Repeated Freeze-Thaw Cycle Effects on Soil Compaction in a Clay Loam in Northeastern Montana

Soil Physics

In recent years, there has been an increased global concern regarding the J. D. Jabro* impact of soil compaction on crop production and soil quality in modern W. M. Iversen mechanized agricultural farming systems. Farm equipment is heavier than R. G. Evans ever before, and many farmers have resorted to energy intensive deep tillage B. L. Allen to alleviate compaction. Freeze-thaw processes influence the physical proper- W. B. Stevens ties of soil, primarily soil compaction and structure. A 3-yr field study was Northern Plains Agricultural Research Lab. established in fall 2009 to investigate the effects of the dynamics of freeze- USDA-ARS thaw cycles (FTCs) on soil compaction in a clay loam. Results showed that 1500 N. Central Avenue frequent FTCs over the winter generally alleviated soil compaction at the 0- Sidney, MT 59270 to 30-cm depth. During the winter of 2009–2010, soil penetration resistance (PR) in compacted treatments that were subject to freezing and thawing conditions was significantly reduced by 73, 68, and 59% at depths of 0 to 10, 10 to 20, and 20 to 30 cm, respectively. In compacted soils that were not subject to freezing, PR was significantly reduced by approximately 50, 60, and 46% at the same respective depths of the soil profile presumably due to the biology of soil and disruptive effects of shrink-swell cycles caused by frequent wetting-drying processes. These results demonstrate that repeated FTCs can alleviate soil compaction and alter soil physical quality. We conclude that FTCs associated with typical winter weather conditions are the most effective and economical way to alleviate soil compaction and improve soil structure through the dynamics of FTCs.

Abbreviations: CEC, cation exchange capacity; CI, cone index; FTCs, freeze-thaw cycles; NGP, northern Great Plains; PR, penetration resistance.

oil compaction is defined as the packing effect of applied external forces that include heavy farm equipment (i.e., tractors, trucks, and harvesters), live- stock, and raindrops which can cause a reduction in soil bulk volume and largeS pores. Soil compaction adversely impacts soil physical properties that affect crop production directly through effects on soil structure and aggregation, large pore space, roots growth, aeration, and water and nutrient movement (Hamza and Anderson, 2005; Batey, 2009). Soil compaction is a global problem due to mechanized modern agricultural systems and is considered one of the most widespread types of soil degradation affecting agricultural land, soil quality, and crop production. It affects approximately 68 million hectares of land worldwide due to vehicu- lar traffic alone (Flowers and Lal, 1998). Soil compaction due to farming operations is an acknowledged problem in the northern Great Plains (NGP) of the United States (Jabro et al., 2009a, 2011; Montana, North Dakota, and Minnesota producers, personal communications, 2007, 2011, 2012) and in millions of hectares around the world. Deep tillage is

Mention of trade names, proprietary products, or specific equipment is intended for reader information only and constitutes neither a guarantee nor warranty by the ARS-USDA, nor does it imply approval of the product named to the exclusion of other products. Soil Sci. Soc. Am. J. 78:737–744 doi:10.2136/sssaj2013.07.0280 Received 12 July 2013. *Corresponding author ([email protected]). © Soil Science Society of America, 5585 Guilford Rd., Madison WI 53711 USA 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. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

