SSSAJ Grow Your Career with SSSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

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 (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by House, M. L.
Right arrow Articles by Fekersillassie, D.
Right arrow Search for Related Content
PubMed
Right arrow Articles by House, M. L.
Right arrow Articles by Fekersillassie, D.
Agricola
Right arrow Articles by House, M. L.
Right arrow Articles by Fekersillassie, D.
Related Collections
Right arrow Mechanical Impediance
Right arrow Hydraulic Conductivity
Right arrow Spatial Distribution
Soil Science Society of America Journal 65:1376-1384 (2001)
© 2001 Soil Science Society of America

DIVISION S-1 - SOIL PHYSICS

Spatial Analysis of Machine-Wheel Traffic Effects on Soil Physical Properties

Marcia L. Housea, William L. Powersb, Dean E. Eisenhauer*,c, David B. Marxd and Daneal Fekersillassiec

a Dep. of Biological Systems Engineering, 138 L.W. Chase Hall, Univ. of Nebraska, Lincoln, NE 68583-0726
b Dep. of Agronomy, 279 Plant Science, Univ. of Nebraska, Lincoln, NE 68583-0915
c 223 L.W. Chase Hall, Dep. of Biological Systems Engineering, Univ. of Nebraska, Lincoln, NE 68583-0726
d Dep. of Biometry, 103 Miller Hall, Univ. of Nebraska, Lincoln, NE 68583-0712

* Corresponding author (deisenhauer1{at}unl.edu)


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
Infiltration in irrigation furrows exhibits spatial variation from furrow to furrow within a field. One major contributing factor is the effect of multiple levels of machine-wheel traffic on soil physical properties. The purpose of this study was to determine the effect of machine-wheel traffic levels within equipment passes on a field basis. Variations in satiated hydraulic conductivity (Ks), penetrometer resistance (Rp), and bulk density ({rho}b) due to nine, eight-row equipment passes were studied in three transects, crossing 72 furrows perpendicular to crop rows, on a Hord silt loam (fine-silty, mixed, mesic, Pachic Haplustoll). Mean values, spatial patterns, and regression relationships between properties were determined. Spectral analysis was used to fit cosine curves to property data that showed significant periods at 2.7, 8.0, and 72 furrows. An additional period of 24 furrows was seen in the Rp and {rho}b data. All properties tested showed significant mean differences due to wheel traffic from equipment passes. Within the equipment passes it was possible to further separate treatment means for all soil properties. Results in wheel tracked furrows were different from all other treatments. Linear regression of log Ks and Rp in a 72-furrow transect shows 58% of log Ks variability is explained by changes in Rp. Predicted vs. measured log Ks in two transects shows predictions somewhat high, although the slope of the linear regression is 0.95, nearly parallel to a 1:1 line.


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
AN UNDERSTANDING of infiltration variability on a field basis is needed for efficient management of surface irrigation systems. Primary sources of this variability are soil infiltration characteristics, which are both temporally and spatially varied, and the decreasing infiltration opportunity time with water advance along furrows (Tarboton and Wallender, 1989; Letey et al., 1984). Maximum application efficiency is inversely related to the magnitude of furrow infiltration variability; differences in infiltration are usually highest during the first irrigation event.

Furrow-to-furrow infiltration variability causes nonuniform water infiltration, furrow stream advance rates, and runoff rates. Both infiltration rate and the infiltration opportunity time determine infiltration depth at any location. Large spatial variability in infiltration has been established in many studies; coefficients of variation (CV) commonly range between 20 and 60%. The consequences of furrow-to-furrow inflow and infiltration variability are excessive runoff and deep percolation, while a portion of the field receives inadequate water (Trout, 1990).

One cause of variation in soil infiltration rates is wheel traffic patterns (Voorhees, 1977; Allen and Musick, 1997). Hillel (1980) stated that soil compaction in modern agriculture has been most commonly caused by machinery wheels, tracks, and soil-engaging tools. Research has shown that it is common for water infiltration in wheel-tracks to be reduced to approximately 50% of the infiltration without traffic (Lindstrom and Voorhees, 1980; Young and Voorhees, 1982; Ankeny et al., 1990; Allen and Musick, 1997). Kemper et al. (1982) measured reductions in infiltration rates from 12 to 80%. Kemper et al. (1982) and Allen and Musick (1997) found water content of the soil at the time of compaction had a significant impact on infiltration, as did the compacting loads. Ankeny et al. (1990) concluded that compaction primarily destroys the large pores. Allen and Musick (1997) found water advance rates in traffic furrows were twice as large as rates in nontraffic furrows, requiring extra management to avoid excessive runoff.

