Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Soil disturbance and quantification of machine traffic soil compaction associated with pushover logging Redfern, Lawrence Stacey 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1998-0296.pdf [ 4.27MB ]
Metadata
JSON: 831-1.0099298.json
JSON-LD: 831-1.0099298-ld.json
RDF/XML (Pretty): 831-1.0099298-rdf.xml
RDF/JSON: 831-1.0099298-rdf.json
Turtle: 831-1.0099298-turtle.txt
N-Triples: 831-1.0099298-rdf-ntriples.txt
Original Record: 831-1.0099298-source.json
Full Text
831-1.0099298-fulltext.txt
Citation
831-1.0099298.ris

Full Text

SOIL DISTURBANCE, AND QUANTIFICATION OF MACHINE TRAFFIC SOIL COMPACTION ASSOCIATED WITH PUSHOVER LOGGING by LAWRENCE STACEY REDFERN B.Sc. (Agr.), University of British Columbia, 1992 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF SOIL SCIENCE We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA APRIL 1998 © Lawrence S. Redfern , 1998 In presenting this thesis in partial fulfillment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly puiposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Soil Science The University of British Columbia Vancouver, Canada March 25th, 1998 Abstract Pushover logging is a timber harvest system that involves pushing over trees with an excavator, thereby cantilevering the tree roots out of the soil. Pushover logging has the potential to cause above average levels of soil disturbance because of the soil disturbance associated with the uprooting process, and the increased trafficking of the harvest area by both the excavator and skidding machines. The purpose of this study was to quantify and describe the soil disturbance associated with pushover logging. Soil disturbance surveys were completed on two pushover-logged cut blocks in the Nelson Forest Region, near the towns of Golden (Colepitts), and Invermere (Mud Creek). The sites were chosen as representative of sites considered particularly sensitive to pushover logging harvest systems because of high compaction hazards. Soil core samples were taken for analysis of changes in soil physical properties, in four machine traffic disturbance types. Measurements of bulk density, total porosity, aeration porosity, and available water storage capacity were made. Soil disturbance is categorized as potentially detrimental if a negative impact on tree productivity, or on off-site forest resource values can be anticipated. Based on soil survey and laboratory results respectively, mean percent of harvest area occupied by potentially detrimental soil disturbance ranged between 11 % and 23 % at Colepitts, and 17 % and 2 % at Mud Creek. Determination of soil compaction in machine traffic areas, by hand-ii checking for changes in soil physical properties during the soil disturbance survey, was not a good indicator of compaction level as determined in the laboratory. Stump holes greater than 30 cm deep, created by the uprooting procedure, occupied 9.5 % of the Colepitts area, and 2.1 % of the Mud Creek area. This difference is attributed to stand and individual tree differences between the two sites. There were no significant differences (a = 0.05) found between disturbance types for bulk density, total porosity, and available water storage capacity measures. Aeration porosity measures showed significant differences between disturbance types at the 2 - 4 cm depth at both sites, and at the 6 - 8 cm depth at Mud Creek. At the Colepitts site, decreases in aeration porosity were associated with increases in machine traffic; at Mud Creek the highest aeration porosity values were at the surface in the heaviest machine traffic disturbance type. A lack of information on specific conditions at the time of harvest at Mud Creek make interpretation of the unexpected results difficult. At Colepitts, the levels of compaction measured are not of sufficient magnitude to forecast a reduction in tree productivity at this site in the future. iii Table of Contents Abstract ii Table of Contents iv List of Tables vi List of Figures viii Acknowledgements 1 X 1 Introduction 1 2 Literature Review 4 2.1 Measurement and Quantification of Soil Disturbance 5 2.2 Measures of Soil Compaction 11 2.2.1 Bulk Density 12 2.2.2 Porosity 13 2.3 Quantification of Forest Soil Compaction 20 2.4 Growth Impacts of Forest Soil Compaction 29 2.5 Longevity of Forest Soil Compaction 35 2.6 Critical Levels of Soil Compaction 37 3 Methods 40 3.1 Research Block Selection 40 3.1.1 Mud Creek 41 3.1.2 Colepitts 43 3.2 Soil Disturbance Survey 46 3.3 Sampling 48 3.4 Laboratory Procedures 49 3.4.1 Bulk Density 50 3.4.2 Particle Size Distribution 50 3.4.3 Particle Density 50 3.4.4 Water Retention 51 3.5 Data Analysis 52 4 Results 53 4.1 Soil Disturbance Survey 54 4.2 Particle Size Distribution 56 4.2.1 Mud Creek 56 4.2.2 Colepitts 57 iv 4.3 Bulk Density 57 4.4 Total Porosity 58 4.5 Aeration Porosity 59 4.5.1 Mud Creek 60 4.4.2 Colepitts 64 4.5 Available Water Storage Capacity 69 5 Discussion 70 5.1 Soil Disturbance 70 5.2 Physical Properties 76 5.2.1 Mud Creek 76 5.2.2 Colepitts 78 6 Conclusions 82 Bibliography 86 Appendix 1. Soil disturbance types and survey assessment point classification. 94 Appendix 2. Soil water potential and equivalent pore sizes. 101 Appendix 3. Complete soil disturbance survey results. 102 Appendix 4. Critical threshold tests. 103 V List of Tables 2.1 Site characteristics for Froehlich's (1978) study of bulk density changes following machine traffic 22 3.1 Results of assessments of soil sensitivity to disturbance for Mud Creek. . 42 3.2 Results of assessments of soil sensitivity to disturbance for Colepitts.. . 44 3.3 Brief descriptions of soil disturbance types which were sampled for physical property analysis 47 3.4 Number of samples used in statistical analysis for each disturbance type, core size, and sample depth at each study site 50 3.5 Cited gauge pressures and equivalent pore diameter of air-filled pores. . . 51 4.1 Summary of soil disturbance levels at Colepitts and Mud Creek for selected machine traffic disturbance types, and the undisturbed 55 4.2 Percent potentially detrimental soil disturbance by disturbance type at the Mud Creek and Colepitts study areas as defined at the time of survey by the British Columbia Ministry of Forests Soil Conservation Guidelines 56 4.3 Mean bulk density (95 % confidence interval) in megagrams per cubic meter at Mud Creek 57 4.4 Mean bulk density (95 % confidence interval) in megagrams per cubic meter at Colepitts 58 4.5 Mean total porosity (95 % confidence interval) as a percent of total core volume at Mud Creek 59 4.6 Mean total porosity (95 % confidence interval) as a percent of total core volume at Colepitts 59 4.7 Mean aeration porosity (95 % confidence interval) as a percent of total core volume for pores > 50 um equivalent pore diameter at Mud Creek. 60 4.8 ANOVA table for aeration porosity (> 50 jam equivalent pore diameter) at 2 - 4 cm at Mud Creek 62 4.9 ANOVA table for aeration porosity (> 50 um equivalent pore diameter) at 6 - 8 cm at Mud Creek 62 4.10 Results of Duncan multiple range test for 2 - 4 cm aeration porosity (> 50 um equivalent pore diameter) at Mud Creek 62 4.11 Results of Duncan multiple range test for 6 - 8 cm aeration porosity (> 50 urn equivalent pore diameter) at Mud Creek 63 4.12 Mean aeration porosity (95 % confidence interval) as a percent of total core volume at Mud Creek (> 30 um equivalent pore diameter) 63 4.13 Mean aeration porosity (95 % confidence interval) as a percent of total core volume at Colepitts (> 50 um equivalent pore diameter) 64 vi 4.14 ANOVA table for aeration porosity (> 50 urn equivalent pore diameter) at 2 - 4 cm at Colepitts 66 4.15 Results of Duncan multiple range test for 2 - 4 cm aeration porosity (> 50 um equivalent pore diameter) at Colepitts 66 4.16 Mean aeration porosity (95 % confidence interval) percent of total core volume at Colepitts (> 30 um equivalent pore diameter) 66 4.17 ANOVA table for aeration porosity (> 30 um equivalent pore diameter) at 2 - 4 cm at Colepitts 68 4.18 Results of Duncan multiple range test for 2 - 4 cm aeration porosity (> 30 um equivalent pore diameter) at Colepitts 68 4.19 Mean available water storage capacity (95 % confidence interval) as a percent of total core volume at both sites 69 vii List of Figures 4.1 Mean aeration porosity and 95 % confidence interval, for pores with equivalent diameters > 50 jam, for the 2 - 4 cm core samples at the Mud Creek site 61 4.2 Mean aeration porosity and 95 % confidence interval, for pores with equivalent diameters > 50 um, for the 2 - 4 cm core samples at the Colepitts site 65 4.3 Mean aeration porosity and 95% confidence interval, for pores with equivalent diameters of > 30 um, for the 2 - 4 cm core samples at the Colepitts site 67 5.1 Mean potentially detrimental soil disturbance levels (percent of area) at Colepitts and Mud Creek pre- and post-inspection of laboratory heavy machine traffic compaction data 72 viii Acknowledgments There was much support for this project from the Ministry of Forests, Nelson Forest Region, including initial financial support of the author, and funding of all lab work not completed by the author. Gerry Davis initially spearheaded the project, and inspired me with her enthusiasm for the work, and her professionalism. Dr. Mike Curran served on my thesis committee and provided support and encouragement through this long process; his constructive reviews were of tremendous help. Don Norris and John Pollack also contributed to, and supported this work. Peter Ott's input on statistical analyses was also appreciated. I owe thanks to my UBC committee members: Dr. Tim Ballard, Dr. Art Bomke, and Dr. Val LeMay. Their patience, helpful reviews, and encouragement were instrumental in the completion of this thesis. I acknowledge the contribution and support of my current employer, Crestbrook Forest Industries Ltd., and in particular Craig Lodge, Dennis Rounsville, and Dave Basaraba, who provided a supportive work environment, and access to computer and other resources, all of which made completion of this thesis much less painful. Finally, I owe great thanks to my friend and partner in life, Sheena, who encouraged me when I lost heart, put up with prolonged absences from our home, and when the going got thick, dug soil pits all over the Nelson Region, and helped process many of the core samples in that hectic eight month race to complete the laboratory work at UBC. ix Chapter 1 Introduction In British Columbia, soil disturbance resulting from timber harvesting activities is a potentially serious problem. Site productivity losses in British Columbia, due to timber harvesting-related degrading soil disturbance, have been estimated by Utzig and Walmsley (1988) at 400,000 cubic metres annually for the ten year period 1976 - 1985. This study was initiated because pushover logging has the potential to cause above-average levels of degrading soil disturbance because of the stump uprooting procedure and the widespread trafficking of the block by the stumping and skidding equipment. Soil disturbance can be viewed as either beneficial, neutral, or detrimental. Beneficial soil disturbance is generally disturbance that promotes early seedling establishment and growth without causing any negative growth effects in the long term, and without causing degradation of other forest resource values. Detrimental soil disturbance is also referred to as soil degradation, potentially detrimental soil disturbance, or degrading soil disturbance, and is generally soil disturbance that results in negative impacts on soil productivity, reductions in tree growth, and/or detrimental off-site impacts. The purpose of measuring soil disturbance is generally to define levels of either beneficial or potentially detrimental soil disturbance. At the time of study initiation, the 1993 Ministry of Forests Interior Soil Conservation Guidelines for Timber Harvesting (Anonymous 1993) and the Interior Soil Conservation Guidelines for Mechanical Site Preparation (Anonymous 1994) recommended maximum allowable levels of potentially detrimental soil disturbance, including machine traffic, created by either timber harvesting or mechanical site preparation. The objective of both guidelines was to promote good planning and execution of field operations to ensure protection of the soil, and other, resources. The guideline levels, which constrained timber harvesting activities, were set based on soil disturbance surveys and expert opinion of the significance of the various disturbance types. There were concerns that errors in the guidelines may impede the efficiency of harvesting operations unnecessarily, or precipitate an unforeseen degradation of the soil resource and reduce expected timber and other resource values in the future. Until 1993, pushover logging-related soil disturbance had only been cursorily investigated through soil disturbance surveys. At the time of initiation of this study, the British Columbia Ministry of Forests was also involved in research projects aimed at quantifying compaction associated with machine traffic. As the use of bulk density and soil porosity measures was planned for these, this study also used these measures to examine potential compactive effects associated with the four machine traffic soil disturbance types selected. 2 This study involved soil disturbance surveys of two harvested blocks, which were logged using pushover logging techniques, and sampling of four machine traffic soil disturbance types, and undisturbed areas, for analysis of changes in soil physical properties associated with soil compaction by machine traffic. The objectives were to record levels and characteristics of soil disturbance associated with pushover logging, and to quantify the soil compaction associated with four machine traffic disturbance types. Results are presented for soil disturbance surveys, and for bulk density, aeration porosity, total porosity and available water storage capacity. It was anticipated that the results of this study might aid in the development of strategies to contain soil disturbance associated with pushover logging within acceptable levels. A second benefit would be information on the efficacy of the soil disturbance survey techniques employed in terms of their ability to adequately capture and describe soil disturbance associated with pushover logging operations, and machine traffic disturbance generally. 3 Chapter 2 Literature Review Soil disturbance associated with timber harvesting activities has long been recognized as having potentially damaging consequences for future site productivity, and for off-site resource values such as water quality and associated fisheries. Among the potentially detrimental types of soil disturbance, soil compaction has been the subject of much work. Numerous studies have documented soil disturbance levels (Dyrness 1965, Bockheim et al. 1975, Martin 1988, Thompson and Osberg 1992, Smith and Wass 1994, Davis and Wells 1994); quantified compaction (Froehlich 1978, Jakobsen and Moore 1981, Greene and Stuart 1985, Lenhard 1986, Incerti et al. 1987, Shetron et al. 1988, Davis 1992, Utzig and Thompson 1992, Smith and Wass 1994); documented reductions in tree growth (Youngberg 1959, Foil and Ralston 1967, Froehlich 1979, Wert and Thomas 1981, Mitchell et al. 1982, Lockaby and Vidrine 1984, Cochran and Brock 1985, McNabb and Campbell 1985, Wasterlund 1985, Donnelly and Shane 1986, Froehlich et al. 1986, Helms et al. 1986, Helms and Hipkin 1986, Clayton et al. 1987, Corns 1988, Van Damme et al. 1992, Smith and Wass 1994), and longevity of compaction (Dickerson 1976, Froehlich et al. 1985). Some critical levels have also been postulated for the various physical properties (Daddow and Warrington 1983, Theodorou et al. 1991) and in some cases incorporated into forest policy (USDA 1996, Howes et al. 1983). Comprehensive reviews of compaction of forest soils have also been prepared (Greacen and Sands 1980). The variety of survey methods, sampling methods, sampling depths, soil types, moisture contents, etc. is considerable when all the literature reviewed for this study is surveyed. For the purposes of this thesis, the majority of the methods encountered will be mentioned, but discussion will center on results where conditions were similar to those in this study. The reader is referred to the cited references for comprehensive discussions of the various papers reviewed. 2.1 Measurement and Quantification of Soil Disturbance In one of the earliest published papers encountered on soil disturbance measurement, Dyrness (1965) used point sampling along randomly located transects to survey the extent of soil disturbance in four disturbance classes on one tractor-logged and three high-lead harvested areas in the western Cascades. The four disturbance types surveyed were undisturbed, slight disturbance (shallow scalps, shallow forest floor and mineral soil mixing, shallow mineral soil deposits), deeply disturbed (surface soil removed), and compacted areas. No areal extent was specified for any of the specific soil disturbance types. 5 The study found 18.9 % to 24.5 % of the area was slightly disturbed in high-lead units and 26.4 % in the tractor unit; 5.8 % to 13.6 % deeply disturbed in high-lead units and 8.9 % in the tractor unit; and 7.1 % to 10.7 % compacted in high-lead units and 26.8% in the tractor unit. Dyrness also noted that the compaction in the tractor logging unit (where it was caused by machine traffic) was more severe than in the high-lead unit (where it was caused by tree skidding), and this was the disturbance type considered of greatest concern on these blocks. Bockheim et al. (1975) examined thirteen high-lead, two tractor-logged, and one helicopter-logged area (all clearcuts) in the Vancouver Forest Region and reported the area distribution of soil disturbance in four disturbance classes. Disturbance classes examined included: undisturbed, forest floor disturbance (but not removal), shallow soil disturbance (essentially the same as Dyrness' (1965) "slight" disturbance class), and deep soil disturbance (B horizon exposed, or mineral soil deposits greater than 5 cm). No areal extent was specified for any of the soil disturbance types. The survey technique involved locating transects in mid-block areas and assessing disturbance at one meter to three meter intervals to provide approximately one hundred survey points per block. A summary of surveyed disturbance for these areas indicated that tractor logging averaged 11 % shallow and 58 % deep soil disturbance, high-lead logging averaged 18 % shallow and 11 % deep soil disturbance, and the helicopter block had 2 % shallow and 3 % deep soil disturbance. Soil disturbance types associated with compaction were not discussed. 