Soil Science Society of America Journal often used to alleviate compaction, but farmers in the NGP area physical and hydraulic properties (Sillanpaa and Webber, 1961; question whether or not this is necessary because they have ob- Benoit, 1973; Asare et al., 1997, 1999; Lehrsch, 1998; Sahin et al., served large changes in soil structure from its compacted state 2008; Edwards, 2013; Fouli et al., 2013). in the fall to a much more mellow state in the spring. The few To date the effectiveness of the freeze-thaw mechanism for studies that have been done on the effects of freezing on compac- alleviating soil compaction is not clearly understood. The paucity tion were done in much warmer locations and concluded that of information on the effect of repeated FTCs on soil compac- there was little effect. In the NGP, most areas experience several tion and the ambiguous nature of this type of research suggest months with an average air temperature well below freezing. This the need for additional studies of FTCs’ effect on soil compac- freezes the soil to depths typically exceeding one meter. We have tion. Thus, a 3-yr study was conducted to evaluate the dynamics seen large changes in soil structure over the winter but have not of repeated FTCs on soil compaction in a clay loam soil in the quantified these changes. By lessening soil compaction, growers NGP region. In this study, soil compaction was also evaluated will not only increase their bottom line but also progress toward under unfrozen soil conditions. increased global food production at a time when that is becom- ing increasingly important. MATERIALS AND METHODS One of the processes affecting soil compaction is repeated Experimental Design cycles of FTCs over winter. Freeze-thaw cycles have been defined A split-split plot arrangement was used for this study. Four in different ways and differed considerably among studies (Baker replications each consisting of four plots were laid out. Within and Ruschy, 1995; Lehrsch, 1998; Ho and Gough, 2006; Henry, each replication two of the plots were kept above freezing while 2007; Edwards, 2013). Ho and Gough (2006) calculated FTCs the other two were allowed to freeze. One of the plots subject to using the daily maximum and minimum temperature above 0°C freezing and one of the plots protected from freezing were com- and below 0°C, respectively. pacted and the remaining two were left uncompacted. Thus, each Repeated seasonal cycles of freeze-thaw can naturally alle- replication had a frozen compacted, frozen uncompacted, unfro- viate soil compaction in which water in the pore space expands zen compacted, and unfrozen uncompacted treatment (Fig. 1). during freezing and contracts during thawing processes. Unger Soil temperatures were monitored to verify that the soils desig- (1991) and Henry (2007) indicated that reduction in soil pen- nated as those protected from freezing did indeed remain above etration resistance and bulk density observed in agricultural soils freezing, and it allowed us to quantify the number of FTCs over winter was attributed to the disruptive effects of frequent at each depth in the soils subject to freezing conditions. The FTCs on soil structure and particle configuration. amount of compaction was quantified with a soil penetrometer. Other soil processes such as periodic cycles of wetting-dry- The soil moisture measurements were monitored to ensure that ing processes can also reduce the effectiveness of soil compac- there were no significant differences in moisture between treat- tion due to shrink-swell dynamics. This mechanism is more pro- ments that could influence the penetrometer measurements. No nounced particularly in clayey soils with 2:1 lattice expansible significant differences in soil moisture were observed, so these clay minerals such as smectite and montmorillonite (Parker et data are not presented. al., 1982; Abou Najm et al., 2010). Research studies conducted to investigate the effect of freeze- Soil and Site Description thaw processes on soil physical properties and primarily soil com- A 3-yr soil freeze-thaw field experiment was initiated in paction and structure have yielded highly inconsistent results. fall 2009 at the Montana State University Eastern Agricultural Numerous studies indicated that repeated FTCs break down soil Research Center (EARC) located approximately 2 km north structure and reduce aggregate stability (Leo, 1963; Bisal and of Sidney, MT, United States (latitude 47.7255 N, longitude Nielsen, 1967; Asare et al., 1999; Oztas and Fayetorbay, 2003). 104.1514 W, and altitude approximately 650 m). The soil at Other previous studies showed that freeze-thaw processes reduce the study site was classified as Savage clay loam (fine, smectitic, soil compaction, enhance aggregation, and improve other soil frigid Vertic Argiustolls), consists of deep, well drained, nearly level soils formed in alluvial parent material. The soil con- tains an expansible 2:1 smectite clay mineral. Soil particle size distribution indicated the textural class of the surface layer (0 to 30 cm) to be consistently within the clay loam classifi- cation. The amount of sand, silt, and clay in the soil at 0- to 30-cm depth ranged from 200 to 210, 410 to 430, and 370 to 380 g kg-1, respectively. The soil contained 1.6% organic matter, pH ranged from 6.2 to 6.8, and cation exchange capacity (CEC) ranged from 22.7 to 24.7 cmol kg–1 soil at the 0- to 30-cm depth. The study area was planted to sugarbeet Beta( vulgaris Fig. 1. Randomized treatments and locations of thermocouple and soil L.) in 2008 and barley ( L.) in 2007. The last moisture sensors. Hordeum vulgare

