UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Effect of alternative site preparation treatments on soil chemistry, physical properties, climate and… Yole, D. 1996

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

Item Metadata

Download

Media
831-ubc_1996-0102.pdf [ 6.41MB ]
Metadata
JSON: 831-1.0086975.json
JSON-LD: 831-1.0086975-ld.json
RDF/XML (Pretty): 831-1.0086975-rdf.xml
RDF/JSON: 831-1.0086975-rdf.json
Turtle: 831-1.0086975-turtle.txt
N-Triples: 831-1.0086975-rdf-ntriples.txt
Original Record: 831-1.0086975-source.json
Full Text
831-1.0086975-fulltext.txt
Citation
831-1.0086975.ris

Full Text

E F F E C T OF ALTERNATIVE SITE PREPARATION TREATMENTS ON SOIL CHEMISTRY, PHYSICAL PROPERTIES, CLIMATE AND SEEDLING GROWTH ON A MESIC SITE IN THE NORTHERN INTERIOR OF BRITISH COLUMBIA by DAVID WALTER DOUGLAS YOLE B.Sc.(Ag.), The University of British Columbia, 1980 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF T H E REQUIREMENTS FOR T H E 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 Jan. 1996 © David Walter Douglas Yole, 1996 In presenting this thesis in partial fulfilment 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 purposes 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 The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT The influence of four commonly used site preparation methods (disc trenching, pile-and-burn, broadcast burn, no treatment) on soil microclimate, soil nutrients, bulk density, and early survival and growth of hybrid interior spruce (Picea glauca [Moench] Voss x engelmannii Parry ex. Engelm.) and lodgepole pine (Pinus contorta Dougl. ex. Loud.) seedlings were examined in plots established in the Engelmann Spruce - Sub alpine Fir (ESSF) zone in the northwest interior of British Columbia. Soil moisture was monitored in 13 potentially plantable microsite types at weekly intervals over the first two growing seasons after site preparation. Although growing season soil moisture in 1994 was significantly effected by the microsite types under study, soil moisture was not believed limiting to seedling growth in any microsite studied and soils did not reach slightly to moderately dry conditions (> 2 bars tension) until late July, and only in three microsites, disc trench berm, disc trench hinge, and pile-and-burn-scalp (mineral soil exposed). Soil and air temperature were strongly affected by the four site preparation treatments and nine microsites studied, particularly in early spring periods between mid-May and mid-June. Soils (10-cm depth) warmed sooner, and for longer duration, in treatments which promoted soil drainage. Treatments which removed or incorporated insulating forest floor layers resulted in greater cumulative hours of soil temperature greater than 8°C in early spring periods, giving a distinct growth advantage over planting sites having intact forest floors. Untreated soils (slash left) were consistently colder for longer periods during the growing season. Broadcast burning decreased the frequency of frost events 20 cm above the ground surface during the 1994 growing season by 43 and 121%, relative to pile-and-burn and no treatment plots, respectively. i i Site preparation had strong effects on coarse fragment free bulk density of near surface (2-7 cm) and rooting zone (0-20 cm) layers one season after treatment. The coarse fragment free bulk density of the disc trench and hinge microsites 0-20 cm layer were 50 and 24% less, respectively, as compared to pre-treatment density. The disc trench berm and hinge microsites had significantly (p<0.05) smaller 0-20 bulk density one season after treatment relative to other microsites studied. Both the disc trench trench-bottom and pile-and-burn under-excavator-track microsites had greatest near surface (2-7 cm) soil density. Several soil nutrients in decayed wood, forest floor and 0-20 cm mineral layers were significantly affected by site preparation treatments. With the exception of total S, exchangeable K, mineralizable N in most treatments and total N in the pile-and-burn burned-pile microsite, soil nutrients and p H in decayed wood materials increased from pre-treatment to one season post-treatment. Nutrients of forest floor materials revealed greater microsite effects than decayed wood materials and may reflect greater biological activity in this uppermost layer. Nine and 12 months after pile and broadcast burning, respectively, there were significant increases in forest floor pH and exchangable and total Ca and Mg relative to untreated soils. Site preparation treatments which promote mixing (i.e. at the berrh and hinge) resulted in significant increases in many soil nutrients in the 0-20 cm layer, most notably total C and N . Greatest one-season losses of total N , S, and C from all three soil layers studied occurred in the pile-and-burn burned-pile microsite. Superior second-year root collar diameter and height increment growth of both planted species occurred on the broadcast burn and disc trench hinge microsites. Poorest growth of both species occurred on the pile-and-burn and untreated plots and was believed at least partially associated with higher soil moisture content and slow soil warming in early spring periods in this moist, cold biogeoclimatic subzone (ESSFmc). i i i T A B L E O F C O N T E N T S Page Abstract i i Table of Contents iv List of Tables v i List of Figures i x List of Appendices x A C K N O W L E D G E M E N T S xi 1 B A C K G R O U N D 1 2 M E T H O D S 6 2.1 Description of Study Site 6 2.1.1 Location of study site 6 2.1.2 Site selection 6 2.1.3 Site history and harvesting of study site 6 2.1.4 Site and soil description 7 2.1.5 Site sensitivity 10 2.2 Experimental Design and Layout 11 2.3 Plot Establishment 12 2.4 Site Preparation Treatments 12 2.4.1 Characterization of microsites created by site preparation 13 2.4.1.1 Description of disc trench microsites 15 2.4.1.2 Description of pile-and-burn microsites 18 2.5 Detailed Sampling 19 2.5.1 Woody fuel and forest floor depth assessments 19 2.5.2 Fire weather information 20 2.5.3 Ecosystem descriptions 20 2.5.4 Bulk density sampling , 21 2.5.4.1 0-20 cm layer 21 2.5.4.2 Near surface 2-7 cm layer 21 2.5.5 Soil nutrient sampling.. 22 2.5.5.1 Sampling before site preparation 22 2.5.5.2 Sampling after site preparation 24 2.5.6 Soil climate measures 24 2.5.6.1 First year after logging 24 2.5.6.2 First and second year after site preparation 24 2.5.6.2.1 Soil moisture 24 iv Page 2.5.6.2.2 Soil and air temperature 25 2.5.7 Planting 26 3 R E S U L T S A N D DISCUSSION 28 3.1 Woody Fuel and Prescribed Fire Assessments 28 3.2 Soil Physical Properties 31 3.3 Soil Nutrient Changes Following Site Preparation 38 3.3.1 Nutrients in decayed wood 42 3.3.2 Nutrients in forest floor materials 50 3.3.3 Nutrients in 0-20 cm mineral layer 58 3.3.4 Total soil nutrients 66 3.4 Changes in Soil Moisture 70 3.5 Soil Temperature 80 3.5.1 Air temperature 82 3.5.2 Soil temperature 84 3.6 Tree Seedling Response 90 3.6.1 Seedling condition over the first two years 90 3.6.2 Height and root collar diameter growth 93 4 S U M M A R Y A N D M A N A G E M E N T IMPLICATIONS 98 5 L I T E R A T U R E CITED 102 v LIST O F T A B L E S Page Table 1. Fire weather indices (FWI) during the day of the broadcast burn 14 Table 2. Summary of sampling done in mechanical treatments and associated microsite types 15 Table 3. Some characteristics of disc trench microsites 16 Table 4. Average fuel consumption in windrow piles 19 Table 5. Characteristics of planting stock 27 Table 6. Average fuel loading and depth of burn for broadcast burn and other treatments 29 Table 7. A N O V A effects on pre-treatment fuel loading 30 Table 8. Bulk density statistics of 2-7 cm layer one season after site preparation treatments 32 Table 9. A N O V A for treatment effects on 0-20 cm mineral bulk density before site preparation 34 Table 10. Average bulk density of 0-20 cm soil layer by treatment prior to site preparation 36 Table 11. Average bulk density of 0-20 cm soil layer after site preparation 36 Table 12. A N O V A for effects of microsite types on 0-20 cm mineral bulk density one season after site preparation 37 Table 13. A N O V A for pre-treatment differences in forest floor, decayed wood and mineral nutrient variables 41 Table 14. A N O V A for effects of treatments and microsite types within treatments on nutrient variables one season after site preparation 41 Table 15. Probability levels for decayed wood nutrient concentrations and contents for treatment effect before site preparation 44 Table 16. Mean decayed wood nutrient concentrations before site preparation 45 Table 17. Mean decayed wood nutrient content (kg/ha) prior to site preparation 46 vi Page Table 18. Probability levels for decayed wood nutrient concentrations and contents for treatment and microsite effect one season after site preparation 47 Table 19. Mean decayed wood nutrient concentrations for treatments and microsites one season after site preparation 48 Table 20. Mean nutrient content (kg/ha) of decayed wood materials one season after site preparation 49 Table 21. Probability levels for forest floor nutrient concentrations and contents for treatment effect before site preparation 50 Table 22. Mean forest floor nutrient contents (kg/ha) prior to site preparation 52 Table 23. Probability levels for forest floor nutrient concentrations and contents for treatment and microsite effect one season after site preparation 53 Table 24. Mean forest floor nutrient contents (kg/ha) one season after site preparation 54 Table 25. Mean forest floor nutrient concentrations for treatments and microsites one season after site preparation 55 Table 26. Probability levels for mineral soil (0-20 cm) nutrient concentrations and contents for treatment effect prior to site preparation 58 Table 27. Mean nutrient concentrations in 0-20 cm mineral layer prior to site preparation 60 Table 28. Mean nutrient contents (kg/ha) in 0-20 cm mineral layer prior to site preparation 60 Table 29. Probability levels for mineral soil (0-20 cm) nutrient concentrations and contents for treatment and microsite effect one season after site preparation 61 Table 30. Mean nutrient concentrations in 0-20 cm mineral layer one season after preparation 62 Table 31. Mean nutrient contents in 0-20 cm mineral layer of various treatments and microsite types one season after site preparation 63 vii Page Table 32. Cumulative totals of soil nutrients by layer for each measurement period 68 Table 33. Comparison of monthly precipitation totals over three years at the Regan Creek study site 74 Table 34. A N O V A used for determining effect of microsite on soil moisture over a 14-week period in the 1994 growing season 75 Table 35. Average soil moisture tension between June 23 - Sept. 22, 1994 as a proportion of total observations in each of three soil moisture classes 77 Table 36. Seedling condition and damage codes for pine after one growing season and the following spring after one winter 91 Table 37. Seedling condition and damage codes for spruce after one growing season and the following spring after one winter 94 Table 38. A N O V A for treatment effects on pine seedling growth over the first two years 94 Table 39 A N O V A for treatment effects on spruce seedling growth over the first two years : 94 Table 40. Average height increment (cm/yr) of seedlings in the second year after site preparation 96 Table 41. Average root collar diameter by treatment for pine and spruce after two growing seasons 97 vii i LIST O F F I G U R E S Page Figure 1. General location of study area 7 Figure 2. Detailed map of study site, ecosystem and treatment units 8 Figure 3. Cross sectional diagram of disc trench treatment showing main microsite types 17 Figure 4. Cross sectional diagram of pile-and-burn treatment showing main microsite types 18 Figure 5. Total woody fuels estimated from fuel transects prior to site preparation in each treatment unit 30 Figure 6. Comparison of near surface 2-7 cm bulk density in several microsites and treatments one season after site preparation 33 Figure 7. Depth of decayed wood (DW), forest floor (FF) and total forest floor (FF Tot) for each of the treatment units of the study 43 Figure 8. Change in forest floor pH before and one season after treatment 56 Figure 9. Mineral soil p H (water) one season after site preparation treatments 64 Figure 10. Change in total C content 0^g/ha) in 0-20 cm soil layer before- and one season after site preparation treatments 65 Figure 11. Change in total N content (kg/ha) in 0-20 cm soil layer before- and one season after site preparation treatments 65 Figure 12. Change in whole soil total N (kg/ha) one season after site preparation treatments 69 Figure 13. Change in whole soil exchangeable K (kg/ha) one season after site preparation 70 Figure 14. Change in soil tension for four selected treatments/microsites over the 1993 growing season 73 Figure 15. Change in soil tension for four selected treatments/microsites over the 1994 growing season for well drained treatment units 75 Figure 16. Total number of weekly observations (all replicates combined) at several levels of soil tension between disc trench and other microsites from June 23 - Sept. 22, 1994 78 ix Page Figure 17. Total number of weekly observations (all replicates combined) at several levels of soil tension between pile-and-burn and other microsites from June 23 - Sept. 22, 1994 79 Figure 18. Minimum air (20 cm and 1 m) and soil (10 cm) temperatures associated with a frost event occurring soon after planting 83 Figure 19. Minimum air temperatures during the 1994 growing season 85 Figure 20. Daily average soil (10-cm depth) minimum and maximum temperature over the 1994 growing season 87 Figure 21. Hourly average soil temperatures at the 10-cm depth for a 24-hour period during a warm day in July, 1994 88 Figure 22. Cumulative daily hours greater than 8 degrees at the 10-cm soil depth during a 2-week spring period (May 17-31) in 1994 89 Figure 23. Percentage of seedlings in good condition averaged for each of four treatments and three measurement periods during the first two growing seasons 92 Figure 24. Percentage of seedlings having snaky stem form after the first winter for each treatment 93 Figure 25. Mean height increment of pine over the first two growing seasons 97 Figure 26. Mean height increment of spruce over the first two growing seasons 98 LIST O F A P P E N D I C E S Page Appendix 1. Habitat features of treatment units at Regan Creek 108 Appendix 2. Analytical procedures used for Regan Creek soil samples (Ministry of Forests, Forest Sciences Lab, Victoria, B.C.) 110 Appendix 3. Mean forest floor nutrient concentrations after harvest and prior to site preparation I l l x A C K N O W L E D G E M E N T S Many people helped in various stages of this thesis for which I am indebted. Dr. T. Ballard provided helpful supervision and advise throughout the project. Dr. V . Lemay assisted greatly with statistical analysis of data. Dr. A. Black was helpful in providing advise in several areas of soil climate research. A. Macadam was involved with all aspects of this study from initial project formulation and