CRP treatments.

Management practices mostly affect the number and area of large elongated pores (Pachepsky et al., 1996). Shallow roots of seasonal crops of the CS treatment may not have penetrated the subsoil and had little effect in developing CT-measured pores and macropores, especially in the subsurface depths. Restored prairie and CRP tend to improve CT-measured macropores and pores, respectively. However, such changes were prominent only in the surface 30-cm depth as compared to the crop soil. Literature shows that soils under permanent vegetation have more pores than agricultural areas due to more roots and organic matter (Chan and Mead, 1989; Seobi et al., 2005; Udawatta et al., 2006). Forbs and grass species under restored prairie had more influence than the CRP grass on the development of macropores in the subsurface soil. Longevity of roots and a longer active growing season may have promoted more pores. Greater root development and fauna activity under permanent vegetation also may have attributed to a larger number of pores (van Noordwijk et al., 1991b; Bharati et al., 2002; Rachman et al., 2005). Soil data support that deep roots of the NP site had more soil carbon in the surface and Bt horizons than other sites (Table 1). Studying soil carbon dynamics at Tucker Prairie (the same NP site), Buyanovsky et al. (1987) stated that prairies had more litter and structural biomass which reflects the perennial nature of roots. The greater number of pores found under the native and restored prairies as well as CRP grass areas in this study can be attributed to greater root development and subsequent root decay, addition of soil organic matter, and improvement in soil physical properties due to permanent vegetation as compared to seasonal crops.

CT-Measured Macroporosity and Mesoporosity
CT-measured macroporosity (diam. > 1000 µm) as influenced by the prairies and CRP was significantly higher as compared to the CS treatment (P < 0.01; Fig. 4A ; Table 2). CT-measured macroporosity values averaged across all scan depths were 0.027, 0.016, 0.021, and 0.005 m3 m–3 for NP, RP, CRP, and CS treatments, respectively (Table 2). The CS macroporosity was 18, 31, and 24% of the NP, RP, and CRP profile porosity. Native prairie macroporosity was 69, 29, and 440% greater than RP, CRP, and CS macroporosity, respectively. The CRP treatment had 31% greater macroporosity as compared to the RP treatment. This was mainly due to the higher macroporosity in the surface 0- to 10-cm depth of the CRP treatment (Fig. 1). The greatest differences among the treatments were found for the 0 to 10 cm depth zone. The CRP treatment had 1.4, 1.6 and 5.9 times greater macroporosity than NP, RP, and CS treatments, respectively, for this depth zone. Macroporosity declined sharply from the first to the second depth zone with the descending order of CRP (7.8 times smaller) > RP (4.4) > NP (1.9) > CS (1.7). The NP treatment maintained the highest macroporosity throughout the profile while CS had the lowest. The difference between RP and CRP was not significant for 10 to 40 cm depth zones. No CT-measured macroporosity was observed beyond 27 and 36 cm scan depths for CS and CRP treatments, respectively. The macroporosity of the fourth depth zone of the NP treatment (0.015 m3 m–3) was better than the macroporosity of the surface depth zone of the CS treatment (0.012 m3 m–3). Also, the macroporosity of the fourth depth zone of the NP treatment was only smaller than the 0 to 10 cm depth zone macroporosity of the RP (0.044 m3 m–3) and CRP (0.067 m3 m–3) treatments.


Figure 4
View larger version (25K):
[in this window]
[in a new window]

  		Fig. 4. Computed tomography (CT)-measured macroporosity (A) and mesoporosity (B) for native prairie (NP), restored prairie (RP), Conservation Reserve Program (CRP), and corn–soybean rotational (CS) treatments by depth (n = 6). Bars indicate LSD (0.05) values (average CT-measured macroporosity and mesoporosity of the NP treatment for the 40–50 cm depth zone were 0.010 and 0.005 m3 m–3, respectively).