738 Soil Science Society of America Journal tillage was performed in the fall of 2008 following the sugarbeet distance between the thermocouples could be accurately con- harvest. On 24 Oct. 2008 the area was disked (JD640, John trolled. The holes created by the penetrometer measurements Deere, Moline, IL), roller harrowed on 31 Oct. 2008 (Brillion, were used for the thermocouples which were installed at depths Brillion, WI), and leveled (Eversman 4512, Eversman, Denver, of 5, 10, 15, 30, and 61 cm in the center of each plot. CO) once on 04 Nov. 2008. Weeds were controlled with applica- Hydra Probes (Stevens Water Monitoring Systems, Portland, tions of glyphosate applied with an all-terrain vehicle. OR) measured soil moisture and temperature in four of the plots at one probe per treatment (see Fig. 1 for sensor placement de- Compaction Process Description tails). Soil moisture content and temperature were monitored ev- A farm truck with a single rear axle was loaded to simulate a ery 2 h. The sensors were connected to three CR10x data loggers truck at harvest loaded to the maximum legal road weight (Fig. though an AM416 and an AM16/32b multiplexer (Campbell 2). The truck was equipped with two tires 22.9 × 50.8 cm (9.00 × Scientific, Logan, UT). The loggers recorded the data every 30 20 nominal Society of Automotive Engineers size) with a mea- min and relayed it to a computer located in the research loca- sured 101.6 cm outside diameter on the front axle and four of tion’s office building via a Bluetooth radio transmitter (SD202, the same size tires mounted as duals on the rear axle. The tires Initium Co., Sungnam-shi, Korea). were inflated to 587 kPa (85 psi). The loaded truck weighed The plots were covered on 8 Oct. 2009, 24 Oct. 2010, and 12,628 kg. The rear axle weight was 9308 kg and the front was 5 Oct. 2011 and were uncovered on 7 June 2010, 17 May 2011, 3320 kg. The area was compacted by reversing the truck onto and 17 May 2012. Heating blankets commonly used on concrete the plot starting at the east edge of the plot (Fig. 1). The truck (Multi-Duty 7.0 × 3.4 m Thaw and Cure, Powerblanket, Salt was then driven forward and to the west the width of the dual Lake City, UT) were installed on the plots designated to be kept wheels on the rear axle and then backed onto the plot again. This frost free (Fig. 1). They were controlled by a thermistor buried 1.7 process was repeated until the truck was at the west edge of the cm deep and placed 20 cm from the corner of the blanket. The plot. The truck was then moved half of a tire width to the east so controllers (DuroStat 102720, FarmTek, Dyersville, IA) were set the strip of soil that was not compacted between the duals on the to supply power to the blanket when the thermistor registered first pass would be compacted by the passes that the truck made 2°C and shut the power off when the temperature reached 4°C. as it made the second pass to the east edge of the plot (Fig. 1). Each blanket had its own temperature controller and was wired There were still some ridges present in the plots where the soil to a separate electrical circuit breaker. The plots that were not squeezed up between the duals, so 2 d after the truck passes were heated were covered with laminated polyethylene tarps to pro- made a steel roller was used to flatten the ridges. The roller was vide conservation of moisture as in the unfrozen (heated) treat- 1.2 m in diameter and weighed 1887 kg. This caused only addi- ments that were covered with blankets. The blankets and tarps tional compaction and facilitated placing the thermocouples at are impermeable [<57 perm SI (57 kg s-1 m-2 Pa-1), <1 perm US appropriate depths and prevented excessive air movement under (57.2135 ng s-1 m-2 Pa-1)]. Plots were uncovered after the last the heating blankets (Multi-Duty 7.010 × 3.353 m Thaw and frost of each year (Fig. 3). Cure, Powerblanket, Salt Lake City, UT). Volumetric moisture content in soil before compaction process was near field capacity, approximately 34 to 35% (Table 1). The soil was compacted only in the first year of the study (fall 2009).

