| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Forest Resources, Alberta Research Council, P.O. Bag 4000, Vegreville, AB, Canada, T9C 1T4
* Corresponding author (dhmcnabb{at}arc.ab.ca)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Compaction of forest soils has been a concern in North America for >40 yr. Considerable research has been done to quantify the effects of compaction on soil and trees, the natural rate of decompaction, comparison of different equipment, tillage of compacted soil, and management of forest operations to minimize soil compaction (Froehlich and McNabb, 1984). Several guidelines or regulations have been proposed or are used to reduce soil compaction by ground-based machines, but these are based on only a general understanding of equipment, engineering properties of soil, and their interaction. For example, many early guidelines to reduce soil compaction focused on regulating equipment on the basis of their ground pressure, or rating soil as to their susceptibility to compaction on the basis of texture or bulk density (Boyer, 1979; Howard et al., 1981; Lewis and Carr, 1989). Unfortunately, simply calculating the static ground pressure of most forest harvesting equipment is not indicative of the dynamic pressure that machines exert on soil during skidding (Lysne and Burditt, 1983), and may not be related to the amount of compaction that occurs (Froehlich et al., 1980). But low pressure or wide rubber tires on skidders have been found to reduce compaction when the soil is wet (Greene and Stuart, 1985).
Assessing the susceptibility of a soil to deformation by compaction is confounded by differences in initial conditions, such as texture, bulk density, and soil water content, and the fact that the value of these properties are interrelated. Measuring the shear strength and compressibility of forest soils is improving our knowledge of the soil deformation process and the relationship among soils (McNabb and Boersma, 1993, 1996). For example, differences in the deformation of forest soils are not as well related to bulk density and texture as previously assumed. As a result, direct comparisons of bulk density among soils is unreliable for ranking the susceptibility of soil to compaction. Soil wetness is another important factor determining the magnitude of the increase in bulk density whenever a load is applied to soil (Terzaghi and Peck, 1967), but in laboratory compaction tests of different soils, the optimum soil water content for compaction and maximum density is inversely related. Soils with a high undisturbed bulk density have a lower optimum water content at maximum density than do soils of lower undisturbed bulk density (Howard et al., 1981). The effect of soil wetness on the compactibility of forest soils has not been well established beyond these general relationships. Nevertheless, compaction of wet forest soils is a serious concern (Moehring and Rawls, 1970; Greene and Stuart, 1985; McNabb, 1994), but attempts to link the compaction of forest soils in laboratory compaction tests to soil compaction in the field have not always been successful (Froehlich et al., 1980). However, the compressibility of forest soils decreases substantially at soil water contents drier than field capacity (McNabb and Boersma, 1996), and the compressive strength of dry clay soil can be higher than that of coarse textured soil (Terzaghi and Peck, 1967).
Given the importance of changes in soil wetness on the compressibility of partly saturated soils, particularly at soil water contents around those of field capacity and wetter (Larson and Gupta, 1980; McNabb and Boersma, 1996), further investigation to more accurately quantify the effects of soil wetness on the operational compaction of forest soil is justified. The compaction of wet soil in the field, however, may be restricted by low air-filled porosity of the soil at the time of trafficking because this type of short-term loading of the soil is insufficient for consolidation by drainage of the soil to occur. Thus, changes in the solid–liquid–gaseous phases of soil must be assessed simultaneously. Our objective was to determine the effects that soil wetness, particularly at water potentials around field capacity, and the number of skidding cycles had on the compaction of boreal forests, as measured by changes in bulk density and air-filled porosity at the time of skidding.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
|
Where the whole-tree skidding system was used, machine travel across the control was minimized. Feller-bunchers made about three passes, about 15 m apart, across the skidding corridors. Skidders retrieved logs from the two outside skidded corridors and some of the trees in the control by backing into the corridor and pulling trees out from the side. Areas within the control where machines had obviously traveled were not sampled. Skidders backed down the center corridor to retrieve logs from the 12-cycle treatment corridor and adjacent portion of the control. The center corridor was the most heavily trafficked treatment and therefore was less sensitive to deviations in the number of skidding cycles than the other treatments (Froehlich et al., 1980). When skidders backed at least two-thirds of the distance into a block, the trip was counted as one of the cycles. At the two cut-to-length sites, the forwarder was partly loaded as it traveled into the block, continued through the block, and loading was completed as it reentered the corridor. A member of the research staff was always present during the clearing of the blocks and skidding cycles to ensure consistency in the establishment of each installation.