 
Native prairie and CRP had similar mesoporosity (Fig. 4B; Table 2). Restored prairie and CS mesoporosity values averaged across depths were 57 and 29% of the NP or CRP mesoporosity. Within the 0 to 10 cm depth zone, CRP had significantly higher mesoporosity than the other three treatments and it was 2.6 times larger than the NP and RP mesoporosity (Fig. 4B). Mesoporosity decreased with depth. For the 10 to 40 cm depth zone, NP had consistently higher mesoporosity than the other three treatments. The mesoporosity of CS cores was the lowest for all sampled depths.

Literature shows that permanent vegetation improves CT-measured soil porosity as compared to agricultural crops (Rachman et al., 2005; Udawatta et al., 2006). Rachman et al. (2005) and Udawatta et al. (2006) observed differences in CT-measured mesoporosity and macroporosity below grass or trees compared to row crop management. They attributed these differences to roots, duration of the growing season, agricultural activities, and organic matter. We attributed the lower macroporosity of the crop treatment in the current study to less root activity and associated biological activity.

Pore space structure is important in enhancing the ability to transport water (Pachepsky et al., 2000), especially macropores which not only reduce nutrient runoff losses but also runoff volume (Cadisch et al., 2004). In the NP, RP, and CRP treatments, CT-measured macroporosity was 78% of the CT-measured porosity (macroporosity plus mesoporosity), whereas it was only 71% for the CS. Although the ratio was similar to the other three treatments, the absolute porosity was only 26% of the other three treatments. Macropores can contribute up to 89% of the flow for well structured soil (Lin et al., 1996). In their study, water flow was primarily controlled by root channels, vertical fissures, and slickensides. In support of water flow in soil and the role of roots, Rasse et al. (2000) found that alfalfa (Medicago sativa) root systems increased total porosity by 1.7%, macroporosity by 1.8%, and saturated hydraulic conductivity by 57%. Their work suggests that dead roots increase connectivity of macropores rather than the macropore volume as the relative increase in hydraulic conductivity is not proportional to the increase in porosity. van Noordwijk et al. (1991a) observed that tree roots change pore size distribution as decaying roots were found to increase macropore flow in a previously forested site. This study did not measure temporal variability to evaluate pore connectivity or pore volume as influenced by roots of the permanent vegetation. Results show that native prairies improved macroporosity and mesoporosity throughout the profile and restored prairies may do the same in due course. However, the role of CRP appears to be marginal in terms of improving CT-measured porosity in deeper horizons. Results of this study suggest that restoration of prairies improve soil physical properties at a much greater rate than can be achieved with CRP or conventional cropping practices.

CT-Measured Pore Circularity
Computed tomography-estimated macropore circularity values were significantly larger for the CS treatment compared to the other three treatments (P < 0.01; Fig. 5 ; Table 2). Profile average circularity values were 0.88, 0.89, 0.91, and 0.93 for NP, RP, CRP, and CS treatments, respectively. The circularity values of the CS treatment (highest mean, 0.93) varied between 0.87 and 0.97, whereas circularity of NP (lowest mean, 0.88) varied between 0.82 and 0.92. The RP treatment had values similar to the NP except for one scan depth (32.8 cm) which had a circularity of 0.94. Compared to the other three treatments, the three surface scan depths of the CRP had the lowest circularity values indicating more elongated and larger pores (Fig. 1). This resulted in the lowest circularity (0.83) among treatments and depth zones. Circularity increased with depth. NP and RP treatments maintained lower circularity (0.84 to 0.88) within the 0 to 20 cm depth zones. Circularity increased (>0.91) below the 20-cm depth. However, circularity of CRP and CS treatments increased below 10 cm corresponding to a reduction of larger pores.


Figure 5
View larger version (13K):
[in this window]
[in a new window]

  		Fig. 5. Pore circularity values for native prairie (NP), restored prairie (RP), Conservation Reserve Program (CRP), and corn–soybean rotational (CS) treatments by depth (n = 6). The bar indicates the LSD (0.05) value (average circularity of the NP treatment for the 40–50 cm depth zone was 0.91).