Temperature and Moisture Sensor Installation and Monitoring Thermocouples were used to monitor soil temperatures in the plots. Type T 0.52 mm2 (20 AWG) Teflon insulated wire was cut to the length required for each combina- tion of plot, depth, and data logger. The thermocouple junctions were formed with a spot welder (Hot Spot 1, DCC Corporation, Pennsauken, NJ). The thermocouples were taped to a 9.5-mm-diameter fiberglass rod so that the depth of insertion and Fig. 2. Wheel tracks from a farm truck during the compaction process. www.soils.org/publications/sssaj 739 measured in Pascals and is described as CI = F/π (d/2)2, where F is total force needed to push the penetrometer into the soil in Newtons (N), and d is diameter of the cone (Randrup and Lichter, 2001). At the time of soil PR measurements, soil water content was deter- mined for each of the 16 plots using a digital TDR soil moisture meter (Field Scout, TDR 300 Soil Moisture Meter, Spectrum Technologies). This device was equipped with 20-cm- long rods spaced 3.3 cm apart that are mounted on a 90-cm-long han- Fig. 3. Plots layout, immediately after removal of thermal blankets and tarps in spring 2011. dle. The rods are pushed into the soil by means of the handle, the soil The blankets covering the plots over the winter were an- moisture value is displayed on the LCD readout, logged to internal chored at each grommet with stakes constructed from 9.5-mm- memory, and the process is repeated at the next location. diameter steel rods that measured 41 cm long. The stakes were Dates of soil PR and average volumetric moisture content sharpened at one end and a flat washer was welded 10 cm from measurements from the surface to 30-cm depth are given in the other end to provide complete contact with the grommet, Table 1. All PR measurements and compaction processes were and a 10-cm piece of the same rod was welded at 90° to the rod performed at moisture contents near soil field capacity (34– to provide a handle to facilitate extraction. 35%), estimated by Jabro et al. (2009b).

Soil Penetration Resistance Measurements Freeze-Thaw Cycles The trailer-mounted penetrometer sensor (Veris In our study, freeze-thaw cycles were calculated using a daily Technologies, 2002) was used to obtain initial PR measurements maximum soil temperature above +0.5°C and a daily minimum in fall 2009 due to the high density of the soil created by the soil temperature less than -0.5°C. Table 2 shows the number of compaction process at moisture contents near field capacity in FTCs at the depths of 5, 10, 15, and 30 cm. Soil temperature clay loam. measurements showed that soil froze to a depth of 61 cm and It was unfeasible to measure PR at these higher levels of soil several FTCs occurred at depths exceeding 30 cm. compaction using a hand-held digital penetrometer. Conversely, the trailer-mounted penetrometer sensor was not used to mea- Statistics sure PR for other sampling dates to avoid soil disturbance and Statistical analyses were performed using the mixed mod- compaction from a trailer-mounted device and the pickup truck el of SAS software, with each sampling date (fall or spring) as under wet soil conditions. Soil PR readings were recorded in 2.5- a repeated measures variable of SAS (SAS Institute, 2011). cm increments to a depth of 30 cm. Soil PR measurements were Treatments were considered as the fixed effect and replication or made by pushing a hand-held digital cone penetrometer (Field blocks as the random effect. Means separations were done using Scout, SC 900 Soil Compaction Meter, Spectrum Technologies, the least square means test in SAS (Littell et al., 1996), differ- Plainfield, IL) into the soil at the center of the plots (one mea- ences among treatments were reported as significant atP = 0.05. surement per plot). Penetrometer measurements are usually reported as the cone Table 2. Freeze-thaw cycles for frozen compacted and frozen index (CI) which is the shear resistance of the soil. Cone index is uncompacted soils at depths of 0–5, 5–10, 10–15, and 15–30 cm in Savage clay loam.

Table 1. Volumetric soil moisture content (0–20 cm) at the Soil freeze-thaw cycles time penetration resistance measurements. Year 0–5 cm 5–10 cm 10–15 cm 15–30 cm Soil water content, % Frozen compacted treatment Date of sampling Season Mean Maximum Minimum 2009–2010 20 13 12 20 26 Aug. 2009 Fall09 35.3 37.5 34.2 2010–2011 21 22 23 19 14 June 2010 Spr10 34.1 36.8 33.9 2011–2012 33 20 16 15 27 Sept. 2010 Fall10 34.0 36.7 33.4 Frozen uncompacted treatment 08 June 2011 Spr11 34.7 36.2 33.4 2009–2010 19 10 8 13 02 Sept. 2011 Fall11 34.8 36.1 33.9 2010–2011 22 23 20 22 06 June 2012 Spr12 34.2 36.7 33.1 2011–2012 37 24 16 14