6 Martin (1988) surveyed soil disturbance using Dyrness' (1965) transect method in three harvested blocks in the northeast United States. A total of eighty possible disturbance types were assessed by tallying ten disturbance types in eight depth/thickness classes. The disturbance types examined included: undisturbed, depressed (forest floor depressed but not displaced; includes depressions caused by falling trees), organic scarification (lateral displacement of forest floor with no mineral soil exposure and no evidence of compression by any means), mineral scarification (total removal of forest floor), organic mounds (mounds of soil covered by organic material such as adjacent to wheel ruts), mineral mounds (mounds of mineral soil or organics covered by mineral soil created during harvesting), organic ruts (wheel ruts still lined with organics), mineral ruts (ruts with exposed mineral soil), dead wood, and rocks (larger than 10 cm on transect line). Depth/thickness categories included 0 cm and then 1-10 cm, 11 - 20 cm, etc. up to a maximum of 70 cm. Martin (1988), working in one spruce (Picea rubens Sarg.) - fir (Abies balsamea (L.) Mill.) and two hardwood forests (Acer saccharum Marsh, Betula alleghaniensis Britton, Fagus grandifolia Ehrh., Fraxinus americana L., Acer rubrum L., and Populus tremuloides Michx.) in the northeast United States, considered any exposure of mineral soils to be potentially detrimental to seedling establishment and growth due to low fertility and impediment of root egress through potential surface crusts. At the three sites, between 48 % and 81 % had some compaction, but if organic ruts less than 10 cm were excluded, then the "potentially serious" compaction was felt to occupy 23 %, 31 %, and 35 %, of the three sites. Martin (1988) concluded that all rutting should be avoided by selecting appropriate equipment and site conditions, that trails should be pre-determined to limit areal extent of compaction, and that all scarification should be planned, and then used only when expected to be beneficial to tree establishment and growth. Thompson and Osberg (1992) compiled data for surveys of ground-based harvesting disturbance in one hundred and sixteen cutblocks in British Columbia using the then standard British Columbia Ministry of Forests soil disturbance survey methodology (Curran and Thompson 1991). The blocks were selected to represent currently attainable levels of soil disturbance given favourable planning and harvesting practices. Ten soil disturbance types were distinguished including skidroad cutbanks or running surfaces and skidroad berms, and machine traffic in six classes based on materials (organics, cohesive and non-cohesive mineral) and depths (<5 cm, 5-10 cm, >10 cm, >15 cm, and >15 cm). The two remaining disturbance types are undisturbed and other; the other disturbance category included log-butt disturbance and natural disturbance such as windthrow. They discussed results in terms of detrimental skidding disturbance (both skidroad types and all machine traffic disturbance, except the < 5 cm type), detrimental disturbance (as for detrimental skidding, but including landings and unapproved haulroads, and total disturbance (detrimental disturbance plus the < 5 cm "light disturbance" type). The 8 factors of season of logging, terrain conditions (complex or uniform, and steepness), and harvest system were concluded to be critical in predicting levels of site disturbance. For summer logged blocks on uniform (non-complex) terrain and slopes less than 15 %, detrimental skidding disturbance averaged about 6 % and total disturbance averaged about 46 %; for winter-logged blocks in the same category, detrimental skidding disturbance averaged about 3 % and total disturbance about 22 %. The average detrimental skidding disturbance for all seasons and terrain types on slopes less than 15% was 5.5%. Smith and Wass (1994) reported soil disturbance levels following stump uprooting on a 26-ha cutblock in the Golden Timber Supply Area. They recognized unique disturbance types associated with stump uprooting gouges and deposits, fireguards, skidroads, and undisturbed areas. Gouges and deposits were further broken down based on the depth of the disturbance as shallow (< 5 cm), deep (5 - 25 cm), and very deep (> 25 cm). They classed 86 % of the area as disturbed soil, of which 72 % was disturbed by the stump uprooting operation and 14 % mainly by fireguards and skidroads not masked by the stumping operation. They found that 57 % of the block was occupied by deep or very deep stump uprooting disturbance, of which 24 % was gouges and 33 % was deposits. Very deep gouges, those deeper than 25 cm, occupied 4 % of the block. Of the 72 % disturbance associated with stump uprooting, 30 % was attributed to tractor tracks. 9 Davis and Wells (1994) completed an exhaustive study of soil disturbance associated with nineteen cut blocks in the Nelson Forest Region, which had either been pushover logged, or had undergone a post-harvest stump uprooting treatment. The primary emphasis of their study was the disturbance associated with the created stump holes, but data on all soil disturbance was collected, including machine traffic. Machine traffic disturbance was surveyed according to Curran and Thompson (1991), with some changes in the way compaction was assessed approximately half way through the study. Davis and Wells (1994) present results for a wide array of soil disturbance types, and for an array of site conditions, harvest seasons, stump uprooting techniques, and soil survey result interpretations. Summaries for detrimental machine traffic and stump holes were presented. For the various seasons and stump uprooting techniques the following results were obtained for mean fraction of the land area affected by potentially detrimental machine traffic disturbance: 12.4 % for summer pushover logging, 2.7 % for winter pushover logging, 2.9 % for summer post-harvest stump uprooted, and 3.6 % for winter post-harvest stump uprooted blocks. Generally, potentially detrimental machine traffic disturbance consisted of bladed or excavated skidroads, main skidtrails, and ruts deeper than 5 cm. Stump holes occupied between 5.1 % and 19.6 % of the harvested area, and averaged 8.3 % for summer harvested areas, and 10.5 % for winter blocks. 10 In summary, surveyed soil disturbance levels can be seen to vary quite widely depending on what is considered disturbance under the survey criteria. With the exception of Davis and Wells (1994), none of the disturbance types in any of the surveys discussed closely correspond with the disturbance types utilized in this study (see Appendix 1). Dyrness' (1965) compact disturbance category may relate to all four of the disturbance types assessed; the deep disturbance category of Bockheim et al. (1975) may equate to this study's heavy traffic disturbance types, but also to the gouge categories; Martin's (1988) depressed and organic ruts categories could also equate to any of the four disturbance types examined; Thompson and Osberg's (1992) and Davis and Wells (1994) < 5 cm (light machine traffic) class potentially compares to the light traffic disturbance types, while the other skidtrail types may compare with the heavy traffic disturbance types. In fact, the soil disturbance type classification system used in this study has not been replicated since to the author's knowledge. The current British Columbia Ministry of Forests soil disturbance survey types (Anonymous 1997) do, however, closely correspond to the types utilized, but no published results of completed surveys are known to the author. 2.2 Measures of Soil Compaction Of the various types of potentially detrimental soil disturbance, soil compaction is one of the most studied. Ground-based harvesting systems have the greatest potential to cause severe and extensive compaction. Soil compaction causes increases in soil strength and results in pore size loss, or redistribution, leading to reduction in water infiltration rates 11 and soil hydraulic conductivity, and reduced gas exchange. For each physical parameter affected by compaction, there is a range of tests which can be performed to determine the changes the compacted soil has undergone relative to an undisturbed control. In the forest soil research literature reviewed, the most common types of soil compaction tests were changes in bulk density and pore size distribution. Bulk density measures are relatively simple and inexpensive to do, which probably accounts for their popularity. Since changes in bulk density are due to reductions in total porosity, they are also well related to physical parameters of direct relevance to plant growth such as aeration and water storage; bulk density values have also been related to soil strength values which inhibit root egress. 2.2.1 Bulk Density Bulk density is defined by Brady (1990) as "the mass of dry soil per unit of bulk volume, including air space. The bulk volume is determined before drying to constant weight at 105 °C." This universally-accepted definition facilitates comparison of results between different researchers, however some care must still be taken where the strict definition provided by Brady (1990) is not followed. For example, Childs and Flint (1990) make a case for considering total and fine fraction bulk densities separately. Dependent on bulk sample size, the presence of large amounts of organic matter (such as coarse roots) could also be a possible source of error in bulk density determination. Terry et al. (1981) 12 found that corrections for inclusion of organic material were not necessary in their study specifically looking at the problem, because the percentage of the bulk sample occupied by organic material was only 2 - 3 % of sample volume. Blake and Hartge (1986a) list four methods for determining bulk density: core, excavation, clod, and radiation. In the forest soil compaction literature reviewed the core, excavation, and radiation methods were all encountered; the radiation and core method most commonly. 2.2.2 Porosity Cary and Hayden (1973) stated "of all the physical measurements which have been used to characterize various aspects of soil structure, the pore-size distribution is one of the most pertinent, so far as plant growth is concerned." They make this assertion because of the critical role of pore size distribution in such critical processes as water infiltration, water storage capacity, and aeration status of soil. The complexity of a soil pore system defies easy characterization. Various techniques have been employed in efforts to overcome the difficulties in characterizing a medium of irregular particle and pore shape and size. Though no method is considered perfect, each has a useful application subject to its own constraints. Not all pore sizes are of interest when considering importance to plants; the particular range of pore sizes usually considered most critical generally corresponds to those larger 13 than about 0.2 um (equivalent pore diameter) as these are the pores involved in rapid water infiltration, soil aeration, and storage of water potentially available to plants (Marshall and Holmes 1988). Only some of the methods of measuring pore size distribution are suited to investigations within this range on fine-textured soils. The ultimate utility of any method in forest soil research will depend on how well it is related to performance characteristics of the soil system; in forestry applications, plant productivity. Four techniques for measuring pore size distribution were reviewed: acoustical (Sabatier et al. 1990, Moore and Attenborough 1992), nitrogen sorption (Sills et al. 1973, Lawrence 1977), micrometric (Olsson 1986, Newman and Thomasson 1979), mercury intrusion (Gregg and Sing 1967, Ponec et al. 191 A, Lawrence et al. 1979, Newman and Thomasson 1979, Thompson et al. 1985), and water desorption (Newman and Thomasson 1979, Bullock and Thomasson 1979, Klute 1986). The acoustical technique is in its developmental infancy and had not been applied in any forest soil compaction studies. Nitrogen sorption is frequently encountered as a technique for measurement of soil pores, however its application in forest soil compaction studies is limited as the pore sizes most suitable for this measurement technique (0.001 um to 0.02 um equivalent pore diameter (Sills et al. 1973)) are at least two orders of magnitude smaller than the range usually of interest. Lawrence (1977) 14 suggests it be used for measurement of specific surface areas and pore size distributions below 0.02 um equivalent pore diameter. Micrometric, water desorption, and mercury intrusion techniques are considered in more detail below. Micrometric studies Direct measurement of pore size distribution has been carried out using microscopy techniques, but this investigative approach has not been common in forestry. A possible advantage offered by direct measurement techniques is the opportunity to assess pore shape and orientation as well as pore size distribution. Olsson (1986) found that compaction resulted not only in a reduction in pore volume but also in a reorientation of pores along primarily horizontal planes. This reorientation shift is of considerable importance to plant productivity when such pore-controlled phenomena as gas exchange and water infiltration are considered. Bullock and Thomasson (1979) compared micrometric and water desorption techniques in studies of soil structure. They found that micrometric methods (pore impregnation with a fluorescent polyester resin, thin sectioning, photography, and analysis with an image analyzing computer) yielded consistently larger macroporosities. This difference was attributed to the importance of pore neck size in water desorption techniques and their irrelevance in micrometric techniques: in the micrometric technique the dimensions of the impregnated pore were measured and the pore classed as a macropore ("macro" > 60 um diameter), whilst if it were connected by a neck of sufficiently small size to prevent drainage at the prescribed macropore tension (5 kPa) it would be ascribed 15 "micro" status by water desorption methods. There are other issues also relevant in comparisons of the two methods: sample preparation methods and sample volume. Samples for both measurement techniques were collected together in the spring. For the water desorption technique the samples were then wetted to saturation, a process the authors felt may have resulted in a soil structure altered from the sampled field state. No such alteration was felt to have occurred with the micrometric techniques which involved drying by acetone replacement of water and subsequent resin impregnation. Sample volumes also varied considerably: 222 cm3 for the water desorption versus thin sections for the micrometric image analysis. Questions regarding the number of fields that need to be scanned in order to ensure adequate characterization of bulk soil using microscopy techniques were raised by Stoops (1973) and Lawrence (1977). Water desorption and mercury intrusion In determining the water desorption curve for a soil, a wetted sample is subject to a series of incremental changes in soil water potential and a measure of water drained at each potential is made. The calculation of the pore size distribution is then made based on the theory that soil pores act like capillary tubes, with water being released from pores of a certain diameter at a critical water potential. Since soil pores are not perfect cylinders an effective rather than actual pore size is obtained. Appendix 2 provides some information on determination of equivalent pore size distribution. 16 The water desorption method is suggested by Danielson and Sutherland (1986) as a reasonable method to use provided certain conditions hold. One condition is an assumption of no change in soil structure with change in water content. For finer-textured soils with a large clay component, the validity of this assumption is questionable (Marshall and Holmes 1988), so the technique may be restricted in its applicability and range of use on these soils. Bullock and Thomasson (1979) note that results will also be impacted by the continuity of the pore system and the size of the connections, particularly where large pores may be connected to the system by pores of smaller diameter. This latter concern is known as the hysteresis effect, and is why specification of sorption or desorption measurement techniques is important; desorption techniques are more common. The mercury intrusion method also involves the use of pressure - pore size relationships, but on a sorption rather than desorbing curve. Since mercury is a non-wetting liquid (contact angle > 90°), an external pressure is required to force it into a pore. The same considerations of pore geometry that apply with water desorption techniques apply here too, and so an effective, rather than actual pore size distribution is calculated. The necessary assumption of no change in soil structure with change in water content also applies to mercury intrusion investigations, as evacuation of fluids is called for prior to the start of analysis. 17 While there are some similarities between the two methods, there are also some important differences. Mercury intrusion measurements are commonly done on natural soil aggregates whereas water desorption is commonly done on undisturbed soil cores. Aggregate volume in mercury intrusion techniques is generally quite small: the largest aggregate encountered in the literature reviewed was 3.6 cm (Newman and Thomasson 1979) while other researchers used aggregates in the 0.5 cm - 3.0 cm range (Thompson et al. 1985, Lawrence et al. 1979). Water desorption measurements are typically done on much larger cores, for example 222 cm (Newman and Thomasson 1979). Klute (1986) suggests that cores of up to 3500 cm could be used but since time to reach equilibrium is proportional to the square of the height, shorter cores are probably preferred with volumes around 100 cm3 being quite practical. Both Stoops (1973) and Lawrence (1977) raised the question of how many samples needed to be examined to get an accurate description of bulk soil when small sample volumes, such as in the micrometric and mercury intrusion techniques, are utilized. The investigations of Lawrence et al. (1979) emphasized to them the importance of interaggregate pores or fissures, which arise from the shrinkage of aggregates, in governing such important processes as conduction of air and water in clay soils. Their investigations also emphasize the limitations of small aggregate investigations when soil structure on a larger scale is of such importance. 18 In his review of methods of measuring soil aeration, Lindstrom (1990) also emphasizes the importance of macrostructure on determining changes in aeration following compaction. Interaggregate pores are also those most commonly affected in the early stages of soil compaction and may be the only pores affected in cases of mild compaction (Tim Ballard, personal communication). As these interaggregate pores may account for most, if not all, of the macropores, measurement techniques which include them are critical. While in some cases water desorption samples may not be large enough to accurately capture all soil macrostructural features, they are much larger than samples used in nitrogen sorption and micrometric techniques. A potential problem with the water desorption technique is changes in sample structure that may occur during the measurement process, or in preparation of the sample for measurement. These potential problems may arise from the drying of field samples prior to analysis or during water desorption analysis, or during the wetting of samples prior to the desorption analysis, or at all of these times. In soils that contain considerable clay-sized fraction, a change in water content may be accompanied by a change in volume. The shrinkage of a soil on drying, or its swelling with wetting, can have marked effects on soil structure (Marshall and Holmes 1988). While the exact mechanisms of shrinking and swelling are not understood, Marshall and Holmes (1988) conclude that the shrinking and swelling behavior of soils depends on the proportion of the sample in the clay fraction particle size, the cation exchange capacity of 19 the clay minerals, and the type of exchangeable cations. Swelling is greater with smectite and with illite than with kaolinite, and with monovalent exchangeable cations than with divalent or trivalent cations (Marshall and Holmes 1988). Some researchers have discounted the importance of shrinking and swelling behavior in investigations of pores of larger size. For example, Newman and Thomasson (1979) felt that the larger pores (> 60 um equivalent diameter) found in their Rothamsted study of soils, with greater than 34 % clay fraction, were probably of biologic origin could be accurately characterized as cylindrical. They found little change in these pores as drying progressed, leading them to conclude such pores were stable enough to withstand surface tension forces. Bullock and Thomasson (1979) felt that measurement of pores smaller than 30 um (equivalent pore diameter) was potentially unreliable due to possible shrinkage effects. Marshall and Holmes (1988) have questioned the absolute validity of a 0 0 contact angle for desorbing water in some situations. They were mainly concerned with dry sandy soils but also had some reservations about soils enriched with organic matter. No other mention of this concern was found. 