Grapple skidders were used for skidding most of the sites (Table 1). All of the wheeled-skidders were equipped with tires between 0.8 and 1.1 m wide. The empty weight of skidders was between 14 and 17 Mg and they were capable of carrying 4 to 6 Mg of trees. Site 5 was skidded with a Caterpillar (Peoria, IL) D4H TSK (crawler) with a grapple. The two cut-to-length sites were forwarded with a Valmet 540 (Site 2), or a Timberjack 520A (Site 13).
Soil Measurements
Soil was sampled within a week of skidding, and most sites were sampled within three days. Samples were collected from four locations in the control and each of the skidding treatments. Undisturbed cores for the determination of bulk density, air-filled porosity, and soil water content were collected from the midpoint of the 5-, 10-, and 20-cm depths. These cores were collected in rings of thin-walled stainless steel tubing, 3 cm in height and 7.6 cm in diameter. Rings were pushed into the soil with a driver that fit the ring. Immediately after the core was removed from the soil and trimmed, the air-filled porosity was measured using an air pycnometer (Wooldridge, 1968). The air pycnometer method measured the volume of air-filled pores in the sample on the basis of a calibrated drop in pressure. Following this test, the soil was removed from the ring, sealed in a plastic bag in the field, stored at 4°C until weighed, and oven-dried at 105°C. These data were used to calculate bulk density and gravimetric water content of each sample. Bulk density was not corrected for coarse fragment content because the content was either low or, occasionally, prevented the collection of any undisturbed cores when pushing the ring into the ground. A total of 192 cores were collected from each site.
Soil water potential was also measured with a hand-held tensiometer (Soil Moisture Equipment Corp., Santa Barbara, CA) in the soil adjacent to the bulk density sample if soil water potential was higher than -60 kPa. On sites where the soil water potential was lower than the range of the tensiometer, the potential was estimated from the relationship between water potential and water content measured in the laboratory from another set of undisturbed cores collected from the block with the most modal soil (Startsev and McNabb, 2001). Organic carbon was determined by wet combustion (Nelson and Sommers, 1982).
Statistical Analyses
Bulk density and air-filled porosity data were analyzed as a repeated analysis of variance where depth was the repeated measure (SAS Institute Inc., 1991; Gumpertz and Brownie, 1993; McNabb, 1994). Gravimetric water content was used as a covariant in analysis of variance. Relationships among soil and site variables were analyzed by step-wise regression analysis.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Most of the soils were relatively wet at the time of harvesting (Table 1). June and July are normally the wettest months of the year, with precipitation decreasing in subsequent months (Environment Canada, 1993). All sites were harvested between July and early October, allowing a range of soil water contents and potentials to be sampled. Five of the sites had average soil water potential higher than field capacity, defined as a water potential of -10 kPa. Another three sites had a soil water potential between -15 and -10 kPa. As a result, our study focuses on the compaction of wet soils much more than the study by Froehlich et al. (1980).
Bulk density increased significantly with increased number of skidding cycles (Tables 2 and 3). The largest increase in bulk density occurred after 3 cycles, and had essentially reached a maximum bulk density for these machines by 12 cycles (Fig. 3). The increase in bulk density was affected by significant differences among sites (Table 3). Inclusion of soil properties in the analysis of variance was limited to soil water content because it was directly measured on the same samples as bulk density and air-filled porosity. Therefore, gravimetric water content was used as a covariant in all the analyses of variance. However, gravimetric water content accounted for much of the variation attributed to differences due to site. For bulk density, including the water content of soil at the time of skidding as a covariant in the analysis reduced the mean squared error from 0.00583 to 0.00213.
|
|
|
When we analyzed bulk density separately for each site, the increase in bulk density was significant after 3 cycles of skidding for at least one depth at 8 of the 14 sites (Table 2). Skidding at 7 and 12 cycles did not cause an additional significant increase in bulk density except at Site 3, where the difference between 3 and 7 cycles of skidding was significant. These analyses confirm that most soil compaction occurs within the first few skidding cycles (Froehlich et al., 1980; Froehlich and McNabb, 1984). Although additional trafficking of these soils is unlikely to increase bulk density, the continued trafficking of adjacent soil when wet did cause some rutting. Some rutting was also observed at Site 3, which had soil at the highest water potential sampled (Table 1), and at some other sites when skidders were turning to enter the corridor. The significant increase in bulk density measured at Site 3 may reflect the sampling of more dense subsoil in the bottom of ruts because the depth was measured from the surface of the soil in the bottom of the depression.