planning, field assistance, and offered much needed advise in mini-crisis situations which appear regularly in site preparation research. D. Coates helped by reading an initial draft of this thesis and provided many useful suggestions. I am truly grateful to several workers who assisted during summer months including R. Trowbridge, B . Colquhoun, M . Duerst, S. Munroe, and particularly, M . Lavigne for field and lab help over the entire length of the project. R. Barbuto helped me cope with many office-related details. The Ministry of Forests Glyn Road Laboratory, Victoria, B.C. carried out the soil chemical analysis. Pacific Inland Resources Ltd. provided operational support during piling treatments and were cooperative in all phases of the study. Houston Forest Products cooperated with providing the disc trenching equipment. The Ministry of Forests, Bulkley Forest District, provided seedlings and financial support for disc trenching and broadcast burning treatments. Funding was provided by the British Columbia Ministry of Forests Integrated Resource Management Branch under the Canada-British Columbia Forest Resource Development Agreement II (1991-1995), Victoria, B.C. xi EFFECT OF ALTERNATIVE SITE PREPARATION TREATMENTS ON SOIL CHEMISTRY, PHYSICAL PROPERTIES, CLIMATE AND SEEDLING GROWTH ON A MESIC SITE IN THE NORTHERN INTERIOR OF BRITISH COLUMBIA 1 B A C K G R O U N D There are several site preparation methods used following harvesting of forested sites in the northern interior of British Columbia (B.C.). These methods most commonly include broadcast burning, disc trenching, mounding, windrowing, and other piling methods with or without burning of slash piles. No site preparation (i.e. direct planting) is also practiced in certain situations where logging slash is limited and forest floor material is not excessively deep. It is expected that planting without site preparation is likely to be less common in the future as logging increases in higher elevation forests (e.g. Engelmann Spruce - Subalpine Fir (ESSF) zone (Banner et al. 1993) where deep forest floors and slash accumulation after harvesting often prohibit direct planting. High-elevation interior forests typically contain mature (i.e. 200 to 300-year-old) subalpine fir (Abies lasiocarpa [Hook.] Nutt.) and hybrid interior spruce (Piceaglauca [Moench] x engelmannii Parry ex. Engelm.) trees, which, following harvesting by traditional logging methods (feller-buncher/skidder or hand fall/skidder), often leave accumulation of considerable fresh slash and dead wood material on the soil surface. This slash/decayed wood material often requires treatment prior to planting. Manipulation of the slash/duff during site preparation will likely impact the soil resource. Increasing public concern about environmental issues, including the potential health risks associated with smoke pollution, have resulted in a gradual decline in the use of broadcast burning for site preparation in many regions of British Columbia. Public pressure on the Ministry of Forests to limit smoke pollution is of particular concern near population centres already identified as being sensitive to air pollution 1 (e.g. Bulkley Valley near Smithers). Smoke management guidelines, introduced fairly recently in the Province by the Protection Branch of the Ministry of Forests have generally resulted in narrowing the burning "window" to a very short period of time in the year when weather conditions are appropriate for smoke dispersal. During the last 5 years the use of partial cutting methods of harvesting have increased, which, in turn, leads to reduced site preparation by broadcast burning. A survey of local logging companies and Ministry of Forests District offices by the author in interior portions of the Prince Rupert Forest Region (PRFR) found that broadcast burning has declined as a site preparation method. Between 1988 and 1991, the total area broadcast burned in the PRFR has declined from 12000 to 4000 ha/yr. During the same three year period, the total area mechanically treated has ranged between 6000 and 10000 ha/yr and total treated area in 1990-91 was greater for mechanical than burning methods (J. Dunbar pers. comm). Despite this trend, broadcast burning still remains an important site preparation method on certain sites and biogeoclimatic zones with up to 30% of the total harvested area per year still planned for broadcast burning by some licensees (M. Dunbar, pers. comm). The advantages of providing good planter access in addition to improvements in early seedling establishment and growth, following ecologically sound broadcast burning, has long been observed by local silviculturalists from the Ministry of Forests and major licensees. As the total area broadcast burned declines, both the Ministry and licensees are broadcast burning sites only when (1) extreme slash conditions exists, (2) sites are away from smoke sensitive areas or, (3) few other site preparation options exist (e.g. very steep terrain with abundant woody fuel). Prior to slash burning being accepted in the Pre-Harvest Silviculture Prescription (PHSP), sites proposed for burning must first meet acceptable standards for site sensitivity (Lewis and Carr 1993) to manage for soil disturbance following prescribed burning to ensure that burning wil l not degrade the 2 long-term productivity of the site. As a result of increased restrictions to burning as a site preparation option, there has been increasing interest in mechanical or other (e.g. chemical) site preparation treatments as alternatives. The pros and cons of this shift away from broadcast burning to mechanical methods in terms of short and long-term changes in site and soil productivity in high elevation forests of the northern interior have yet to be fully studied or understood. Broadcast burning in northern clearcuts can, under some circumstances, result in significant losses of site nitrogen (Macadam 1987; Taylor and Feller 1987) as well as other nutrients such as sulphur (Blackwell 1989). At the same time, slashburning can commonly lead to a short term increase in soil p H and exchangeable cations the first growing season after burning (Macadam 1987, Taylor and Feller 1987, Feller 1982) which may contribute to improved seedling growth response in replanted sites, at least in the short term. Changes in soil moisture and temperature following broadcast burning of mesic sites in the north central interior of B.C. have not been studied and may be important factors influencing growth response following burning. The potential for mechanical site preparation (MSP) treatments to impact positively or negatively on site quality has yet to be fully evaluated particularly when comparing a variety of treatment options at one site. Soil compaction, displacement of forest floor and surface soil materials, and increased surface erosion are potential concerns. The potential for nutrient depletion or improvement in nutrient availability (e.g. burned piles) associated with MSP treatments depends on the nature and extent to which soils are disturbed and changes in soil micro-climate in the newly formed microsites. Piling of slash can occur in many ways using a variety of machinery, but the end result is a large proportion of the woody material piled up and then burned or left to decrease fire hazard or increase planter access. Large losses of soil nutrients can occur from the site if the forest floor is removed from productive growing areas during the 3 piling procedure (Ballard and Hawkes 1989) and/or a high intensity fire occurs during burning of piles (Blackwell 1989). If there is no removal of forest floor materials during the piling procedure, the impact on the site nutrient capital may be less consequential. Some workers have found that soil compaction accompanies the piling method (Lopushinsky et al. 1992; Froehlich 1979), possibly affecting the long-term soil productivity and early seedling performance. Such negative effects are strongly dependent on the type of machinery used and the range of environmental and site factors which exist at the time of treatment. Kranabetter (pers. comm. 1995), looking at long-term soil productivity and skid road compaction studies in soils derived from glacial t i l l in the northern interior of B.C., found that uniform compaction of 2.5 cm with or without the presence of forest floor material can result in a significant decrease in aeration porosity and macropore space, which in turn can have negative consequences toward soil drainage, aeration, biology of the soil organisms, and ultimately, tree productivity. Disc trenching is also becoming a popular site preparation method on mesic sites of high elevation areas of the central interior (M. Dunbar, pers. comm.). The effects of this treatment on soil climate and nutrient status has not been studied in detail except in Sweden (Orlander et al. 1990). The hinge position (mid slope position of the material flipped over during the discing procedure) is traditionally planted in the PRFR as this microsite is slightly raised, and is believed to contain greater amounts of available nutrients, and have warmer soil temperatures (Orlander et al. 1990). Such factors have not been well studied for disc trenching in northern British Columbia and are another focus of this thesis. The no treatment option, where seedlings are directly planted on the harvested site, is still a popular method in the northern interior and accounts for approximately 25-30% of the harvested area. The biological consequences of this treatment option as compared to other treatments available are not well understood. 4 No studies have compared the potentially growth limiting factors of soil micro-climate, physical properties, and nutrient status associated with planting microsites from the most commonly used site preparation treatments currently used in the northern interior of B.C. By application of four site preparation treatment options at one site, this study compared the growth performance of lodgepole pine (Pinus contorta Dougl. ex. Loud.) and hybrid interior spruce (Piceaglauca x engelmannii) and examined some of the environmental factor(s) which may be contributing to early seedling growth responses found. My thesis addresses some of the consequences of the shift from broadcast burning to mechanical or no-treatment methods of site preparation. It examines several soil factors believed to affect growth of planted conifers after site preparation. Specifically, the thesis compares the effects of disc trenching, pile-and-burn, broadcast burning, and no site preparation treatments, as well as several microsites within each of the treatments, at one site in the ESSFmc biogeoclimatic subzone (Banner et al. 1993), with respect to: * soil physical properties (primarily soil density) * soil chemical properties * growing season soil climate * survival and initial growth of lodgepole pine and hybrid spruce seedlings 5 2 METHODS 2.1 Description of Study Site 2.1.1 Location of study site The study site is located approximately 54 km east of Smithers (Figure 1) in the Bulkley Forest District of the Prince Rupert Forest Region (PRFR) in northwest interior of British Columbia (B.C.) (54° 49'N, 126° 38'W). Figure 2 details the specific location of the study site (subsequently referred to as Regan Creek), located at 9 km on the "5200" Road accessed 37 km on the Babine Lake Forest Service Road. Major ecosystem units identified in the Pre-Harvest Silviculture Prescription (PHSP) as occurring on the cutblock and the location of experimental units under intensive study are also included in Figure 2. 2.1.2 Site selection Considerable time was spent in the 1991 field season viewing proposed logging blocks in interior Districts of the PRFR in order to find an appropriate site which would accommodate the four treatments. A site was required which was mesic or fresh in terms of soil moisture, and contained topographic (e.g. uniform terrain with slope gradient < 25%) and soil characteristics conducive to implementation of the four site preparation treatments. 2.1.3 Site history and harvesting of study site Harvesting (with a feller-buncher) occurred between November 1991 and early March 1992 under heavy snowpack (generally > 1 m). The site was clearcut logged and trees were moved to landings with rubber-tired skidders. Winter logging and deep snowpack resulted in relatively high stumps (30-70 cm in height) being left on the site. The stand originated following wildfire, and was approximately 155 years of 6 Figure 1. General location of study area. 7 Ii Figure 2. Detailed map of study site, ecosystem and treatment units. 8 age. The cutblock contained above-average log quality and merchantable volume relative to other stands in the area. Stand density was variable, ranging from 600-1800 stems per hectare. The main tree canopy of mesic ecosystems included lodgepole pine, subalpine fir and hybrid spruce. The shrub layer contains predominately subalpine fir regeneration, Vaccinium membranaceum and Menziesia ferruginea. The herb layer, although sparse, contained such species as Cornus canadensis, Clintonia uniflora, Petasites palmatus and Linnaea borealis. High elevation or heavy snowpack indicator species such as Senecio triangularis, Valeriana sitchensis and Barbilophozia spp. were common to moister areas of the site prior to logging. 2.1.4 Site and soil description The study stand, located at approximately 1000 m elevation, is best described as transitional between the SBSmc and ESSFmc biogeoclimatic subzones. For a description of these subzones, see Banner et al. (1993). For convenience of discussion, the study site will be subsequently referred to as being in the ESSFmc (moist, cold) subzone. The ESSF zone climate is characterized by cold winters with deep winter snowpack (1-2 meters) which lasts from November to early May. Frost events are common at any time during the growing season. Summers are warm (15-30°C) and generally moist with most rainfall occurring during June/July and in late fall (Banner et al. 1993). Slopes at Regan Creek are gently rolling to level, ranging from 0-10% and averaging 2%. Parent materials are primarily glacial t i l l having loam to silty clay loam texture in the upper 30 cm of the soil profile. Minor portions of the study area contain water-worked glaciofluvial deposits which contain soils having a greater percentage of sand in the soil profile. Soils are classified as Orthic Dystric Brunisols and Brunisolic Gray Luvisols, 9 with relatively thin (3-8 cm) forest floors classified primarily as Hemimors, with some Mormoders. Every attempt was made to avoid locating treatment units in slightly moister/richer ecosystems containing darker surface soil horizons (higher organic matter content) and/or gleyed horizons at depths < 30 cm. Treatment units (plots) were generally located in well to moderately well drained soils, although some units contain small portions which were slightly moister or imperfectly drained. Coarse fragment content (>2 mm fraction) of mineral soils was extremely variable, ranging from 0-75% containing considerable gravel, cobble and stone content in the upper 30 cm of soil profile. Although moist ecosystems dominate this cutblock overall, the study plots were concentrated on mesic or well drained portions of the cutblock as these ecosystems dominate the landscape of this high elevation subzone. Mesic ecosystems also contain soil conditions and slash loading which can be site prepared by any of the alternatives being considered in this study. Appendix 1 presents detailed environmental features for each of the 12 experimental units under study. 