740 Soil Science Society of America Journal The least square means is the SAS term that refers to the more ing processes, respectively (Kohnke, 1968; Taylor and Ashcroft, common term marginal means, which is sometimes referred to as 1972), where soils in the NGP region typically freeze and thaw estimated marginal means (SAS online documentation: http:// often during winter and early spring in most years. This substan- onbiostatistics.blogspot.com/2009/04/least-squares-means- tial mitigation in soil PR and strength due to natural processes marginal-means-vs.html and http://support.sas.com/onlined- and the dynamic activity of the freeze-thaw phenomenon is con- oc/913/getDoc/en/statug.hlp/glm_sect34.htm). sidered good news to farmers because little or no compaction means more soil pore space, better soil structure, and a healthier RESULTS AND DISCUSSION root environment which are essential to air, water, nutrient move- Compacted and Uncompacted Soils ment, plant growth, and root distribution in soil ecosystems. Compacted treatments had higher PR values than 1.5 MPa at depths greater than 2.5 cm, reaching a maximum of nearly 2.2 MPa at the 5- and 7.5-cm depths (Fig. 4). Uncompacted treatments did not exceed 1.5 MPa, except in the case when PR approached 1.6 MPa at the 5- and 7.5-cm depths. Soils with PR values exceeding 1.5 MPa were considered compacted (Kulkarni, 2003; Copas et al., 2009). Soil PR values for compacted plots were significantly greater than those under uncompacted plots. Compacted soil ex- hibited considerable soil strength due to compressive forces applied to the soil from wheel passes of the truck during the compaction process at near field capacity water content.

Frozen Soil Conditions Soil PR in compacted soils that had been subjected to frozen conditions was significantly reduced by 73, 68, and 59% over one winter (November 2009 to April 2010) at depths of 0 to 10, 10 to 20, and 20 to 30 cm, respectively (Fig. 5a, 5b, and 5c). The average soil PR in compacted frozen treatments decreased by approximately 67% (from 2.10 MPa to 0.69 MPa) at the 0- to 30-cm depth. Frequent FTCs over winter 2009 to 2010 and early spring 2010 alleviated a majority of soil compaction at depths of 0 to 10, 10 to 20, and 20 to 30 cm. The significant reduction in PR of the top 0 to 30 cm of the soil profile was due to dynamic effects of frequent FTCs on soil structure, pore space, and particle con- figurations. This natural mechanism is caused when the volume of soil pore water expands and creates forces that cause soil particles to push and contract repeatedly during freezing and thaw-

Fig. 5. Temporal variations of soil penetration resistance (PR) for compacted and uncompacted soils under frozen conditions at depths of 0 to 10 cm (a), 10 to 20 cm (b), and 20 to 30 cm (c). Same Fig. 4. Initial soil penetration resistance profile of compacted and lowercase letters indicate that means within a treatment are not uncompacted soils measured in fall 2009. Error bars represents two significantly different at P ≤ 0.05. Same uppercase letters indicate that standard errors of the mean. means within a treatment are not significantly different at P ≤ 0.05. www.soils.org/publications/sssaj 741 Penetration resistance of frozen uncompacted soils in spring soil conditions for plant growth, despite the initial value of PR 2010 was not significantly different from other subsequent four 1.34 < 1.5 MPa which was initially classified as uncompacted soil sampling dates for all three sampling depths (Fig. 5a, 5b, and 5c). (Kulkarni, 2003; Copas et al., 2009). This considerable decrease In the frozen uncompacted treatments, soil PR was significantly in soil PR of the upper 0 to 30 cm of the soil profile was likely influenced by FTCs and reduced by approximately 73, 66, and caused predominantly by the effects and dynamics of repeated 49% over winter and early spring of 2009 to 2010 at depths of FTCs on the arrangement and orientation of soil particles as well 0 to 10, 10 to 20, and 20 to 30 cm, respectively (Fig. 5a, 5b, and as formation of new patterns and geometry of pore spaces in the 5c). Averaged across three depths, soil PR decreased by approxi- soil (soil structure). These results were in agreement with those mately 62% (from 1.34 MPa to 0.51 MPa), providing optimum found by Lehrsch (1998), Oztas and Fayetorbay (2003), and Edwards (2013) in terms of soil aggregate stability and porosity as affected by FTCs. No significant differences in soil PR were observed between spring 2010 and subsequent sampling dates throughout the course of this study (Fig. 5a, 5b, and 5c).