Lindstrom et al. (1981) found after 10 yr of wheel traffic, increased bulk density ({rho}b), and associated reduction of soil porosity in wheel traffic interrows resulted in lower saturated hydraulic conductivity. Young and Voorhees (1982) also found wheel traffic compaction can significantly reduce total pore volume and saturated hydraulic conductivity.

Wheel traffic can increase penetrometer resistance (Rp) to a depth of 30 cm (Voorhees et al., 1978, 1986; Voorhees, 1979). Young and Voorhees (1982) found increases in Rp as deep as 45 cm below wheel tracks. Because compaction decreases the total porosity of the soil (Young and Voorhees, 1982; Blackwell et al., 1986), increases in {rho}b occur below wheel traffic areas (Voorhees et al., 1978; Voorhees, 1979; Assouline et al. 1997; Allen and Musick, 1997; Lindstrom et al., 1981; Reicosky et al., 1981).

With each equipment pass, several levels of compaction are introduced to the soil simultaneously in a repetitive pattern across a field. This pattern is important to furrow irrigation management. On a field basis, it is important to understand and describe the variability of soil physical and hydraulic properties in two-dimensional space. While many studies have compared wheel track and non-wheel track soil properties, very little has been done to identify the repeatability of those measurements across a sequence of furrows. Therefore, this study was undertaken to determine the effects of machine-wheel traffic from eight-row equipment on the variability of soil physical properties across furrows on a field basis.

The main objective of this research was to characterize the effect of machine-wheel traffic on soil physical properties on a field basis, for a furrow-irrigated production system. Specific subobjectives were to:

  1. Determine the spatial patterns of Ks, Rp, and {rho}b of a surface soil having varying levels of wheel traffic
  2. Identify differences in soil Ks, Rp, and {rho}b in furrows having varying levels of wheel traffic
  3. Determine the spatial patterns of Rp in different depth intervals below furrows having varying levels of wheel traffic
  4. Evaluate the effect of depth on Rp treatment mean differences
  5. Develop regression relationships between Ks, Rp, and {rho}b in furrows having varying levels of wheel traffic
  6. Determine if satiated hydraulic conductivity (Ks) can be estimated from measurements of Rp


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
Field Site
This study was conducted at the Management Systems Evaluation Area site for water quality research in the Central Platte Valley near Shelton, NE. The soil within the study area is an alluvial Mollisol, 1 to 2 m deep. It overlays sand grading to gravel and is characterized by a single mapping unit; Hord silt loam, terrace, 0 to 1% slopes (fine-silty, mixed, mesic, Pachic Haplustoll). General soil properties for this mapping unit are shown in Table 1.


View this table:
[in this window]
[in a new window]
  		Table 1. Properties of the Hord silt loam. Mapping Unit Hd, terrace, 0 to 1% slopes.{dagger}

 
Irrigation water is delivered to the field with gated pipe, and runoff is recovered downstream with a reuse pit. The field has been graded to a longitudinal slope of 0.0013 m m-1 (Fekersillassie and Eisenhauer, 2000). Corn (Zea mays L.) has been planted continuously for many years. Furrow length is 380 m and crop row spacing is 0.91 m. The width of this experimental area is 72 furrows or nine, eight-row equipment passes.

Experimental Design
Four levels of wheel traffic were induced in the furrows of each eight-row pass: (i) tractor wheels (WN, WS); (ii) outside dual-wheels (DN, DS), not used in all operations; (iii) non-wheel and guess (NWN, NWS, and G), having no traffic; and (iv) centerline (C), having a single pass of the center wheel of a self-propelled, high clearance, field sprayer. The sequence of furrow traffic treatments for each equipment pass from north to south across the field was NWN, DN, WN, C, WS, DS, NWS, and G. Subscripts denote the north (N) and south (S) half of each equipment pass. Furrow numbering begins at the north side of the field with this study area using Furrows 297–368.