2.1.5 Site sensitivity The Ministry of Forests has developed a set of hazard assessment keys for evaluating site sensitivity to soil-degrading processes (Lewis and Carr, 1993). Site sensitivity is determined following assessments of several factors which are believed to influence puddling and compaction, displacement of surface mineral soil layers, forest floor displacement, surface erosion, and mass wasting. This information is used in the PHSP of a forest development plan and can determine the season/method of harvesting and type of site preparation treatments appropriate for a given site. 10 At Regan Creek, site sensitivity was assessed by the author prior to site preparation. The following soil/site hazard ratings for the study site were determined based on the keys provided in Lewis and Carr (1993): Relative Rating Compaction Hazard: Mineral Soil Displacement Hazard: Forest Floor Displacement Hazard: Surface Erosion Hazard Mass Wasting Hazard: L M L M - H L - M The leading site sensitivity hazard is used to indicate overall site sensitivity and limitations to harvesting/site preparation. The leading hazard for mesic portions of the study site was high (compaction hazard) based mainly on fine soil textures, and this factor was given particular attention both during harvesting (winter on snowpack) and site preparation (i.e. mechanical treatments done when soil moisture was relatively dry). Four site preparation treatments are being studied in this project including broadcast burn (BB), disc trench (DT), pile-and-burn (PB) and no treatment (NT) or control. Three replicate plots were established for each of the treatments. A completely randomized experimental design (CRD) was used with plots (e.g. BB1, BB2, BB3) nested within respective treatments (e.g. BB) and following site preparation, newly-formed microsites nested within plots and treatments. "Standard" CRD methodology (Zar, 1988) involves random assignment of plots throughout the entire experimental area. However, completely random plot allocation could not be followed in this study as a result of the isolated broadcast burn area. For logistical reasons and in order to achieve an "operational" type burn, the broadcast burn treatment occurred in a 12 ha portion of the cutblock west of the 2.2 Experimental Design and Layout 11 entering the cutblock (Figure 2). Although the three broadcast burn plots are somewhat segregated from the other plots, the terrain and soils of these two portions of the study area were visually quite similar prior to harvesting. It was therefore assumed that there were no pre-treatment differences in populations parameters being estimated between the burned and unburned portions of the cutblock and standard CRD methodology was therefore followed. A l l MSP and untreated plots were randomly allocated. Detailed ANOVAs (analysis of variance) vary among the different components (soil moisture, nutrients, physical properties and growth characteristics) and stages (before and after site preparation) in the study and are described in each appropriate Results and Discussion section. A l l statistical analyses were carried out using SAS statistical software, version 6.03 (SAS Institute Inc. 1987). 2.3 Plot Establishment Layout of treatment units involved first delineating areas of the cutblock containing mesic ecosystems and having relatively homogenous site characteristics (i.e. similar slope, vegetation complex, moderately well to well drained soils). Of these mesic areas, twelve - 47.5 x 25 m plots were laid out and then randomly assigned to represent the 12 treatment units. Treatment unit dimension was selected to be sufficient in size to accommodate sampling of all components of the study before and after site preparation. 2.4 Site Preparation Treatments Site preparation treatments were to be implemented in a manner consistent with commonly used operational methods. Prior to MSP, tree stumps were cut where necessary in each of the DT and PB plots to heights <30 cm to facilitate proper operation and limit damage to mechanical equipment. 12 Site preparation treatments occurred during the summer of 1992 when soil moisture conditions were sufficiently dry and equipment was available. Disc trenching (DT) occurred during the first week of August using a TTS-Delta power disc trencher powered by a Cat-528 skidder. A V-rake attachment was mounted on the front of the skidder to help deflect slash into areas where seedlings would not be planted. Owing to relatively fine soil textures and high compaction hazard (Curran and Thompson 1991) at this site, machinery was required which would limit ground pressure and minimize soil compaction while piling slash. To this end piling of slash into long windrows (approximately 1-1.5 m high, 2 meters wide and 50m in length) was accomplished with a wide-track Cat-180 excavator fitted with a brush rake. Piling occurred in early October and burning of windrow piles occurred after the first snowfall in last week of October, 1992. The broadcast burn treatment occurred in late June, 1992 when fine fuel moisture (FFMC) and weather conditions were suitable for prescribed burning. The method of ignition used was hand held drip torch and burning in strips (approximately 50-100 meters in width at one time) into the wind. The objectives of the burn were to consume all of the fine fuels, most of the medium fuels and very little duff (approximately 1.0 cm) since duff depths were already relatively thin (4-8 cm) prior to burning. Table 1 outlines the fire weather indices which occurred a the time of the broadcast burn. The no-treatment option occurred where no machinery or site preparation occurred after harvesting. Seedlings would be directly planted into soils of these plots. 2.4.1 Characterization of microsites created by site preparation Several microsites are formed during mechanical site preparation. One 13 aspect of of the study was to collect sod measurements for nutrients, temperature and moisture status associated with some of the more important (on an area basis) microsites resulting from site preparation. Identifying the extent of each microsite would help to better characterize some of the processes occurring in these microsites and/or the contribution of each microsite type to the treatment effects as a whole. Table 1. Fire weather indices (FWI) during the day of the broadcast burn. FWI Code Predicted 1 Actual 2 F F M C 82-86 91 D M C 25-29 36 DC 151-350 176 IGNITION R A N K 4-6 7-8 SPREAD R A N K 4-6 7-8 CONTROL R A N K 4-5 4 IMPACT R A N K 3-4 3 Predicted - fuel moisture and fire weather codes using the Prescribed Fire Planner (Trowbridge et al. 1987) to achieve desired burning objectives. 2 Actual = fuel moisture and fire weather codes recorded at the automated weather station on June 25, 1992. Owing to time and funding limitations, only selected microsites (i.e. microsites which made up an appreciable portion of the total treated area (approximately >5%) and/or were regular enough throughout treatment units to facilitate replicated measurement through systematic or random sampling were measured for different variables. To determine the areal extent (%) of the different microsites associated with each treatment in general, a series of line transects was run in each of the DT and PB plots. Line transects were run perpendicular to trenches in the DT treatment (three lines systematically located per plot), and perpendicular to windrows in the P B treatment (five lines per plot) to arrive at estimates of the areal extent (%) of microsite types formed by the mechanical treatments. More transect lines were required in the 14 PB treatment due to greater microsite variability. The total length of each microsite type (e.g. DT trench) was summed for each transect line and recorded as a percentage of the total transect length. The microsites selected for the purposes of post-treatment nutrient, moisture and temperature sampling are identified in Table 2. Table 2. Summary of sampling done in mechanical treatments and associated microsite types. Treatment/ Chemistry, Moisture Temperature Microsite Bulk density DT hinge Y Y Y DT berm Y Y Y DT trench Y Y Y DT between-trench-forest floor intact Y Y Y PB burned pile Y Y Y PB between-piles-forest floor intact N Y Y PB under-track-forest floor intact N Y N PB under-track, exposed mineral N Y Y PB track-intermixed material N Y N . PB scrape-intermixed material N Y N PB scrape-exposed mineral N Y N PB between-piles-all microsites Y N N 2.4.1.1 Description of disc trench microsites A n additional five 30-m line transects were established in DT plots to determine some general characteristics of microsites formed during the treatment (e.g. berm height, trench depth etc.). These DT characteristics are summarized in Table 3. Transect lines were oriented perpendicular to the direction of trenching activity and were sampled systematically at equal distances along the longest axes 15 of each DT plot. Figure 3 gives a schematic representation of the DT microsites and % areal extent (in brackets) of each of the microsites. The "between-trench" microsite type describes those microsites where forest floor is still intact under the area where the skidder moved. Based on visual assessments, only minor compaction (< 2 cm impression or "light" in Soil Disturbance Guidelines, Curran and Thompson, 1991) resulted from the skidder traffic within these areas, likely due to relatively dry soil conditions when the treatment occurred. Table 3. Some characteristics of disc trench treatment plots. Width Berm height Total No. Hinge composition Trench Plot between (range) trenches Min cap Intermixed Org cap depth passes per plot (m) (cm) (cm) DTI 2.4 10-50 16 10 75 15 0-30 DT2 2.3 10-40 16 10 70 20 5-35 DT3 2.9 5-40 14 48 50 2 5-30 The berm and hinge microsite types were quite variable in terms of material composition, often containing loose mixtures of mineral soil, coarse fragments, forest floor, and slash material. In general, berms were looser than hinge microsites and were elevated 10-30 cm above the undisturbed forest floor. Where soils were coarse textured and/or had shallow forest floor layers less than 5 cm (e.g. DT3), berm and hinge microsites tend to consist of a higher proportion of mineral soil inverted over the original soil surface. In plots containing thicker duff or more decayed wood (i.e. DT2 and DTI), intermixed mineral/organic or pure organic deposits were more common than mineral deposits. Berm height from original surface soon after 16 treatment ranged from 0 to 50 cm with an overall mean height of 25 cm across the 3 D T plots. After one winter, berms had subsided approximately 15-20%. Hinge microsites consisting of >80% organic material (i.e. no minera l or intermixed material) were avoided during a l l phases of soil sampling and plant ing activity. Hinges were the only microsite chosen for planting i n this study as this microsite is routinely planted, operationally, on disc trenched mesic sites of the region. 1. Top of berm (intermixed mineral/organic material) (20%) 4. Between trench (intact forest floor) (30%) 2. Hinge (intermixed, mineral, forest floor) (17%) 5. Mixed slash/ forest floor (16%) 3. Trench (mineral) (17%) between flipped berms Figure 3. Cross-sectional diagram of disc trench treatment showing m a i n microsite types (proportion of each microsite type in brackets). The trench microsite was composed mainly of mineral material . M a n y coarse fragments rolled down from the hinge and berm position and accumulated i n the trench during and after the operation. In the spring following treatment (1993), water ponded i n many trench bottoms on al l D T plots un t i l at least m i d June, and even later i n plots D T I and DT2. Mottles were observed to form i n minera l soil directly under these temporarily saturated microsites. Mixed-between microsites (16% of the DT microsite area) contain significant quantities of slash material and other organic debris and were unsuitable for soil sampling, instrumentation or planting. 17 2.4.1.2 Descr ipt ion of pile-and-burn microsi tes The excavator was successful at moving slash into windrows with minimal soil disturbance, but activity was slow (~ 2.0 ha/day in study and surrounding area). Figure 4 gives a schematic representation of the main microsite types (not all) created during the PB treatment. The microsites sampled for chemistry and microclimate (Table 2) include burned pile, scalped mineral spots caused by excavator rake during piling, excavator track impressions (mineral exposed, intermixed material, organic material exposed), and areas between piles and excavator tracks with forest floor intact and slash largely removed. 1. Burned windrow pile (8%) 4. Scrape or scalp to mineral (3%) 2. Under excavator track, all materials (10%) 5. Unburned windrow pile (5%) 3. Between-piles and tracks, forest floor intact (62%) Figure 4. Cross-sectional diagram of pile-and-burn treatment showing main microsite types (proportion of each microsite type in brackets). Soil displacement and mineral soil exposure after the piling treatment were minimal. Most of the scrapes caused by raking were narrow (<5 cm wide), about 5-10 cm in depth and approximately 1-m in length. The bottom of the scrape microsite was either mineral soil, intermixed mineral/organic, or all organic, although only mineral and intermixed microsites were sampled (Table 2). 18 During piling, excavator tracks created 2- to 10-cm impressions into the mineral soil. In some cases small amounts of forest floor materials were squeezed out and deposited onto the edges of tracks (Figure 4). Ponding of water was observed in tracks during early spring periods in both 1993 and 1994, especially in plots containing finer-textured soils. Table 4 details the mean consumption (%) of each windrow pile following burning based on point estimates of consumption at 5-m intervals along the length of each windrow pile. Over all 3 PB plots, an average of 49% of the total fuels were consumed in windrows. Mineral soil exposure under burned windrow microsites averaged between 3-5%. Ash layers of 1 to 2 cm depth were common where complete consumption of fuels had occurred. Live mosses commonly persisted under much of the ash layer despite complete consumption of fuels < 7cm in diameter. Table 4. Average fuel consumption in windrow piles. MEAN PLOT WINDROW CONSUMPTION % PB1 1 28 2 65 PB2 1 77 PB3 1 36 2 41 ALL PLOTS 49 2.5 Detailed Sampling 2.5.1 Woody fuel and forest floor depth assessments Fuel loading, forest floor depth, and mineral soil exposure were measured for 19 comparison among the 12 plots after harvesting and before site preparation using the triangular intersect method described by Trowbridge et al. (1987). Total fuel mass per hectare (t/ha) for each plot was found by tallying different fuel size classes in a single triangle per plot and applying the line intercept methods and formulas for calculating total fuels as described by Trowbridge et al. (1987). Forest floor depth and mineral soil exposure were measured at 5-m intervals along each triangular segment. The broadcast burn plots were assessed both before and after burning to determine consumption of forest floor and woody fuels. Because the other three treatments involved no change or a re-distribution of fuels after piling/disc trenching, fuels were not re-assessed there after site preparation. 