 
Results of this study support previous findings that pore circularity of permanent vegetation is lower than for crop soils (Rachman et al., 2005; Udawatta et al., 2006). Rachman et al. (2005) observed 10% lower circularity values for soil under grass as compared to soil under crop areas in Iowa while Udawatta et al. (2006) observed 22% more circular pores under crop soils than agroforestry and grass buffers in Missouri. The difference in these two studies and the current study could be attributed to sampling depth, soil type, vegetation maturity, and management.

The larger the pore, the higher the probability that the pore is elongated or planar and the lower the probability that it is round (Mermut et al., 1992; Pachepsky et al., 2000). Results of the current study show that pore shape or form was highly correlated with vegetation management. In this study, pore circularity was the smallest in soil under the prairies and the highest under row crop management. Pore shape, roughness, and circularity may have changed due to soil aggregation, root activity, and macrofauna activities. The results of the current study indicate that restoration of prairies had increased more elongated larger pores in soils as compared to CS or CRP practices.

CT-Measured Pore Fractal Dimension
Fractal theory has been applied to characterize particle and aggregate distributions in soils and fractal dimension presented in this section apply to macropores. Fractal dimension was significantly different among the four treatments (P < 0.01) and dimension decreased with depth (Table 2). Native prairie treatment had the highest fractal dimension throughout the profile and this dimension was different from the other three treatments (Fig. 6 ; Table 2). The fractal dimension of the NP treatment declined marginally as compared to the other three treatments and the differences between the NP and other treatments were larger for deeper depths as compared to the 0 to 10 cm depth zone. Among the four treatments, the CS treatment had the lowest fractal dimension for any given depth zone and values were missing below 27-cm depth as there were no CT-measured pores. The average CS treatment fractal dimension for the upper three 10 cm depth zone increments was significantly lower (P < 0.05) compared to the NP fractal dimension for the same depths.


Figure 6
View larger version (14K):
[in this window]
[in a new window]

  		Fig. 6. Fractal dimension of computed tomography (CT)-measured macroporosity for native prairie (NP), restored prairie (RP), Conservation Reserve Program (CRP), and corn–soybean rotational (CS) treatments by depth (n = 6). The bar indicates the LSD (0.05) value (average fractal dimension of the NP treatment for the 40–50 cm depth zone was 1.336).

 
Fractal dimension closely followed the distribution of CT-measured soil pores (Fig. 1 and 3). Coefficients of determination between fractal dimension and the number of pores were 0.77, 0.57, 0.66, and 0.48 for NP, RP, CRP, and CS treatments, respectively. The fractal dimension of the CS treatment (1.293) was between the values reported by Rachman et al. (2005; 1.305) and Gantzer and Anderson (2002; 1.26) for similar soil management. Differences in soils and sampling depths may have contributed to these differences. Estimation of the fractal dimension is also dependent on image resolution, techniques for estimating fractal dimension, and thresholding. Rachman et al. (2005) reported a fractal dimension of 1.559 for grass hedges in Iowa. The sampling depth for their study was 20 cm. The average fractal dimension for the surface 20 cm depth zone of NP, RP, and CRP was 1.470 in this study. Soil water retention of fine textured soils predicted by fractal dimension showed a good relationship with measured water retention (Huang and Zhang, 2005). In another study, Baveye et al. (1998) observed that stain pattern and measured fractal dimension were highly correlated when a dye was used to study preferential flow. The higher fractal dimension observed in the NP, RP, and CRP treatments may suggest better preferential water flow than for the CS treatment.

