| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Dep. of Crop, Soil, and Environmental Sciences, Univ. of Arkansas, 115 Plant Sci. Bldg., Fayetteville, AR 72701
* Corresponding author (kbrye{at}uark.edu)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: OM, organic matter • EC, electrical conductivity
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Of the rice-producing states in the United States, Arkansas is the leading rice producer accounting for more than 40% of the total United States rice production (ASB, 2005). Consequently, it is estimated that between 28000 and 33000 ha of cropland are land leveled annually primarily for rice production in the row-crop agricultural region of eastern Arkansas alone (Young et al., 2004). The potential agronomic benefits of land leveling, including the ability to flood-irrigate previously non-level soils, increased irrigatable acreage, improved distribution of irrigation waters, soil and water conservation, and improved uniformity of crop growth and yield, have been recognized for more than 50 yr (Whitney et al., 1950), but land leveling also has numerous potential productivity-limiting disadvantages.
Though recognized as a water-conservation practice, land leveling is a severe soil disturbance (Brye et al., 2005). Exposing subsoil can result in major changes in soil surface pH, decreased organic C, and sodium toxicity (Miller, 1990). Growth- and yield-limiting nutrient deficiencies (e.g., N and P) have been shown to be direct results of subsoil exposure following land leveling (Whitney et al., 1950; Eck, 1987; Robbins et al., 1997, 1999). Increased spatial variability of soil properties following land leveling has been implicated in non-uniform productivity of rice in Arkansas (Miller, 1990) and tropical lowland rice in the Philippines (Dobermann et al., 1995, 1997).
Inorganic fertilizers, such as N, P, and Zn, and/or organic soil amendments, such as composted or uncomposted poultry litter, can help restore postleveling productivity to some degree (Miller et al., 1990, 1991), but are not always successful (Miller et al., 1990). Walker et al. (2003) reported rice yields were between 6 and 45% lower in cut compared to filled areas on recently leveled clay soils in Mississippi and attributed some of the yield decline to decreased soil-test P. In addition to affecting productivity and ultimately yield, changes in soil chemical properties as a result of land leveling may also impact soil biology (Bardgett et al., 1996).
In a recent set of studies on a silt-loam Alfisol in east-central Arkansas, it was demonstrated that relatively shallow-cut land leveling activities (i.e., <15 cm of cut and/or fill) significantly altered the magnitudes and spatial distributions of near-surface soil physical (Brye et al., 2003, 2005), chemical (Brye et al., 2004a), and biological properties (Brye et al., 2003) and the relationships between soil physical and chemical properties and microbial biomass (Brye et al., 2003, 2004a). Brye et al. (2005) demonstrated that the hard pan that developed from years of machinery traffic had essentially been relocated closer to the new soil surface following shallow-cut land leveling, which reduced the effective soil rooting depth for subsequent crops by 25%. Consequently, first-year soybean growth and production following land leveling was highly variable (Brye et al., 2004b). Brye et al. (2004c) also showed that even a comprehensive postleveling characterization of soil surface properties, including those directly related to traditional soil fertility, provided insufficient data to accurately predict first-year soybean or second-year rice response following leveling in a silt loam.
Since rice production in the United States is somewhat unusual compared to other cereal grains in that rice is typically grown for an extended period in flooded-soil conditions, clay soils with low surface and subsurface hydraulic conductivities are desirable because the flood is easier to maintain. However, much of the previous research evaluating the effects of land leveling has been conducted on somewhat coarser clay- and silt-loam soils with little attention given to clay-soil response to land leveling in areas of concentrated rice production. In Arkansas alone, approximately 41% of the total rice production occurs on soils with a clay or clay loam surface texture (Wilson and Branson, 2004). On a clay soil in northeast Arkansas, soil bulk density and particle-size distribution have been shown to be significantly altered by relatively deep-cut land leveling (Brye et al., 2006). The frequency of clay-soil land leveling has substantially increased in the past few years and will likely continue into the near future. Therefore, evaluating soil biogeochemical property change and subsequent crop response following land leveling on clay soils is necessary to improve management capabilities to ensure maximum or near-maximum production from graded, fine-textured, alluvial soils.