2.5.2 Fire weather information Fire weather information (FWI) was collected as in Trowbridge et al. (1987) to assess fuel moisture conditions during the 1992 field season and to predict the appropriate time to conduct the broadcast burn treatment. A n automated weather station was set up in an adjacent cutblock (0.5 km away, on a site containing similar slope and aspect to the study site) for continuous measurement leading up to the burn. 2.5.3 Ecosystem descriptions Soil and site descriptions were based on one modal soil pit within each treatment unit using methods outlined by Luttmerding et al. (1990). Relevant habitat features and soil classification for each plot were recorded and are summarized in Appendix 1. Percentage cover estimates of native vegetation were made annually in 5 systematic locations of each treatment unit at a reconnaissance level. Detailed vegetation changes resulting from imposition of treatments (i.e. statistical 20 comparisons) will not be included in this thesis as sampling methods provided reconnaissance information only. 2.5.4 Bulk density sampling 2.5.4.1 0-20 cm layer Ten mineral soil (0-20 cm) bulk density (Db) samples were collected from each treatment unit in August, 1992 following harvesting and, again in August, 1993 after site preparation in each of the DT and PB microsites to be sampled for soil chemistry. Bulk density samples were not re-sampled in 1993 in the broadcast burn or no-treatment plots as it was assumed that no measurable changes in bulk density would have occurred in the 0-20 cm layer without machinery presence (i.e. B B and NT treatments). These 0-20 cm Db values would be used for conversion of soil nutrient values to a mass per area basis. Ten randomly selected samples were collected for each microsite per plot using the 2.5-m planting grid as a reference for sample locations. Samples were collected using the excavation/displacement method outlined in Kl inka et al. (1981). This method involved first removing surface forest floor/slash material and then excavating a cylindrical hole (approximate 12-cm diameter) of mineral soil to 20-cm depth. Excavated mineral soil was placed in plastic bags and subsequently oven-dried in the lab at 105° C for determination of oven-dry total mass. Coarse fragments and other organic debris within the mineral sample were separated and weighed. Total volume of soil was determined in the field by filling the excavated hole (0-20 cm) with silicon chips and measuring the displaced volume with a graduated cylinder. Volume of coarse fragments was determined from dry mass and assumed mineral particle density of 2.65 g/cm^. 2.5.4.2 Near surface 2-7 cm layer 21 After site preparation, water was noticed to pond in the DT trench and the PB track microsites and this observation prompted investigation of near surface soil compaction in 1993, one season after site preparation. Since no pre-treatment 2-7 cm bulk density sampling was done, no statistical comparison can be made before treatment. As a result, descriptive statistics are presented and discussed for 2-7 cm bulk density data in the post-treatment period. Similar sampling procedure as described in Section 2.5.4.1 was used for this layer with slight modification. To obtain a reasonable volume of soil for bulk density determination (i.e. 1 liter minimum) each of the 10 samples collected per microsite was formed by excavating and recording the volumes of two adjacent cylindrical holes (approximately 12 cm in diameter) within the same microsite. The surface 2-cm layer of mineral soil was removed in the field since this layer contained abundant rooting which would likely confound bulk density interpretation. The oven-dry mass of the two samples collected per sampling point was eventually determined in the lab using procedures discussed in Section 2.5.4.1. 2.5.5 Soil nutrient sampling 2.5.5.1 Sampling before site preparation Soils were sampled in each of the plots in June, 1992 after timber harvest and before site preparation, and again, the following summer (July, 1993) one season after site preparation. Thirty-five sampling points within each plot were selected randomly (i.e. randomly selected sampling points selected from a 2.5-m grid pattern representing approximate seedling planting locations and this same approach applied to all 12 plots). From each selected sampling point, forest floor, decayed wood, and 0-20-cm mineral soil were sampled. For each layer, 7 composite samples were formed in the field by bulking 5 randomly selected individual samples. 22 If present, seperate forest floor and decayed wood samples were excavated at each sampling point and within 15 x 15 cm area. Depth of forest floor and decayed wood material (if present) were recorded in the field from each sampling point to determine percentage of each substrate. In the lab, moist composite samples of forest floor and decayed wood were thoroughly mixed, roots and hard (material not easily broken down by rubbing by hand) decaying wood material removed, and total wet weight recorded. Subsamples were taken from each composite and subsequently oven-dried at 105°C for 24 hours after air drying, or until no further weight change occurred. Subsequent determination of water content of moist organic samples allowed final calculation of oven-dry mass of each decayed wood and forest floor composite sample to be used in nutrient conversions on a kg/ha basis. A n additional subsample (approximately 1.5 liters) from each moist organic composite was then put aside for air drying (at 22°C) and chemical analysis. The dried organic samples were ground to pass a 2-mm sieve prior to further chemical analysis. Mineral soils were sampled in the field using a 5-cm diameter soil auger. The 0-20 cm mineral layer was bulked into composite samples in the field, and subsequently air-dried and ground to pass a 2 mm sieve in the lab. A l l composite samples were analyzed for total and mineralizable N , total C and N , available P, exchangeable Ca, Mg, and K, CEC, and p H (CaCl2 and water). Total S, P, Ca, Mg, K, Cu, Zn, and Fe were also determined for forest floor and decayed wood samples to estimate total nutrient pool of forest floor material and subsequent changes before and after treatments. Soil analyses were done at the Ministry of Forests North Road Research Lab in Victoria, B.C. Procedures employed for soil samples are included in Appendix 2. 23 2.5.5.2 Sampling after site preparation Soils were sampled again in July, 1993 one season after site preparation treatments and approximately one month after planting, using the methodology described above except that selected microsite types were sampled in the 0-20 cm layer within a given treatment. Table 2, (Section 2.4.1) identifies the microsite types sampled for soil chemistry. In situations where sample points landed on an unacceptable planting spot (i.e. excessive surface water, stones, or large, undecomposed logging debris) samples were moved a minimal distance to a location representing the same microsite type. Decayed wood materials were not avoided during nutrient sampling if that planting spot could have been screefed and planted during normal operational planting. Broadcast burn and untreated plots had only one microsite type (i.e. broadcast burn and intact forest floor, respectively). 2.5.6 Soil climate measures 2.5.6.1 First year after logging Most plots were very moist to wet (i.e. at or above field capacity) in the upper 30 cm of mineral soil until late July, 1992. For this reason, no attempt was made to assess moisture differences at the site during 1992. Water tables remained within 30 cm of the soil surface for the majority of the growing season in all treatment units. 2.5.6.2 First and second year after site preparation 2.5.6.2.1 Soil moisture Soil water potential was monitored manually by using a soil moisture meter (Model 5910-A, Soilmoisture Equip. Corp., Santa Barbara, CA) at weekly intervals using Delmhorst gypsum resistance blocks installed 10 cm from the soil surface. Table 2, Section 2.4.1 lists the combinations of microsite and treatment, which were 24 chosen for instrumentation for the first two years after site preparation. A l l 12 plots were sampled for soil moisture in both the 1993 and 1994 growing season (May-October). Soil moisture blocks were randomly located in 7 spots in the B B and NT plots. Owing to funding and time limitation, as well as a large number of microsites being monitored in the DT and PB plots, fewer replicates were placed in each of the DT (6 replicates for each of 4 microsites) and PB (5 replicates for each of 7 microsites) treatment units. 2.5.6.2.2 Soil and air temperature Four treatment units were selected for monitoring soil and air temperature over the 1993 and 1994 growing seasons. Three replicate copper wire thermocouples were installed at the 10-cm soil depth in locations believed suitable for planting and representing each of the following microsites: PB under-excavator-track-mineral-soil-exposed, PB burned-pile, PB between-piles-and-tracks-forest-floor-intact, DT berm, DT hinge, DT trench, DT between-trench-forest-floor-intact. Owing to minimal microsite /complexity, 5 replicate thermocouples were placed in the soil for one representative broadcast burn and no treatment plot. Air temperature was measured between two shaded 10 x 10 cm metal plates installed on a free-standing temperature stand located in the center of each instrumented plot and measured continuously at 5- and 30-cm height above the surface for each of the DT microsites and the BB treatment (1994 only). Two replicates each of air temperature were measured at 20 cm and 1 m above the surface in each of the PB, BB and NT plots. Temperature was recorded at 1-minute intervals during the growing season (late May - Sept.). Campbell Scientific CR-21 and 2IX dataloggers (Campbell Scientific, Logan, UT) were programmed to record average hourly temperature for 25 daily mean, maximum, and minimum temperatures. Equipment failures occurred periodically during both 1993 and 1994 but to a lesser extent in 1994. For this reason, most discussion of soil and air temperature wil l be made for the 1994 season. No statistical comparisons of soil temperature data was made without plot replication. 2.5.7 Planting Lodgepole pine (PI) and hybrid interior spruce (Sx) seedlings were planted as plugs in each plot in early June, 1993. Relatively healthy seedlings of uniform height ( ± 2 cm) and caliper (+1 mm) were originally selected. Seedlings were planted at 2.5-m spacing in a 7 row x 7 row pattern giving 49 trees of each species in the center of each treatment unit for subsequent monitoring of growth response. This spacing was maintained, except where minor adjustment was necessary to avoid unacceptable planting microsites (stumps, standing water in planting spot, excessive coarse fragments, etc.). Plot dimensions (i.e. 25 x 47.5 m) were sufficient to allow planting of 49 PI and Sx seedlings plus an additional two buffer rows around each seedling block. An outer row of buffer trees (defining the final plot boundary) were planted at 1-m spacing to provide extra trees for root morphology sampling (not discussed in this thesis except for visual observations). Information about planting stock is given in Table 5. In planting spots where site preparation did not result in exposed mineral soil or intermixed material (e.g. NT, B B plots), a 15x15 cm area of surface forest floor was removed at the base of each seedling as standard planting methodology^. In the DT treatment, only the 1 Province of B.C. 1989. Planting quality inspection - F.S. 704 . Silviculture Branch, Ministry of Forests, Victoria, B.C. 26 hinge position was planted as this is the common operational procedure in the northern interior. In other treatments, microsites were planted as they were encountered in the 2.5 m grid spacing. Unburned or undisturbed woody debris piles in the PB and DT treatments were not planted. Seedlings were grown at Summit Nursery, Telkwa, B.C. Lodgepole pine seedlings appeared healthy in all respects. However, approximately 30% of the 2+0 (grown 2 years in containers in a greenhouse prior to planting) spruce seedlings showed signs of forking or multiple leaders and were planted owing to a shortage of seedlings. This forking likely resulted from frost damage during the first growing season in the nursery (G. Pinkerton and Joe Wong, pers. comm.). Badly forked seedlings were replaced by new stock in spring, 1994 as it was decided that forked seedlings may ultimately confound treatment effects on seedling performance. Fifteen of the worst forked Sx seedlings were selected and replaced in all 12 plots in 1994 by new stock (Table 5) in good condition. Measurement of seedling growth [total height (cm), height increment (cm/yr) and diameter (mm)] from seedlings planted in 1993 (n=34 for Sx seedlings and n=49 for PI) were analyzed for first and second year treatment differences. Where leaders or terminal buds of seedlings had died during the first or second year, seedlings were removed from height growth T A B L E 5. Characteristics of planting stock. Species (Year otd) Stock Type Seedlot Seed Source/Elevation(m) PI ('93) 312 1 1+02 32998 Cummings Ck./960 Sx ('93) 415 2+0 8693 Cabinet Ck./1200 Sx C94) 415 1+0 35035 Cabinet Ck./1200 1 Stock type codes 312 and 415 refer to containerized cavity dimension of plug (3 cm diameter and 12 cm long for pine stock; 4 cm diameter and 15 cm long for spruce stock) 2 Stock type codes 1+0 and 2+0 refers to the number of years grown in the greenhouse (first number) and the number of years grown outside (second number in this case = 0). 27 statistics. Root collar diameter measurements were taken after the second growing season only, to avoid unnecessary damage to sensitive stems of seedlings after only one season of growth. 