2.3 Quantification of Forest Soil Compaction Bulk density is the most common means by which soil compaction is quantified. After bulk density, porosity determinations are the next most common assessment method, followed by penetration resistance and saturated hydraulic conductivity. Where cores are 20 used researchers have often performed multiple analyses on each core (for example some or all of: bulk density, porosity, penetration resistance, saturated hydraulic conductivity, i.e. Jakobsen and Moore 1981, Incerti et al. 1987). Bulk density and porosity are of interest here; a brief review of the results from previous studies where these two measures were utilized and discussed is presented. Of the various bulk density measurement methods available, the core method was encountered most frequently in the reviewed literature (Jakobsen and Moore 1981, Greene and Stuart 1985, Lenhard 1986, Incerti et al. 1987, Utzig and Thompson 1992, and others), followed by radiation (Froehlich 1978, Davis 1992, and others), and excavation methods (Shetron et al. 1988, Smith and Wass 1994). Most of the research reviewed involved examination of bulk density in only the upper 20 cm of soil, and frequently only a single sample depth near the surface; Jakobsen and Moore (1981) had the deepest samples reported at 40 - 45 cm. For determinations of porosity the water desorption method was used exclusively (Jakobsen and Moore 1981, Lenhard 1986, Incerti et al. 1987, Utzig and Thompson 1992), except where only total porosity was calculated (Shetron et al. 1988). Froehlich (1978) quantified soil compaction using bulk density measures carried out with a portable nuclear densitometer. The specific purpose of his research was to quantify the impacts of an FMC 200 Series skidder notable for its low ground pressure, 21 when loaded, of from 49 kPa to 64 kPa. Three sites in western Oregon, each with different soil conditions, were utilized in the study. Table 2.1 provides brief information on each of the three sites. Table 2.1 Site characteristics for Froehlich's (1978) study of bulk density changes following machine traffic. Site % sand % silt % clay textural class L F H depth (cm) moisture content* (%) Mt. Hood 65 17 18 SL 5-20 43 Umpqua 51 30 19 L 13-23 29 Malheur 70 20 10 SL < 1 13 *only a "percent moisture" reported; refers to upper 30 cm of soil profile. Measurements of soil density at 5 cm, 10 cm, 15 cm, and 25 cm were taken prior to any traffic, and after one, three, six, ten, and twenty machine trips over the trail. Measurements were also taken at Mt. Hood and Umpqua of soil density for main trails near landings that were estimated to have received between fifty and sixty, and ninety and one hundred trips, respectively. No statistical analysis was apparently done on any of the results; mean values and trends are reported only. Froehlich (1978) concluded that during the first twenty trips density changes were primarily in the 5 cm to 10 cm depths, that density increased the most during the first few trips, and that density continued to increase in amount and depth, though sometimes very subtly, with additional trips over twenty. These results are illustrated at the Mt. 22 Hood site, which recorded the largest changes due to an initially very low mean bulk density (0.65 Mg m"3), and high moisture content. These combined to give, at the 5 cm depth, after three trips, a mean bulk density of 1.03 Mg m"3, and after twenty trips of 1.12 Mg m"3. For the main trail near a landing (fifty to sixty trips), bulk density was measured at 1.24 Mg m"3. At 15 cm, the differences were less striking with mean bulk density increasing from 0.87 Mg m"3 initially, to 0.89 Mg m"3 after three trips, and 1.01 Mg m"3 after ten trips. There was no further increase between ten and twenty trips and for the main trail near the landing the mean bulk density was 0.96 Mg m" . At 25 cm, the undisturbed bulk density of 0.88 Mg m" does not appear to be any different from the main trail (fifty to sixty trips) bulk density of 0.90 Mg m"3. At the Umpqua site, there were very few changes in bulk density from undisturbed values in the first twenty trips. At 5 cm bulk density increased from an undisturbed value of 1.06 Mg m"3 in the undisturbed, to 1.09 Mg m"3 after three trips, and remained unchanged until ninety to one hundred trips, when it was measured at 1.33 Mg m" . The trend remained the same for the deeper depths, but the magnitude of the change was reduced. At the Malheur site the results were similar, with the largest change being at the surface for the first three trips (bulk density increased from 1.09 Mg m"3 to 1.21 Mg mJ),and no increase measured at the 25 cm depth regardless of trip number (a maximum of twenty trips were measured at Malheur). 23 Shetron et al. (1988) measured bulk density using the excavation method, and (presumably) calculated total porosity from bulk density and particle density measurements. Samples were taken in the upper 5 cm of soil prior to traffic, and after four and twelve skidding passes, for skidders with no load (62 kPa ground pressure), half load (87 kPa), and full load (112 kPa), and immediately after, six months after, and one year after operations. The soil conditions are described only as being texturally a SiL with a moisture content of approximately 50 % (gravimetric), which was felt to be at, or slightly above field capacity. There were no significant differences (a = 0.05) for bulk density as a result of loading, number of passes, or time. Controls were significantly different (a = 0.05) from all treatments, suggesting that the maximum change in bulk density would occur with the first four trips of an unloaded machine. Total porosity is described as being reduced almost 18 %, from 85 % in controls to 70 % in wheel track areas, but there is no further discussion vis a vis the number of trips, or the load on the machine, that resulted in this change. Shetron et al. (1988) concluded that on the basis of their results no adverse effects on plant growth could be forecast. They employed a critical bulk density value of 1.30 Mg m"3 (Foil and Ralston 1967) and porosity value of 50 % (Trouse and Baver 1962) in coming to this conclusion. Jakobsen and Moore (1981) conducted a trial to assess differences between two skidding machines, a D-7 Caterpillar and an FMC 220 CA, in terms of compaction impacts. 24 Among other measures, macroporosity (equivalent pore diameter greater than approximately 60 um) and bulk density were measured using the core method; samples were taken at 0 - 5 cm, 5-10 cm, and 40 - 45 cm. The soil conditions included an A horizon with a CL/SCL texture (45 % sand, 22 % silt, and 33 % clay), and a moisture content of approximately 0.7 kg kg"1 at 5 cm, and 0.5 kg kg"1 at 45 cm. The role of slash cover in preventing compaction under the two machines was also investigated. Mean aeration porosity (60 um equivalent pore diameter) was 26 % for the undisturbed in the upper sample (2 - 7 cm for undisturbed) location, and 16.6 % at the lower (40 -45 cm) location. After seven passes by the FMC the mean aeration porosity for the upper sample had been reduced to 20.3 % for bare soil traffic areas, but increased to 30 % for slash covered areas. For the D-7, the bare soil aeration porosity was reduced to 16.7 % and the slash-covered soil aeration porosity increased to 28 %. The difference in effects between bare soil and slash-covered trails was gone by fifteen passes, as measured at the 5 - 10 cm depth, with the FMC area having values of 10.1 % and 9.2 %, and the D-7 area 10 % and 8.8 %, respectively. Bulk density values followed the same general trend. Initial bulk density at 0 - 5 cm was 0.705 Mg m"3 and was increased to a maximum of 0.738 Mg m"3 by the D-7 on bare soil, and reduced to a minimum of 0.565 Mg m"3 by the FMC on slash-covered trails. At the 25 40 - 45 cm depth, the undisturbed and trafficked area values were essentially the same (0.938 Mg m"3 and 0.929 Mg m"3 respectively). Jakobsen and Moore (1981) reported that the differences in impacts between the two machines were not significant (ot-level not reported). They also concluded that aeration values were considered below critical levels (10 %, no reference), but that biological activity in the surface layers by soil macrofauna may ameliorate the effects of compaction over time. Lenhard (1986) measured bulk density and pore size distribution, following from one to thirty-two machine trips with a rubber-tired skidder (55.2 kPa ground pressure) on a SiL soil with a volumetric water content of 0.11 m3 m"3. Undisturbed soil samples (46 cm3) were extracted from the 3 - 6 cm depth with a core sampler. Aeration porosity was determined at a matric potential of - 5.9 kPa (> 50 um equivalent pore diameter) using a tension table apparatus; a pressure chamber apparatus was used to complete the pore size distribution work at matric potentials of -11.8, -33.3, -101, -304, and -507 kPa. Bulk density values reached a maximum after four trips, but the distribution of pore sizes continued to change after four trips. Lenhard (1986) considered the -5.9 kPa matric potential (i.e. > 50 um equivalent pore diameter) to provide an estimate of air-filled porosity that was synonymous with macroporosity. The volume of macropores was seen to be significantly different (a = 0.05) between zero, four, and eight trips, but not 26 between eight, sixteen or thirty-two trips. The amount of air-filled pore space after sixteen trips and after thirty-two trips was significantly different (a = 0.05) for the 3 um equivalent pore diameter size (-101 kPa matric potential), indicating continuing soil structural changes were occurring as a result of the compactive forces. Incerti et al. (1987) measured bulk density and porosity using soil cores on a SiCL/SiC soil in southeastern Australia. Bulk density in machine traffic areas was 27.1 % higher than in undisturbed areas (significant at a = 0.01), while macroporosity (60 um equivalent pore diameter) was reduced from 28.6 % in the undisturbed to 9.7 % in the machine traffic areas (significant at a = 0.01). Microporosity (< 40 um and > 0.02 um equivalent pore diameters) was significantly different (significant at a = 0.01), and was seen to increase from 30.8 % in the undisturbed to 39.6 % in the machine traffic areas. No information was provided on soil moisture content, or on the number of machine passes, except to say there were more than three machine passes. Utzig and Thompson (1992) studied compaction at three sites in the southeastern interior of British Columbia, where a feller-buncher - random skidding harvesting system had been employed on moist soils. Core samples were taken for bulk density measurements at 2 - 8 cm and 8-14 cm, and for water retention measurements at 2 - 4 cm and 8 -10 cm. Soil disturbance types followed Curran and Thompson (1991). They found impressions greater than 5 cm into the soil were associated with increases in bulk density of 14 - 20 %, which they considered detrimental (the forerunner to the USD A (1996) i 27 publication was presumably used as a guide). At one site, aeration porosity (not defined in terms of an equivalent pore size or water potential) is shown to have been reduced below 15 % for light machine traffic and impression (rut) disturbance types at both 2 -4 cm, and 8 - 10 cm depths. It is not reported whether or not these values were significantly different from the undisturbed values (which had a mean aeration porosity value below 15 % at 8 - 10 cm). In summary, the number of machine passes, or level of machine traffic, was a common aspect of the reviewed studies, where compaction was quantified. The data generated in these research projects generally supports the suggestion that the majority of the compactive effect is realized in the first few trips. However, the data also show that very little change in bulk density may occur on some sites, given specific site conditions such as favourable moisture contents, and that the amount of change in bulk density can be ameliorated where the soil is covered by logging slash. The results of these studies where soil compaction has been quantified are of interest because: 1) they confirm that bulk density and porosity measures can detect compaction following machine traffic on forest soils, 2) they suggest that the number of trips and soil moisture content are important determinants in the amount of compaction that will be detected following machine traffic, 3) they show that, in some cases, the changes in bulk density may be minimal given certain site conditions, 4) they show a continuing change 28 in porosity while bulk density measurements remain constant is possible, and 5) they suggest that machine type and load, and visual estimation of compaction may not be effective predictors of measured impacts. 2.4 Growth Impacts of Forest Soil Compaction Russell (1949) felt that the best way to approach soil structure descriptions was to study soil conditions and document their influence on plant growth. Youngberg (1959), Foil and Ralston (1967), Froehlich (1979), Wert and Thomas (1981), Mitchell et al. (1982), Lockaby and Vidrine (1984), Cochran and Brock (1985), Wasterlund (1985), Froehlich et al. (1986), Helms et al. (1986), Helms and Hipkin (1986), Clayton et al. (1987), Corns (1988), and Van Damme et al. (1992), among others, have all documented the effects of soil compaction on tree growth. For the most part, the studies reviewed were of two types: 1) laboratory, greenhouse, and field studies looking at very early tree growth (i.e. first few weeks to two years out planted), and 2) field studies investigating growth after 5 to 32 years. Van Damme et al. (1992) took a unique look at the use of compaction of the microsite to enhance establishment of direct-seeded Pinus banksiana in northwestern Ontario. They found that seedling establishment was improved by 30 % through manual tamping of scarified seeding sites. One of the earliest studies on tree growth and soil compaction was by Youngberg (1959), who examined height growth of 2-0 root-pruned Pseudotsuga menziesii seedlings out planted at two sites in the spring of 1955. Examined seedlings were planted in tractor 29 roads (assumed to be equivalent to bladed or excavated skidroads), berm areas, and undisturbed cutblock areas. Recognizing the array of factors potentially impacting seedling height growth, Youngberg did a range of soil chemical analyses in addition to physical properties. Reduced height growth was observed at both sites on the tractor road disturbance compared to the undisturbed; tractor roads also had significantly (a = 0.01) reduced height growth compared to berms. At one site there was no difference in height growth between the berm and undisturbed, at the other the height growth of trees in the disturbed berm area was reduced compared to the undisturbed. Bulk densities on the roads were nearly double that of the cutblock areas and berm areas, suggesting that soil compaction was a factor accounting for the reduced growth of seedlings in the road area compared to the berm. Foil and Ralston (1967), Mitchell et al. (1982), Wasterlund (1985), and Corns (1988) all studied early seedling growth in compacted soils. They all found growth reductions in the growth parameters measured relative to undisturbed, or normal, soil density levels. Mitchell et al. (1982), and Corns (1988) found that the degree of negative impact on growth increased as soil compaction increased. Wasterlund (1985) noted that there may be differences between species in terms of early growth on compacted soils; in his seedling study Picea abies L. Karst. was more sensitive to compaction than Pinus sylvestris L. 30 Cochran and Brock (1985) and Lockaby and Vidrine (1984) studied early growth of Pinus ponder osa and P. taeda, respectively, on clearcuts subject to tractor logging and site preparation treatments. Cochran and Brock (1985) concluded that although a reduction in height growth could be correlated with increasing bulk density, less than half the variation in height growth was accounted for by bulk density in the regressions (r2 < 0.50). Other than some plastic cages which had been employed for early seedling protection from animal damage, and had caused some tree damage, no other growth-impacting parameters were mentioned. Lockaby and Vidrine (1984) found that in addition to a 39 % reduction in height growth, there were 88 % fewer trees growing on primary skidroads than non-trafficked areas. They noted no significant difference (a = 0.05) in bulk density between primary skidroads and non-trafficked areas at 2 - 4" (5.1 -10.2 cm) and 4 - 6" (10.2 -15.2 cm) depths; non-trafficked bulk density increased 15 % between the 0 - 2" (0 - 5.1 cm) depth and the two deeper depths. Secondary skidroads were not significantly different (a = 0.05) from non-trafficked areas for bulk density or height, but did show reduced stocking levels. Given even seed distribution during direct seeding, the researchers concluded that survival was negatively impacted by increasing levels of compaction. Froehlich et al. (1986) found a number of variables which better explained differences in P. ponderosa growth than bulk density: age, site index, and basal area. However, they found that diameter growth decreased more than height growth for a given level of bulk density increase. Froehlich et al. (1986) also examined P. contorta for growth effects 31 due to compaction and found no significant relationships between tree growth and bulk density changes. They felt this may be due to an increased ability of P. contorta to root through denser soils as noted by Minore et al. (1969), reduced compaction compared to other sites, and increased moisture availability at the microsite reducing tree moisture stress. Froehlich et al. (1986) noted that natural regeneration of P. ponderosa on the skidtrails was delayed 4-18 years, P. contorta regeneration only 1-4 years. Clayton et al. (1987) studied P. contorta and P. ponder osa along with soil compaction and lateral soil displacement on tractor-logged, and slash-piled and burned, clearcuts in central Idaho. They found penetration resistance was better related to growth reduction than bulk density changes. This they felt was due to thin layers of increased bulk density within the upper 17 cm that their large bulk density samples did not detect (17 cm x 4.7 cm diameter cylindrical cores). Soil displacement was a reliable indicator of tree growth performance, with increased displacement relating to decreased growth. Helms and Hipkin (1986) found height growth of P. ponder osa was significantly related to changes in bulk density (a = 0.01), but diameter growth was not. Stocking and survival were also affected, with areas with the highest bulk density having substantially lower stocking than undisturbed areas or those with slight increase in bulk density. Their study was based on mapping of soil bulk density zones 15 years after logging had occurred. At that time bulk density on an old primary skidtrail was still 130 % of the undisturbed values. 