The objectives of this study were to characterize the short-term effects of land leveling on (i) the magnitudes, variance, and spatial variability and distributions of selected soil chemical properties, and (ii) the relationships between soil chemical and biological properties in a Vertisol commonly used for rice production in the Mississippi Delta region of northeast Arkansas. Similar to previous results of shallow-cut land leveling on a silt-loam Alfisol (Brye et al., 2004a) and for effects on soil physical properties in a clay soil (Brye et al., 2006), it was hypothesized that land leveling significantly alters the magnitude, variance, and spatial variability and distribution of soil pH, EC, organic matter, and extractable soil nutrients and that land leveling significantly alters the relationships between soil chemical and biological properties of a clay-textured Vertisol.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Study Site Manipulations
Land leveling activities occurred over a 3-d period in mid-April 2004. Following initial land leveling activities, the entire field was disked and land planed numerous times to reduce soil clod size to an approximate diameter of <2 cm. Land leveling altered the surface drainage from the original East-to-West to a South-to-North orientation.
Soil Sampling and Measurements
Immediately before and within a 3 d period following land leveling, relative elevation was measured and soil samples were collected from the top 10 cm at each of the 50 grid points in the study area to characterize elevational and soil biochemical property changes as a result of land leveling. Elevational changes were measured manually using a laser level and stadia rod.
One set of 4.8-cm diameter soil cores was collected from the 0- to 10-cm depth within a 20-cm radius surrounding each grid point for soil bulk density determination (i.e., 50 total samples). The soil-core sampling chamber was beveled to the outside to minimize compaction upon sampling. Soil samples were oven dried at 70°C for 48 h, crushed, and sieved through a 2-mm mesh screen for soil chemical analyses (i.e., pH, electrical conductivity (EC), extractable nutrients, soil organic matter (OM), and total soil N and C). Soil pH and EC were determined with an electrode on a 1:2 soil-to-water solution. Dried and sieved soil was extracted with Mehlich-3 extractant solution (Tucker, 1992) in a 1:10 soil-to-extractant-solution ratio and analyzed for extractable nutrients (i.e., P, K, Ca, Mg, Na, S, Fe, Mn, Zn, and Cu) using an inductively coupled argon-plasma spectrophotometer (CIROS CCD model, Spectro Analytical Instruments, MA). Organic matter was determined on sieved soil by weight-loss-on-ignition after 2 h at 360°C (Schulte and Hopkins, 1996). Total soil C and N were determined by high-temperature combustion using a LECO CN-2000 analyzer (LECO Corp., St. Joseph, MI) and used to calculate soil C/N ratios.
A second set of soil samples consisting of 10, 2-cm diameter cores were collected and composited from the 0- to 10-cm depth from within the same 20-cm radius surrounding each grid point. Samples were kept cool and within 1 d of collection were sent to the Soil Foodweb (Soil Foodweb, Inc., Corvallis, OR) for total fungal and bacterial biomass determinations. Total fungal (Ingham and Klein, 1984) and bacterial biomass (Babiuk and Paul, 1970) concentrations were estimated by epi-fluorescent microscopy and sample staining. Additional details describing the microbial biomass determination procedures were reported in Brye et al. (2006).
Extractable soil nutrient and microbial biomass concentrations, expressed on a mass mass–1 basis [µg (g dry soil)–1], and the soil bulk density measured at each grid point were used to calculate extractable nutrient and microbial biomass contents, expressed on a mass area–1 basis (g m–2).
Statistical Analyses
Paired t tests were performed, with or without equal variance, to determine the overall effect of land leveling on soil chemical and biological properties (Minitab Version 13.31, Minitab Inc., State College, PA). Pearson linear correlations were performed to ascertain correlations among selected soil chemical properties, between microbial biomass and soil pH, EC, OM, and total N and C, and whether significant correlations determined before land leveling changed because of land leveling. Based on significant correlations, linear regression analyses were used to determine the relationships among selected soil chemical properties and between soil chemical and biological properties. Analysis of covariance techniques were used to determine if the slope and/or intercept differed between pre- and postleveling linear relationships (SAS Version 8.1, SAS Institute, Inc., Cary, NC).
The effects of land leveling on the spatial variability of soil chemical properties were determined by several methods. Homogeneity of variance was evaluated using Levene's test (Levene, 1960). Geostatistical analyses were also conducted using GS+ (version 5.1, Gamma Design Software, Plainwell, MI). Only isotropic semivariograms were considered and semivariance parameters for spherical, exponential, or linear models are reported for pre- and postleveling pH, EC, extractable soil nutrient contents, OM, and total N and C.