3 RESULTS AND DISCUSSION 3.1 Woody Fuel and Prescribed Fire Assessments Table 6 details amounts of woody fuel before and after burning in the B B plots. For comparative purposes, fuel estimates for the other 9 plots prior to site preparation treatments are also included, based on the same methodology (Trowbridge et al. 1987) used for the BB plots. Table 7 details A N O V A effects on pre-treatment differences (based on a significance level of p<0.05) for total slash loading in each treatment prior to site preparation. No significant differences (p=0.0778) in pre-treatment total fuel loading were found between the four treatments prior to site preparation (Table 7) although fuel quantity (t/ha) at plot BB2 was higher than other plots (Figure 5). Broadcast burn plots BB1 and BB2 had slightly greater mean total fuels (t/ha), particularly for the large (>7 cm) fuel class. From an operational perspective, the quantity of slash and woody debris left at this site after harvesting (not including stumps) can be viewed as moderate for interior forests (overall mean of 48.2 t/ha). Disc trenching was the treatment most limited by large woody debris and the number and height of stumps remaining on the site despite the V-rake attachment on the prime mover and stump re-cutting in the experimental areas. Large quantities of large-size logging debris and abundance of other decayed wood materials on the forest floor are expected to cause treatment limitations especially when disc trenching in similar high elevation cutblocks. 28 Table 6. Average fuel loading and depth of burn for broadcast burn and other treatments. Fuel loading Size class Treatment Pre Post Total % Duff Depth of Duff Duff consumption consumption depth burn remaining consumption cm t/ha t/ha t/ha cm cm cm % 0.1 0.6 • 1.1 • 3.1 • 5.1 • 0.5 1.0 3.0 5.0 7.0 BB >7.1 <7.0 Total 0.45 0.01 0.44 98.5 - - -1.92 0.08 1.84 96.01 - - -4.21 0.39 3.82 90.73 - - -2.37 1.15 1.23 51.69 - - -2.90 1.02 1.89 64.98 - - -53.46 36.27 17.19 32.16 - - - -11.85 2.64 9.22 77.76 - - - -65.31 38.31 26.41 40.43 6.3 1.1 5.1 17.6 0.1 0.6 • 1.1 • 3.1 • 5.1 • 0.5 1.0 3.0 5.0 7.0 PB >7.1 <7.0 Total 0.38 1.9 3.75 1.81 2.77 33.26 10.62 43.88 5.7 5.7 0.1 0.6 1.1 3.1 5.1 0.5 1.0 3.0 5.0 7.0 DT >7.1 <7.0 Total 0.4 1.68 2.99 1.55 1.66 30.01 8.27 38.28 6.8 6.8 0.1 0.6 1.1 3.1 • 5.1 • 0.5 1.0 3.0 5.0 7.0 NT >7.1 <7.0 Total 0.37 1.96 3.53 1.6 2.29 33.78 9.75 43.53 4.6 4.6 29 Table 7. A N O V A effects on pre-treatment fuel loading. Source of Variation df SS MS F value p(F) Treatment 3 1252 417.4 3.32 0.0778 Plot (treatment) 8 Total 11 Virtually all of the fine fuels (<0.1 cm size class) were consumed in the broadcast burn (Table 6) treatment while only 30% larger fuels (>7 cm in diameter) were consumed. A n average of 1.2 cm or 17.6% of the forest floor was consumed and there 11L 1111 j— C\J CO CD CD CD n CD CD CM I— Q Treatment Unit CD CD Q- Q_ Total luel Figure 5. Total woody fuels estimated from fuel transects prior to site preparation in each treatment unit. was a small (1.3%) increase in mineral soil exposure following broadcast burning, suggesting a quick and hot, but low impact fire had occurred. Owing to dry fine-medium woody fuels yet moist to wet forest floor and mineral soil layers, broadcast burning results were seen as ideal from a protection and plantability point of view 30 (i.e. minimal duff reduction, maximum fine slash reduction for plantability). Minimal mineral soil exposure (<1.3% overall) occurred after burning and would help limit possible soil erosion and nutrient losses through leaching. Based on observations during and soon after the burn, very little fly ash was deposited on non-burn experimental units of the study area. 3.2 Soil Physical Properties Changes in soil physical properties resulting from harvesting and/or site preparation activity has been studied by several researchers (Donnelly et al. 1991; Johnson et al. 1991; Cochran and Brock 1985; Heilman 1981; Carr 1988). Soil compaction and increased density of surface mineral soil layers commonly occurs during heavy machinery traffic (e.g. on skid trails) although the degree of compaction is highly dependent on other factors such as soil texture, structure, moisture and soil strength at the time of treatment or harvesting activity. In this study, bulk density was sampled at two soil layers, both believed representative of the early rooting environment of tree seedlings, namely the 0-20 and 2-7 cm mineral layers. Bulk density samples of the 0-20 cm layer were collected in the same microsite positions where nutrient sampling would occur in order to study changes in soil compaction and nutrient content on an area basis. The decision to sample soil density of the 2-7 cm layer came during the 1993 field season after water was observed ponding on the surface of the trench and track impressions for prolonged periods the season following site preparation. As pre-treatment bulk density values for the near surface 2-7 cm layer were not obtained,statistical comparisons before and after treatment cannot be attempted and differences in soil bulk density are compared for the post-treatment period only. Table 8 presents 2-7 cm bulk density averages for each plot and selected microsites one season 31 Table 8. Bulk density statistics of 2-7 cm layer one season after site preparation treatments. Treat Plot Microsite l N2 Mean Stderr M i n Max B B 1 B B 8 1.06 0.05 0.96 1.41 B B 2 B B 10 1.09 0.06 0.78 1.33 B B 3 B B 9 1.17 0.05 0.88 1.36 BBAve B B 27 1.11 0.05 0.87 1.36 DT 1 B 10 0.70 0.05 0.48 0.92 DT 2 B 9 0.49 0.05 0.32 0.79 DT 3 B 9 0.96 0.13 0.44 1.63 DT BMAve B 28 0.72 0.08 0.38 1.11 DT 1 H 10 0.69 0.09 0.34 1.18 DT 2 H 10 0.56 0.05 0.28 0.79 DT 3 H 7 1.31 0.11 0.83 1.79 DT HGAve H 30 0.85 0.08 0.48 1.25 DT 1 T 10 1.19 0.07 0.82 1.48 DT 2 T 10 1.31 0.05 1.12 1.55 DT 3 T 7 1.56 0.12 0.83 1.79 DT TRAve T 27 1.35 0.09 0.92 1.60 DT 1 U 10 0.91 0.09 0.39 1.28 DT 2 U 10 0.88 0.03 0.73 1.03 DT 3 U 8 1.48 0.15 0.77 1.89 DT UNAve U 28 1.09 0.09 0.63 1.40 NT 1 NT 9 1.37 0.06 1.08 1.64 NT 2 NT 9 0.99 0.04 0.81 1.18 NT 3 NT 10 1.03 0.07 0.69 1.30 NTAve NT 28 1.13 0.05 0.86 1.37 PB 1 N T K 10 1.11 0.14 0.52 1.90 PB 2 N T K 9 1.14 0.05 0.94 1.35 PB 3 N T K 8 1.53 0.05 1.15 1.89 PB NTKAve N T K 27 1.26 0.08 0.87 1.71 PB 1 T K 10 1.28 0.04 1.08 1.30 PB 2 T K 9 1.32 0.06 0.93 1.33 PB 3 T K 8 1.44 0.06 0.94 1.71 PB TKAve T K 27 1.35 0.05 0.98 1.60 1Microsite types include: B B = broadcast burn; H = hinge; B = berm; TR = trench; U = between-trenches-forest-floor-intact; NT = forest floor intact; T K = under-escavator-track; N T K = between-track-forest-floor-intact 2Different sample size as a result of samples removed following inappropriate field sampling 32 after treatment. Figure 6 depicts average 2-7 cm density for each treatment and associated microsite type after site preparation. Of note are higher mean 2-7 cm bulk density values associated with the DT trench and PB track microsites (1.35 g/cm^ for both) and lower soil density of the DT hinge and DT berm microsites (0.85, and 0.72 g/cm.3, respectively) relative to untreated (1.13 g/cm^) plots, suggesting a more porous surface rooting environment in the hinge and berm microsites. Greater root growth was visually observed following destructive sampling of some representative seedlings in both PI and Sx seedlings growing in berm and hinge microsites in this study. Orlander et al. (1990) discusses how soil loosening following disc trenching in Sweden was shown to promote root growth of young conifer seedlings. In addition, site preparation and/or mixing treatments can improve the redistribution of organic matter into lower horizons in addition to increasing water and air transport within the planting medium (Morris and Lowery 1988). CO E u 0) i £ u i CM c 0) T3 CD 1.40 1.20 1.00 0.80 0.60 0.40 0.20 0.00 — IHlil i IT' 1 I fMi i l l * V3 • Pill u u ha c 3 v> m _ 8-a 03 01 E 0) u G C O C O H) 10 tl) 10 o 15 CO Q_ Figure 6. Comparison of near surface 2-7 cm bulk density in several microsites and treatments one season after site preparation. 33 Several studies have shown that increased soil density can have a negative impact on tree productivity in young seedlings (Heilman 1981; Cochran and Brock 1985) and older plantations (Minore 1986; Bosworth and Studer 1990) largely by reducing water infiltration, root penetrability and subsequent nutrient availability. Minore et al. (1969) assessed root penetrability of seven conifer species in sandy loam soils of the Pacific Northwest. They found there were differences in species rooting adaptability to different levels of soil compaction. Specifically, they found that lodgepole pine and Douglas-fir (Pseudotsuga menziesii [Mirb.] Franco) roots better adapted to penetration of compacted soils than Sitka spruce (Picea sitchensis [Bong.] Carr.), western hemlock (Tsuga heterophylla [Raf.] Sarg.) and western red cedar (Thujaplicata Donn. ex D. Don), and that the less tolerant species began showing reduced root penetration at soil densities of greater than 1.45 g/cm^. In addition, no species could penetrate a soil with bulk density of greater than 1.59 g/cm^. Based on visual observations of seedling health in this study, seedling growth and rooting habit after the first two growing seasons was poorest for both PI and Sx in the 2-7 cm layer of the DT trench and PB track microsites where mean soil density was 1.35 g/cm^. By using untreated soils for comparision of changes in 2-7 cm bulk density after site preparation, there was a 19% increase in density (fine fraction) of DT trench and PB track microsites. Table 9 summarizes the A N O V A used in analysis of 0-20 cm bulk density data before site preparation treatments were implemented. Table 9. A N O V A for treatment effects on 0-20 cm mineral bulk density before site preparation. Source of Variation Treatment Plot (treatment) Total 1 df 3 8 11 SS 0.433 MS 0.144 F p_>F 1.58 0.2683 ^Based on n=10 samples/plot before site preparation. 34 Table 10 and 11 summarize average values for 0-20 cm bulk density before, and, one season after treatment, respectively. There were no significant treatment differences (p=0.2683) in fine fraction 0-20 cm bulk density prior to site preparation (Table 9) and average values ranged between 1.16 and 1.32 g/cm^ across all treatments (Table 10). Table 12 presents the A N O V A showing results of microsite and treatment effects on 0-20 cm bulk density one season after site preparation. There were significant treatment (p=0.0043) and microsite (p=0.001,2) effects on 0-20 cm bulk density (Table 12) one season after site preparation. Bulk density of the DT berm (mean 0.58 g/cm^) was significantly less than that of the DT hinge (mean of 0.88 g/cm^) and both were significantly less than other microsites studied (Table 11). Bulk density of the 0-20 cm layer decreased 50 and 24% in the berm and hinge microsites, respectively, one growing season after treatment. Few data are available for comparison to this study. There were no significant differences in 0-20 cm bulk density between microsites where forest floor material remained intact (means ranging from 1.18 and 1.32 g/cm^) on the surface after site preparation. These results may indicate the relative capacity of duff layers to protect mineral soils from compaction, despite light machine traffic in some of the treatments (i.e. skidder use in the DT between-trench microsite). Bulk density of the DT trench microsite increased 14% in the period from before to one-season after treatment; however, average soil density of the 0-20 cm mineral layer was never higher than 1.32 g/cm^ (DT trench and BB) in the treatments studied, and is not in the range expected to severely limit root growth, based on studies by Minore et al. (1969). There was very little change (+2%) in 0-20 cm bulk density pre- (1.16 g/cm^) versus one year post-treatment (1.18 g/cm^) in the DT between-trench microsite where 35 Table 10. Average bulk density of 0-20 cm soil layer by treatment prior to site preparation (standard error in brackets). Treatment Mean (s/cm3) Min Max B B 1.32 (0.05) 0.88 1.89 DT 1.16 (0.05) 0.59 1.68 NT 1.26 (0.05) 0.79 1.68 PB 1.21 (0.04) 0.64 1.51 Table 11. Average bulk density of 0-20 cm soil layer after site preparation (standard error in brackets). Treatment Microsite Bdnut 1 Bdfme2 Type (g/cm3) (g/cm3) B B B B 1.21 (0.05) 1.32a3 (0.05) DT B M 0.49 (0.05) 0.58c (0.04) DT H G 0.73 (0.05) 0.88b (0.04) DT TR 1.17(0.05) 1.32a (0.04) DT U 0.99 (0.04) 1.18a (0.03) NT U 0.91 (0.04) 1.26a (0.05) PB N T K 1.06 (0.05) 1.21a (0.05) xBdnut = bulk density of 0-20 cm layer used for nutrient calculations only using: oven dry mass<2mm/total volume. 2Bdfine = coarse fragment free bulk density in 0-20 cm layer for assessment of compaction effect using equation: oven dry mass<2mm/coarse fragment free volume 3Means followed by the same letter are not significantly different (Tukey's Studentized Ranee (USD) test, P<0.05. 36 Table 12. A N O V A for effects of microsite types on 0-20 cm mineral bulk density one season after site preparation. Source of Variation Treatment Plot Microsite (plot x treatment) Sam(micro x plot x treat) df 3 8 9 189 SS MS F p>F 2.239 0.249 4.98 0.0012 1.133 0.378 10.05 0.0043 Total 2091 ^Based on n=10 samples per treatment unit and 7 microsites across all treatments in total including 4 microsites for DT, and 1 microsite each for PB,BB, and NT treatments the skidder travelled. Such little effect of skidder compaction may have been in part due to relatively dry soil conditions at the time of treatment. Others have found that soil compaction can be affected by the harvesting/site preparation method and the frequency of passes over the soil surface (Froehlich 1979). Exposure of denser subsurface soil horizons associated with landings and skid trails following disturbance has been shown to have a negative impact on site productivity (Carr 1988) near Fort St. John, B.C. Although plantable area generally increases with piling treatments, researchers have found that increases in soil compaction and declines in nutrient capital (Bosworth and Studer 1990) can combine to have long-term negative consequences on conifer growth. Although 0-20 cm bulk density was not measured in the PB track microsite of this study, based on 2-7 cm data, it is reasonable to expect that bulk density of the 0-20 cm layer would have increased in the range of 10-20% one growing season after treatment. Of consideration is that the track microsite comprises only 10% of the total surface area of this excavator-piled treatment (Figure 4). Lopushinsky et al. (1992) found a 16% increase in bulk density in the upper 30 cm of the soil after caterpillar piling on a well drained Podzolic soil in southeast Washington. Slightly greater changes in soil density were noted to occur near the soil surface (2-7 cm layer) relative to the entire 0-20 cm layer following modification by site 37 preparation treatments. This may be a consequence of greater changes in macropore space in upper soil layers (Skinner and Bowen 1974) relative to denser, fine textured subsurface horizons where smaller micropores are likely more common. Platy soil structure in the 2-7 cm layer was noticed under excavator track microsites during soil sampling and suggests some degree of soil compaction had occurred. 