Unfrozen Soil Conditions In unfrozen compacted soils, PR was significantly reduced by 50, 60, and 46% over one winter (November 2009 to April 2010) at depths of 0 to 10, 10 to 20, and 20 to 30 cm, respectively (Fig. 6a, 6b, and 6c). Averaged across three soil depths, PR in compacted treatments decreased by approximately 52% (from 2.16 to 1.03 MPa). This reduction in soil PR at the surface of soil profile (0–30 cm) may be due to the biology of soil, increased microbial, mycorrhizal, microfaunal activities, and disruptive effects of shrink-swell cycles caused by frequent drying- wetting processes under warm conditions where soils were prevented from freezing with electrically heated blankets (Henry, 2007). This presumed biological and physical behavior may have caused a reduction in soil compaction and changes in soil structure. Expansive clayey soils such as Savage clay loam (37–38% clay; 1.6% organic matter) used in this study are susceptible to shrink-swell behavior and cracks were visible on the surface under dry conditions. (Abou Najm et al., 2010). These soils absorb water, then water enters into spaces between the soil lattices of 2:1 smectite clay minerals and as more water is absorbed, the lattice plates are forced further apart, leading to an increase in soil pore pressure causing soil structure to change (Basma et al., 1996;

Fig. 6. Temporal variations of soil penetration resistance (PR) for compacted and uncompacted soils under unfrozen conditions at depths of 0 to 10 cm (a), 10 to 20 cm (b), and 20 to 30 cm (c). Same lowercase letters indicate that means within a treatment are not Fig. 7. Soil penetration resistance profiles for compacted soils in fall significantly different at P ≤ 0.05. Same uppercase letters indicate that 2009, compacted unfrozen in spring 2010, and compacted frozen in means within a treatment are not significantly different at P ≤ 0.05. spring 2010. Error bars represents two standard errors of the mean.