The effects of land leveling on the spatial distributions of soil chemical properties were determined by mapping pre- and postleveling soil chemical properties using Surfer 7 (Golden Software, Inc., Golden, CO). Point kriging, using all data and no specified search radius, was used as an unbiased, weighted linear interpolation method that minimizes total parameter variance by incorporating semivariogram functions, to create contour maps (Isaaks and Srivastava, 1989). Only the linear semivariogram function for all parameters was used to facilitate mapping and comparisons among soil chemical properties.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
1 m occur in northeast Arkansas on clay soils, but are less common. Brye et al. (2006) also reported the significant alteration of soil particle-size fractions and bulk density as a result of land leveling in this same field. Sand, silt, and clay fractions averaged 0.21, 0.24, and 0.55 kg kg–1, respectively, before leveling, but averaged 0.18, 0.21, and 0.61 kg kg–1, respectively, after leveling. Soil textural class did not change as a result of land leveling. Average soil bulk density in the top 10 cm increased (P < 0.001) from 1.12 g cm–3, ranging from 0.93 to 1.24 g cm–3, before leveling to 1.25 g cm–3 after leveling, ranging from 1.09 to 1.44 g cm–3. Consequently, land leveling significantly affected the magnitude, variability, and spatial distribution of soil chemical properties.
Soil Chemical Properties
Land leveling significantly affected (P < 0.05) all soil chemical properties measured in the top 10 cm, except for extractable soil Zn, which did not change (Table 1). Soil pH, EC, and Mehlich-3 extractable soil K, Ca, Mg, Na, S, Fe, and Cu contents increased significantly (P < 0.05) following land leveling. In contrast, soil OM, total N and C concentrations, C/N ratio, and extractable soil P and Mn contents decreased significantly following land leveling. Total soil N and extractable soil P has similarly been shown to decrease after land leveling in alluvial soils of Mississippi (Walker et al., 2003). These results are also generally similar to those of Brye et al. (2004a) who reported that soil EC and Mehlich-3 extractable soil K, Mg, Na, S, and Cu contents increased and soil OM decreased significantly following shallow-cut land leveling in a silt-loam Alfisol in east-central Arkansas.
|
Decreased soil OM, total C and N, and C/N ratios are particularly negative effects of land leveling. It is generally well-known that soil OM and total C and N decrease exponentially with soil depth; thus the simple removal of previous topsoil and exposure of subsoil during the land leveling process would result in lowered soil OM and, hence, total C and N. In addition, natural soil fertility is related to soil OM; thus nutrient availability from natural OM mineralization and nutrient cycling would be expected to be negatively affected by decreased soil OM (Rochette et al., 1999). Similarly, since soil OM is the major substrate for soil microorganisms, decreased soil OM is generally related to decrease microbial populations and biomass (Paul and Clark, 1996). Decreases in soil OM have also been shown to adversely affect numerous other soil physical and hydraulic properties, such as soil structure (Franzluebbers, 2002; Caravaca et al., 2001), aggregate stability (Chenu et al., 2000; Caravaca et al., 2001), infiltration (Franzluebbers, 2002), and water-holding capacity (Diaz-Zorita et al., 1999; Haynes and Naidu, 1998).
Since many soil macro- and micronutrients, such as K, Ca, Mg, Na, and Cu, are mobile cations in soil or have their mobility increase under flooded-soil conditions for rice production (i.e., Fe and Mn), it is not surprising that these soil nutrients generally increased after leveling. In contrast to the decline in soil OM and total C and N, exposure of subsoil following land leveling resulted in generally higher nutrient contents, expect for Mn and P.
Aside from altering the magnitudes of numerous soil chemical properties and similar to the result of shallow-cut land leveling a silt-loam Alfisol (Brye et al., 2004a), land leveling also significantly affected soil chemical property variability. Land leveling increased (P < 0.05) the overall variability of soil pH and extractable soil P, K, Na, S, Mn, Cu, and Zn contents, while that of soil EC, OM, total N and C concentrations, C/N ratio, and extractable soil Ca, Mg, and Fe contents did not change (Table 2).
|
|
Based on contour mapping, the spatial distributions of all soil chemical properties measured in the top 10 cm in this study were altered to some degree by land leveling. At least one portion in the middle of the study area with relatively low soil pH before leveling resulted in higher soil pH after land leveling exposed slightly more alkaline subsoil (Fig. 1 ). With the exception of one spot, soil EC was generally uniformly low throughout the entire study area before leveling, but soil EC increased relatively uniformly after leveling (Fig. 1). Similar spatial distribution changes occurred with soil OM, total N and C, and C/N ratio (Fig. 2 ) and extractable soil nutrient contents (Fig. 3 , 4, and 5) . The most visually noticeable changes occurred with the spatial distributions of total soil N (Fig. 2) and extractable soil Na and S contents (Fig. 4).