A disk–plant tillage system was employed, which is typically used by the producers of this region. Machine-wheel traffic was confined to the same furrows, beginning with the planting operation. Prior fall-ripping was performed with a deep-shank chisel plow at a 10 to 15° angle to crop rows. Field operations occurring in this annual cycle are:

Soil sampling and soil resistance data collection began the last week of June 1997, directly after the ridging operation and prior to the first irrigation event. Three transects were established across the sampling area, perpendicular to crop rows, at 65, 200, and 335 m from the upper end, and will be referred to as Transects 1, 2, and 3, respectively. Transect 1 was well downslope of the equipment turnaround zone inside the field boundaries. All soil samples and Rp measurements were taken from the center of each furrow along these transects.

Sample Collection and Analysis
Spatial patterns of Ks, Rp, and {rho}b, relative to varying wheel traffic, were examined using spectral analysis. These data were generated from soil samples collected from Transect 1 and the corresponding Rp field measurements.

Test results from undisturbed soil cores (150 by 50 mm), taken from the furrow surface to a depth of 50 mm, in all furrows of Transect 1 were used to calculate Ks. Additionally, in each of Transects 2 and 3, soil cores for Ks determinations were taken in one wheel track and one non-wheel furrow, in three of the nine equipment passes.

Sampling rings were cut to 70-mm lengths, to allow space for an initial 20-mm head of water above the soil in these rings for Ks testing. Specific field and laboratory procedures for sampling and core preparation are detailed in Trompke (2000). The cores were protected on the bottom by burlap fabric and wire mesh screens, held above the floor of the water-bath pan with rubber stoppers and satiated from below for a period of 24 h (48 h for the wheel track cores). Water was drained from the pan to 10 mm above the base of the core and time for the falling head of water to move 20 mm, from ring top to the soil surface, was recorded (Flannery and Kirkham, 1963). The equation to quantify Ks from the falling head test (Klute and Dirksen, 1986), is

(1)
where a is the cross-sectional area of the standpipe (L2), L is the sample length (L), A is the cross-sectional area of the sample (L2), t is time (T), and H1 and H2 are the hydraulic head differences across sample of length L at the beginning and end of the measurement period, respectively.

To account for any changes in viscosity and density of the percolating fluid, Ks was corrected to a 20°C standard with this equation:

(2)
where subscripts t and 20 denote test value and standard value, respectively; {eta} is the dynamic viscosity of the fluid; and {rho} is the density of the fluid.

Biological poison was not used as the time to satiate and test these cores did not exceed 3 d. Wheel track cores required additional pressure head to provide a measurable flux of water. Two additional rings were attached to the top of each test core with 50-mm sections of tire inner tubes, and again the falling head method (Klute and Dirksen, 1986) was used to determine Ks. In this case, small manometers constructed from bent glass rod and flexible tubing were hung over the top of the rings and taped to the outside of the rings, so that a drop in the water level could be measured easily. Following Ks determinations, oven-dry weights and the known volume of these cores were used to calculate {rho}b.

A cone penetrometer was used to measure Rp in the field at the furrow surface and at 150-mm intervals to a depth of 450 mm along all three transects. Maximum gauge readings were recorded for each interval. The measurements were taken with a gauged, hand penetrometer using a 12.9-mm2 (0.2 in2) cone, according to the procedure of ASAE standard: S313.2 Dec. 94 (American Society of Agricultural Engineers, 1997). All readings of Rp were obtained by the same two-person team, one operator and one recorder.

Statistical Analysis
Two different approaches were used to analyze the Ks, Rp, and {rho}b data. First, harmonic analysis provided a qualitative description of the cyclic pattern imposed across furrows by different levels of wheel traffic. Distance in single furrow intervals along Transect One was used as the independent variable. Significant periods (furrows) were not assumed a priori; instead, the normal distributions of Ks, Rp, and {rho}b data were fitted to a spectral density function. Period (2{pi}/frequency) lengths or the number of furrows associated with maximum spectral densities were selected from this procedure for use in the harmonic analysis. Amplitudes and phase angles of cosine curves were fitted to the data, and periods were tested for statistical significance.