The probability that the increase in bulk density became significant increased as the gravimetric water content increased (Fig. 4). At a gravimetric water content greater than about 0.30 kg kg-1, the increase was statistically significant. A significant increase in bulk density was also related to the soil water potential at the time of skidding. Significant compaction after three cycles only occurred when the soil water potential was greater than about -15 kPa. Decreasing soil water potential is directly related to an increase in the effective shear strength of soil (Bishop and Blight, 1963; Snyder and Miller, 1985). Therefore, soil water potential is a more scientific measure of the effects of soil wetness on soil strength than is a gravimetric measure of soil water content.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Soil wetness is a well established factor affecting the compaction and compressibility of soils (Proctor, 1933; Tergazhi and Peck, 1967). The compressibility of soil decreases as soil dries because of two processes. The effective stress increases as soil water potential decreases (Bishop and Blight, 1963), and direct contact between soil particles increases as the thickness of the water films around clay particles becomes thinner (McNabb and Boersma, 1996). As a result, the compressibility of soils with a higher clay content decreases faster than in coarser-textured soils as soil water potential decreases (Larson and Gupta, 1980). Laboratory compaction of forest soils has also confirmed that soils containing smectitic clays are more sensitive to changes in soil water content than soils containing less expandable clay minerals (Froehlich and McNabb, 1984). Froehlich et al. (1980) were unable to establish that soil wetness was a significant factor affecting the compaction of forest soils in northern California, but they generally worked on sites where the soil was much drier than field capacity. Our data shows that soil wetness, in conjunction with the level of trafficking, dominates the compaction process of soils when wide-tired skidders operate on soils that are at a gravimetric water content near field capacity or wetter (Fig. 4).
When boreal forest soils are wet, the increase in bulk density is limited by the amount of air-filled porosity remaining in the soil after trafficking (Fig. 5). Soil will compact down to an air-filled porosity of about 0.07 m3 m-3 (Table 2). Further decreases in soil volume as a result of compaction is not possible because the remaining soil air is trapped in the soil pores (Xu et al., 1992). In our soils, all of the changes in air-filled porosity occurred in pores larger than those drained at about -20 kPa; changes in smaller pores apparently did not occur because the water retention curves were not affected by trafficking at lower soil water potentials (Startsev and McNabb, 2001). But, trafficking of these soils when air-filled porosity was low most likely resulted in positive air and water pressures. These positive pressures reduced the effective stresses associated with trafficking, hence soil strength decreased (Bishop and Blight, 1963). A decrease in effective stresses increases the probability of a bearing capacity type of soil failure (Tergazhi and Peck, 1967), which manifests itself in the rutting of soil (McNabb, 1992). Shallow ruts developed in the soil at Site 3, where the air-filled porosity was lowest (Table 2). We speculate that only the careful operation of wide-tired skidders and forwarders in the straight corridors of the experimental blocks minimized the rutting of these soils when they were very wet.
We had few options in the types of equipment to evaluate in this study (Table 1). Most of the companies had already changed to wider tires on skidders or other machines to reduce their impact on soil. As a result, we were only able to include two forwarders and one wide-track crawler. Based on the probability of an increase in bulk density or a decrease in air-filled porosity as a function of soil water potential, there is no obvious difference among these machines (Fig. 4). If the soil was wetter than field capacity, it was significantly compacted regardless of the type of machine used.
Tire size is an important option for reducing the compaction of wet soils (Greene and Stuart, 1985). Unfortunately, we were not able to provide collaborative data of the value of wide tires in reducing soil compaction in this study. However, air-filled pore space will limit the amount of compaction when a soil is wet regardless of tire size (Fig. 5), although narrower conventional tires are likely to cause more ruts than wider tires (McNabb, 1993). When soils are less wet, that is, the water potential is lower, conventional tires will cause more soil compaction than wider tires carrying the same load (Greene and Stuart, 1985). As a result, conventional tires are expected to cause statistically significant soil compaction at a lower water potential than do wider tires (Fig. 4, Froehlich et al., 1980). From an operational perspective, the time period that soils are susceptible to significant compaction following precipitation will be longer for conventional tires compared to wider tires because it will take longer for the soil to dry as a result of evapotranspiration.