|
|
|
|
|
Relationships among Soil Chemical Properties
Since soil nutrient concentrations and availability are often controlled by soil pH and OM, pre- and postleveling relationships among soil pH and OM and extractable soil nutrient contents were examined to identify further specific effects of land leveling on soil chemical properties. Before land leveling, soil pH was positively correlated with OM, total C, C/N ratio, and extractable soil Na content, but negatively correlated with extractable soil P, S, Fe, and Mn (Table 4). There was no significant relationship between soil pH and soil EC or extractable soil Ca, Mg, and Zn contents. These results for a clay soil differ somewhat compared to results for a silt-loam soil (Brye et al., 2004a) where preleveling soil pH was also positively correlated with extractable soil Na, but not with OM, and soil pH was positively correlated with extractable soil Ca and Mg, but not Zn or EC.
|
Numerous changes in relationships among soil chemical properties occurred as a result of land leveling. After leveling, soil pH was still positively correlated with extractable soil Na and negatively correlated with extractable soil P, S, Fe, and Mn, but became positively correlated with extractable soil Ca and negatively correlated with soil OM, total C and N and extractable soil K (Table 4). Similarly, after leveling, soil OM was positively correlated with total soil C and N and negatively correlated with extractable soil Fe, but became positively correlated with soil EC and extractable soil K, S, and Mn and negatively correlated with soil C/N ratio. Several changes also occurred after leveling for correlations among total soil C and N, C/N ratio, and extractable soil nutrients (Table 4). Numerous changes in relationships among soil chemical properties, although not necessarily the same ones as were evident in this study for a clay soil, were also reported for a shallow-cut silt-loam soil (Brye et al., 2004a).
Changes in linear correlations among soil chemical properties are further substantiated by differences in either the slope, intercept, or both when linear regression relationships before and after land leveling were compared. Both the slope and intercept changed after leveling for the relationships between soil pH (as independent variable) and OM, total C and N, extractable soil K, S, Fe, and Mn, while only the slope changed in the relationship between soil pH and extractable soil Ca (Table 5). Similarly, the slope, intercept, or both changed after leveling for the relationships between soil OM, total C, and/or total N (as independent variables) and soil pH, soil OM, total soil N, C/N ratio, and extractable soil P, K, Na, S, Mn, and Zn (Table 5).
|
Relationships between Soil Chemical and Biological Properties
Despite the many significant preleveling correlations among selected soil chemical properties, only extractable soil Cu was positively correlated with bacterial biomass content before leveling (r = 0.31; P < 0.05). Neither fungal biomass content nor fungal-to-bacterial biomass ratio were related to any soil chemical properties before leveling in the Sharkey clay soil. In contrast, both fungal biomass content and fungal-to-bacterial biomass ratio were positively correlated with soil pH and extractable soil Ca, Mg, S, Fe, Mn, and Cu and negatively correlated with extractable soil K before shallow leveling activities in a silt-loam soil (Brye et al., 2004a).
After leveling, bacterial biomass content was positively correlated with soil EC, total C and N, and extractable soil S content, while fungal biomass content was positively correlated with total C and extractable soil P, Mn, and Zn contents (Table 4). Both fungal and bacterial biomass content were negatively correlated with soil pH, and fungal biomass content was negatively correlated with extractable soil Na content after leveling (Table 4). Soil fungal-to-bacterial biomass ratio was unrelated to any measured soil chemical property after leveling. In contrast, Brye et al. (2004a) reported no postleveling correlations between bacterial biomass content and soil pH, EC, OM, or extractable soil nutrient, but that fungal biomass content was positively correlated with extractable Mn and Zn contents and negatively correlated with extractable soil Na content after shallow leveling activities in a silt-loam soil in south-central Arkansas. Only the slope between bacterial biomass (dependent variable) and total soil C (independent variable), and both the slope and intercept between bacterial biomass (dependent variable) and extractable soil Cu content (independent variable), differed significantly as a result of land leveling (Table 5). Only the slope between fungal biomass (dependent variable) and extractable soil S content (independent variable), and the intercepts between fungal biomass (dependent variable) and total soil C, C/N ratio, and extractable soil P content, differed significantly as a result of land leveling (Table 5). The effects of land leveling on linear relationships among soil chemical properties and between soil chemical and biological properties in a clay soil appear to be fewer than those from a similar study conducted in a silt-loam soil in south-central Arkansas (Brye et al., 2004a).