Second, classic statistical methods were used to provide quantitative analysis of these data. Differences of Ks, Rp, and {rho}b at the furrow surfaces, relative to varying levels of wheel traffic, were found by comparing mean treatment values. Analysis of variance was used to find parameter means and to test machine-wheel traffic levels for significant differences within the eight-row equipment passes. Fisher's least square difference (LSD or lsmeans) test provided further separation of wheel traffic treatment means within a single equipment pass.

Spectral analysis techniques were used again to examine the spatial patterns of Rp with depth along all transects. Further, the effects of wheel traffic with depth on Rp treatment means were compared using an analysis of variance for predetermined depth increments below the furrows along all three transects. Differences were separated using Fisher's LSD test.

Linear regression relationships for {rho}b vs. Rp, log Ks vs. {rho}b and log Ks vs. Rp were calculated. Regression analysis used on these data allowed comparisons between all properties relative to the varying levels of wheel traffic. Mass and volumetric water contents ({theta}m and {theta}v) are included for referencing Rp measurements. Samples for {theta}m were taken for the 0- to 50-mm interval of each furrow and in all three transects.

The predictability of Ks from measurements of Rp was found with the regression of Ks as a function of Rp in Transect 1. That relationship was used to predict Ks in Transects 2 and 3 in the 12 furrows where Rp and the sampling of Ks occurred.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
SUMMARY AND CONCLUSIONS
 REFERENCES
 
Data Distributions
Statistical tests are based on the assumption that a data set exhibits normality (Sisson and Wierenga, 1981). All transect data sets were transformed and tested using the Shapiro-Wilk W statistic until normal distributions of each property were found.

Property distributions from furrow surfaces in Transect 1 were evaluated. Both Ks and Rp data were lognormally distributed, while {rho}b data had a normal distribution. Table 2 contains a listing of these properties along with corresponding units, sample size number (N) and value of the W statistic. Other quantified base parameters of these data are also listed in Table 2.


View this table:
[in this window]
[in a new window]
  		Table 2. Parameter values of the ln Ks, ln Rp, and {rho}b data sets.

 
Spatial Patterns of ln Ks, ln Rp, and {rho}b
Results of the qualitative analysis begin with Fig. 1, 2, and 3 ; the spectral density graphs of ln Ks, ln Rp, and {rho}b data, respectively, from Transect 1. Spectral density analysis provides an estimate of the proportion of the variance in the distance series that is accounted for by a particular frequency band. Peaks, smoothed by a technique using weighted nearest neighbor averaging, describes major periodic components of the data (Warner, 1998). Frequency has been converted to periods (furrows) in these figures for utility.



View larger version (14K):
[in this window]
[in a new window]
  		Fig. 1. Periods (furrows) of spectral density peaks—ln Ks.

 


View larger version (14K):
[in this window]
[in a new window]
  		Fig. 2. Periods (furrows) of spectral density peaks—ln Rp.

 


View larger version (16K):
[in this window]
[in a new window]
  		Fig. 3. Periods (furrows) of spectral density peaks—{rho}b.

 
These data show peak spectral densities occurring at periods of 2.7, 8.0, and 72.0 furrows for all properties tested. A peak also exists at a period of 24 furrows in the RP and {rho}b data. Based on these peaks, periods of 2.7, 8.0, 24.0 (Rp and {rho}b only), and 72.0 were tested for significance. Corresponding P values are reported in Table 3. The model constants for the cosine curves are also output from this procedure.


View this table:
[in this window]
[in a new window]
  		Table 3. Significant periods and cosine model constants.

 
The property data obtained from Transect 1, actually point data but plotted as lines in Fig. 4, 5, and 6 , are shown with their corresponding fitted curve model. The models were developed by summing all cosine curves having significant periods. In general, the models fit observed periodic spatial trends in these data well, although both maximum and minimum amplitudes tend to be under predicted. The repeatability of the periodic or cyclic spatial trends across the field due to wheel traffic can be clearly seen with this method of analysis. Fitted curves graphed together (not shown) indicate either a direct or inverse relationship exists between all three of the soil properties tested.



View larger version (45K):
[in this window]
[in a new window]
  		Fig. 4. Observed and fitted Ks along Transect 1.