3.3 Soil Nutrient Changes Following Site Preparation Silvicultural activities including site preparation can alter nutrient content directly through slash and soil displacement, and indirectly through compaction and changes to soil moisture availability. Some treatments, (i.e. mounding, discing) can improve nutrient availability to conifers by potentially increasing the organic matter content and volume of soil (and nutrients) available for exploitation by seedling roots, decreasing vegetation competition (Coates et al. 1991), increasing soil warming in the rooting zone, or improving soil drainage and/or soil porosity which may lead to enhanced initial seedling performance and survival (Macadam 1991, Bassman 1989, Orlander et al. 1990). Much of the nutrient capital of northern forests is organically bound in forest floor, decayed wood and vegetation and is generally only slowly available to seedling roots. Increased nutrient availability following site preparation in northern ecosystems may be partly due to increased biological activity in the soil brought about by improved access to incoming solar radiation, soil warming, adequate but not excessive moisture, and increased access to humus and decayed wood substrates for utilization by soil organisms. Site preparation may be used as a tool for accelerating the process of organic decomposition and mineralization of nutrients by improving some or all of the factors mentioned above (Orlander et al. 1990). Several studies have looked at soil nutrient status following burning treatments in the northern interior of B.C. (Feller 1982; Blackwell 1989; Macadam 1987). Short-38 term changes in soil chemistry following broadcast burning largely depend on the intensity/severity of the burn, but commonly include increases in p H and concentration of "available" nutrients (e.g. exchangeable Ca, K, Mg) in the mineral soil and forest floor (Feller 1982). Nutrient losses in surface forest floor/decayed wood materials can be considerable. Macadam (1987) found moderate impact slashburns in mesic sites in the Sub-boreal Spruce Zone (Banner et al. 1993) of the PRFR resulted in substantial losses of total soil N up to 376 kg/ha. At the same time, Macadam found increases in available P (37-157 kg/ha), pH, and exchangeable Ca and Mg in forest floor materials. Several other workers have reported substantial losses of nutrients by volatilization (especially total N , S and, to a lesser extent, P) (DeBell and Ralston 1970; FCimmins and Feller 1976; Blackwell 1989) from forest floor, woody slash, surface mineral horizons. Site losses of nutrients contained in fly ash (e.g. Mg, K and Ca) have been reported (Feller 1982) although these losses will depend on the nature and depth of the ash created, and the degree of convection created during burning and winds present during and immediately after the burn. The extent and duration of change in soil nutrient status following burning are dependent on a variety of site factors (e.g. moisture regime, slope position, slash loading and species) in addition to the intensity of the prescribed fire and the amount of organic material consumed (Feller 1992). Lopushinsky et al. (1992) found that extractable soil nutrients initially increased in broadcast burn and burned pile treatments but returned to unburned levels within 3 years. Availability of many macronutrients in forest floor, and, to a lesser extent, mineral soil, will increase concurrently with increase in p H after burning. Some micronutrients can become less available at higher soil p H values (particularly Fe, Cu, Zn and Mn). On sensitive sites (i.e. having coarse mineral soils with minimal forest floor depth), burning can permanently remove organic materials which often contain much of the nutrient reserves of a forest soil. To minimize permanent damage to sensitive soils or ecosystems by broadcast or pile burning, the 39 Ministry of Forests has developed burning guidelines for various forest regions in B.C. to identify criteria for evaluating soil features (i.e. L F H depth, soil texture, soil depth) which could affect sensitivity of a forest site to fire impact. Some believe that the N losses by volatilization in organic layers are compensated by an increase in available forms of N to the soil system (Mroz et al. 1980) although this will depend on leaching losses out of the soil system and nutrient retention as influenced by C E C of the soil after burning, as well as on mineralization and immobilization. Piling treatments are often practiced where slash loadings are high (e.g. moist/cold high elevation forests of northern B.C.). The re-distribution of nutrients, removal of duff layers and compaction of mineral soil layers during the piling procedure have been shown to have negative and possibly long-term consequences on nutrient reserves in the forestry site (Stone 1983) at the expense of improved plantability of the site overall and improved tree growth often associated with the burned windrow area (Ballard 1985). Variability in machine operation during piling treatment can influence the degree of site degradation (nutrient displacement, compaction) which occurs. It is therefore important not to generalize effects of treatments such as piling unless soil and site conditions and machinery type are known ahead of site preparation activity. Minore (1986) found measured height growth of 5 year-old Douglas-fir seedlings to be better on broadcast burned vs. pile-and-burn sites. He believed differences may have been attributed to soil compaction or loss of surface top soil layers during the piling procedure. Most research has shown that maintaining some degree of organic matter on the surface following timber harvest and site preparation is important when maintaining long-term forest productivity (Harvey et al. 1987; Weetman 1987). Acidic environments commonly to occur under decayed wood-dominated forest floor material and this can influence soil forming processes in addition to decomposition and mineralization of organic materials by soil microbes (Krajina 1969; Harvey et al. 1987). 40 The A N O V A in Table 13 was used to assess the pre-treatment differences for each soil nutrient variable. Following site preparation, the A N O V A in Table 14 was used to assess microsite differences in addition to treatment effects on soil nutrients. There were a total of 8 microsites sampled for mineral soil and 5 microsites sampled for each of the forest floor and decayed wood layers across the four treatments in this study, accounting for the different degrees of freedom for microsite in Table 14. Table 13. A N O V A for pre-treatment differences in forest floor, decayed wood and mineral nutrient variables. Source of Variation Treatment Plot(Treatment) Sample(plot x treatment) Total iBased on n=7 samples per plot or treatment unit Table 14. A N O V A for effects of treatments and microsite types 1 within treatments on nutrient variables one season after site preparation. Forest floor, decayed wood 0-20 cm mineral Source of Variation df df Treatment 3 3 Plot(treatment) 8 8 Microsite (plot x treatment)1-2 3 12 S ample (micro x plot x treatment) 135 216 Total 149 239 'Nested microsite types for organic materials include: broadcast burn; PB burned-pile and PB between windrows; untreated; DT between trenches (forest floor intact) 2Nested microsite types (within treatments) for 0-20 cm mineral soil include: DT hinge, berm, trench and between-trench; PB burned-pile and PB between-windrow piles; untreated; broadcast burn df 3 8 72 83 1 41 3.3.1 Nutrients in decayed wood Over the 12 treatment units in this study, decayed wood (soft) was an important component of this high elevation site and rotting wood was visually recorded at the surface of 21.6% of the total area sampled (as determined during woody fuel transects described earlier). The depth of decayed wood materials (average of 2.2 cm over the 12 plots) was slightly less than forest floor material (average of 2.7 cm) but still makes up an important component of the total forest floor depth at this site (Figure 7). On a mass per area basis, prior to site preparation, decayed wood (i.e. well rotted logs and rotten woody debris) made up 52% of the total dry mass of organic materials averaged over all 12 plots. There were no significant differences in treatment means of decayed wood nutrient concentrations and contents prior to site preparation. Table 15 details probability levels for treatment effects on decayed wood nutrients expressed on a concentration and content basis. Tables 16 and 17 detail average nutrient concentrations and contents, respectively, of decayed wood materials prior to site preparation. Decayed wood material tend to be slightly more acidic hpH(water) 4.0 - 4.2] than comparable forest floor samples tpH(water) 4.3 - 4.4]. Many macronutrient elements tend to be less available in acid soil conditions (Brady 1980) and this became evident during nutrient sampling as there were relatively few feeder roots observed in decayed wood relative to non-decayed wood forest floor materials. Decayed wood materials were generally in a more advanced stage of decomposition at this site, resulting in higher total C concentration and content, and CEC relative to less decomposed forest floor materials. Table 18 presents probability of treatment and microsite effect on decayed wood nutrients one season after site preparation treatments. The treatments and microsites studied had relatively little effect on decayed wood nutrients. Microsite effect was 42 significant (p<0.05) for exchangeable Mg and total Cu on a concentration basis, and on PHH2O andpHCaCl2 despite that a relatively short time had elapsed (9-12 months) 7 - r 6 4-5 4 Figure 7. Depth of decayed wood (DW), forest floor (FF) and total forest floor (FF Tot) for each of the treatment units of the study. since treatments were implemented. Tables 19 and 20 present average nutrient concentrations and contents, respectively, of decayed wood materials, one season after site preparation. With the exception of total S in all treatments, and total N in the PB burned pile microsites, all nutrient concentrations and contents (kg/ha) in decayed wood material increased between the two measurement periods, two complete seasons following harvesting. An increase in decayed wood nutrients following harvesting has been observed and described in other nutritional studies involving forest floor materials (Weetman 1987) as the "assart effect" where increased availability of light, heat, nutrients and moisture following removal of trees often allows improved 43 Table 15. Probability levels for decayed wood nutrient concentrations and contents for treatment effect before site preparation. Nutrient (prob>F) Concentration Content Exch Ca 0.8729 0.3665 Exch K 0.2722 0.7916 Exch Mg 0.8432 0.3112 CEC 0.2897 n/a Min N 0.2777 0.5849 Avail P 0.2772 0.5553 pH(H 2 0) 0.3872 n/a pH(CaCI 2) 0.6400 n/a Tot C 0.4688 0.3705 Tot N 0.5028 0.2445 C/N 0.1817 n/a Tot Ca 0.6792 0.7736 Tot K 0.7617 0.7384 Tot Mg 0.3814 0.6273 Tot P 0.7570 0.7690 Tot S 0.7131 0.7648 Tot Zn 0.4433 0.4070 Tot Mn 0.6039 0.5049 Tot Cu 0.5810 0.8488 Tot Fe 0.5820 0.7483 conditions for microbial activity and mineralization of nutrients. Some nutrient increases in decayed wood, particularly exchangeable cations, may have resulted from increased solubility of elements from overlying forest floor layers followed by strong retention capacity (CEC ranging from 88-97 cmolc/kg) of the decayed wood layer. The greatest loss of total S (11.7 kg/ha) and total N (86 kg/ha) in decayed wood (Table 17 and 20) occurred in the burned pile microsite 9 months after treatment, which may be partially explained by volatilization losses which are common even at relatively low burning temperatures (Feller 1982; Lindeburg 1990). A relative low intensity fire is suspected to have occurred in the windrow burns of this study, as less than 2 cm of forest floor was consumed during the burn. Lesser losses of total S (5.7 kg/ha) 44 Table 16. Mean decayed wood nutrient concentrations before site preparation (standard error in brackets). Variable Units Broadcast Burn T R E A T M E N T 3isc Trench Untreated Pile & Burn Exch Ca (cmoLAg) 16.3(0.97) 19.4(1.65) 16.3(1.13) 19.5(1.99) Exch K (cmolc/kg) 1.6(0.12) 1.5(0.12) 1.6(0.09) 1.3(0.08) Exch M g (cmoLAg) 3.0(0.18) 3.3(0.30) 3.0(0.15) 4.4(0.21) C E C (cmoLAg) 110(2.7) 108(3.7) 96(2.4) 109(3.6) M i n N (mg/kg) 238(15.1) 227(26.3) 194(23.5) 203(16.0) Avai l P (mg/kg) 44.2(4.3) 46.7(2.7) 52.6(4.09) 38.5(2.5) p H ( H 2 0 ) 4.03(0.04) 4.16(0.06) 4.03(0.04) 4.25(0.07) pH(CaCl 2 ) 3.40(0.04) 3.56(0.07) 3.41(0.04) 3.61(0.08) T o t C (cg/g) 53.5(0.58) 53.3(0.56) 54.3(0.46) 54.1(0.73) Tot N (cg/g) 0.77(0.03) 0.79(0.04) 0.72(0.03) 0.83(0.04) C/N 72.4(3.57) 67.4(4.05) 78.0(3.49) 65.7(3.28) Tot Ca (cg/g) 0.42(0.05) 0.52(0.08) 0.42(0.03) 0.50(0.06) Tot K (cg/g) 0.077(0.101) 0.078(0.008) 0.078(0.006) 0.073(0.010) Tot M g (cg/g) 0.055(0.009) 0.061(0.008) 0.053(0.004) 0.072(0.017) To tP (cg/g) 0.065(0.005) 0.063(0.006) 0.062(0.003) 0.068(0.007) TotS (cg/g) 0.100(0.005) 0.100(0.005) 0.088(0.004) 0.094(0.007) Tot Zn (mg/kg) 33.3(11.1) 21.2(3.3) 17.1(2.2) 24.7(3.3) Tot M n (mg/kg) 270(40.6) 308(64.6) 326(69.7) 258(69.0) 45 Table 17. Mean decayed wood nutrient content (kg/ha) prior to site preparation (standard error in brackets). Variable Broadcast burn Disc trench T R E A T M E N T Untreated Pile & burn Exch Ca 82.2 (7.5) 97.4 (12.6) 64.3(5.1) 86.7 (12.1) Exch K 14.9 (1.5) 13.4(1.3) 16.5 (1.0) 11.1 (1.0) Exch M g 9.0 (0.6) 9.4 (1.0) 7.2 (0.4) 9.0 (0.9) M i n N 5.6 (0.47) 5.8 (0.76) 5.6 (0.78) 6.5 (0.78) Avai l P 1.1 (0.12) 1.2 (0.09) 1.4(0.11) 1.1 (0.07) To tC 13584 (2791) 15077 (1944) 14942 (1663) 18291 (2984) Tot N 183 (13.0) 212 (20.3) 189(8.1) 222 (10.9) Tot Ca 102 (9:8) 126 (15.4) 83 (6.7) 107 (13.7) Tot K 18.6 (2.02) 18.0 (1.50) 15.1 (0.94) 15.2 (1.29) Tot M g 12.8 (1.28) 14.7 (1.88) 10.4 (0.63) 14.8 (1.88) To tP 15.3 (1.57) 14.5 (1.65) 12.2 (0.76) 14.0 (1.47) TotS 23.3 (1.34) 25.9 (2.22) 23.1 (1.18) 25.0 (1.73) Tot Cu 0.13 (0.009) 0.16 (0.021) 0.18 (0.078) 0.17 (0.019) Tot Fe 45.6 (5.89) 61.7 (13.2) 42.4 (4.52) 66.2 (11.2) Tot Zn 0.81 (0.25) 0.49 (0.05) 0.38 (0.03) 0.56 (0.07) Tot M n 6.7 (0.76) 6.6 (0.82) 12.9 (6.47) 5.4 (0.94) 46 Table 18. Probability levels for decayed wood nutrient concentrations and contents for treatment and microsite effect one season after site preparation. Nutrient (prob>F) Concentration Content Treatment Microsite Treatment Micrositt Exch Ca 0.6324 0.0684 0.6968 0.3759 Exch K 0.6799 0.3637 0.4351 0.8198 Exch Mg 0.2213 0.0031 0.7551 0.3769 C E C 0.4935 0.9837 n/a n/a Min N 0.3426 0.1879 0.3463 0.2613 Avail P 0.0011 0.8068 0.4090 0.6898 pH(H 2 0) 0.0643 0.0075 n/a n/a pH(CaCI 2) 0.1404 0.0066 n/a n/a Tot C 0.1259 0.0827 0.5274 0.0001 Tot N 0.9169 0.6067 0.6621 0.0024 C/N 0.6050 0.9856 n/a n/a Tot Ca 0.5419 0.0809 0.6948 0.3746 Tot K 0.7734 0.3172 0.7827 0.5767 Tot Mg 0.3712 0.2656 0.0989 0.0067 Tot P 0.3593 0.4542 0.6756 0.6262 Tot S 0.3474 0.3066 0.6121 0.0047 Tot Zn 0.0958 0.7842 0.7648 0.1034 Tot Mn 0.3958 0.8306 0.6885 0.2712 Tot Cu 0.0073 0.0248 0.2030 0.0107 Tot Fe 0.5325 0.1073 0.8833 0.4461 were found in the broadcast burn treatment after a one-year period, suggesting possible lower burning temperature in this treatment relative to burned piles. There were no comparable studies which looked at nutrient changes in decayed wood materials as seperated from forest floor materials after site preparation. Generally forest floor samples are collected as a composite of decayed wood/forest floor material and reported as such. Owing to the large proportion of decayed wood material occurring at this site (i.e. in some cases comprising 100% of the forest floor material present at the surface), observations and collection of separate decayed wood samples were seen as important in a study of nutrient changes occurring after site preparation. 