742 Soil Science Society of America Journal Andrieux et al., 2011; Fouli et al., 2013). Further, the dynamics ly 62% (from 1.34 to 0.51 MPa) thus providing better soil con- of shrink-swell due to drying-wetting processes can result in conditions for plant growth. In unfrozen compacted soils, PR was siderable changes in soil structure and porosity. Our results con- significantly reduced by approximately 52% (from 2.16 to 1.03 curred with those found by Abou Najm et al. (2010) on Savage MPa) in the top 0 to 30 cm of the soil profile presumably due to clay loam. When dry soils are wetted after rainfall or irrigation the biology of soil and disruptive effects of shrink-swell cycles events, they expand or swell; conversely, a wet soil shrinks on a caused by frequent wetting-drying processes. drying period creating cracks, and the cracks disappear when soil The results from this study could save growers considerable expands or swells after irrigation and rainfall (Abou Najm et al., time, money, and energy currently required to alleviate soil com- 2010). However, the compacted soil under unfrozen conditions paction using other methods such as subsoiling and deep tillage. was returned to optimum soil conditions for plant growth due Our findings demonstrated that repeated freeze-thaw cycles can to the effects of periodic drying-wetting processes over winter of alleviate soil compaction, improve soil structure, alter soil physi- 2009 to 2010 and early spring of 2010. No significant differences cal quality, and create optimal soil conditions required for the in soil PR were observed between spring 2010 and subsequent profitable growth of agricultural crops. Some naturally occur- sampling dates. ring weather patterns can provide mechanisms to reverse soil Similarly, in unfrozen uncompacted soils, PR was signifi- compaction and enhance soil structure through the dynamics of cantly reduced by 72, 54, and 33% over winter and early spring freeze-thaw cycles that occur in soils in Montana and other parts (November 2009 to April 2010) at depths of 0 to 10, 10 to 20, of the country. and 20 to 30 cm, respectively (Fig. 6a, 6b, and 6c). Averaged across the three soil sampling depths, PR in uncompacted treat- ACKNOWLEDGMENTS ments decreased by approximately 53% (from 1.33 MPa to 0.63 We sincerely thank Mr. Dale Spracklin for his technical support with MPa). The magnitude in soil PR reduction in both compacted sampling, installation of temperature and moisture sensors, and for maintaining and monitoring the experiment throughout the course and uncompacted treatments under unfrozen conditions was of study. The authors thank Dr. Mark West, a statistician from the almost identical. The reduction in soil PR in the uncompacted Northern Plain area in Fort Collins, CO, for his assistance in statistical treatment was caused by reasons mentioned in the preceding analyses. We thank Dr. Ymène Fouli at the environmental consulting section (i.e., soil microbial activity and shrink-swell behavior in in Calgary, Canada, and Dr. Mario Flores Mangual at the Department expansive clayey soil). The low values in soil PR of the top 0 to of Crops and Agro-Environmental Sciences, University of Puerto Rico-Mayagüez, for their constructive comments and suggestions. The 10 cm may be caused by loosened soil particles from the presence helpful suggestions and constructive comments of the Associate Editor of a thick layer of algal growth on the soil surface in the spring of Wolfgang Durner and six anonymous reviewers are greatly appreciated. 2010 (Fig. 3). Figure 7 illustrates PR profiles to a depth of 0 to 30 cm of REFERENCES initially compacted soil in fall 2009 (fall09), following frozen Abou Najm, M.R., J.D. Jabro, W.M. Iversen, R.H. Mohtar, and R.G. Evans. 2010. conditions and measured in spring 2010 (spr10) and unfrozen New method for characterization of 3D preferential flow paths at the field. Water Resour. Res. 46:W02503. doi:10.1029/2009WR008594. conditions and measured in spring 2010 (spr10). Significant re- Andrieux, C., M. Chretien, A. Denis, R. Fabre, and J.-F. Lataste. 2011. Shrinkage ductions in soil PR were observed between initially compacted and swelling of clay soils. Eur. J. Environ. Civ. Eng. 15:819–838. soil in fall 2009, and frozen compacted and unfrozen compacted Asare, S.N., R.P. Rudra, W.T. Dickinson, and G.J. Wall. 1997. Frequency of freeze- thaw cycles, bulk density and saturation effects on soil surface shear and soil in spring 2010 over winter of 2009 to 2010 and early spring aggregate stability in resisting water erosion. Can. Agric. Eng. 39:273–279. of 2010. These reductions in PR were due to soil freeze-thaw and Asare, S.N., R.P. Rudra, W.T. Dickinson, and G.J. Wall. 1999. Effect of freeze- shrink-swell mechanisms in clay loam soil. Our results concurred thaw cycle on the parameters of the Green and Ampt infiltration equation. with those found by Unger (1991), Henry (2007), Parker et al. J. Agric. Eng. Res. 73:265–274. doi:10.1006/jaer.1999.0415 Baker, D.G., and D.L. Ruschy. 1995. Calculated and measured air and soil freeze- (1982), and Abou Najm et al. (2010). However, the compacted thaw frequencies. J. Appl. Meteorol. 34:2197–2205. doi:10.1175/1520- soil under frozen conditions was returned to optimum soil con- 0450(1995)034<2197:CAMAAS>2.0.CO;2 ditions for plant growth due to the effects of FTCs over win- Basma, A.A., A.S. Al-Homoud, A.I.H. Malkawi, and M.A. Al-Bashabsheh. 1996. Swelling-shrinkage behavior of natural expensive clays. Appl. Clay ter 2009 to 2010 and early spring of 2010 (Fig. 7). These results Sci. 11:211–227. doi:10.1016/S0169-1317(96)00009-9 demonstrated that repeated FTCs can alleviate soil compaction, Batey, T. 2009. Soil compaction and soil management—A review. Soil Use alter soil physical quality, and contribute to a healthy soil ecosys- Manage. 25:335–345. doi:10.1111/j.1475-2743.2009.00236.x Benoit, G.R. 1973. Effect of freeze-thaw cycles on aggregate stability and tem for root growth. hydraulic conductivity of three soil aggregate sizes. Soil Sci. Soc. Am. Proc. 37:3–5. doi:10.2136/sssaj1973.03615995003700010007x CONCLUSIONS Bisal, F.K., and F. Nielsen. 1967. Effect of frost action on the size of soil aggregates. This study evaluated the effect of FTCs on soil compaction Soil Sci. 104:268–272. doi:10.1097/00010694-196710000-00006 Copas, M.E., A.J. Bussan, M.J. Drilias, and R.P. Wolkowski. 2009. Potato yield in a clay loam in the semiarid NGP. Soil PR in frozen compacted and quality response to subsoil tillage and compaction. Agron. J. 101:82– treatments was decreased by 67% over one winter (November 90. doi:10.2134/agronj2007.0031 2009 to April 2010) at the 0- to 30-cm depth due to dynamic Edwards, L.M. 2013. The effects of soil freeze-thaw on soil aggregate breakdown and concomitant sediments flow in Prince Edward Island. A review. Can. effects of frequent FTCs on soil structure. Similarly, in frozen J. Soil Sci. 93:1–14. uncompacted treatments, soil PR was decreased by approximate- Flowers, M., and R. Lal. 1998. Axle load and tillage effect on soil physical