Correlation of Pore Parameters and Saturated Hydraulic Conductivity
In this analysis, seven CT-measured pore parameters [number of pores, number of macropores, total porosity (macroporosity plus mesoporosity), macroporosity, mesoporosity, pore circularity, and fractal dimension] and bulk density were correlated using regression equations with saturated hydraulic conductivity. Regression relationships between CT-measured pore parameters with Ksat showed that the number of macropores predicted 43% of the variation (Table 3 ). The number of pores was the second best single variable explaining 39% of the variation in Ksat. The number of macropores and bulk density were the best two parameter combinations and accounted for 45% of the variation in Ksat. Increasing the number of parameters did not improve the coefficient of determination. Soil bulk density as a single variable explained only 22% of the variation in Ksat. However, none of the combinations explained more than 45% of the variation in Ksat.


View this table:
[in this window]
[in a new window]

  		Table 3. Relationships between CT-measured pore parameters and bulk density with saturated hydraulic conductivity.

 
The pore data used in regression relationships with Ksat were determined using scanning and image analysis techniques as compared to traditional data from water retention methods (Anderson et al., 1990). In this analysis, the means of the five scans from each depth zone were used as the core parameter for that property for the regression. Those properties were not measured throughout the core using adjacent scans and therefore pore parameters did not indicate exact pore volume, nature of the pore, pore throats, or pore connectivity. However, results showed that these pore parameters which increase with increasing porosity were positively correlated with Ksat. The number of macropores and number of pores ranked the best. Increased macroporosity should increase infiltration and reduce sediment transport capacity of water (Dosskey et al., 2007). In contrast, pore circularity which decreases with larger soil pores correlated negatively with Ksat. A pore with the same cross-sectional area but lower circularity will have more resistance to flow due to intermolecular attractions between the fluid molecules and solid wall (Hillel, 1982). Circularity values of one indicate the least cross-sectional surface and thus the lowest pore wall area that has contact with a moving media. Although conservation practices have improved the number of pores, number of macropores, total porosity, and macroposity, these practices have reduced the circularity of macropores which will exert more resistance to flow as compared to pores with a circularity of one.

Results of this analysis suggest that CT-measured pore parameters can be used to compare management effects on soil properties, water movement, and gas movement. Although traditional methods can evaluate influences of management on changes in soil properties, they do not provide spatial information at a pore scale which illustrates treatment effects important for water and gas transport processes. Another main advantage is potentially better parameterization for models that more realistically represent processes associated with soil pores and their distribution.


   Conclusions
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results and Discussion
 Conclusions
REFERENCES
 
The purpose of the study was to examine changes of CT-measured pore parameters as influenced by prairie restoration and compare these with CRP and row crop management. Measured pore parameters and bulk density were used to understand relationships with saturated hydraulic conductivity. Computed tomography-measured total number of pores, number of macropores, macroporosity, mesoporosity, pore circularity, and fractal dimension of macroporosity were found to be significantly different among the treatments. Differences were larger in the surface 10 cm compared to deeper depths. The natural prairie soils showed differences throughout the profile as compared to soils being row cropped. The results also showed that restored prairies and CRP management improved soil pore parameters. However, improvements were more prevalent in the subsurface horizons of native and restored prairies compared to CRP management. The number of macropores and pores appeared to be the best parameters that correlated with measured hydraulic conductivity. Soil bulk density was negatively correlated with Ksat and explained only 22% of the variation. Results of the study show that restoration of prairies improve CT-measured pore parameters. However, the time needed to reach the conditions similar to that of natural prairies was not determined as temporal variations were not examined. Further research may be needed to estimate how restoration of prairies changes with time.


   ACKNOWLEDGMENTS
 
The authors gratefully acknowledge the financial support from the Prairie Fork Trust Fund.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results and Discussion
 Conclusions
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.


   REFERENCES
TOP
 NOTES
 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 PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Udawatta, R. P.
Right arrow Articles by Garrett, H. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Udawatta, R. P.
Right arrow Articles by Garrett, H. E.
Agricola
Right arrow Articles by Udawatta, R. P.
Right arrow Articles by Garrett, H. E.
Related Collections
Right arrow Ecosystem Management
Right arrow Ecosystem Restoration
Right arrow Best Management Practices

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
Soil Science Society of America Journal Journal of Plant Registrations The Plant Genome