47 Table 19. Mean decayed wood nutrient concentrations for treatments and microsites one season after site preparation (standard error in brackets). Variable Units Broadcast burn Disc trench TREATMENT Untreated Pile & burn Pile & burn BB 1 U MICROSITE TYPE U 1 BP U Exch Ca (cmolc/kg) 22.5 (1.9) 23.2 (2.3) 16.0 (0.9) 30.0 (2.0) 17.4 (1.5) Exch K (cmolc/kg) 1.4(0.11) 1.2 (0.10) 1.2 (0.08) 1.0 (0.12) 0.9 (0.06) Exch Mg (cmolc/kg) 4.2 (0.32)ab2 4.3 (0.41)ab 2.9 (0.08)b 6.5 (0.64)a 3.8 (0.28)ab CEC (cmolc/kg) 97 (2.1) 99 (1.6) 95 (1.9) 88 (4.1) 94 (2.1) MinN (mg/kg) 197 (16.3)ab 221(15.4)a 164 (9.6)bc 124 (13.2)c 182 (14.4)ab Avail P (mg/kg) 53 (3.3) 35 (2.6) 30 (2.3) 129 (17.0) 47 (3.3) pH(H20) 4.47 (0.06)bc 4.38 (0.09)bc 4.07 (0.03)c 5.05 (0.06)a 4.37 (0.06)bc pH(CaCl2) 3.92 (0.09)b 3.87 (0.1 l)b 3.60 (0.04)b 4.64 (0.07)a 3.86 (0.07)b Total C (cg/g) 48.6 (0.7) 48.9 (0.5) 50.1 (0.7) 45.5 (0.9) 44.0 (0.9) Total N (cg/g) 0.70 (0.03) 0.73 (0.03) 0.65 (0.02) 0.67 (0.04) 0.67 (0.03) C/N (cg/g) 70.9 (4.3) 65.9 (3.5) 73.8 (3.4) 67.0 (3.9) 67.2 (3.2) Total Ca (cg/g) 0.61 (0.059) 0.59 (0.061) 0.41 (0.022) 0.83 (0.055) 0.46 (0.048) Total K (cg/g) 0.07 (0.005) 0.06 (0.003) 0.06 (0.004) 0.07 (0.006) 0.06 (0.002) Total Mg (cg/g) 0.067 (0.005) 0.069 (0.007) 0.052 (0.003) 0.119 (0.007) 0.086 (0.006) Total P (cg/g) 0.07 (0.004) 0.07 (0.004) 0.06 (0.004) 0.10 (0.008) 0.08 (0.003) TotalS (cg/g) 0.061 (0.006) 0.087 (0.004) 0.077 (0.002) 0.066 (0.004) 0.070 (0.004) Total Cu (mg/kg) 5.4 (0.34)b 6.1 (0.81)ab 4.2 (0.24)b 9.4 (0.88)ab 8.0 (0.72)a Total Fe (mg/kg) 1964 (193) 2791 (453) 1613 (202) 4403 (497) 5585 (472) Total Zn (mg/kg) 33.5 (2.60) 23.4 (1.55) 20.2 (1.27) 41.8 (2.60) 28.9 (1.50) Total Mn (mg/kg) 437 (48.7) 318 (35.1) 267 (30.4) 798 (103) 365 (31.5) 1 Microsite types: BB=broadcast burned; U=forest floor intact; BP=burned pile. 2 For different microsite types, means followed by the same letter are not significantly different using Tukey's Studentized Range (HSD) test, P<0.05 level. 48 Table 20. Mean nutrient content (kg/ha) of decayed wood materials one season after site preparation (standard error in brackets). Variable Broadcast burn Disc trench T R E A T M E N T Untreated Pile & burn Pile & burn B B U MICROSITE T Y P E 1 U BP U Exch Ca 125.6 (11.7) 168.2 (27.9) 100.5 (13.0) 123.3 (12.1) 85.3 (4.7) Exch K 15.3(1.71) 13.0 (0.88) 14.2 (1.70) 7.7 (0.73) 10.5 (0.13) Exch M g 14.9(1.52) 18.4 (2.91) 10.9(1.19) 15.0 (1.07) 11.6(0.67) M i n N 5.6 (0.56) 7.1 (0.75) 5.0 (0.49) 2.6 (0.28) 4.9 (0.51) Avai l P 1.5 (0.13) 1.2 (0.17) 0.9(0.10) 2.6 (0.41) 1.2 (0.12) TotalC 14262 (868)a2 14913 (1227)a 15040 (9194)a 9194 (648)c 11954 (1069)b Total N 199 (14.3)ab 233 (22.9)a 195 (18.4)ab 136 (11.6)c 181 (15.8)b Total Ca 169 (15.4) 215 (35.9) 130 (15.5) 171 (16.1) 111 (6.0) Total K 19.1 (1.73) 18.7 (1.77) 17.77 (2.02) 14.2 (1.21) 14.8 (1.33) Total M g 19.2 (1.48)bc 23.9 (4.36)bc 15.5 (1.82)c 25.2 (2.62)a 21.4(1.71)b Total P 19.5 (1.49) 22.2 (2.80) 17.4(1.75) 21.6 (2.43) 20.3 (1.48) Total S 17.6 (1.95)bc 27.4 (2.54)a 23.6 (2.28)a 13.3 (1.19)c 19.3 (2.30)b Total Cu 0.16(0.012)b 0.22 (0.046)a 0.13 (0.013)c 0.20 (0.025)a 0.20 (0.014)a Total Fe 59 (7.3) 103 (23.2) 53 (9.6) 95 (13.7) 144 (17.2) Total Zn 0.95 (0.089) 0.80(0.109) 0.63 (0.063) 0.82 (0.067) 0.75 (0.058) Total M n 12.1 (1.20) 10.4 (1.69) 7.8 (0.94) 15.9(2.10) 9.9 (1.35) 1 Microsite types: BB=broadcast burned; U=forest floor intact; BP=burned pile. 2 For different microsite types, means followed by the same letter are not significantly different (Tukey's Studentized Range (HSD) test, P<0.05 level. 49 3.3.2 Nutrients in forest floor materials Non-decayed wood forest floor material comprises the dominant component (55% or 2.7 cm) of the total duff material at this site (Figure 7). Prior to site preparation, treatment means were found to be significantly different (Table 21) for forest floor nutrient concentrations of available P (p=0.0051), total P (p=0.0398) and total K (p=0.0024). These differences were sufficiently weak that when these three nutrient variables were converted to a kg/ha basis, significant differences were no longer present. It is for this reason that subsequent comparisons of these nutrients during the two measurement periods wil l be based on nutrients expressed on a nutrient content basis only. Appendix 3 presents pre-treatment forest floor nutrient concentrations. Table 21. Probability levels for forest floor nutrient concentrations and contents for treatment effect before site preparation. Nutrient (prob>F) Concentration Content (kg/ha) Exch Ca 0.7164 0.6005 Exch K 0.1549 0.6044 Exch Mg 0.7374 0.6635 CEC 0.6089 n/a Min N 0.7263 0.6468 Avail P 0.0051 0.8684 pH(H 2 0) 0.5459 n/a pH(CaCI 2) 0.7135 n/a Total C 0.2232 0.4994 Total N 0.5552 0.4631 C/N 0.2587 n/a Total Ca 0.7506 0.7736 Total K 0.0024 0.7824 Total Mg 0.6543 0.8162 Total P 0.0398 0.8608 Total S 0.8462 0.4341 Total Zn 0.9902 0.8376 Total Mn 0.5949 0.5049 Total Cu 0.8637 0.8787 Total Fe 0.2938 0.6848 50 Average pre-treatment nutrient contents (kg/ha) of forest floor materials are summarized in Table 22. Both before and after site preparation, forest floor materials tend to have relatively higher mean nutrient content than decayed wood materials from the same treatment and/or microsite type. This stresses the importance of maintaining some of the forest floor layer following site preparation to help sustain long-term productivity of the site, particularly on sites such as the present study, where relatively thin forest floor layers (i.e. 5-7 cm) exist. Microsite had a significant effect on many more forest floor nutrient concentrations and contents than decayed wood materials sampled from the same location one season after site preparation (Table 23). On a nutrient content basis , all forest floor nutrients except total K, total P, and total M n were significantly effected by microsite type within one year. Exchangeable cations, total N , mineralizable N and total S content decreased in all treatments and microsites one-season after treatment (Table 22 and 24). Part of this decrease may be related to increased mineralization and subsequent leaching following change of elements into more water soluble forms in addition to improved microclimate and conditions for microbial activity. Soon after burning treatments, a large portion of nutrient cations are believed to be converted into more available or water soluble forms (Feller 1982) and are transported into lower lying forest floor (including decayed wood) or mineral soil layers, or may completely lost from the site by leaching. This trend appears partially true in the present study with an apparent increase in exchangeable cations in decayed wood and 0-20 mineral layers (discussed later) following pile and broadcast burning treatments. Some of the total N and S losses in burning treatments may also have been the result of volatilization of these elements during burning as discussed by many other workers (e.g. Macadam 1987; Blackwell 1989; Feller 1982; Lindeburg 1990). 51 Table 22. Mean forest floor nutrient contents (kg/ha) prior to site preparation (standard error in brackets). Variable Broadcast burn Disc trench TREATMENT Untreated Pile & burn Exch Ca 111.7 (12.2) 137.9 (21.5) 73.4 (9.0) 104.8 (16.4) Exch K 37.1 (3.30) 30.8 (3.83) 27.0 (1.97) 25.6 (1.61) Exch Mg 12.5 (1.44) 14.3 (1.91) 9.1 (0.83) 11.3 (1.20) Min N 16.0 (1.35) 14.6 (1.98) 10.8 (0.84) 12.9 (1.11) Avail P 3.4 (0.40) 3.2 (0.32) 3.1 (0.23) 2.7 (0.24) Total C 12716 (956) 12027 (1291) 8915 (700) 10189 (778) Total N 331 (24.9) 337 (41.2) 214(15.9) 271 (20.9) Total Ca 176 (15.7) 193 (24.8) 131 (10.7) 156(9.9) Total K 46.0 (3.00) 47.3 (4.50) 45.6 (2.88) 36.5 (1.25) Total Mg 22.6 (2.47) 26.5 (2.80) 18.7 (1.33) 22.4 (0.88) Total P 37.9 (2.51) 39.5 (3.54) 35.4 (1.95) 43.5 (1.60) Total S 34.2 (3.00) 33.2 (4.02) 22.2 (1.79) 26.6 (2.33) Total Cu 0.26 (0.02) 0.27 (0.030) 0.23 (0.02) 0.27 (0.01) Total Fe 90 (11.0) 102 (12.2) 77 (8.4) 112 (8.9) Total Zn 1.46 (0.12) 1.28 (0.13) 1.13 (0.14) 1.29 (0.08) Total Mn 21.9 (2.45) 31.2 (6.09) 23.3 (3.11) 32.9 (5.19) 52 Significantly higher forest floor pH (water and CaCl2) (Table 25) values were recorded 9 months after pile burning and 12 months after broadcast burning relative to other treatments and microsites, as shown in Figure 8. A similar rise in p H of forest floor material one season following burning has been described by other workers (Macadam 1987; Feller 1982; Taylor and Feller 1989) and may be explained by the deposition of base-rich ash material, including oxides on the surface of forest floor following burning. Following wetting, hydrolysis of base oxide compounds results in Table 23. Probability levels for forest floor nutrient concentrations and contents for treatment and microsite effect one season after site preparation. (prob>F) Nutrient Concentration Content Treatment Microsite Treatment Microsite Exch Ca 0.0065 0.0001 0.0216 0.0003 Exch K 0.0435 0.0991 0.3474 0.0251 Exch Mg 0.0112 0.0001 0.1961 0.0001 CEC 0.6311 0.0547 n/a n/a Min N 0.0183 0.0005 0.8691 0.0001 Avail P 0.0063 0.0001 0.0007 0.0009 pH(H 2 0) 0.0009 0.0027 n/a n/a pH(CaCI 2) 0.0001 0.0005 n/a n/a Total C 0.0199 0.1949 0.3899 0.0176 Total N 0.1669 0.0110 0.4222 0.0018 C/N 0.0265 0.3355 n/a n/a Total Ca 0.0030 0.0001 0.0010 0.0004 Total K 0.9474 0.1696 0.6429 0.1322 Total Mg 0.0092 0.0035 0.0057 0.0005 Total P 0.0089 0.0239 0.0440 0.0685 Total S 0.0600 0.0122 0.7233 0.0156 Total Zn 0.0020 0.0001 0.0014 0.0039 Total Mn 0.0365 0.1315 0.0365 0.1315 Total Cu 0.0338 0.1707 0.0464 0.0062 Total Fe 0.0121 0.0161 0.0039 0.0125 53 Table 24. Mean forest floor nutrient contents (kg/ha) one season after site preparation. Variable Broadcast burn Disc trench TREATMENT Untreated Pile & burn Pile & burn BB U MICROSITE TYPE1 U BP U Exch Ca 207 (11.8)a2 105 (11.5)c 68.4 (7.3)d 170 (10.1)b 102 (15.1)c Exch K 13.8 (0.81) 13.8 (1.15) 13.5 (1.08) 8.6 (0.81) 9.9 (0.58) Exch Mg 16.8 (1.33)a 10.9 (0.97)b 12.5 (1.10)b 15.9 (1.01)a 11.8 (1.59)b Min N 6.3 (0.30)b 8.4 (0.71)a 6.6 (0.55)b 2.5 (0.24)c 8.6 (0.95)a Avail P 4.4 (0.31 )a 1.4 (0.10)b 1.4 (0.11)b 4.0 (0.15)a 1.8 (0.12)b Total C 11276 (539)a 8328 (501 )bc 8449 (564)bc 6141 (322)d 8305 (457)c Total N 291(12.2)a 225 (20.0)b 207 (13.1)b 161 (11.6)c 227 (14.7)b Total Ca 280 (,19.4)a 135 (14.4)c 99 (11.4)d 248 (19.6)a 133 (18.5)c Total K 21.9 (1.08) 17.4 (1.22) 17.6 (1.58) 20.3 (1.40) 17.2 (1.16) Total Mg 26.2 (1.56)b 16.8 (1.89)bc 12.5 (1.10)c 30.8 (1.63)a 22.9 (2.46)b Total P 36.6 (1.26) 23.2 (1.67) 22.0 (1.39) 32.9 (1.83) 27.6 (1.93) Total S 21.1 (1.31)b 21.0 (1.70)b 19.6 (1.43)b 11.3 (1.12)a 18.7 (1.49)b Total Cu 0.29 (0.009) 0.16 (0.015) 0.15 (0.013) 0.28 (0.016) 0.26 (0.020) Total Fe 72.7 (4.48)a 77.2 (13.2)b 58.2 (7.8)b 112 (8.9)a 163 (16.1)b Total Zn 2.15 (0.12)a 0.89 (0.08)c 0.79 (0.07)c 1.66 (0.09)b 0.91 (0.05)c Total Mn 51.5 (5.03) 20.2 (3.04) 23.7 (4.57) 46.9 (1.83) 19.4 (1.75) 1 Microsite types: BB=broadcast bvirned; U=forest floor intact; BP=burned pile. 2 For different microsite types, means followed by the same or no letter are not significantly different (Tukey's Studentized Range (HSD) test, p<0.05 level. 54 Table 25. Mean forest floor nutrient concentrations for treatments and microsites one season after site preparation (standard error in brackets). Variable Units Broadcast burn Disc trench TREATMENT Untreated Pile & burn Pile & burn BB U MICROSITE TYPE U 1 BP U ExCa (cmolc/kg) 39.8 (1.3)b2 26.3 (2.1)c 18.4 (0.9)d 50.8 (2.7)a 22.2 (2.2)cd ExK (cmolc/kg) 1.39 (0.04) 1.86 (0.09) 1.94 (0.05) 1.38 (0.14) 1.23 (0.06) ExMg (cmolc/kg) 5.5 (0.2)b 4.8 (0.3)b 4.0 (0.1 )b 8.0 (0.5)a 4.4 (0.4)b CEC (cmolc/kg) 88 (2.0) 97 (2.7) 94 (1.4) 80 (4.6) 90 (3.7) MinN (mg/kg) 248 (10)c 437 (19)a 352 (14)b 153 (13)d 409 (26)ab AvailP (mg/kg) 173 (7.3)b 80 (4.4)c 77 (6.0)c 247 (11.2)a 87 (5.9)c pHH 20 5.48 (0.06)b 4.76 (0.08)c 4.31 (0.05)c 5.97 (0.07)a 4.68 (0.05)c pHCaCI2 5.09 (0.07)b 4.02 (0.08)c 3.82 (0.06)c 5.70 (0.07)a 4.23 (0.07)c Total C (cg/g) 44.8 (0.6) 44.4 (0.5) 45.7 (0.4) 37.3 (1.5) 40.5 (0.9) Total N (cg/g) 1.16 (0.02) 1.22 (0.02) 1.12 (0.03) 0.97 (0.05) 1.12 (0.03) C/N 38.4 (0.6) 36.2 (0.6) 41.1 (0.7) 37.8 (1.5) 36.4 (0.6) Total Ca (cg/g) 1.12 (0.05)b 0.69 (0.05)c 0.53 (0.03)c 1.52 (0.12)a 0.60 (0.05)c Total K (cg/g) 0.09 (0.003) 0.09 (0.005) 0.09 (0.002) 0.13 (0.012) 0.08 (0.004) Total Mg (cg/g) 0.11 (0.003)ab 0.09 (0.015)c 0.07 (0.004)c 0.19 (0.009)a 0.11 (0.010)b Total P (cg/g) 0.15 (0.004)b 0.12 (0.003)c 0.12 (0.004)c 0.20 (0.011)a 0.14 (0.005)b Total S (cg/g) 0.09 (0.003)ab 0.11 (0.004)a 0.11 (0.002)a 0.07 (0.006)b 0.09 (0.004)a Total Cu (mg/kg) 11.7 (0.29) 8.8 (0.56) 8.0 (0.25) 17.4 (0.89) 12.0 (0.64) Total Fe (mg/kg) 3056 (201 )ab 4138 (509)c 3047 (252)c 6885 (569)b 7645 (637)a Total Zn (mg/kg) 86.6 (5.6)ab 45.4 (2.0)c 40.1 (1.8)c 101.4 (5.5)a 43.5 (1.5)c Total Mn (mg/kg) 1958 (203) 1143 (205) 1166 (166) 2834 (288) 916 (70) Microsite types: BB=broadcast burned; U=forest floor intact; BP=burned pile. o For different microsite types, means followed by the same letter are not significantly different (Tukey's Studentized Range (HSD) test, P<0.05 level. 