 


View larger version (40K):
[in this window]
[in a new window]
  		Fig. 5. Observed and fitted Rp along Transect 1.

 


View larger version (38K):
[in this window]
[in a new window]
  		Fig. 6. Observed and fitted {rho}b along Transect 1.

 
The significant period of 2.7 furrows has no physical meaning; however, it is explained by the calculations performed with this method of analysis. Within each of the eight-row equipment passes, models show a sequence of three repeating periods, of 2, 3, and 3 furrows. Analysis placed a peak spectral density at the calculated average of these periods (8/3 = 2.7). A significant period of 24 furrows also exists in Rp and {rho}b data and is unexplained. The significant 72.0-furrow period appears to be unimportant to machine-wheel traffic effects on physical properties and may be model noise, as it is the maximum separation distance in these data.

Analysis of Variance
Quantitative differences in treatment means were determined next. Each of the eight furrows of a single equipment pass were tested independently even though non-wheel, dual-wheel, and wheel are replicated on the north and south sides. Values of P <= 0.0001 for all three properties along Transect 1 indicate significant differences in means across the eight-row set of furrows; that is, at least one mean in an eight-row pass is significantly different from other treatments due to the effects of machine-wheel traffic. Property means and standard deviations are: ln Ks = 2.83 ± 0.801 (Ks base units are mm h-1), ln Rp = 7.620 ± 0.370 (Rp base units are kPa), and {rho}b = 1.44 ± 0.078 Mg m-3.

Mean separation using the protected Fisher's LSD test was performed to find similarities and differences in property means of the individual furrows. Actual traffic treatment mean values for each property are reported in Table 4; however, the Ks and Rp tests were run on the transformed-normal data sets. The letter designations following the mean values separate traffic treatments into similar groups.


View this table:
[in this window]
[in a new window]
  		Table 4. Effect of wheel traffic on soil physical properties.

 
Clearly, for each of the properties, wheel-traffic furrow means are statistically alike and quite different from all other furrow treatments. This was apparent during both field sampling and lab testing as well. Although letter designations are somewhat different, both non-wheel and guess furrows show the three lowest mean values for {rho}b and Rp, and correspondingly the highest for Ks values. The mid-group of three mean values in all cases again are the outside dual-wheel and sprayer-wheel (centerline) furrow. Volumetric water content data has been listed in Table 4 as well.

Comparison of the measured Ks values with the known range of 16 to 51 mm h-1 for this soil (USDA, 1974) shows that three of the means fall within the expected range. Variation of Ks within furrows due to traffic levels better defines this range. Assuming no other changes occur, as wheel traffic causes Ks to decline and Rp and {rho}b to increase, infiltration rates will decrease within furrows and downstream runoff in an irrigation event will begin sooner.

Data obtained in 1998 from this field, taken independently of this study, show the range of {rho}b to be 1.33 to 1.53 Mg m-3 for the 0- to 300-mm interval. Soil survey reports show maximum dry density for the Hord silt loam soil to be 1.65 Mg m-3 in the 0- to 200-mm interval. Wheel traffic furrow means of {rho}b approach this value and all values fall closely within the range found in 1998.

Spatial Patterns of Rp with Depth
The observed and fitted Rp data along Transect 1, at the surface and in 150-mm depth intervals to 450 mm, are shown in Fig. 7 . Spatial patterns from wheel traffic effects on Rp at the surface diminish with depth. The number of furrows with maximum Rp values decline and are less ordered with depth. Minimum values of Rp are higher in every depth interval, relative to surface measurements. This probably results from soil loosening by tillage. The model found the 8.0-furrow period significant in the 0- to 150-mm interval; however, it was not significant below 150 mm. The 300- to 450-mm interval had significant periods of 3.0, 24.0, and 72.0 furrows.



View larger version (34K):
[in this window]
[in a new window]
  		Fig. 7. Maximum Rp at furrow surface and in 150-mm depth intervals to 450 mm along Transect 1.

 
Analysis of Variance of Rp with Depth
The effects of depth and wheel traffic on Rp were determined using analysis of variance, where each depth along each transect was tested separately. The results along Transect 1 are shown as P values in Fig. 7. There was a significant difference (P <= 0.10) in treatment means at the surface and in the 0- to 150-mm interval, but no significant differences due to machine-wheel traffic were found below 150 mm. Results for Transects 2 and 3 are the same and are not shown.