32 Wert and Thomas (1981) completed a study that looked at growth of Pseudotsuga menziesii on compacted soils. The 4.33-ha study area in the Coast Range of Oregon had been clearcut logged in 1947 and left to regenerate naturally; the five seed trees left were removed in 1950. In the study, the cutblock area was mapped, based on repeated traverses, until the original skidroad pattern was established based the on remaining visual evidence. They examined trees growing on old skidroads, those in three-meter-wide "transition zones" on either side of skidroads, and remaining areas (classed as "undisturbed"). They found that total volume per hectare on skidroads, transition zones, and undisturbed areas was 34.1, 97.2, and 128.9 m3 respectively. Skidroads produced 74 % less volume and had 41 % fewer trees, transition zones 25 % less volume and 17 % fewer trees. These growth reductions translated to an 11.8 % volume loss over the entire area. They noted that as much as 20 % of the "undisturbed" areas were found to have ) quite high bulk densities (> 1.20 Mg m"3) and felt they may have misclassified some "undisturbed" areas based on these results (undisturbed mean bulk density, with the suspect 20 % of samples included, was 1.10 Mg m"3). They felt poor early seedling survival, and related delayed early height growth of skidroad trees, was critical in observed volume differences. Trees were observed to grow at the same rate once a certain age had been reached. They also observed that in the upper 15 cm, the soil bulk density in skidroad areas was not significantly different from the undisturbed, but that at 20-cm and 30-cm depths significant differences (a = 0.05 and 0.01, respectively) still existed 30 years post-harvest. 33 Froehlich (1979) assigned residual young Pinus ponderosa to one of three soil density classes, based on a hypothesized proportion of the tree's roots growing in impacted soil. The average age of the residual trees was 64 years and it had been 16 years since the partial cutting entry had occurred. Subsequent tree growth was shown to be most impacted by variables other than soil density: diameter at time of partial cutting, crown volume, and a competition index. Froehlich (1979) concluded with questions regarding why soil density was not causing observable growth reductions: were other stresses more limiting? are roots able to spread to non-impacted areas because of tree spacing? Due to the persistence of the compaction (density at 7.6 cm and 15.2 cm still 18 % higher than undisturbed soils), use of the same trails for any subsequent entry was recommended. Helms et al. (1986) documented the only case encountered where compaction was suggested as possibly helping tree growth. Studying P. ponderosa growth in a plantation in the Sierra Nevada of California, they found that following a two-year drought, 16-year-old trees growing on higher bulk density soils experienced less growth reduction than those growing in low bulk density soils. They suggested that this could be explained by higher stocking levels and more competition for water on less compacted sites, but they also suggested that increase in bulk density from 0.8 Mg m" to 1.1 Mg m" in this loam soil may have enhanced soil water retention and capillary flow. 34 In summary, a review of the literature, while finding exceptions here and there, will generally conclude that increases in soil compaction, as determined by bulk density measurements, results in reduced tree productivity potential on the particular compacted land area. Impairment of naturally-seeded stand establishment and retarded early growth explains some of the volume reductions observed, but it is also apparent that, in the many cases, individual trees growing on compacted soils will not be as productive as those in undisturbed areas. The depth to which compaction occurs and its longevity also varies. 2.5 Longevity of Forest Soil Compaction While the results of some of the previously reviewed studies involved assessment of compaction years after logging and, therefore, an assessment of compaction longevity, Dickerson (1976), and Froehlich et al. (1985) have specifically investigated the persistence of compaction in forest soils. Dickerson (1976) studied soil compaction recovery over a 5-year period for the upper 5 cm of soils ranging in texture from loamy sand to silty clay loam that had been subject to seven skidder passes when relatively wet (12 % for sandy soils to 35 % for silt loams, volume basis). After 5 years, bulk density was 11 % greater in the skidder track areas than in the undisturbed. If the recovery rate observed over the first 5 years continued it would take 12 years for bulk densities to return to undisturbed levels. The degree of compaction still present at 5 years was felt sufficient to still interfere with seedling establishment and early growth. 35 Froehlich et al. (1985) examined soil recovery from compaction by measuring bulk density changes at three depths on skidroads and undisturbed areas along a chronosequence of logging sites on loamy and loamy-sand textured soils. A total often study sites were located in partial cut areas in west-central Idaho spanning a time period of 25 years. The loamy-sand soil showed highly significant differences (a = 0.01) between skidroads and undisturbed for all depths and age classes, except for the surface (5.1 cm) depth for the age classes older than 10 years. For the loamy soil, there were highly significant differences (a = 0.01) for all depths and age classes, except the oldest age class at the surface (which was significant at a = 0.05) and at the 30.5-cm depth (which was not significant). The rate of recovery of the surface soils was found to be faster on both sites than for deeper depths and the rates of recovery for the two soil types were found not to differ from each other. While none of the reviewed literature documented compaction persisting throughout a rotation, compaction lasting long enough to become cumulative with multiple entries associated with partial cuts and/or thinning operations in some stands was documented, and the possibility of compaction persisting throughout rotation on some sites exists. Surface soils (top 5 cm) are seen to be more heavily impacted, and to potentially recover more quickly from compaction over time. Soil texture is a factor in the likelihood of potentially detrimental compaction occurring (Anonymous 1995b), although possibly not in the rate of recovery. For sites with shallow compaction, surface tillage was suggested as a possible way of ameliorating negative effects (Dickerson 1976, Froehlich et al. 1985, Greacen and Sands 1980, Wert and Thomas 1981). 36 2.6 Critical Levels of Soil Compaction Critical levels of compaction are suggested in a number of literature articles (Lull 1959, Mitchell et al. 1982, Daddow and Warrington 1983, Pritchett and Fisher 1987, Theodorou et al. 1991), and in some cases incorporated into official government forest management policy (USDA 1996, Howes et al. 1983). The USDA has established specific soil parameters that determine whether or not a soil has been degraded (USDA 1996). Some of the specific parameters are: 1) a 15 % increase in bulk density over the undisturbed except for volcanic ash or pumice soils where a 20 % increase is required, 2) a 50 % reduction in macropore space, or 3) less than 15 % macropore space (USDA 1996, Howes et al. 1983). "Macropore" is not defined in either publication, which is unfortunate as considerable confusion exists regarding the use of such undefined terminology (see Danielson and Sutherland 1986, Luxmoore 1981, Bouma 1981, Greenland 1977). The depth at which the measurement is to be done is also not specified. The USDA (1996) requires that the changes in macropore volume be determined by measurement with an air permeameter. Geist et al. (1989) evaluated the potential for using a portable air permeameter as a rapid way to diagnose compacted conditions. They found no potential for its use on the volcanic ash soils on which they were working as the correlations with bulk density measures were very poor (r = 0.130 to 0.342). No 37 other papers were encountered where an air permeameter was used to quantify compaction, nor where air permeameter measures were compared to more common measures. There have been a number of studies completed that identified possible critical or threshold levels of compaction, above which impacts on tree productivity could be expected. Lull (1959), in a review of soil compaction on forest and range lands, stated that roots cannot penetrate highly compacted soils. He cites Raney et al. (1955) as saying that, for finer-textured soils, root growth is restricted for soils with bulk densities above 1.4 Mg m"3, and for coarse-textured soils, above 1.6 Mg m"3. Pritchett and Fisher (1987) suggest that penetration of tree roots is impeded in compacted sands with bulk densities above 1.75 Mg m"3, and in clays with bulk densities exceeding 1.55 Mg m"3. Mitchell et al. (1982) found, in a greenhouse study of seedling growth, that root growth was reduced at bulk densities above 1.4 Mg m"3, and terminated completely at 1.8 Mg m"3. The range of bulk densities required to impede root penetration for various soil textures was highlighted in a literature review by Daddow and Warrington (1983). They concluded that coarse-textured soils generally required greater bulk density for root growth curtailment than finer-textured soils. They constructed a textural triangle of growth-limiting bulk density "isodensity" lines based upon their review of the literature. 38 Unfortunately the papers reviewed were primarily based on research conducted in agricultural fields. They were also not able to incorporate potentially relevant factors such as soil moisture content or plant species. 39 Chapter 3 Methods 3.1 Research Block Selection Research blocks were selected by first visiting all recently (less than two years old) pushover harvested blocks in the Nelson Forest Region. Blocks were then short-listed to include only those that had High or Very High Compaction hazards, non-complex terrain, less than 30 % slope, less than 20 % coarse fragment content in the upper 20 cm, and at least 3 ha meeting these criteria. A preference was also given to winter-harvested blocks to reflect realistic operating conditions for the sensitive soils (high compaction hazard). Other criteria, such as ease of access, and problems with growth of brush and vegetation obscuring soil disturbance and interfering with sampling, were also considered. These criteria were in place because: 1) these were the types of blocks where concerns about soil degradation were greatest, and 2) the low coarse fragment content would allow use of the available laboratory methods for porosity measurements. The two blocks selected were the only two found in the Nelson Forest Region that met all the criteria, and that would be available during the survey and sampling windows; both were harvested during the winter. In 1990, due to concerns about soil degradation, the Nelson 40 Forest Region enacted a Standard Operating Procedure that "red flagged" blocks such as these for further consideration and planning prior to approval of harvest plans; this Standard Operating Procedure has since been updated (Norris 1995). The two blocks selected are referred to from here on as Colepitts and Mud Creek. The Colepitts block is within the Golden Forest District and is situated approximately 23 km north-northwest of Golden. The Mud Creek block is within the Invermere Forest District and is situated approximately 3.75 km east of Canal Flats. Both sites are in the Rocky Mountain Trench bench lands. 3.1.1 Mud Creek The Mud Creek block is approximately 5 ha and falls within the Kootenay Dry Mild Interior Douglas-fir biogeoclimatic variant 01 site series; soil nutrient regime is medium to poor, soil moisture regime mesic (Braumandl and Curran 1992). The study area is flat. Sensitivity to soil disturbance was determined at the time using the methods outlined in British Columbia Ministry of Forests Land Management Handbook Field Guide Insert 8 (Lewis and Carr 1993), and recently using the British Columbia Forest Practices Code procedures outlined in the publication "Hazard Assessment Keys for Evaluating Site Sensitivity to Soil Degrading Processes Guidebook" (Anonymous 1995b). Table 3.1 outlines the results of these assessments. 41 Table 3.1 Results of assessments of soil sensitivity to disturbance for Mud Creek. Soil-degrading process 1993 assessment procedure 1995 assessment procedure Soil compaction and puddling High High Displacement Moderate Moderate Forest floor displacement Very High High Surface soil erosion Low Moderate Mass wasting Low Low A soil description according to the Canadian System of Soil Classification (Agriculture Canada Expert Committee on Soil Survey 1987) was also completed. The soil is classified as an Orthic Eutric Brunisol. The forest floor classification follows Klinka et al. (1981). depth (cm) 1.5-0 horizon description LFH Orthixeromoder; abrupt, wavy boundary 0-29 Bm1 dark yellowish brown (10YR 4/4 m); SiL; 10 % c.f. gravels; weak, fine to coarse subangular blocky; friable; very sticky, plastic; clear, smooth boundary; moderate effervescence in 10 % HC1. 29 - 40+ Cca light brownish yellow (10YR 6/4 m); SiCL; 15 % c.f. gravels; weak, medium platy; firm; slightly sticky, plastic; strong effervescence in 10 % HC1. ' The B horizon should have been described as two horizons, Bm and Bmk, at the time of soil description. A separate description of these horizons is not possible, based upon collected field ionformation, at the time of writng. 42 Depth to carbonates, based on effervescence in 10 % HC1, was 19 cm. Pommier (1995) described the forest cover in the research area as 40 % Pseudotsuga menziesii, 30 % Pinus contorta, 20 % Larix occidentalis Nutt, and 10 % Pinus ponderosa. For the merchantable trees , the average height was 15.7 m, average diameter at breast height was 24.5 cm, and average volume per tree was 0.3 cubic meters. Total volume per hectare was estimated at 65 cubic meters. This information applies the entire 305.3 ha cut block area. The author's observations are that the stand in the entire area was relatively homogenous, and the gross cut block area stand description area is a reasonably accurate portrayal of the research area characteristics. 3.1.2 Colepitts The Colepitts block is approximately 5.5 ha and falls within the Columbia - Shuswap Moist Warm Interior Cedar - Hemlock biogeoclimatic variant 01 site series; soil nutrient regime is poor to medium, soil moisture regime mesic to subhygric (Braumandl and Curran 1992). The study area is predominantly gently sloping, with a slope of 5 - 10 % and a west - northwest aspect (there are a few very short slopes, < 10 m total slope length, steeper than 10 %, within the study area). Sensitivity of the site to disturbance was also determined, at the time using the methods outlined in British Columbia Ministry of Forests Land Management Handbook Field Guide Insert 8 (Lewis and Carr 1993), and recently using the British Columbia Forest Practices Code procedures 2 Merchantable trees had a diameter (outside bark) at 30 cm stump height of 15 cm for P. contorta, and 20 cm for all other species. 43 outlined in the publication "Hazard Assessment Keys for Evaluating Site Sensitivity to Soil Degrading Processes Guidebook" (Anonymous 1995b). Table 3.2 outlines the results of these assessments. Table 3.2 Results of assessments of soil sensitivity to disturbance for Colepitts. Soil-degrading process 1993 assessment procedure 1995 assessment procedure Soil compaction and puddling High High Displacement Moderate Moderate Forest floor displacement High High Surface soil erosion Moderate High Mass wasting Low Low A soil description according to the Canadian System of Soil Classification (Agriculture Canada Expert Committee on Soil Survey 1987) was completed. The soil is classified as an Orthic Eutric Brunisol3. The forest floor classification follows Klinka et al. (1981). 3 Dr. L. M. Lavkulich's (personal communication) contribution in determining the best classification of this soil, based on the limited morphological data collected at the time of field work, is acknowledged. 44 depth (cm) horizon description 8.5-0 LFH Orthihemihumimor; abrupt, wavy boundary 0-25 Bf yellowish red (5YR 5/8 m); SiL; 10 % c.f. gravels; weak, medium to very coarse subangular blocky; very friable; slightly sticky; non- to slightly plastic; clear, smooth boundary. 25 - 54 + Ck brownish yellow (10YR 6/6 m); SiCL; moderate, medium platy; firm; sticky; plastic; moderately effervescent in 10 % HC1. Note: a discontinuous Ae was occasionally observed 0 - 3.5 cm during survey and sampling, but was not common. Depth to carbonates, based on effervescence in 10 % HC1, was 25 cm. Thornley (1990) provided stand description and forest cover information for the cut block area. The forest cover was described as 60 % Pseudotsuga menziesii, 30 % Picea engelmanii Parry, and 10 % a mixture of Thujaplicata Donn., Abies balamea (L.) Mill., and Populus trichocarpa Torr. and Gray (information presumably a summary of cruise data for the cut block area). For the merchantable trees4, the average height was 25.4 m, 4 Merchantable trees had a 17.5 cm diameter at breast height, except for P. contorta, which had a diameter at breast height of 12.5 cm. Though not specified, it is presumed these would be outside bark diameters. 45 average diameter at breast height was 31.4 cm, and average volume per tree was 0.74 cubic meters. Total volume per hectare was estimated at 609 cubic meters. 3.2 Soil Disturbance Survey The blocks were surveyed using a systematic survey point location method (as per areas < 10 ha, Anonymous 1997). Transects were located perpendicular to the majority of the main skidroads, and ran parallel to each other. Transects were spaced eight meters apart at Mud Creek and nine meters apart at Colepitts; soil disturbance was assessed every two meters along each transect. A wider transect spacing was selected at Colepitts to reduce the number of assessment points (due to time constraints). The first point of the surveys was located by randomly selecting distance to transect start, and distance to first assessment point. Soil disturbance at each point was keyed out according to the key in Appendix 1. A total of 43 distinct disturbance types were utilized in keying out the soil disturbance at each survey point; most of these were simply divisions of operational soil disturbance survey classes made to allow greater resolution of disturbance types for the purposes of research (for example, deposits were separated into depth classes). A larger number of points were surveyed than would be conventional for small blocks (i.e. approximately 500, Anonymous 1997) so as to ensure adequate points in each disturbance class for later sampling. A total of 1717 points were assessed at Colepitts and 2452 points at Mud Creek. 