www.soils.org/publications/sssaj 743 properties and soybean grain yield on a mollic ochraqualf in northwest Soil Sci. 163:63–70. doi:10.1097/00010694-199801000-00009 Ohio. Soil Tillage Res. 48:21–35. doi:10.1016/S0167-1987(98)00095-6 Leo, M.W.M. 1963. Effect of freezing and thawing on some physical properties Fouli, Y., B.J. Cade-Menun, and H.W. Cutforth. 2013. Freeze-thaw cycles and of soils as related to tomato and barley plants. Soil Sci. 96:267–274. soil water content effects on infiltration rate of three Saskatchewan soils. doi:10.1097/00010694-196310000-00007 Can. J. Soil Sci. 93:1–12. Littell, R.C., G.A. Milliken, W.W. Stroup, and R.D. Wolfinger. 1996. SAS Hamza, M.A., and W.K. Anderson. 2005. Soil compaction in cropping systems: systems for mixed models. SAS Inst., Cary NC. A review of the nature, causes and possible solutions. Soil Tillage Res. Oztas, T., and F. Fayetorbay. 2003. Effect of freezing and thawing processes on soil 82:121–145. doi:10.1016/j.still.2004.08.009 aggregate stability. Catena 52:1–8. doi:10.1016/S0341-8162(02)00177-7 Henry, H.A.L. 2007. Soil freeze-thaw cycle experiment: Trends, methodological Parker, J.C., D.F. Amos, and L.W. Zelazny. 1982. Water adsorption and weakness and suggested improvements. Soil Biol. Biochem. 39:977–986. swelling of clay minerals in soil systems. Soil Sci. Soc. Am. J. 46:450–456. doi:10.1016/j.soilbio.2006.11.017 doi:10.2136/sssaj1982.03615995004600030002x Ho, E., and W.A. Gough. 2006. Freeze thaw cycles in Toronto, Canada in a Randrup, T.B., and J.M. Lichter. 2001. Measuring soil compaction on changing climate. Theor. Appl. Climatol. 83:203–210. doi:10.1007/ construction sites: A review of surface nuclear gauges and penetrometers. s00704-005-0167-7 J. Arboric. 27:109–117. Jabro, J.D., R.G. Evans, Y. Kim, and W.M. Iversen. 2009a. Estimating in situ soil- Sahin, U., I. Angin, and F.M. Kiziloglu. 2008. Effect of freezing and thawing processes water retention and field water capacity measurements in two contrasting on some physical properties on saline-sodic soils mixed with sewage sludge of soil textures. Irrig. Sci. 27:223–229. doi:10.1007/s00271-008-0137-9 flay ash. Soil Tillage Res. 99:254–260. doi:10.1016/j.still.2008.03.001 Jabro, J.D., U.M. Sainju, W.B. Stevens, A.W. Lenssen, and R.G. Evans. 2009b. SAS Institute. 2011. The SAS system for windows. Version 9.2. SAS Inst., Cary, NC. Long-term tillage influences on soil physical properties under dryland Sillanpaa, M., and R.L. Webber. 1961. The effect of freezing–thawing and conditions in northeastern Montana. Arch. Agron. Soil Sci. 55:633–640. wetting-drying cycles on soil aggregation. Can. J. Soil Sci. 41:182–187. doi:10.1080/03650340902804316 doi:10.4141/cjss61-024 Jabro, J.D., W.B. Stevens, W.M. Iversen, and R.G. Evans. 2011. Bulk density and Taylor, S.A., and G.L. Ashcroft. 1972. Physical edaphology. The physics of hydraulic properties of a sandy loam soil following conventional or strip irrigated and non-irrigated soils. W.H. Freeman and Company, San tillage. Appl. Eng. Agric. 27:765–768. doi:10.13031/2013.39576 Francisco, CA. Kohnke, H. 1968. Soil physics. McGraw-Hill, New York. Unger, P.W. 1991. Overwinter changes in physical-properties of no-tillage soil. Soil Sci. Kulkarni, S. 2003. Soil compaction modeling in cotton. Summaries of Arkansas Soc. Am. J. 55:778–782. doi:10.2136/sssaj1991.03615995005500030024x Cotton Research. AAES Res. Ser. 521:68–70. Veris Technologies. 2002. Veris Technologies. www.veristech.com (accessed Lehrsch, G.A. 1998. Freeze-thaw cycles increase near-surface aggregate stability. 29 May 2013).

744 Soil Science Society of America Journal