Treatment means and significant grouping of those means within an equipment pass are presented in Table 5 for Rp at the surface and 0- to 150-mm interval. As treatment differences were not significant for the 150- to 300- and the 300- to 450-mm intervals, only treatment means are listed.


View this table:
[in this window]
[in a new window]
  		Table 5. Mean Rp comparisons with LSD tests for all depth intervals.

 
Mean values of Rp in the 150- to 300-mm interval are relatively lower than means in the 0- to 150-mm interval. While mean differences for treatments within the equipment pass are no longer significant at lower depths, mean values of Rp are consistently higher in the 150- to 300-mm interval than the 300- to 450-mm interval in every treatment.

Regressions of Soil Properties
Relationships between the soil physical properties along Transect 1 were determined using regression analysis. The linear regressions of {rho}b vs. Rp, log Ks vs. {rho}b, and log Ks vs. Rp are shown in Fig. 8, 9, and 10 , respectively. Confidence intervals shown in all three figures are for a 95% probability.



View larger version (29K):
[in this window]
[in a new window]
  		Fig. 8. Regression of {rho}b vs. Rp along Transect 1 at 95% confidence intervals (CI).

 


View larger version (41K):
[in this window]
[in a new window]
  		Fig. 9. Regression of log Ks vs. {rho}b along Transect 1 at 95% confidence intervals (CI).

 


View larger version (42K):
[in this window]
[in a new window]
  		Fig. 10. Regression of log Ks vs. Rp along Transect 1 at 95% confidence intervals (CI).

 
Fifty-two percent of the variability in {rho}b can be explained by the changes in Rp. Fifty-eight percent of the variability in log Ks is caused by changes in {rho}b. It follows that a relationship exists between log Ks and Rp. The regression of those two properties shows 58% of the variability of log Ks is explained by differences in Rp. The range of the prediction interval for Ks spans just under two orders of magnitude at any {rho}b or Rp, yet this improves predictions in comparison to using averages.

Predictability of Ks from Rp
Predictability of Ks from Rp measurements in Transects 2 and 3 were tested using the regression equation obtained from Transect 1. Predictions are for the six furrows (12 points) where Ks samples were taken in each of Transects 2 and 3. Figure 11 shows the results of linear regression analysis for Ks predicted vs. Ks measured, and includes the 1:1 relationship and the 0.99 confidence interval. Predictions are somewhat high for this small sampling, the linear regression implies a bias factor near 2; however, it nearly parallels the one to one line (slope is 0.95). Soil texture and organic matter are not factors in this regression. Table 6 shows particle-size analysis from the samples taken in all three transects of the fifth equipment pass (both wheel and non-wheel furrows). Differences in textures and organic matter are insignificant.



View larger version (50K):
[in this window]
[in a new window]
  		Fig. 11. Log Ks predicted vs. log Ks measured along Transects 2 and 3.

 

View this table:
[in this window]
[in a new window]
  		Table 6. Particle-size analysis for all transects for Equipment Pass 5.

 

   SUMMARY AND CONCLUSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
REFERENCES
 
Machine-wheel traffic effects on soil physical properties were studied in irrigation furrows on a field basis for a Hord silt loam. Soil samples for hydraulic conductivity and bulk density, and field measurements of cone penetrometer resistance were taken immediately after the ridging (hilling) operation and prior to the first irrigation. The experimental area included three, 72-furrow transects, perpendicular to crop rows, or nine, eight-row equipment passes.

The distributions of ln Ks, ln Rp, and {rho}b were found to be normal, with means of 2.83 (base units of Ks, mm h-1), 7.62 (base units of Rp, kPa), and 1.44 Mg m-3, respectively. Properties show repeating spatial patterns along Transect 1, fitting a cosine model with significant periods of 2.7, 8.0, and 72.0 periods (furrows). Within an eight-row equipment pass, a sequential pattern of three repeating periods can be seen: 2, 3, and 3 furrows. The model placed the significant period at the average value (8/3 = 2.7). The 8.0-furrow period coincides with the equipment width. The 72.0-furrow significant period appears to be unimportant to wheel traffic effects on physical properties and may be model noise as it is the maximum distance of separation in these data. An additional significant period of 24 furrows also exists in the Rp and {rho}b data and is unexplained. Comparisons of the fitted curves for these properties show Rp and {rho}b data directly related to each other, and Ks inversely related to both Rp and {rho}b along a surface transect with varying levels of wheel traffic.