46 Points along survey transects were periodically marked in the field to allow easier relocation. Soil disturbance at each point was coded to allow description of all soil disturbance present at a point, and not just the leading disturbance type. For example, a point landing on a compact heavy skidtrail with an intact forest floor was tallied as such, but if a 20-cm mineral deposit was also present, it too was recorded at the point. The reason for proceeding in this manner was to allow examination of different hierarchical soil disturbance keys, and to accurately record complex disturbance types. A brief description of the specific soil disturbance types sampled for physical property analysis is provided in Table 3.3. Table 3.3 Brief descriptions of soil disturbance types which were sampled for physical property analysis (complete details in Appendix 1). Disturbance type symbol Description of soil disturbance type HF Heavy machine traffic with forest floor intact and compaction detected by digging and hand-checking at survey assessment point. LF Light machine traffic with forest floor intact and compaction detected by digging and hand-checking at the survey assessment point. HF Heavy machine traffic with forest floor intact and no compaction detected by digging and hand-checking at survey assessment point. LF Light machine traffic with forest floor intact and no compaction detected by digging and hand-checking at survey assessment point. U Areas undisturbed by any harvesting activity. The determination of whether an observed disturbance was in the heavy machine traffic or light machine traffic disturbance types was based on areal criteria (described in detail in Appendix 1). If there was any doubt about machine traffic at a point (i.e. surveyor making a decision between light machine traffic and undisturbed disturbance types), it was always assumed to have experienced the light machine traffic. This was done to ensure the undisturbed disturbance type was totally free of any machine impacts. 3.3 Sampling For each of the five soil disturbance types of interest (four disturbance types plus the undisturbed) all points within the disturbance type were grouped and a random sample of twenty points was selected5. Each of these one hundred points was then sampled at three depths for water retention measurements, and at two depths for bulk density measurements. Water retention cores were taken at 2 - 4 cm, 6-8 cm, and 14 - 16 cm depths; bulk density cores were taken at 2 - 8.7 cm and 10 - 16.7 cm depths. Samples were taken using a hammer-driven core sampler. Water retention cores were 36.19 cm (4.8 cm diameter x 2.0 cm depth), and bulk density cores were 495.12 cm (9.7 cm diameter x 6.7 cm depth). If relocation of a sample point was needed (i.e. if too many large coarse fragments), then the next point of the same disturbance type along the transect was sampled. Three points were relocated at Mud Creek, and five at Colepitts, using this method. All samples at a point were taken in close proximity to each other (not more than 15 cm separated core samples). Small cores were extracted with a maximum amount of soil left intact around the sampling cylinder (samples were roughly baseball-sized); these samples 5 Only 19 samples were taken in one of the disturbance types at Colepitts; the soil disturbance survey located only 19 points in total for this disturbance type. 48 were carefully wrapped in the field using plastic wrap and masking tape. Small core sampling locations were staggered to prevent sampling of upper cores disturbing soil where lower core samples would be taken. Small cores were handled as little as possible, and were transported in boxes packed with Styrofoam chips and rolled newspaper. Large cores were shaved to volume in the field and deposited in a double plastic bag. 3.4 Laboratory Procedures All bulk density samples were processed by the author at the University of British Columbia, Soil Science Department laboratories. Particle size distribution analyses, completed on bulk density core samples, was done at Griffin Laboratories Corporation in Kelowna, British Columbia. The 2 - 4 cm small cores were processed by the author at the University of British Columbia, Soil Science Department laboratories. The 6-8 cm, and 14 - 16 cm cores, were processed at Soilcon Laboratories Ltd. in Richmond, British Columbia. Some small cores were found unsuitable for laboratory work once excess soil from the field was shaved off to get the exact sample volume. Typical problems included the presence of large rocks or roots protruding out of the bottom of the core. Some large cores were also not suitable for use due to loss of sample volume during transport or processing. Table 3.4 provides details on the number of samples in each disturbance type, at each depth, and at each site, that were used in the statistical analysis for the large and small core data. 49 Table 3.4 Number of samples used in statistical analysis for each disturbance type, core size, and sample depth at each study site. Co epitts Mud Creek core HF LF HF LF U HF LF HF LF U 2 - 8.7 cm 16 20 20 20 20 20 19 20 20 20 10-16.7 cm 16 19 20 20 19 18 19 18 20 18 2-4 cm 15 19 18 18 18 19 20 18 18 17 6-8 cm 16 20 20 20 20 19 20 20 20 20 14 - 16 cm 16 19 20 20 20 18 18 17 19 19 3.4.1 Bulk Density Bulk density was determined by drying to constant mass at 105 °C (Blake and Hartge 1986a). Total bulk density is reported. 3.4.2 Particle Size Distribution For the fine fraction, particle size distribution was determined, after pretreatment for removal of carbonates and organic matter, using the pipet method (Gee and Bauder 1986). Coarse fraction (2 mm separation) was determined by dry sieving following mechanical treatment of sample. 3.4.3 Particle Density Particle density was determined for a sub-sample of the fine fraction of each small core at the 2 - 4 cm depth using the pycnometer methods of Blake and Hartge (1986b). For the 6 - 8 cm and 14 - 16 cm depths, fine fraction particle density was determined on grouped sub-samples by Soilcon Laboratories Ltd., Richmond, British Columbia using the same methods. 3.4.4 Water Retention A porous plate extractor method was utilized on the intact small cores (Klute 1986). For the 2 - 4 cm cores, gauge pressures of 5.9 kPa, 9.8 kPa, 33 kPa and 1500 kPa were used. For the 6 - 8 cm and 14 -16 cm cores, gauge pressures of 9.8 kPa, 33 kPa, and 1500 kPa were used. Table 3.5 provides the equivalent diameter of pores that will be air-filled at the various gauge pressures used. Total porosity for each small core was determined by dividing the core's total bulk density by its determined mean particle density, and subtracting from one. Table 3.5 Cited gauge pressures and equivalent pore diameter of air-filled pores. Gauge pressure (kPa) Equivalent pore diameter drained (um) 5.9 >50 9.8 >30 33 >9 1500 >0.2 During field sample collection, close examination of the cores is not possible due to the presence of the additional, adhered soil, which is retained with the field sample to help protect it from physical changes during transport. With shaving to exact core volume in the laboratory, some cores were found to be unusable, due to the presence of large rocks protruding out of the core, or the presence of an extraordinary amount of coarse woody root material. If the amount of core damage exceeded about 5 % of total core volume 51 (approximately 1.8 cm3), these cores were excluded from analysis. If less than 5 % of core volume was impacted then soil was gently placed in the void and the core processed. 3.5 Data Analysis An ANOVA (analysis of variance) was completed for each site, at each depth, to determine if there were significant differences among disturbance types for each physical property. If the ANOVA showed that differences were significant (a = 0.05), then a multiple range test using Duncan's procedure was conducted. These statistical analyses were carried out using SPSS for Windows Student Version, Release 6.1 (SPSS 1994). Two t-tests were used to check measured physical property values in the compact heavy machine traffic disturbance class at Colepitts, against critical values identified in the literature. The details of these tests are in Appendix 4. The hypothesis to test for whether macroporosity in the compact heavy machine traffic disturbance type is below 15 % is: Ho: U H F = 0.15 Ha: U H F < 0-15 To test whether the macroporosity in the compact heavy machine traffic is less than half of the macroporosity of the undisturbed areas, a transformation of undisturbed data was completed. The macropore volume of each sample taken in undisturbed areas was multiplied by 0.5, and a new data set created. The hypothesis for the test is: Ho: (J,HF = Hu transformed Ha: U H F < u-u transformed No other critical threshold tests were completed. 52 Chapter 4 Results Soil disturbance survey results are reported first. Then results for physical properties are presented seperately for bulk density, total porosity, aeration porosity, and available water storage capacity. Summaries of all disturbance class means and 95 % confidence intervals for physical properties are presented in tables; in three cases the results are also presented graphically. If the ANOVA showed that differences among disturbance types for the particular physical property were significant (a = 0.05) for a certain depth at a particular site, the results of the ANOVA are presented. These are followed by the results of the Duncan multiple range test. For the Colepitts site, tests were conducted against critical threshold values from the literature. One-way ANOVA with the disturbance class as the factor, and the physical property of interest as the dependent variable, were used throughout. Sokal and Rohlf (1973) discuss the assumptions integral in the ANOVA process. Each of the assumptions they listed: additivity, randomness, independence, normality, and homogeneity of variances, was tested for each data set. 53 Additivity. The ANOVA model: Yij = M + Ti + Eij where: Yij is the measured value for a particular sample j in disturbance type /', M is the expected mean if there is no effect due to disturbance type, Ti is the effect of disturbance type i, and Eij is the random and normally distributed "error" within the disturbance type that is attached to sampling unit /'. The model is additive and is a reasonable representation of the relationship between the samples from the various disturbance types. There is no specific test for additivity of the model (Bergerud 1982). Randomness. Methods of sample selection have been detailed; all samples were selected in a random fashion. Independence. Samples were selected randomly. The assumption of spatial independence can therefore reasonably expected to have been met. Normality. Using SPSS (1994) a Normal P-P Plot of data was completed. Examination of these plots revealed only very minor departures from normality. Homogeneity of variance. SPSS (1994) provides a Levene Test for homogeneity of variance; there were no significant tests (a = 0.05). An Fmax-test (Sokal and Rohlf 1973) was also completed for the 2 - 4 cm aeration porosity data sets. 4.1 Soil Disturbance Survey At each sample point the type of soil disturbance was determined according to the key in Appendix 1. Complete survey results, according to this key, are in Appendix 3. Results for some machine traffic disturbance types, including all disturbance types which were sampled in this study, and the deep gouge disturbance type, are provided in Table 4.1. Table 4.1 Summary of soil disturbance levels at Colepitts and Mud Creek for selected machine traffic disturbance types, and deep gouges and the undisturbed. Disturbance type6 Soil disturbance (percent of surveyed area) Colepitts Mud Creek HE 2.5 1.6 HF 9.8 11.6 HF 1.1 14.8 LE 6.5 0.5 LF 59.6 21.4 LF 1.5 14.3 Gl 9.5 2.1 undisturbed 2.3 22.6 HF, HF, LF, LF, and U disturbance types were sampled for physical property analyses in this study. At the time of the soil disturbance survey, potentially detrimental soil disturbance for pushover logging was defined by the publications: Soil Conservation Guidelines for Timber Harvesting - Interior British Columbia (Anonymous 1993), and the Interim Soil Conservation Guidelines for Mechanical Site Preparation - Interior British Columbia (Anonymous 1994). For the two study areas, a total of eight potentially detrimental soil disturbance types were identified: bladed skidroads, heavy machine traffic, machine ruts, deep gouges, wide gouges, long gouges, and wide and very wide scalps (see Appendix 1 for complete definitions of each of these disturbance types). Four of these potentially detrimental disturbance types 6 HE - heavy machine traffic with forest floor removed and without compaction at assessment point; HF -heavy machine traffic with forest floor intact and without compaction at assessment point; HF - heavy machine traffic with forest floor intact and with compaction at assessment point; LE - light machine traffic with forest floor removed and without compaction at assessment point; LF - light machine traffic with forest floor intact and without compaction at assessment point; LF - light machine traffic with forest floor intact and with compaction at assessment point. Not all disturbance types tallied during survey are included in table. Complete details in Appendix 1. 55 were found in the two study areas. Table 4.2 provides details on the extent of each of these four disturbance types, in each of the two blocks, according to the soil disturbance survey. Table 4.2 Percent potentially detrimental soil disturbance by disturbance type at the Mud Creek and Colepitts study areas as defined at the time of the survey by the British Columbia Ministry of Forests Soil Conservation Guidelines. Soil disturbance type Colepitts Mud Creek Heavy machine traffic 1.1 14.9 Wide gouge <0.1 <0.1 Deep gouge 9.5 2.1 Wide scalp <0.1 0 Total potentially detrimental soil disturbance 10.7* 17.1* * column totals differ from cited totals as a result of rounding. 4.2 Particle size distribution 4.2.1 Mud Creek Particle size analysis was conducted for seven randomly selected sample points for the 2 - 8.7 cm large cores, and for ten points for the 10 - 16.7 cm large cores. For the 2 -8.7 cm samples, all were classed as silt loam, with particle size distributions ranging from 20 % to 30 % sand, 54 % to 65 % silt, and 13 % to 22 % clay. A typical sample had 23 % sand, 60 % silt, and 17 % clay. For the 10 - 16.7 cm samples, six were classed as silt loams and four as loams. Particle size distributions ranged from 19 % to 35 % sand, 43 % to 73 % silt, and 8 % to 24 % clay. A typical sample had 28 % sand, 53 % silt, and 19 % clay. The coarse fragment content was determined, on a mass basis, for all large cores. The 2 - 8.7 cm cores had an average coarse fragment content of 3.4 %, and ranged between 0 and 18 %; the 10 - 16.7 cm cores averaged 5.7 %, and ranged between 0 and 31 %. The coarse fragments were predominantly fine gravels (Day 1983). 56 4.2.2 Colepitts Particle size analysis was conducted for ten randomly selected sample points for the 2 -8.7 cm and 10 - 16.7 cm large cores. For the 2 - 8.7 cm samples, all were classed as silt loam, with particle size distributions ranging from 11 % to 31 % sand, 55 % to 71 % silt, and 7 % to 21 % clay. A typical sample would be 23 % sand, 62 % silt, and 15 % clay. For the 10 - 16.7 cm samples, eight were classed as silt loams and two as loams. Particle size distributions ranged from 19 % to 50 % sand, 41 % to 69 % silt, and 9 % to 19 % clay. A typical sample would be 23 % sand, 62 % silt, and 15 % clay. The coarse fragment content was determined, on a mass basis, for all large cores. The 2 - 8.7 cm cores had an average coarse fragment content of 4.5 %, and ranged between 0 and 33 %; the 10 - 16.7 cm cores averaged 9 %, and ranged between 0 and 40 %. The coarse fragments were predominantly fine gravels (Day 1983). 4.3 Bulk Density Bulk density was determined at two depths: 2 - 8.7 cm and 10 - 16.7 cm. The data for both depths at both sites is presented in Tables 4.3 and 4.4 below. Table 4.3 Mean bulk density (95 % confidence interval) in megagrams per cubic meter at Mud Creek. Disturbance class Core depth HF LF HF LF U 2 - 8.7 cm 1.28 (1.23, 1.32) 1.32 (1.26, 1.37) 1.23 (1.17, 1.29) 1.25 (1.21, 1.30) 1.24 (1.19,1.29) 10-16.7 cm 1.29 (1.24, 1.35) 1.29 (1.24, 1.34) 1.27 (1.22, 1.33) 1.27 (1.22, 1.31) 1.22 (1.15, 1.28) 57 The ANOVAs did not show any significant differences in mean bulk density (a = 0.05) among disturbance types at either depth at Mud Creek. Table 4.4 Mean bulk density (95 % confidence interval) in megagrams per cubic meter at Colepitts. Disturbance c ass Core depth HF LF HF LF U 2 - 8.7 cm 0.81 (0.70, 0.93) 0.82 (0.74, 0.90) 0.95 (0.84, 1.07) 0.94 (0.85, 1.04) 0.87 (0.80, 0.95) 10- 16.7 cm 1.16 (1.04, 1.28) 1.12 (1.01, 1.22) 1.18 (1.06, 1.30) 1.06 (0.98, 1.15) 1.11 (0.99, 1.23) The ANOVAs did not detect any significant differences in mean bulk density (a = 0.05) among disturbance types at either depth at Colepitts. 4.4 Total Porosity Total porosity was determined for all disturbance types, at all depths, at both sites. Data summaries are presented by site, with mean and 95 % confidence intervals reported for each disturbance type and depth. Tables 4.5 and 4.6 present mean total porosity data for the two sites. 58 Table 4.5 Mean total porosity (95 % confidence interval) as a percent of total core volume at Mud Creek. Disturbance class Core depth HF LF HF LF U 2-4 cm 54.8 53.8 53.1 55.1 53.3 (52.9, 56.7) (51.9,55.6) (50.1,56.2) (52.5, 57.7) (51.0,55.6) 6 - 8 cm 48.8 51.1 48.8 50.6 49.3 (46.7, 50.8) (48.9, 53.2) (46.4,51.2) (48.4, 52.9) (47.3,51.2) 14 - 16 cm 49.4 50.2 51.5 52.0 51.7 (48.1,50.6) (48.1,52.4) (49.7, 53.3) (49.6, 54.4) (49.1, 54.2) The ANOVA did not show any significant differences (a = 0.05) in mean total porosity among disturbance types at any of the depths at Mud Creek. Table 4.6 Mean total porosity (95 % confidence interval) as a percent of total core volume at Colepitts. Disturbance c ass Core depth HF LF HF LF U 2 - 4cm 66.5 65.8 65.1 65.1 67.2 (62.1,70.1) (62.7, 68.9) (62.1,68.1) (62.0, 68.2) (64.7, 69.8) 6 - 8 cm 63.2 63.1 59.8 59.3 60.6 (59.4, 66.9) (59.8, 66.4) (55.4, 64.3) (56.8,61.7) (57.5, 63.7) 14 - 16 cm 53.1 57.1 56.0 59.6 55.0 (48.6, 57.6) (53.6, 60.5) (52.2, 59.7) (57.3,61.8) (50.6, 59.5) The ANOVA did not show any significant differences (a = 0.05) in mean total porosity among disturbance types for any of the depths at Colepitts. 4.5 Aeration Porosity Aeration porosity was determined for all depths at 5.9 kPa gauge pressure, and additionally for the 2 - 4 cm depth at 9.8 kPa gauge pressure. Data summaries are presented with disturbance type mean, and 95 % confidence intervals, reported on a depth basis. If the ANOVA showed that differences were significant, the ANOVA results are presented, followed by the results of the Duncan multiple range test. 4.5.1 Mud Creek Table 4.7 provides mean aeration porosity and 95 % confidence interval information for all depths at the Mud Creek site. Figure 4.1 illustrates graphically the mean and 95 % confidence intervals for the 2-4 cm core depth. Table 4.7 Mean aeration porosity (95 % confidence interval) as a percent of total core volume for pores > 50 um equivalent pore diameter at Mud Creek. Disturbance class Core depth HF LF HF LF U 2-4 cm 20.8 (18.1,23.5) 15.6 (14.0, 17.3) 15.9 (12.7, 19.1) 18.3 (15.2,21.4) 14.3 (11.9, 16.6) 6-8 cm 15.2 (13.0, 17.5) 16.0 (13.5, 18.6) 11.2 (9.0,13.5) 14.4 (12.4, 16.5) 12.6 (11.0, 14.2) 14 - 16 cm 16.6 (15.1, 18.0) 14.5 (12.3, 16.7) 13.8 (12.4, 15.3) 15.1 (13.0, 17.3) 15.2 (13.3, 17.1) 60 .30T .Z)f .101 N= -1— 17 u 18 LF 18 H~ 2D LF 19 HF Soil dstirbanoetype Figure 4.1 Mean aeration porosity and 95 % confidence interval, for pores with equivalent diameters > 50 jam, for the 2 - 4 cm core samples at the Mud Creek site. Porosity is given as cubic meters of pores with equivalent diameters > 50 um per cubic meters of bulk soil. 61 The ANOVAs for the 2 - 4 cm and 6 - 8 cm depths are summarized in Tables 4.8 and 4.9. Table 4.8 ANOVA table for aeration porosity (> 50 um equivalent pore diameter) at 2 - 4 cm at Mud Creek. Source d.f. S.S. M.S. F ratio F prob. Between 4 0.0490 0.0123 4.3288 0.0031 Within 87 0.2463 0.0028 Total 91 0.2953 Table 4.9 ANOVA table for aeration porosity (> 50 um equivalent pore diameter) at 6 - 8 cm at Mud Creek. Source d.f. S.S. M.S. F ratio F prob. Between 4 306.3387 76.5847 3.6543 0.0082 Within 94 1969.9877 20.9573 Total 98 2276.3264 The results of the Duncan multiple range test for these two groups are summarized in Tables 4.10 and 4.11. Table 4.10 Results of Duncan multiple range test for 2 - 4 cm aeration porosity (> 50 um equivalent pore diameter) at Mud Creek. The "*" indicates a significant difference. The test was run with a = 0.05. U LF HF LF HF U LF HF LF * j_jp * * * 62 Table 4.11 Results of Duncan multiple range test for 6 - 8 cm aeration porosity (> 50 um equivalent pore diameter) at Mud Creek. The "*" indicates a significant difference. The test was run with a = 0.05. HF U LF HF LF HF U LF * HF * LF * * Summary of results for 2 - 4 cm cores at aeration porosity (> 30 um equivalent pore diameter) are presented in Table 4.12. Table 4.12 Mean aeration porosity (95 % confidence interval) as a percent of total core volume at Mud Creek (> 30 um equivalent pore diameter). Disturbance c ass Core depth HF LF HF LF U 2 - 4cm 21.7 (19.0, 24.3) 18.2 (16.5,20.0) 19.1 (15.5,22.8) 20.1 (16.8, 23.3) 16.6 (13.9, 19.3) The ANOVA did not show any significant differences (a = 0.05) in aeration porosity among disturbance types. 