Management Implications
With the significant alteration of magnitudes, variabilities, and spatial distributions of soil chemical properties as a result of land leveling, it is likely that crop responses would also be negatively affected even shortly after leveling. Norman et al. (2003) suggested that the potential productivity decline after land-leveling activities is less in alluvial Vertisols of the Mississippi River Delta region of eastern Arkansas, such as the Sharkey clay, due to similar topsoil and subsoil characteristics compared to other rice soils where topsoil and subsoil characteristics may differ more drastically. However, it has also been suggested that postleveling productivity may be uncertain due, at least in part, to disrupted and increased spatial variability of soil properties (Norman et al., 2003). Walker et al. (2003) reported rice yields were between 6 and 45% lower in cut compared to filled areas on recently leveled clay soils in Mississippi and attributed some of the yield decline to decreased extractable soil P. Although, in contrast to this study, Walker et al. (2003) did not sample immediately before land leveling, but only afterward in areas known to have been cut and filled.
In eastern Arkansas, N, P, and S are common yield-limiting factors for rice grown on clay soils (Norman et al., 2003). Since rice N requirements are typically supplied by inorganic fertilizer N, most commonly urea-N in Arkansas (Norman et al., 2003), a decrease in total soil N concentration following leveling is of little agronomic concern.
In this study, extractable soil S, K, Ca, Mg, Fe, and Cu contents and soil pH actually increased following leveling indicating that, from a strict and immediate soil nutrient availability standpoint, soil fertility is perhaps enhanced to some degree following leveling in the Sharkey clay soil. The exposure of K-rich subsoil by land leveling would potentially benefit the K nutrition of subsequent crops following leveling, such as rice and soybean (Norman et al., 2003). Although, Miller et al. (1990) indicated that K fertilization did not increase rice grain yields on some land-leveled soils in eastern Arkansas. However, any short-term enhancement of soil fertility will likely be short-lasted because of the significant decline in soil OM as the long-term source of inherent, natural soil fertility. Therefore, use of organic soil amendments, such as poultry litter, are commonly recommended for restoring productivity to recently leveled soils (Miller et al., 1990, 1991).
In contrast to soil N and S, the significant decrease in extractable soil P following leveling is of greater agronomic concern. Soil, in general, fixes P, but soil's P-fixation capacity tends to be enhanced in the presence of clay minerals (Stevenson and Cole, 1999). Thus, for a soil such as the Sharkey clay, which has a high montmorillonitic-clay fraction, any decrease in soil-test P, such as from excessive P mining by crops without the input of fertilizer P or by the physical removal of P-rich topsoil by land-leveling activities, may increase fixation of subsequent P additions. Increasing the P-fixation capacity of a clay soil after leveling will likely result in higher P-fertilization requirements to achieve the same pre-leveling yield response. Furthermore, the resulting increased extractable soil Fe after leveling may also contribute to an increased P-fixation capacity because of potential precipitation of insoluble Fe-phosphates under low soil pH conditions (Stevenson and Cole, 1999).
Time of year when land leveling is performed also likely affects the resulting magnitude of soil property change. In this study, land leveling was conducted in the spring a few weeks before planting. Had land leveling been conducted later in the year when the field could be left fallow for several months or less intensively managed with a winter cover crop, the magnitude of soil property change would likely have been less because of the additional organic matter from weed or crop roots, and the extra time allowed for microorganisms to acclimate to the new physical and chemical environment of the surface soil.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Previous results, along with the results of this study, indicate that relationships among soil biogeochemical properties are neither simple nor consistent. The more subtle effects of altered soil chemical properties, beyond those that are easily correctable with inorganic fertilizers, such as the effect on soil biological properties, will likely have immediate effects on postleveling crop productivity. Restoring pre-disturbance productivity to land-leveled soils may require precision agriculture technologies, such as variable-rate fertilization and/or site-specific organic soil amendments.
Received for publication October 16, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. E. Haggard, D. R. Smith, and K. R. Brye Variations in Stream Water and Sediment Phosphorus among Select Ozark Catchments J. Environ. Qual., November 1, 2007; 36(6): 1725 - 1734. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 | |||
Elev, m), and estimated soil volume manipulated during land leveling (Vs).
0.05 for the slope (ß1) and/or intercept (ß0) parameter indicates land leveling caused the pre-leveling relationship to change significantly.