Quantitatively, our results confirm results of other studies regarding the effects of machine-wheel traffic on soil physical properties. Machine-wheel traffic reduces Ks (Lindstrom et al., 1981; Voorhees et al., 1986; Young and Voorhees, 1896), increases Rp at the surface and with depth (Allen and Musick, 1997; Blackwell et al., 1986; Reicosky et al., 1981; Voorhees, 1979; Voorhees et al., 1978, 1986; Young and Voorhees, 1982), and increases {rho}b, decreasing total pore volume (Allen and Musick, 1997; Assouline et al., 1997; Blackwell et al., 1986; Hillel, 1980; Lindstrom et al., 1981; Reicosky et al., 1981; Voorhees, 1979; Voorhees et al., 1978, 1986).

Each of the property distributions, ln Ks, ln Rp, and {rho}b, have significant differences in means within an eight-row equipment pass due to the varying levels of wheel traffic, P <= 0.0001. Further separation of treatment means within eight-row equipment passes with Fisher's LSD test showed the wheel-track means (WN, WS) were similar and quite different from all other traffic treatments. Non-wheel and the guess furrows (NWN, NWS, and G) have the lowest Rp and {rho}b means and correspondingly the highest Ks means in the sequence. Finally, both dual-wheel and sprayer wheel traffic furrows (DN, DS, and C), with lighter and/or less frequent traffic have property means between those of maximum wheel traffic and the non-wheel furrows. These results agree with previous research conclusions; both the distribution of equipment weight transferred to the soil surface and the number of repetitions of furrow traffic will affect changes in soil physical properties (Allen and Musick, 1997; Lindstrom et al., 1981; Voorhees et al., 1986; Young and Voorhees, 1982).

Repeating spatial patterns of Rp with depth, due to varying levels of wheel traffic, are apparent at the surface and are still significant at the 0.10 level in the 0- to 150-mm interval. Below 150 mm, mean differences due to wheel traffic are no longer significant.

Regression analysis shows 52% of the change in {rho}b can be explained by changes in Rp, 58% of the variation in log Ks is due to variations in {rho}b and 58% of the variability in log Ks is explained by differences in Rp.

A regression analysis of predicted vs measured values of Ks, generated from the relationship between log Ks and Rp in Transect 1, shows predictions somewhat higher than measured values. However, the regression line nearly parallels the 1:1 line, the slope is 0.95. Results suggest a need for incorporating machine-wheel traffic effects on soil physical properties into furrow irrigation management, rather than using mean values of Ks, Rp, and {rho}b.


   ACKNOWLEDGMENTS
 
Funding and advisory support: USDA Cooperative State Research Extension and Education Service, Nebraska MSEA Project, cooperators; Darrell G. Watts, and Derrel L. Martin, Department of Biological Systems Engineering, and James S. Schepers, Department of Agronomy, University of Nebraska-Lincoln. Joseph M. Skopp, Associate Professor of the School of Natural Resources, University of Nebraska-Lincoln. Technical support: Alan L. Boldt, Department of Biological Systems Engineering and Wallace W. Troyer, Department of Agronomy, University of Nebraska-Lincoln. Community support: Mark Thurber, Grubb & Ellis/Pacific Realty Group of Lincoln, field sample cold storage.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
Journal Series no. 13122 of the Agricultural Research Division, Univ. of Nebraska, Lincoln.

Received for publication September 5, 2000.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND 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 (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by House, M. L.
Right arrow Articles by Fekersillassie, D.
Right arrow Search for Related Content
PubMed
Right arrow Articles by House, M. L.
Right arrow Articles by Fekersillassie, D.
Agricola
Right arrow Articles by House, M. L.
Right arrow Articles by Fekersillassie, D.
Related Collections
Right arrow Mechanical Impediance
Right arrow Hydraulic Conductivity
Right arrow Spatial Distribution

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