63 4.5.2 Colepitts Table 4.13 presents mean aeration porosity and 95 % confidence interval information for all depths at the Colepitts site. Figure 4.2 illustrates graphically the mean and 95 % confidence intervals for the 2 - 4 cm core depth. Table 4.13 Mean aeration porosity (95 % confidence interval) as a percent of total core volume at Colepitts (> 50 um equivalent pore diameter). Disturbance class Core depth HF LF HF LF U 2-4 cm 17.2 25.2 19.7 21.4 29.4 (12.4,21.9) (22.0, 28.3) (16.1,23.2) (19.1,23.6) (25.3, 33.6) 6 - 8 cm 10.9 14.2 12.8 16.2 13.8 (8.3,, 13.5) (12.0, 16.5) (9.9, 15.6) (13.7, 18.7) (11.0, 16.6) 14 - 16 cm 9.3 13.7 11.8 12.8 10.4 (6.4, 12.2) (11.2, 16.1) (9.0, 14.6) (10.2, 15.4) (8.2, 12.7) 64 .40T ^.30 t : N= 18 18 18 19 15 U LF hF LF HF Soil dstutanoetype Figure 4.2 Mean aeration porosity and 95 % confidence interval, for pores with equivalent diameters > 50 um, for the 2 - 4 cm core samples at the Colepitts site. Porosity is given as cubic meters of pores with equivalent diameters > 50 um per cubic meter of bulk soil. 65 Significant ANOVA and multiple range test results for mean aeration porosity in the 2 - 4 cm depth are summarized in Tables 4.14 and 4.15; at other depths the ANOVA did not show any significant differences (a = 0.05) among disturbance types. Table 4.14 ANOVA table for aeration porosity (> 50 um equivalent pore diameter) at 2-4 cm at Colepitts. Source d.f. S.S. M.S. F ratio F prob. Between 4 0.1588 0.0397 7.8292 0.0000 Within 83 0.4209 0.0051 Total 87 0.5797 Table 4.15 Results of Duncan multiple range test for 2 - 4 cm aeration porosity (> 50 um equivalent pore diameter) at Colepitts. The "*" indicates a significant difference. The test was run with a = 0.05. LF LF U A summary of results for 2 - 4 cm cores aeration porosity (> 30 um equivalent pore diameter) is presented in Tables 4.16 and 4.17. Figure 4.3 illustrates graphically the mean and 95 % confidence intervals for the 2 - 4 cm core depth. Table 4.16 Mean aeration porosity (95 % confidence interval) percent of total core volume at Colepitts (> 30 um equivalent pore diameter). Disturbance c ass Core depth HF LF HF LF U 2 - 4cm 18.0 (13.3,22.6) 26.7 (23.6, 29.8) 21.7 (17.7, 25.6) 24.4 (22.0, 26.9) 31.6 (27.2, 35.9) HF HF LF LF U HF HF 66 .40T N= 18 18 18 19 15 U LF hF LF hF Soil dstutenoetype Figure 4.3 Mean aeration porosity and 95% confidence interval, for pores with equivalent diameters of > 30 um, for the 2 - 4 cm core samples at the Colepitts site. Porosity is given as cubic meters of pores with equivalent diameters of > 30 um per cubic meter of bulk soil. 67 Table 4.17 ANOVA table for aeration porosity (> 30 um equivalent pore diameter) at 2 -4 cm at Colepitts. Source d.f. S.S. M.S. F ratio F prob. Between 4 0.1771 0.0443 8.1268 0.0000 Within 83 0.4523 0.0054 Total 87 0.6295 Table 4.18 Results of Duncan multiple range test for 2 - 4 cm aeration porosity (> 30 um equivalent pore diameter) at Colepitts. The "*" indicates a significant difference. The test was run with a = 0.05. LF U The compact heavy machine traffic disturbance type approaches critical levels identified by the USDA (1996) for detrimental compaction. The USDA (1996) defines, for sites with similar soils as Colepitts, a reduction of macroporosity below 15 % as detrimental. Determination of detrimental compaction can also be made if macroporosity is reduced below 50 % of the undisturbed value. Macroporosity is not defined, in terms of an equivalent pore diameter, in the USDA (1996) publication. The tests conducted here considered pores > 30 um equivalent pore diameter as macropores. For the test of reduction of macroporosity below 15 %, the analysis showed that the null hypothesis cannot be rejected. There was no evidence that macroporosity was reduced below 15 % in the compact heavy machine traffic disturbance type. For the test of reduction HF HF LF LF U HF HF LF 68 of macroporosity below 50 % of undisturbed values, the analysis showed that the null hypothesis can not be rejected. There is no evidence that macroporosity in the compact heavy machine traffic disturbance type is less than half that of the undisturbed areas. No other critical threshold level tests were completed. 4.6 Available Water Storage Capacity Available water storage capacity was determined for the 2 - 4 cm depth. It is reported as the percentage of the total core volume that is occupied by pores that release water between gauge pressures of 33 kPa and 1500 kPa. Table 4.19 Available water storage capacity (95 % confidence interval) as a percent of total core volume at both sites. Disturbance class Core depth HF LF HF LF U 2-4 cm 17.9 19.3 19.1 17.8 20.4 Mud Creek (16.0, 19.9) (17.7, 20.9) (16.6,21.6) (15.4, 20.2) (17.2, 23.5) 2 - 4 cm 23.0 19.9 22.4 21.5 21.2 Colepitts (20.0, 26.0) (17.1,22.7) (20.0, 24.9) (20.0, 23.1) (18.9, 23.4) The ANOVA did not show any significant (a = 0.05) differences among any of the disturbance types at either site. 69 Chapter 5 Discussion 5.1 Soil Disturbance At the time of the soil disturbance surveys, potentially detrimental soil disturbance for pushover logging was defined in two publications: Soil Conservation Guidelines for Timber Harvesting - Interior British Columbia (Anonymous 1993), and the Interim Soil Conservation Guidelines for Mechanical Site Preparation - Interior British Columbia (Anonymous 1994). The results of the surveys completed as part of this study are not strictly comparable to these two publications as the criterion used to judge compaction here was more sensitive than that contained in the guidelines. In this study, a point was considered compacted if the surveyor had any suspicion that it may be. In contrast, the standard protocol in operational surveys at the time was to find some physical evidence that suggested compaction, before deeming a survey point compacted. The expected result is higher levels of heavy machine traffic disturbance being surveyed in this study than would be found by an operational soil disturbance survey conducted on the same blocks, at the same time. 70 However, an examination of the laboratory data suggests that hand testing was an unreliable indicator of compaction. This raises questions about how the soil disturbance data should be interpreted. An interpretation based strictly on the soil survey results, where potentially detrimental disturbance is assigned to disturbance types as per the Ministry of Forests' Soil Conservation Guidelines (Anonymous 1993, 1994), could be seen to overestimate heavy machine traffic at Mud Creek, and to underestimate it at Colepitts. Based on the soil disturbance survey results alone, only 1.1 % of the Colepitts block is considered to have potentially detrimental soil disturbance in the heavy machine traffic disturbance type. After an examination of laboratory measures and statistical analyses potentially detrimental heavy machine traffic at Colepitts is 13.5 %. After examining the data for Mud Creek, the potentially detrimental disturbance in the heavy machine traffic disturbance type would drop from 14.9 % to zero. Figure 5.1 illustrates the effects on total potentially detrimental soil disturbance at each site that inspection of the laboratory data has. At Colepitts, total potentially detrimental disturbance is of two main types: heavy machine traffic, and deep gouges (stump holes). The total potentially detrimental disturbance for the Colepitts area (based on an assessment after examining laboratory data) is 23 %, of which 13.5 % is heavy machine traffic, and 9.5 % is deep gouges. Davis and Wells (1994) recommended that acceptable levels of soil disturbance for pushover logging operations be 20 %, with a requirement for rehabilitation activities post-harvest. They felt that rehabilitation of machine traffic disturbance could result in a net level of potentially detrimental soil disturbance below the desired level, of 13 %, set 71 out in the two Soil Conservation Guidelines (Anonymous 1993, 1994). Other researchers have also suggested that for shallow compaction impacts, surface tillage may ameliorate negative effects (Dickerson 1976, Wert and Thomas 1981, Froehlich et al. 1985). If the rehabilitation measures suggested by Davis and Wells (1994) were applied to the Colepitts study area, the total potentially detrimental soil disturbance level would potentially be reduced below the 13 % acceptable level. H d e e p gouges m heavy machine traffic Colepitts - Colepitts - Mud Cr. - Mud Cr. -survey laboratory survey laboratory Figure 5.1 Mean potentially detrimental soil disturbance levels (percent of area) at Colepitts and Mud Creek pre- and post-inspection of laboratory heavy machine traffic compaction data. Disturbance types are described in detail in Appendix 1. At Mud Creek, heavy machine traffic occupied 28 % of the site, but test results suggest there was no potentially detrimental compaction associated with this disturbance type. It is possible that the initially dense soils, coupled with frozen ground conditions, prevented potentially detrimental soil compaction from occurring at Mud Creek. If the recommendations of Davis and Wells (1994) were followed at Mud Creek, tillage operations might be carried out in error, and over an extensive area. 72 Regardless of the final impacts the heavy machine traffic at Mud Creek may or may not have had, the density of trails is approximately twice that of Colepitts. Stand characteristics and traffic management strategies influenced the density of heavy machine traffic in each area. At Colepitts the density of main trails was lower partly because the average tree was 10 m taller than at Mud Creek. Also at Colepitts, stumps bucked off on the block would have tended to concentrate traffic onto the few routes which were cleared of stumps. The harvest plan at Colepitts called for the use of designated trails; these were to occupy 14 % of the harvest area. The survey suggests this goal was achieved. The use of designated trails, to predict and control machine traffic disturbance, has been recommended by soil conservation professionals working in the Nelson Forest Region (Redfern 1994). At Mud Creek, where trails were not designated, the few large stumps dispersed over the block would have no effect on traffic patterns, and a higher density of dispersed trails would be expected (Redfern 1994). The use of designated trails at Mud Creek would lower the density of the heavy machine traffic disturbance type; however, due to the shorter tree length a higher trail density would still be expected, unless other disturbance-reducing strategies were adopted. Given the apparent lack of compaction on main trails, a strategy of further dispersing traffic, rather than concentrating more traffic on fewer trails, may have been the best strategy at Mud Creek. 73 Deep gouges (associated exclusively with stump holes) occupied only 2.1 % of the Mud Creek block, while at Colepitts the number was 9.5 %. The difference in the amount of deep gouges at the two areas is explained by the different stand characteristics. Many more large trees at Colepitts meant many more large stump holes. At Colepitts, large stump holes were common, and were evenly distributed across the area. At Mud Creek, large stump holes were infrequent and dispersed. The role of tree size and density, as a determinant of potentially detrimental disturbance attributable to stump holes, was recognized by Davis and Wells (1994). In a study on post-harvest de-stumping near Golden, Smith and Wass (1994) found 4 % of their study area occupied by very deep gouges (in their study gouges > 25 cm were classified as very deep). Stand characteristic data were not reported. Davis and Wells (1994) found between 5.1 % and 19.6 % of their nineteen study areas to be occupied by stump holes. For winter pushover logging operations they found an average of 10.5 % of study areas occupied by stump holes; for summer operations the number was 8.3 %. Davis and Wells (1994) did not report variables such as average log size, diameter at breast height, volume per hectare, or average tree length in their study. The area occupied by stump holes at Colepitts (9.5 %) is clearly within a range that is normal for winter pushover operations. The Mud Creek percentage is considerably lower, most likely due to the unique stand characteristics at this area (predominantly mature Pinus contorta with dense Pseudotsuga menziesii thickets, and a few widely-spaced P. menziesii, Larix occidentalis and Pinus ponderosa veterans). The generally smaller tree size and wider spacing would be expected to result in fewer, and smaller, stump holes. 74 Based on harvest inspection reports at Colepitts, and anecdotal evidence from people working at Mud Creek the winter of harvest, snow and weather conditions varied considerably between them. At Colepitts there was approximately 0.7 m of dry snow at the start of harvest and approximately 1 m of total snow depth within two weeks of the start; at Mud Creek there was about 0.3 m at the start of harvest and this depth did not change substantially throughout operations. Anecdotal evidence from personnel working at the Mud Creek site suggests that the temperatures fluctuated around zero degrees Celsius during portions of the harvest operation. The deeper snow depths at Colepitts would be expected to reduce the frequency with which machine traffic would lead to compaction, and to reduce the magnitude of the compaction. At Mud Creek, the warmer temperatures and shallower snow pack would be expected to result in greater compaction. The data suggest that neither of these expectations were necessarily met. Another factor that would be expected to reduce the frequency of compaction in main skidtrail areas at Colepitts was the depth of slash built up on the trails. Slash depths of 20 cm were not uncommon on main skidtrails at Colepitts. Presumably, during operations, this slash was mixed with snow, resulting in a slash-snow mixture over the intact forest floor that would be expected to reduce machine and skidded log impacts on the mineral soil below. At Mud Creek there was rarely any slash build up on trails, and where it was present, it was only a few centimeters thick. 75 5.2 Physical Properties 5.2.1 Mud Creek The results for the Mud Creek site are difficult to interpret. No significant differences were found among disturbance types at any depths for bulk density, total porosity, available water storage capacity, or for aeration porosity associated with pores greater than, or equal to, 30 um equivalent pore diameter. For aeration porosity associated with pores greater than, or equal to, 50 um equivalent pore diameter, significant differences (a = 0.05) were found among disturbance types at the 2 - 4 cm and 6 - 8 cm depths. However, the trends were opposite to what would be expected: for example, at the 2 -4 cm depth, the highest mean aeration porosity (20.8 %) was in the compact heavy machine traffic disturbance type, and the lowest mean aeration porosity (14.3 %) in the undisturbed. Figure 4.1 illustrates the results for aeration porosity associated with pores greater than, or equal to, 50 um equivalent pore diameter for the 2 - 4 cm cores at Mud Creek. One possible explanation for the higher aeration porosity in the compact heavy machine traffic disturbance type is freeze-thaw cycles. Freeze-thaw cycles are more likely to occur in the compact heavy machine traffic disturbance type, as these areas would typically become snow-free earlier than adjacent non-trail areas due to displacement of snow from these areas during skidding. This would potentially lead to a situation where snowbanks along trails melted during the warmer days and delivered water to trails where surface soil melting had occurred. Slow, and shallow, percolation of water into 76 the soil surface would occur until the evening freeze cycle began again. This phenomenon would not be expected in undisturbed or light machine traffic areas, as they would be snow-covered, and somewhat insulated, up until the last of the snow cover melted away. These areas could therefore be subject to fewer freeze-thaw cycles, as coincident with exposure to the freeze-thaw cycles (due removal of snow cover), there would be a reduction in water available for percolation (snow source gone). However, this explanation does not satisfactorily explain why the non-compact heavy machine traffic disturbance type does not also have a higher aeration porosity. In fact, these two disturbance types were found to be significantly different (a = 0.05). There is no documentation of the role of freeze-thaw processes in soil decompaction in British Columbia, nor have similar effects to those proposed here been documented in any other similar work conducted in British Columbia (Mike Curran, personal communication). Other inconsistencies also exist in the data, for example, there is no common trend for the 2 - 4 cm, and 6 - 8 cm aeration porosity (pores greater than 50 (am equivalent diameter) data. Another unexplainable difference is at the 2 - 4 cm depth between the undisturbed and the non-compact light machine traffic class. It would be expected that these two classes would be the least different, as the survey methodology meant a number of possibly undisturbed points were included in the non-compact light machine traffic disturbance type. However, at both sites the multiple range test showed a significant difference (a = 0.05) between these disturbance types. 