Our data have confirmed the importance of soil water potential in the compaction process under standardized levels of soil trafficking (Fig. 4). When these results are coupled with a better understanding of the shear strength and compressibility of forest soils (Froehlich and McNabb, 1984; McNabb and Boersma, 1993, 1996), managing soil wetness becomes the most important factor in rating soils according to their susceptibility to soil compaction and other forms of soil modification (McNabb, 1993).
In addition to using wide tires to reduce soil compaction, forest managers and logging supervisors have considerable opportunity to manage soil wetness to reduce soil compaction (McNabb, 1993). Soil wetness can be managed in several ways, such as scheduling poorly drained soils and soils with lower hydraulic conductivity for drier seasons of the year or when the soil is more likely to be frozen. An important part of managing soil wetness is to minimize the period between felling and skidding. This increases the opportunity for trees to transpire soil water and decrease soil water potential, and minimizes the risk that precipitation will recharge soil water prior to skidding. Once the trees are cut, most soils will only drain to a soil water potential of approximately -10 kPa, which still leaves the soils susceptible to significant compaction (Fig. 4). The potential for the types of machines used in our study to significantly compact the soil can be readily assessed in the field with a hand-held tensiometer. While a tensiometer is a convenient tool to use in the field, the assessment of the potential for soil to be significantly compacted could be more easily assessed in the field by establishing a relationship between soil consistence and soil water potential. For example, standard field measures of soil consistence for different textures could be related to soil water potential.
The susceptibility of soil to compaction is a relatively straightforward process involving the interaction of machines with soil that depends on the shear strength and compressibility of soil and characteristics of the machines. The consequences of compaction on ecosystem processes and function is much more complex because it is ecosystem specific (McNabb and Campbell, 1985; McNabb, 1995). As a consequence, not all compaction reduces the growth of trees (Greacen and Sands, 1980; Miller et al., 1996) or affects soil biological process the same way. For example, compaction reduces soil respiration but increases decomposition in our soils when compacted because the soil biota have apparently adapted to a partly anaerobic environment in the forest prior to harvest (Startsev et al., 1998). The implications that soil compaction in the western boreal forest have on tree performance remains to be determined on these sites. Some compacted soils in the region can recover from compaction relatively quickly (Corns, 1988; McNabb, 1994). Soil compaction does reduce the density and height growth of young aspen in western Alberta (Greenway and McNabb, unpublished data), but the effects of compaction on conifer growth is more problematic. Most of our sites are prepared for planting by machine, which we hypothesize will negate some of the short-term effects that soil compaction might have on conifer growth. Therefore, the effects of compaction on the growth of planted conifer seedlings and the rate of decompaction continues to be measured on nine of these study sites.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
When soils are wet, compaction will increase bulk density and decrease air-filled porosity until only trapped air remains in the soil pores. Under these conditions, trafficking results in a significant increase in bulk density. But these soils are only susceptible to a significant increase in bulk density from wide-tired skidders if the soil water potential is at or above (wetter than) field capacity. Field measurement of soil water potential provides an effective method of assessing when soil is most susceptible to compaction by wide-tired skidders. This water potential can be easily measured in the field, using a hand-held tensiometer or by estimating it based on soil consistence. In some forest ecosystems, managing soil wetness is an effective method for reducing soil compaction; soil wetness in the range of field capacity can be managing by allowing soil time to drain naturally and by retaining tree cover to transpire water from the soil until just prior to skidding.
|
ACKNOWLEDGMENTS |
|---|
Received for publication December 7, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. J. Shestak and M. D. Busse Compaction Alters Physical but Not Biological Indices of Soil Health Soil Sci. Soc. Am. J., January 1, 2005; 69(1): 236 - 246. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Prevost Predicting Soil Properties from Organic Matter Content following Mechanical Site Preparation of Forest Soils Soil Sci. Soc. Am. J., May 1, 2004; 68(3): 943 - 949. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Teepe, R. Brumme, F. Beese, and B. Ludwig Nitrous Oxide Emission and Methane Consumption Following Compaction of Forest Soils Soil Sci. Soc. Am. J., March 1, 2004; 68(2): 605 - 611. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Block, K. C. J. Van Rees, and D. J. Pennock Quantifying Harvesting Impacts using Soil Compaction and Disturbance Regimes at a Landscape Scale Soil Sci. Soc. Am. J., September 1, 2002; 66(5): 1669 - 1676. [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 | |||