77 The explanation that Froehlich (1978) provided for some anomalous data is a distinct possibility at Mud Creek. This explanation reflects the fact that the soil sample points were not located at the exact location of the soil disturbance survey assessment point. Sampling locations had to be located approximately 15 - 20 cm away from assessment points due to the destructive nature of the original assessment which involved excavation of soil. This feature, coupled with the way in which compaction at the original assessment point was determined, may have resulted in some pairs of sampling and assessment points belonging to different disturbance types. This would increase variability and decrease the probability of detecting differences, given the sample size. As both sites had soil disturbance surveys, and point assessments, completed by the same individuals, any surveyor errors are expected to be the same at each site. A final explanation is previous harvest activity in the Mud Creek area, and the extensive cattle grazing that has also occurred there in recent years. Although neither of these explanations seems likely to be the sole cause of the data anomalies, combined with some of the other possibilities previously mentioned, they could have played a role. 5.2.2 Colepitts Bulk density, total porosity, and available water storage capacity were not shown to be significantly different (a = 0.05) among disturbance types at any depth. No significant differences (a = 0.05) among disturbance types were found for either aeration porosity measured at 6 - 8 cm, or 14 - 16 cm depths. At the 6 - 8 cm depth the data trend was similar to that at 2 - 4 cm, at the deeper depth no trend is discernible. 78 At the 2 - 4 cm depth, significant differences (a = 0.05) were found among disturbance types, for both aeration porosity measures. Data trends are the same for both measures, with the trend in aeration porosity associated with each disturbance type generally as expected: undisturbed > light machine traffic > heavy machine traffic. Figures 4.2 and 4.3 illustrate mean aeration porosity (equivalent pore diameters > 50 um, and > 30 um, respectively) for the two aeration porosity measures, for the 2 - 4 cm core depth, at Colepitts. The question of the ability of soil disturbance surveyors to accurately assess soil compaction by hand is again relevant at this study site. At Colepitts, both the compact heavy machine traffic disturbance types, and non-compact heavy machine traffic types were shown to be significantly different (a = 0.05) from the undisturbed aeration porosity, but not from each other. This apparent lack of ability to distinguish compaction, by the methods used in this study, suggest a change in the technique is needed. With compaction established in the upper 2 - 4 cm of the soil, the question of its importance, in terms of soil degradation and related future tree productivity, arises. The two heavy machine traffic disturbance types occupied 13.5 % (mean) of the cutblock; this represents a slightly higher level of potentially detrimental disturbance than the 13% considered acceptable at the time (Anonymous 1993, 1994). The surface tillage treatment, suggested by Dickerson (1976), Wert and Thomas (1981), and Froehlich et al. (1985), for areas with shallow compaction impacts, appears a reasonable approach at 79 Colepitts. If no treatment were applied, some researchers have found that surface soil compaction will be naturally ameliorated with time (Dickerson 1976, Froehlich et al. 1985), and this too may ameliorate potential negative impacts over time. If no treatments were applied and there was no natural amelioration of compaction with time, are the measured levels of compaction sufficient to lead to tree growth productivity impacts? Bulk density measures were not significantly different (a = 0.05) at either depth, and are well below the 1.4 Mg m"3 cited as a critical level by various studies for soils similar to the Colepitts site (Raney et al. 1955, Mitchell et al. 1982, Daddow and Warrington 1983). The literature reviewed, that documented a tree growth impact, found that bulk densities in reduced growth areas (historic machine traffic areas) were significantly different (a = 0.05 or 0.01) from undisturbed areas. The lack of significant differences (a = 0.05) here could be taken as an indication that the compaction at these study areas was not as severe. Standards for assessment of detrimental compaction, based on impacts on macropore space, have been established by the USDA (1996). As was previously pointed out, the USDA (1996) does not define a pore size for macropore, and the measurement of macropore volume is to be done with an air permeameter (depth of measurement is also not specified but is presumably at the surface based on the comments of Geist et al. (1989)). For soils such as those at the Colepitts site, a decrease in macropore volume of 50 % or more, or a reduction below 15 % macropore volume, would be considered detrimental if USDA (1996) criteria were applied. 80 The statistical tests were completed using the pore volumes greater than or equal to 30 am equivalent pore diameter to increase the chance of a detrimental effect being detected. 81 Chapter 6 Conclusions Potentially detrimental soil disturbance resulting from pushover logging does not necessarily exceed identified, acceptable levels as evidenced by these two sites. Factors of tree density and size are likely determinants of the frequency at which potentially detrimental deep gouges will occur. Further research should examine this relationship using the areas studied here, and those included in the comprehensive review of stumping and pushover logging by Davis and Wells (1994). Machine traffic disturbance associated with pushover logging is highly variable, and depends on the same factors as for other conventional operations: initial soil density and porosity, soil moisture condition, ground protection, and the number of machine passes. (There was insufficient site-specific information for the study areas about these factors at the time of harvest to draw firm conclusions about their relative importance to measured soil impacts here.) The soil disturbance survey method of assessing compaction at survey points by hand, was an unreliable method of assigning compaction to a point. In this study, compaction was assigned to a point if there was any doubt in the surveyors mind as to whether or not 82 the point might be compact. Even using this method, which should result in an over-estimate of compaction, as much as 92 % of the machine traffic disturbance at the Colepitts site was misclassified as non-compact. The notion that soil disturbance appearance is not proportional to actual impacts was put forth by Jakobsen and Moore (1981), it now appears that soil compaction determined by hand, is not necessarily proportional to actual compaction. Given that current British Columbia Ministry of Forests soil disturbance survey techniques utilize a hand test as part of the criteria for assigning compaction (along with visual evidence), it is suggested that further research into this technique be carried out, or a totally different technique be developed. The measures of porosity used in this study are more sensitive measures of changes in soil structure, caused by compaction, than measures of bulk density. This is due to compaction causing changes in pore size distribution, but not in total pore volume, and to shallower compactive effects than the bulk density measures were able to detect. The heavy machine traffic disturbance types at the Colepitts site were compacted in the surface 2-4 cm, however the magnitude of the compaction can not be shown to exceed critical threshold values identified in the literature. As pointed out be Clayton et al. (1987), and discussed earlier, recovery rates from compaction are an unknown factor, and this uncertainty makes any estimates of growth impacts at rotation highly speculative. At Colepitts, given the apparent shallow depth of the compacted layers, the 83 lack of significant differences detected for bulk density measures, and considering the evidence on natural decompaction forces, it is impossible to conclude there will be a net loss of productivity at the Colepitts site due to machine traffic compaction. There is a lack of information on the possible effects freeze-thaw cycles may have on ameliorating compaction in surface soils over time. For winter-logged blocks, where only shallow surface compaction may occur due to deep penetration of frost into trail areas, this is of particular interest as it may be that shallow surface tillage rehabilitation treatments are not necessary. The efficacy of shallow surface tillage of main skidtrails as a means of restoring favourable soil physical characteristics should be researched further. Such a practice has been advocated by a few notable researchers, but at least one study of the results of such treatments calls into question their economic sensibility (Laing and Howes 1988). Research to establish relationships between the various methods of measurement of soil physical characteristics is needed. In particular, relationships between air permeameter and water desorption methods for measurement of soil macropores, and radiation and core methods for measurement of soil bulk density should be established. If relatively easy field measurements of compaction are possible, perhaps with the air permeameter technique, for example, then the identified problems associated with determining compaction by hand would be moot. 84 Tools to predict the likelihood of potentially detrimental compaction at a site have an important role to play in guiding choices made by harvesting managers relating to maximum dispersion of traffic, or confinement of traffic to designated trails. Well dispersed machine traffic can reduce silviculture costs by acting as a site preparation technique (Redfern 1994). Initial work to develop predictive tools is near completion (To 1998), and follow-up work is ongoing (Mike Curran, personal communication). 85 Bibliography Agriculture Canada Expert Committee on Soil Survey. 1987. The Canadian System of Soil Classification. 2 n d ed. Agriculture Canada Publ. 1646. 164 pp. Anonymous. 1993. Soil Conservation Guidelines for Timber Harvesting. British Columbia Ministry of Forests, Victoria, British Columbia. 6 pp. Anonymous. 1994. Soil Conservation Guidelines for Mechanical Site Preparation. British Columbia Ministry of Forests, Victoria, British Columbia. 10 pp. Anonymous. 1995a. Soil Conservation Guidebook. British Columbia Ministry of Forests/ British Columbia Environment, Victoria, British Columbia. 17 pp. Anonymous. 1995b. Hazard Assessment Keys for Evaluating Site Sensitivity to Soil Degrading Processes Guidebook. British Columbia Ministry of Forests/ British Columbia Environment, Victoria, British Columbia. 24 pp. Anonymous. 1997. Soil Conservation Surveys Guidebook. British Columbia Ministry of Forests/ British Columbia Environment, Victoria, British Columbia. 73 pp. Bergerud, W. 1982. Brief overview of experimental design. BC Min. For., Research Branch, Victoria, BC. 54 pp. Blake, G.R. and K. H. Hartge. 1986a. Bulk Density. In Klute, A. (ed.), Methods of Soil Analysis, Part 1, Physical and Mineralogical Methods, Second edition, Agronomy Number 9, American Society of Agronomy/Soil Science Society of America, Madison, Wisconsin, pp. 363 - 375. Blake, G.R. and K. H. Hartge. 1986b. Particle density. In Klute, A. (ed.), Methods of Soil Analysis, Part 1, Physical and Mineralogical Methods, Second edition, Agronomy Number 9, American Society of Agronomy/Soil Science Society of America, Madison, Wisconsin, pp. 383 - 409. Bockheim, J.G., T. M . Ballard, and R. P. Willington. 1975. Soil disturbance associated with timber harvesting in southwestern British Columbia. Can. J. For. Res. 5: 285 - 290. Bouma, J. 1981. Micro-, meso-, macroporosity and channeling flow phenomena in soils. Comments and letters to the editor. Soil Sci. Soc. Am. J. 45: 1244 - 1245. 86 Brady, N. C. 1990. The Nature and Properties of Soils, tenth edition. MacMillan. 621 pp. Braumandl, T and M. P. Curran 1992. A field guide for site identification and interpretation for the Nelson Forest Region. Land Manage. Handbook Num. 20, British Columbia Ministry of Forests, Victoria, British Columbia. 311 pp. Bullock, P. and A.J. Thomasson. 1979. Rothamsted studies of soil structure II: measurement and characterization of macroporosity by image analysis and comparison with data from water retention measurements. J. Soil Science 30: 391 - 413 Cary, J.W. and C.W. Hayden. 1973. An index for soil pore size distribution. Geoderma 9: 249 - 256. Childs, S. W. and A. L. Flint. 1990. Physical properties of forest soils containing rock fragments. In Sustained Productivity of Forest Soils, 7th North American Forest Soils Conference, Univ. B.C., Vancouver, Canada, pp. 95 - 121. Clayton, J.L., G. Kellogg, and N. Forrester. 1987. Soil disturbance-tree growth relations in central Idaho clearcuts. Res. Note INT-372, Intermountain Research Station, USDA For. Ser., 6 pp. Cochran, P.H. and T. Brock. 1985. Soil compaction and initial height growth of planted ponderosa pine. Res. Note PNW-434, Pacific Northwest Forest and Range Experiment Station, USDA For. Ser.. 4 pp. Corns, I.G.W. 1988. Compaction by forestry equipment and effects on coniferous seedling growth on four soils in the Alberta foothills. Can. J. For. Res. 18:75 - 84. Curran, M.C. and S. R. Thompson. 1991. Measuring soil disturbance following timber harvesting. Land Manage. Field Guide Insert #5. Ministry of Forests. Victoria, BC. 25 pp. Daddow, R.L. and G. E. Warrington. 1983. Growth-limiting soil bulk densities as influenced by soil texture. Watershed Systems Development Group Report WSDG-TN-00005, USDA Forest Service, Ft. Collins, CO. 17 pp. Danielson, R. E. and P. L. Sutherland. 1986. Porosity. In Methods of Soil Analysis, Part 1 - Physical and Mineralogical Methods, 2nd edition, A. Klute (ed.), American Society of Agronomy/Soil Science Society of America, Agronomy Num. 9 (Part 1). pp. 443 - 461. Davis, S. 1992. Bulk density changes in two central Oregon soils following tractor logging and slash piling. West. J. Appl. For. 7(3): 86 - 88. 87 Davis, G. and W. H. Wells. 1994. Stumping and pushover logging in the Nelson Forest Region: 1992 and 1993 soil disturbance surveys. Tech. Rep. TR - 009, British Columbia Ministry of Forests, Victoria, British Columbia. 45 pp. Day, J.H. 1983. The Canada Soil Information System. LRRI Contribution No. 82 - 52, Agriculture Canada, Ottawa, Ontario. 97 pp. Dickerson, B. P. 1976. Soil compaction after tree-length skidding in northern Mississippi. Soil Sci. Soc. Am. J. 40: 965-966. Donnelly, J.R. and J.B. Shane. 1986. Forest ecosystem responses to artificially induced soil compaction. I. Soil physical properties and tree diameter growth. Can. J. For. Res. 16: 750-754. Dyrness, C. T. 1965. Soil surface condition following tractor and high-lead logging in the Oregon Cascades. J. For. 63: 272-275. Foil, R. R. and C.W. Ralston. 1967. The establishment and growth of loblolly pine seedlings on compacted soils. Soil Sci. Soc. Am. Proc. 31: 565-568. Froehlich, H.A. 1978. Soil compaction from low ground-pressure, torsion suspension logging vehicles on three forest soils. Research Paper 36, Forest Research Lab, Oregon State University School of Forestry, Corvallis, Oregon. 13 pp. Froehlich, H.A. 1979. Soil compaction from logging equipment: effects on growth of young ponderosa pine. J. Soil Water Cons. 34: 276-278. Froehlich, H.A., D.W.R. Miles, and R.W. Robbins. 1985. Soil bulk density recovery on compacted skid trails in central Idaho. Soil Sci. Soc. Am. J. 49: 1015-1017. Froehlich, H.A., D.W.R. Miles, and R.W. Robbins. 1986. Growth of young Pinus ponderosa and Pinus contorta on compacted soil in central Washington. For. Ecol. Manage. 15:285-294. Gee, G.W. and J.W. Bauder. 1986. Particle-size analysis. In Klute, A. (ed.), Methods of Soil Analysis, Part 1, Physical and Mineralogical Methods, Second edition, Agronomy Number 9, American Society of Agronomy/Soil Science Society of America, Madison, Wisconsin, pp. 383-409. Geist, J. M., J. W. Hazard, and K. W. Seidel. 1989. Assessing physical conditions of some pacific northwest volcanic ash soils after forest harvest. Soil Sci. Soc. Am. J. 53: 946 - 950. Greacen, E. L. and R. Sands. 1980. Compaction of Forest Soils. A Review. Aust. J. Soil Res. 18: 163 - 189. 88 Greenland, D. J. 1977. Soil damage by intensive arable cultivation: temporary or permanent? Phil. Trans. R. Soc. Ser. B. 281: 193 - 208. Gregg, S. J. and K. S. W. Sing. 1967. Adsorption, surface area and porosity. Academic Press, London. Greene, W. D., and W.B. Stuart. 1985. Skidder and tire size effects on soil compaction. South. J. Appl. For. 9(3): 154-157. Helms, J. A. and C. Hipkin. 1986. Effects of soil compaction on tree volume in a California ponderosa pine plantation. West. J. Appl. For. 1(4):121-124. Helms, J. A., C. Hipkin, and E.B. Alexander. 1986. Effects of soil compaction on height growth of a California ponderosa pine plantation. West. J. Appl. For. 1(4): 104-108. Howes, S., J. Hazard, and J.M. Geist. 1983. Guidelines for sampling some physical conditions of surface soils. R6-RWM-146-1983. USDA For. Ser., Pacific Northwest Region. 34 pp. Incerti, M., P.F. Clinnick, and S.T. Willatt. 1987. Changes in the physical properties of a forest soil following logging. Aust. J. For. Res. 17: 91-98. Jakobsen, B.F. and G. A. Moore. 1981. Effects of two types of skidders and of a slash cover on soil compaction by logging of Mountain Ash. Aust. For. Res. 11: 247-255. Klinka, K., R. N. Green, R. L. Trowbridge, and L. E. Lowe. 1981. Taxonomic classification of humus forms in ecosystems of British Columbia, first approximation. Land Manage. Rep. Num. 8, British Columbia Ministry of Forests, Victoria, British Columbia. 54 pp. Klute, A. 1986. Water retention: laboratory methods. In Methods of Soil Analysis, Part 1 - Physical and Mineralogical Methods, 2nd edition, A. Klute (ed.), American Society of Agronomy/Soil Science Society of America, Agronomy Num. 9 (Part 1). pp. 443-461. Laing, L.E., and S. W. Howes. 1988. Detrimental soil compaction resulting from a feller buncher and rubber-tired skidder timber harvest operation: a case study. In Degradation of Forested Land: Forest Soils at Risk, Proceedings of the 10th B.C. Soil Science Workshop, February, 1986, pp. 191 - 195. Land Manage. Rep. Num. 56. British Columbia Ministry of Forests, Victoria, British Columbia. 89 Lawrence, G. P. 1977. Measurement of pore sizes in fine-textured soils: a review of existing techniques. J. Soil Science 28: 527-540. Lawrence, G. P., D. Payne and D. J. Greenland. 1979. Pore size distribution in critical point and freeze dried aggregates from clay subsoils. J. Soil Science 30: 499-516. Lenhard, R. J. 1986. Changes in void distribution and volume during compaction of a forest soil. Soil Sci. Soc. Am. J. 50: 462-464. Lewis, T. and W. Carr. 1993. Hazard assessment keys for evaluating site sensitivity to soil-degrading processes - Interior sites. Land Manage. Handbook, Field Guide Insert 8, British Columbia Ministry of Forests, Victoria, British Columbia. Lindstrom, J. 1990. Methods of measurement of soil aeration. Dept. of Soil Science, Swedish Univ. Agric. Science, Uppsala, Sweden. 53 pp. Lockaby, B.G. and C.G. Vidrine. 1984. Effect of logging equipment traffic on soil density and growth and survival of young loblolly pine. South. J. Appl. For. 8: 109-112. Lull, H. W. 1959. Soil compaction on Forest and range lands. USDA For. Ser., Washington, D. C. 33 pp. Luxmore, R. J. 1981. Micro-, meso, and macroporosity of soil. Comments and letters to the editor. Soil Sci. Soc. Am. J. 45: 671 - 672. Martin, C.W. 1988. Soil disturbance by logging in New England - review and management recommendations. North. J. Appl. For. 5: 30-34. Marshall, T. J. and J. W. Holmes. 1988. Soil Physics, 2nd edition. Cambridge University Press, Cambridge. 374 pp. McNabb, D. H. and R.G. Campbell. 1985. Quantifying the effects of forest activities on soil productivity. A paper presented at the Society of American Foresters 1985 convention, Ft. Collins, CO. pp. 116 - 120. Minore, D., C. E. Smith, and R. F. Woolard. 1969. Effects of high soil density on seedling root growth of seven northwestern tree species. USDA For. Ser. Res. Note PNW- 112. 6 pp. Mitchell, M. L., A.E. Hassan, C. B. Davey, and J. D. Gregory. 1982. Loblolly pine growth in compacted greenhouse soils. ASAE Paper No. 81-1040. pp. 304 - 312. Moore, H. M. and K. Attenborough. 1992. Acoustical determination of air-filled porosity and relative air permeability of soils. J. Soil Science 43: 211-228. 90 Newman, A. C. D. and A. J. Thomasson. 1979. Rothamsted studies of soil structure III: pore size distributions and shrinkage processes. J. Soil Science 30: 415-439. Norris, D. 1995. Nelson Forest Region Standard Operating Procedures: Armillaria Root Disease. Nelson Forest Region Standard Operating Procedures Num. 6 - 1. British Columbia Ministry of Forests, Victoria, British Columbia. 37 pp. Olsson, M. T. 1986. Micromorphometric evaluation of artificial compaction of fine sandy till. For. Ecol. Manage. 17: 109-117. Pommier, G. 1995. Logging plan - cruise summary for CP 113 Block 72. Crestbrook Forest Industries, Canal Flats Division. 4 pp. Ponec, V., Z. Knor and S. Cerny. 1974. Adsorption on solids. Butterworth and Co. Ltd., London. English translation edited by Smith, D. and N. G. Adams, Dept. of Space Research, University of Birmingham. Pritchett, W. L. and R. F. Fisher. 1987. Properties and Management of Forest Soils, Second Edition. John Wiley and Sons, New York. 494 pp. Raney, W. A., T. W. Edminster, and W. H. Allaway. 1955. Current status of research in soil compaction. Soil Sci. Soc. Am. Proc. 19: 423 - 428. Redfern, L. 1994. Soil disturbance surveys. Unpublished internal report for Crestbrook Forest Industries Ltd., Cranbrook, British Columbia. 307 pp. Russell, M. B. 1949. Methods of measuring soil structure and aeration. Soil Sci. 68: 25 -35. Sabatier, J.M., H. Hess, W. P. Arnott, K. Attenborough, M. J. M. Romkens, and E. H. Grissinger. 1990. In situ measurements of soil physical properties by acoustical techniques. Soil Sci. Soc. Am. J. 54: 658 - 672. Shetron, S.G., J. A. Sturos, E. Padley, and C. Trettin. 1988. Forest soil compaction: effect of multiple passes and loadings on wheel track surface soil bulk density. North. J. Appl. For. 5: 120-123. Sills, I. D., A. A. G. Aylmore, and J. P. Quirk. 1973. An analysis of pore size in illite-kaolinite mixtures. J. Soil Sci. 24: 480 -490. Smith, R. B. and E. F. Wass. 1994. Impacts of a stump uprooting operation on properties of a calcareous loamy soil and on planted seedling performance. Inf. Rep. BC-X-344, Canadian Forest Service, Pacific Forestry Centre, Victoria, British Columbia. 19 pp. 91 Sokal, R. R. and F.J. Rohlf. 1973. Introduction to Biostatistics. W.H. Freeman and Company, San Francisco. 368 pp. SPSS. 1994. SPSS 6.1 for Windows Student Version. Copyright 1994 by SPSS Inc., Chicago, USA. Stoops, G. 1973. Optical and electron microscopy. A comparison of their principles and their use in micropedolgy. In G.K. Rutherford (ed.), Soil microscopy, Proc. 4th Working-Meeting on Soil Micromorphology, Kingston, Ont., August 27th to 30th, 1973. pp. 101-118. Terry, T. A., D.K. Cassel, and A.G. Wollum. 1981. Effects of soil sample size and included root and wood on bulk density in forested soils. Soil Sci. Soc. Am. J. 45: 135 -138. Theodorou, C , J.N. Cameron, and G.D. Bowen. 1991. Growth of roots of different Pinus radiata genotypes in soil at different strength aeration. Aust. For. 54(1): 52-59. Thompson, M. L., J. F. McBride and R. Horton. 1985. Effects of drying treatments on porosity of soil materials. Soil Sci. Soc. Am. J. 49: 1360-1364. Thompson, S. R. and P.M. Osberg. 1992. Soil disturbance after logging in British Columbia: 1991 results. British Columbia Min. For., Victoria, BC. 24 pp. Thornley, D. 1990. Pre-harvest silvicultural prescription. Lower Colepitts. Ministry of Forests Small Business Forest Enterprise Program. Ministry of Forests, Nelson Region. 9 pp. To, J. H. 1998. Determining the use of simple hand tools for the purpose of assessing soil trafficability. B.S.F. thesis (in progress), Faculty of Forestry, University of British Columbia, Vancouver, BC. Trouse, A. C. and L. D. Baver. 1962. The effect of soil compaction on root development, pp. 258 - 263. In Transactions of a joint meeting of Comm. IV and V; Int. Soc. Soil Sci. Int. Soil Conf. Bur. P. B. Lower Hurt, New Zealand. USDA (United States Dept. of Agriculture). 1996. USDA Forest Service. Forest Service Manual. Portland, Oregon. Title 2500 - Watershed Management. R-6 Supplement No. 2500-96-2. Chapter 2520 - Watershed Protection and Management. 5 pp. Utzig, G.F., and M.E. Walmsley. 1988. Evaluation of soil degradation as a factor affecting forest productivity in British Columbia. FRDA Rep. 025. Canadian Forestry Service. Victoria, BC. I l l pp. 92 Utzig, G.F. and S. R. Thompson. 1992. Soil compaction from random skidding in southeastern British Columbia. Project D20, Forest Sciences Section, Nelson Forest Region, BC Min. For., Nelson, BC. 3 pp. Van Damme, L.W., L. Buse, and S. Warrington. 1992. Microsite soil compaction enhances establishment of direct-seeded jack pine in northwestern Ontario. North. J. Appl. For. 9(3): 107-112. Walpole, R. E. 1972. Introduction to Statistics (Eighth printing). MacMillan Co., Toronto, Ontario. 365 pp. Wasterlund, I. 1985. Compaction of till soils and growth tests with Norway spruce and Scots pine. For. Ecol. Manage. 11: 171 - 189. Wert, S., and B. R. Thomas. 1981. Effects of skid roads on diameter, height, and volume growth in Douglas-fir. Soil Sci. Soc. Am. J. 45: 629 - 632. Youngberg, C. T. 1959. The influence of soil conditions, following tractor logging, on the growth of planted Douglas-fir seedlings. Soil Sci. Soc. Am. Proc. 23: 76 - 78. 93 Appendix 1 Soil disturbance types and survey assessment point classification. Table A. 1 Definition of main soil disturbance types utilized during soil disturbance surveys at Colepitts and Mud Creek. Disturbance code Definition O Disturbance at point is due to old logging or other activities of people. R Bladed skidroad. A continuous bladed structure constructed to facilitate ground-based skidding operations. W, W Wide gouge. An area 1.8 m x 1.8 m square within which 80 % of the area has been gouged to a depth of 5 cm or greater. If wide gouge is not created by MSP the code is underlined. HE, HE, HF, HF Heavy machine traffic. It is possible to include the sample point in a 2 m diameter circle within which 75 % of the area show that heavy machine traffic has created coarse, platy soil structure and / or dense or puddled soil relative to the undisturbed soil on the site, and one or more of the following conditions: 1) forest floor and fine slash crushed and / or incorporated into the surface mineral soil; 2) woody debris crushed down into the running surface of the skid trail; or, 3) extensively scalped organic layers resulting in exposed mineral soil. The E code indicates exposed mineral soil; F intact forest floor. If the soil at the assessment point was considered compact by hand testing the code is underlined. Hand testing involved physical manipulation of a sample of soil to determine if structure appeared to have been altered in any way, from the undisturbed. If there was any doubt in the surveyor's mind about whether or not a structural change had occurred, it was assumed one had and the sample point would be considered compacted. 94 Table A. 1 (continued) TE, TE, TF, TF Ruts greater than 5 cm deep into mineral soil and greater than 30 cm wide and 2.0 m long. The E code indicates exposed mineral soil; F intact forest floor. If the soil at the assessment point was considered compact by hand testing the code is underlined. Hand testing involved physical manipulation of a sample of soil to determine if structure appeared to have been altered in any way, from the undisturbed. If there was any doubt in the surveyor's mind about whether or not a structural change had occurred, it was assumed one had and the sample point would be considered compacted. IE, IE, IF, IF Impressions greater than 5 cm deep into mineral soil. The E code indicates exposed mineral soil; F intact forest floor. If the soil at the assessment point was considered compact by hand testing the code is underlined. Hand testing involved physical manipulation of a sample of soil to determine if structure appeared to have been altered in any way, from the undisturbed. If there was any doubt in the surveyor's mind about whether or not a structural change had occurred, it was assumed one had and the sample point would be considered compacted. ME, ME, MF, Alternate machine traffic. It is possible to include the sample point in MF a 80 cm x 2.0 m rectangle within which 100 % of the area show that heavy machine traffic has created coarse, platy soil structure and / or dense or puddled soil relative to the undisturbed soil on the site, and one or more of the following conditions: 1) forest floor and fine slash crushed and / or incorporated into the surface mineral soil; 2) woody debris crushed down into the running surface of the skid trail; or, 3) extensively scalped organic layers resulting in exposed mineral soil. The E code indicates exposed mineral soil; F intact forest floor. If the soil at the assessment point was considered compact by hand testing the code is underlined. Hand testing involved physical manipulation of a sample of soil to determine if structure appeared to have been altered in any way, from the undisturbed. If there was any doubt in the surveyor's mind about whether or not a structural change had occurred, it was assumed one had and the sample point would be considered compacted. 95 Table A. 1 (continued) LE, LE, LF, LF Light machine traffic. Machine traffic disturbance not captured by any other disturbance type. The E code indicates exposed mineral soil; F intact forest floor. If the soil at the assessment point was considered compact by hand testing the code is underlined. Hand testing involved physical manipulation of a sample of soil to determine if structure appeared to have been altered in any way, from the undisturbed. If there was any doubt in the surveyor's mind about whether or not a structural change had occurred, it was assumed one had and the sample point would be considered compacted. NOTE: in order to ensure the 'LP disturbance type was free of any machine traffic disturbance, points which were suspected of possibly having experienced machine traffic were added to the light machine traffic disturbance type. This may have resulted in some points, which were actually undisturbed - but could not be determined so with certainty - being added to the light machine traffic disturbance type. OT Natural feature scalp or gouge. P Pile. Piles of debris not associated with landings which are larger than 3.0 m x 3.0 m x 0.5 m deep. These have less than 33 % mineral soil. B l , B2, B3 Slash. Collections of slash not associated with landings which are larger than 3.0 m x 3.0 m in area. These have no mineral soil incorporated. If 15 - 30 cm deep add ' 1' to code, if 31 - 80 cm deep add '2' to code, if deeper than 80 cm add '3' to code. XI, X2, X3 Mixtures. Mixtures of soil and woody material which would generally not be considered good planting medium. Mineral soil content is greater than 33 %. If 5 - 15 cm deep add ' 1' to code, if 16 -30 cm deep add '2' to code, if deeper than 30 cm add '3' to code. DI, D2, D3 Mineral deposits. Very little slash and woody debris incorporated, would generally be considered acceptable planting medium. If 5 - 15 cm deep add T to code, if 16 - 30 cm deep add '2' to code, if deeper than 30 cm add '3' to code. Gl Deep gouge. Assessment point is in a gouge that is greater than 30 cm into mineral soil. G2 Class 2 deep gouge. Assessment point is in a gouge that is greater than 20 cm into mineral soil and is within 50 cm of a gouge greater than 30 cm into mineral soil. 96 Table A. 1 (continued) LG Long gouge. A gouge at least 1.0 m x 3.0 m x 5 cm deep. C Continuous gouge. A gouge greater than 5 cm deep along a continuous length of 5 m. V Very wide scalp. Forest floor removed at the assessment point and over 80 % of a 3.0 m x 3.0 m area that includes the point. S Wide scalp. Forest floor removed at the assessment point and over 80 % of a 1.8 m x 1.8 m area that includes the point. OS Other scalp. Scalps and gouges not captured in any other scalp or gouge category. U Undisturbed. On the following pages is the hierarchical key that was used to determine the type of soil disturbance at each survey assessment point. 97 Hierarchical key used to classify soil disturbance at the assessment point. At each assessment point the surveyor would move through this key until a specific disturbance type was indicated. A hierarchical key is required to ensure consistent classification of soil disturbance as some assessment points may have multiple disturbance types present. Additional information regarding the definitions attached to each of these disturbance types is listed above. 1. Any disturbance, or a deposit greater than or equal to 5 cm? N - undisturbed Y - 2 2. Any suggestion of machine traffic? N-3 Y - 17 3. Scalped or gouged? N - 4 Y - 7 4. Pile? N-5 Y - "p" 5. Is it dominantly mineral material? N - 6 Y - "D" (measure depth: Dl= 5-15cm; D2 = 15-30cm; D3 = 30cm+) 6. "X" (measure depth: XI, X2, X3 as for "D" above) 7. Natural feature? N - 8 Y - "OT" 8. Gouge greater than 30cm into mineral soil? N - 9 Y-"G1" 9. Gouge greater than 20cm into mineral soil and within 50cm of a 30cm or deeper gouge? N - 10 Y - "G2" 10. Is the gouge greater than 5cm deep into the mineral soil and over 80% of a 1.8m X 1.8m square that includes the survey point? N - 12 Y - 11 11. Was the gouge a result of MSP? N - "W" Y - "W" 12. Is the gouge 5cm or more into mineral soil at its maximum depth and greater than 1.0m X 2.5m in dimension? N - 13 Y - "LG" 98 13. Is the gouge greater than 3.5m in length? N - 14 Y - "C" 14. Is the forest floor removed at the point and over 80% of a 3.0m X 3.0m square that includes the point? N - 15 Y - "V" 15. Is the forest floor removed at the point and over 80% of a 1.8m X 1.8m square that includes the point? N - 16 Y - "S" 16. "OS" 17. Is it a bladed skidroad? N - 18 Y - "R" 18. Is it equivalent to bladed? (i.e. is it a wide gouge?) N - 19 Y - 11 19. Is it heavy machine traffic? Does 75% of the area within a 2m diameter circle (oriented in any direction but including the point) and the point itself show evidence of one (or more) of the following: 1) forest floor and fine slash crushed or incorporated into the surface mineral soil, 2) woody debris crushed down into the running surface of the skidtrail, or 3) extensively exposed mineral soil? N-20 Y - primary code is "H". Go to 24. 20. Is the point within a rut that is > 5cm deep into the mineral soil and greater than 0.3m wide and 2.0m long? N-21 Y - primary code is "T". Go to 24. 21. Is the point within an impression that is greater than 5cm deep into mineral soil? N-22 Y - primary code is "I". Go to 24. 22. Are any of the features 1), 2), or 3) described in 19. above present over 100% of a 0.8m x 2.0m area? N-23 Y - primary code is "M". Go to 24. 23. the primary code is "L". Go to 24. 24. Is the forest floor intact at the point? N - add "E" to the primary code. Go to 25 Y - add "F" to the primary code. Go to 25. 99 25. Is the point potentially compacted? (As evidenced by hand testing a small soil sample for coarse platy structure and/or increased density relative to the undisturbed. Hand testing involved physical manipulation of a sample of soil to determine if structure appeared to have been altered in any way, from the undisturbed. If there was any doubt in the surveyor's mind about whether or not a structural change had occurred, it was assumed one had and the sample point would be considered compacted.) N - leave code as is. Y - circle code. 100 Appendix 2 Soil water potential and equivalent pore sizes The basis for establishing an equivalent diameter for soil pores, is based on soil water potential, specifically the matric potential, and the phenomenon of capillarity. The matric potential is related to both capillarity and adhesion phenomena. Due to the dynamic equilibrium that exists between the two factors their individual effects cannot be separated. The attraction of water to the surface of soil particles results in a lowering of the free energy of the adsorbed water. The attraction of water molecules to each other is manifest in the surface tension of water. These attractive forces together result in the capillary effect. The capillary effect is illustrated by the capillary rise equation: H - 2 x cos (j) / r p w g where, H is the height of rise, at equilibrium, of a column of water in a tube of radius r, subject to the force of gravity, g, and with a density of water of p w , a surface tension of water against air of x, and a contact angle between the water and the tube of (j). For a temperature of 20°C the density of water can be taken as 0.9982 Mg m"3, and the surface tension as 0.07275 N m"1. The contact angle, (j), can be taken as 0° for most soils (an exception being soils exhibiting signs of hydrophobicity), so the term, cos §, is equal to one. The equation, now solved for r, shows that the radius of a pore within which water will rise to a height of one meter at equilibrium, is approximately 14.9 um. A i m head of water at 20°C exerts a pressure of 9.8 kPa. At a soil water potential of -9.8 kPa only pores less than = 30 urn equivalent diameter will remain water filled. Table A2.1 Relationship between cited gauge pressure and pore diameter. Pressure (kPa) Equivalent pore diameter (urn) 5.9 49.5 9.8 29.7 33 8.8 1500 0.2 101 Appendix 3 Complete soil disturbance survey results for the study areas (as determined using hierarchical key in Appendix 1). Colepitts Mud< Creek Disturbance type number of points percentage number of points percentage o 0 0 12 0.5 R 0 0 0 0 W 0 0 0 0 w 2 0.1 4 0.2 HE 43 2.5 39 1.6 HF 169 9.8 284 11.6 HE 0 0 2 0.1 HF 19 1.1 362 14.8 TE 1 0.06 0 0 TF 0 0 0 0 TE 0 0 0 0 TF 0 0 0 0 IE 2 0.1 0 0 IF 1 0.06 0 0 IE 0 0 0 0 IF 0 0 1 0.04 ME 2 0.1 3 0.1 MF 4 0.2 32 1.3 ME 0 0 0 0 MF 0 0 48 2 LE 112 6.5 13 0.5 LF 1024 59.6 525 21.4 LE 1 0.06 2 0.1 LF 25 1.5 351 14.3 OT 9 0.5 23 1 P 9 0.5 5 0.2 XI 0 0 36 1.5 X2 0 0 0 0 X3 0 0 0 0 Dl 2 0.1 21 0.9 D2 4 0.2 12 0.5 D3 7 0.4 6 0.2 Gl 163 9.5 52 2.1 G2 22 1.3 33 1.3 LG 0 0 0 0 C 0 0 0 0 V 0 0 0 0 S 1 0.06 0 0 OS 19 1.1 31 1.3 u 39 2.3 555 22.6 B3 2 0.1 0 0 B2 26 1.5 0 0 Bl 9 0.5 0 0 102 Appendix 4 Critical threshold tests 1) Test to determine if macroporosity (volume of pores > 30 um equivalent pore diameter) in the heavy compact machine traffic disturbance type at the Colepitts site, at the 2 - 4 cm sample depth, was less than 15 % (Walpole 1972). Ho: IO-HF = 0.15 Ha: U - H F < 0.15 a = 0.05 Critical region: tcriticai < - ta , where t has degrees of freedom with n^F - 1. Test statistic: t c rj ti c ai = ^ H F - 0.15 / (s / n 0 5) = 1.3657 where, s is the standard deviation, and n the number of samples, for the compact heavy machine traffic disturbance type. 2) Test to determine if macroporosity (volume of pores > 30 um equivalent pore diameter) in the heavy compact machine traffic disturbance type at the Colepitts site, at the 2 - 4 cm sample depth, is less than 50 % of the undisturbed macroporosity (Walpole 1972). A transformation of the sample data for the undisturbed was done prior to running the test: each individual sample value was multiplied by 0.5 (Peter Ott, personal communication). Ho: 14-HF - Hu(trans) = 0 Ha: U - H F - Uu(trans) < 0 a = 0.05 Critical region: tcriticai < - ta , where t has degrees of freedom v. 2 2 2 2 2 2 2 V = ( S Hf In H F + S U(trans) /n U(trans)) /{ [(s H F / N _ H F ) /(n H F " 1 )]+[(s U(trans) /n U(trans)) /(n U(trans) " 1)]} v^20 test Statistic: t' = [ ( U H F - HUftrans)) - 0]/[( S H£ 2 /n H F ) + ( S U(trans)2/n U(trans))]° 5 = 0.901 103 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0099298/manifest

Comment

Related Items