55 H Pre-treatment B One season post-treatment a i n t a c t i n t a c t Figure 8. Change in forest floor pH before and one season after treatment. [Means followed by the same letter are not significantly different (Tukey's Studentized Range test, p<0.05 level)] increased p H and an increase of cations (including C a + + ) on the exchange sites at the expense of H + ions. Generally a rise in pH of acidic forest soils (common to medium productivity sites of the central interior) is favourable for macronutrient availability, at least in the short term (i.e. 2-3 years). A rise in soil p H after burning can negatively affect availability of several micronutrient elements (e.g. Fe, Cu, Mn), and can potentially limit tree performance. Other studies in the central interior of B.C. have shown deficient foliar micronutrient concentrations (Ballard 1985; Yole and Brockley 1989) occurring in young conifer stands regenerating after fire history. The p H values reported in this study after burning (i.e. <6) are not believed high enough to cause a serious decrease in availability of micronutrients however. Available P and exchangeable Ca and Mg content were significantly greater in the two burning treatments relative to unburned treatments (Table 24). Small increases in forest floor available P (0.9-1.0 kg/ha) occurred following the two burning treatments. These results agree with those of Macadam (1987) and Taylor and Feller 56 (1987) who found increases in available forms of P and exchangeable cations in forest floor soon after burning in the central interior of B.C. Taylor and Feller found increases in available P in forest floor to be short-lived, and P content dropped to pre-burn levels only 9 months after broadcast burning on a mesic site in the SBS zone near Prince George, B.C. In unburned treatments, where duff was left intact, there was a decrease in available P after one season (Table 22 and 24). Using foliar analyses of lodgepole pine and white spruce (Picea glauca [Moench] Voss) in interior stands of B.C. several researchers have found that N and S are commonly deficient, particularly in stands exposed to a continuous fire history (Brockley and Yole 1985; Ballard 1985; Blackwell 1989). The PB burned-pile and DT between-piles (forest floor intact) microsites had the greatest losses of total N in the forest floor (110 and 112 kg/ha, respectively) and S (15.3 and 13.4 kg/ha, respectively), in the period from before to one-season after site preparation (Table 22 and 24). The relatively high losses of forest floor total N and S from burned piles is believed caused by greater volatilization of these elements brought about by a more intense fire (containing a greater concentration of fuels) relative to the lower losses encountered in the broadcast burn treatment where fuel loading and burn intensity were likely less. Relatively high losses of total N from the DT between-piles microsite may be due in part to improved drainage and others factors contributing to increased mineralization. Broadcast burning at this study resulted in a 40 kg N/ha loss of forest floor total N (not including decayed wood), notably less than N losses reported from moderate intensity burns from similar northern ecosystems and slash types reported by Macadam (1987). Macadam found a significant squared correlation (r^=0.56) between total N loss and forest floor consumption in the sites from the SBSmc subzone in the Rupert Forest Region. Harvested sites having undergone repeated fire history (including the present study) may sustain long-term nutrient shortages following additional severe burning or other 57 harsh forms of site preparation (considerable forest floor displacement), neither of which occurred at this site. Micronutrient contents of forest floor total Cu, Zn and M n material showed a general increase one season after both burning treatments and decreased in treatments where forest floor remained intact including untreated plots (Table 22 and 24). An overall decrease in organic matter content (total C) of forest floor materials in treatments where forest floor remained intact at the surface after site preparation may indicate an increase in microbiological activity and/or mineralization brought about by changes in microclimate, and/or as a result of combustion of organic material in the two burning treatments. 3.3.3 Nutrients in 0-20 cm mineral layer There were no significant treatment differences in mineral soil nutrients prior to site preparation (Table 26). Table 26. Probability levels for mineral soil (0-20 cm) nutrient concentrations and contents for treatment effect prior to site preparation. Nutrient (prob>F) Concentration Content Exch Ca 0.6682 0.8130 Exch K 0.3925 0.7925 Exch Mg 0.7024 0.7546 CEC 0.2921 n/a Min N 0.5985 0.6091 Avail P 0.4524 0.2025 pH(H 2 0) 0.8843 n/a pH(CaCI 2) 0.6194 n/a Total C 0.1258 0.1954 Total N 0.1986 0.4507 C/N 0.4533 n/a 58 Mean nutrient concentrations and contents for the 0-20 cm layer before site preparation treatment are given in Table 27 and 28, respectively. One season following site preparation, many nutrients from different microsite positions of the 0-20 cm mineral or intermixed layers were significantly affected by site preparation treatment (Table 29). Highly significant microsite effects were found for all mineral nutrient concentrations except exchangeable Mg. Mean nutrient concentrations and contents of the 0-20 cm layer one season after treatment are summarized in Table 30 and 31, respectively. Greatest overall changes in mean nutrient concentrations (Table 27 and 30) and contents (Table 28 and 31) in the 0-20 cm layer were noticed in the DT berm and hinge microsites which had accumulated greater contents of organic matter (total C) relative to other microsites under study. This reflects the incorporation of organic materials during the trenching activity. Prior to site preparation, total C concentration of mineral soil across the four treatments ranged from 1.92 - 2.76 cg/g (Table 27). The first season following treatment, mean total C in the DT berm and hinge microsite had climbed to 10.0 cg/g and 4.3 cg/g, respectively, with both treatments significantly different from each other and both having significantly greater total C than other microsites studied (ranging from 1.3 to 2.2 cg/g) (Table 30). Organic matter contains a high percentage of the soils total nutrient capital including microbial populations which convert nutrient elements from organic to more plant available forms. As a result, concentrations of other nutrients, including total N , exchangeable cations, and mineralizable-N (Table 30) were also higher in berm and hinge microsites one season after site preparation. These results agree with those of Orlander et al. (1990) where a "composting" effect was described following mounding treatments in Sweden, with increased mineralization of N , K and P in raised, inverted and intermixed mounds for at least 3 years after treatment. 59 C M CO o X O x PH be I ra t>c| 12 »l I 6 be o I be bo i<3 bo be | 2 ~5b -5" a CD g « CO" 0? o o O P g g g d CO CO o o Ci •<* CO CO in" c? co" o o o p g g g d . CD CD 00 o CO CO 00 CO ^ • * 00 CM" s-CO 1 > CO CM CO CO IO *—' 00 CO CO CM ^ H CM CO IO CO CO CO ^ CM CO CO si co' '—• CO t - CO CO CO CM • * CM CM CO o S " 6? t - 00 CO p g d - H CD TP CO t> CO CM CO "~H CM CM f ^ CM CO O o 05 CO • * ^ H O o * H d g g g o •* CD t - cn d d d d , ^ r ^ CO ' O o o o o o o g d d d o CM c-IN CM CM CM d d d d CD CO CO CO CD d g g d • — ' o co 00 CM co' co' IO "* s y—* ID cn p CO d d d *—- •—' CM o ca d OS CM CM CO" IO O O p p o d g g d o »-< CO -* —H i-H i-H *—i d d d d CM1 S " o" O CM c- CM • - H ^ H d g g d CM CM CO CO CM CM CM CM £ a & ii O rO SH PQ CD •8 2 a D r^H o C CD 3 P-l _o '+3 03 S H 03 ft o> ft 03 co O -!-> U o 'C ft S H CD >> O o • O be <^ 03 & <D - u d o o CD a a c3 CD X CN CD be a , V CO 1—t —' [11-2) (11.2) '—, CO 2 -I 1 CO r-f t> m 00 in o c~ y—-IO '—v CD in CO CO CX3 iri f—1 CO CO CM 00 CO o f in co S i co t—i co CO •—' co o CN 00 m i—i o o CM CO CM 00 CX) o CT3 CM in CO CO CM 00 m C O co in m cj CO o 1-1 m ^3 ^ ^ 1—1 i-H o CO in 1—1 i—1 ~— N ^ <J3 i f 00 05 r-H i-H -—\ ^ ^ ^ ^ ^ CO o CM i-H CO i-H cn CM i-H CO i-H -—' —^- '—' CO 1> O m 00 o CM OS CD i-H i-H i-H i-H 00 T3 CD CM CM CM CO CM o a CD SH CO 00 CO CO Oi CM —^> 00 CD 00 CD o CM, CO 00 CD in t> O i-H co t- CM 03 CD O CO in 3 H O CD 60 Table 29. Probability levels for mineral soil (0-20 cm) nutrient concentrations and contents for treatment and microsite effects one season after site preparation. Nutrient (prob>F) Concentration Content Treatment Microsite Treatment Microsite Exch Ca 0.6901 0.0245 0.7860 0.5193 Exch K 0.4513 0.0001 0.7086 0.0025 Exch Mg 0.8322 0.0730 0.8898 0.4597 CEC 0.4942 0.0001 n/a n/a Min N 0.8416 0.0001 0.9500 0.0001 Avail P 0.9587 0.0166 0.8897 0.8268 pH(H 2 0) 0.4715 0.0001 n/a n/a pH(CaCI 2) 0.5172 0.0002 n/a n/a Total C 0.0013 0.0001 0.0823 0.0001 Total N 0.8908 0.0001 0.6872 0.0034 C/N 0.1832 0.0001 n/a n/a After burning treatments, increases in mineral soil pH after 9-12 months were far less pronounced relative to organic layers and may reflect the greater buffering capacity of these relatively fine textured mineral soils. Mineral soil p H was significantly higher (5.43 water) in the DT trench microsite, relative to all other treatments/microsites (Figure 9) one season after treatment. This change in p H may be explained by greater occurrence of base en-riched subsurface mineral horizons exposed in this microsite, and common to soils of this area (Banner et al. 1993). Lowest p H values (4.51) were recorded in the DT berm microsite as shown in Figure 9, likely a consequence of increased incorporation of more acidic organic material, particularly decayed wood, into the soil matrix. 61 0 CO CP 1 CN o _ S 3 O o a -<=• o IN X « w P. OH 00 g bf 3 V DO •9 00 be ° 1 60 X o H g bo c e 00 a O c 2 co C a CS CO o ID g in CO o p PQ PQ CO re ee PQ c SH 3 43 ® O ci 2 03 3 J 3 W F " CO o o CO C3 CO u 4H c A 43 5 § « 3 g CO CO o d 03 ?q CO g o d 43 CO CN O s C? o g 00 CO 04)bc 05)bc .05)d .04)c .05)a .05)b 02(0. yotu. .51(0 .83(0 .43(0 0)90' ID ID ID .6)a .8)a ,5)b o O 43 S3. CO 22.5(5. 37.4i 25.21 38.61 26.4( 13.9( 22.5(5. CN" ID CO o CN o" CO ID ID cj t-CO, (D, 00 I—\ t- t> t—' 00 CO d CO ID CN .6)c .4)c o CN" 0)b 2)c 18(1.5)ab g 15(0 36(2. 23(2. 18(1.5)ab CO CN 9" oq g 00 g g g g ^H g, CM d d .01)d ,01)d .03)a .02)c 02)d 32)bc g oq o of g S-CO g CN d CN d d d d d d g o CO g CO T3 o d o CO g ID ID TP O d, CO o PQ K CJ CJ CJ CJ C c C C CO CO CO CO u SH SH CJ CJ CJ CJ co CO CO co Q Q Q s 43 43 X d © 00 —I CO 00 00 03 d. g C3 4= 5q o d CO d g 00 CO d o o CP c 3 CO PH £2 =H 54)b o ID" o 9" u o g CO g CO d p g of g CN CN —1 CN c S-l 3 eg eg CP PH T3 o> C Si 3 II PH PQ a o o -Q cj C co SH +J II P4 cu ox .a 43 II o a SH CO 4 2 -C CO CO '3 SH II CJ CO SH O CO <D SH ir in" O O CO CO SH «S T 3 03 c SH 3 4 2 II PQ PQ CO o 31 c S CO nd 03 "3 cu .a o ? I: 62 3 * v o c co CP C co CO CD C O CP co C3 3 s be CO o X o s S +J CO CD tH EH CD CD ^ — V I O © ' — N 00 LO LO CO N — ' — ' 00 o CO cj C J © © C O C O 1—-LO C M 1—1 C O C O r ^ t—1 LO C O LO i-H C O C N i-H C O —^' *—' C M o C O IO C M 1—1 C M - d be c o - 9: c o C O i-H I—1 i-H o LO iST r-H »—H 00 C M i-H C M C N LO C O CO i-H o LO C M < M a s LO LO C O PQ PQ CO _, o fl cO 5 8 S h 8-° C PQ p co C M CD 00 *—-LO C D c-C D < M C M i-H ^—' L O C O C D CD C M CD i-H o o CO CO C D LO c o C D C O O PQ -8 o LO • LO LO o 00 C M T3 C J i-H C M , C D © I— C O i-H oo C M 00 00 CO i-H CD C D C O o X C D C O 1—i C O C J 00 C M 00 r H C O T 3 - — N C O -*' i-H — ' o C D 00 00 *—' C M CO 65" o i—i c-LO C M 6 ? c ~ C O LO O C O *—' ' c- oo LO c o C D C D i-H C O 00 C M C N C M C D 00 C M C D C O C O C M oo o c o PH EH X X C J C J C J c C C 03 a> CD SH SH U -u C J C J C J CO CO CO • »H Q Q Q LO C O C D C N 65-C N ^—' C D t -C N C J —^' C O C J fi 03 SH C J CO s O T-H ^—^ LO O LO C N LO C N O O C N LO u C O ^—-00 C O ^—^ C N C D 00 00 LO C N , i-H ^ ' o IO LO i-H LO C N C O C N C N C J • CO -a o i-H I—i LO d C N , C N C N , ^—* C D o o C D oo C O i-H C N C M LO C N C D C O C M o C O C J C J C J 00 S © i-H o o i-H o C D s C N C N C O C D C M C N C D i-H oo t -C M 00 c o !=> BP a C u SH P X X =3 CD PH PH CD tSXl .s x. n O a SH 03 H = T 3 03 CO •FH CO fr PQ C J CO .s SH O CO Q3 SH SH O i3 CO 03 SH c 2 § 1 SH >, II g PQ £ PQ T3 . . 03 O S co PQ 03 B c 5 T 3 ^ 5 CO _r« ® O co S P SH SH " J i 5 P>H LO o C3 V PH CO 03 Q 03 c co PH T3 03 _N '-3 C 03 T3 +^  CO ">» 03 s c 03' SH .03 c CO C J HH c bp 'co o C 03 SH CO co SH 03 03 S M % C J CO 03 s .3 M *^ r S a % ® co ^3 03 >-< .-B _cO co O C 'H J " 63 5.6 Untreated Broadcast DT-berm DT-hinge DT-trench PB-burn burned pile Figure 9. Mineral soil pH (water) one season after site preparation treatments. Treatments with the same or no letter are not significantly different (Tukey's Studentized Range test, p<0.05). On a kg/ha basis, there was a general increase of 0-20 cm mineral nutrients from pre-treatment (Table 28) to one season post-treatment (Table 31) in mineral soil microsites where forest floor materials remained intact above mineral layers including untreated soils (excluding burning treatments). This increase in nutrients in the 0-20 cm layer in all treatments, including the control, is similar to the results found in forest floor materials and may be explained by increased nutrient availability soon after harvesting brought about by improved microclimate and decreased competition for nutrients (Orlander et al. 1990). The two burning treatments resulted in declines in total C (Figure 10) and N (Figure 11) content after one season, which is likely a consequence of volatilization losses, particularly in uppermost mineral soil layers. It is generally accepted that most C and N losses and transformations occur in the surface (i.e. upper 5-10 cm) layers of the mineral soil (Macadam 1987; Lopushinsky et al. 1992) except where severe soil 64 120000 100000 -c 80000 ~ 60000 1 40000 o r-20000 0 OLJL 11 Pre-treatment • One season post-treatment Untreated Broadcast DTberm DT hinge DT trench PB bum burned-pile Figure 10. Change in total C content (kg/ha) in 0-20 cm soil layer before- and one season after site preparation treatments. Treatments with the same or no letter are not significantly different (Tukey's Studentized Range test, p<0.05). 3500 ^ 2500 CP 2000 Z 1500 ° 1000 500 0 H Pre-treatment • One season post-ire a trie nt Untreated Broadcast DTberm DT hinge DT trench PB burned-bum pile Figure 11. Change in total N content (kg/ha) in 0-20 cm soil layer before- and one season after site preparation treatments. Treatments with the same or no letter are not significantly different (Tukey's Studentized Range test, p<0.05). 65 displacement or erosion follows treatment. Very little mineral soil was displaced during site preparation/harvesting at this site and soil erosion is expected to be minimal owing to very gentle slope gradient and predominately intact forest floor. Available P content decreased slightly over the measurement period in the 0-20 cm mineral layer across all treatments including burning (Table 28 and 31). This is in contrast to reviews (Feller 1982; Lindeburg 1990) and other studies in similar ecosystem units (Macadam 1989), where moderate impact broadcast burning resulted in marked increases in P availability in 0-15 cm mineral layer for up to 21 months after broadcast burning. In general, treatments which result in mechanical mixing or organic/mineral layers would appear to increase total, exchangeable and available nutrients in the short term. Other positive attributes of the disc trench berm and hinge microsites are increased organic pools of of C and N , ultimately resulting in improved environments for seedling roots, microbial populations and activity, and mineralization of nutrients (Orlander 1990, Weetman 1987). Burning treatments resulted in a slight increase in exchangeable nutrients in mineral layers, one season following burning treatments (Table 31). 3.3.4 Total soil nutrients Soil nutrients and changes following site preparation treatments can be looked at as cumulative totals (kg/ha) of all soil layers (forest floor, decayed wood and mineral). Although undecomposed woody slash at the surface can provide be an important contribution to total nutrient capital of the site (Blackwell 1989) these materials in addition to vegetation were not studied or included in nutrient totals in this project. The research of Blackwell (1989), which involved burning of lodgepole pine slash in a similar climate and ecosystem (mesic SBSmc) in the central interior of B.C., determined 66 that the quantities of nutrients lost from slash (including N , S, P, Mg, Ca, K, and Na) were generally greater than those lost from the forest floor. It was evident from visual assessments of excavated roots of seedlings after the second growing season that most of the newly formed root biomass was concentrated in the forest floor and upper 5 cm of mineral soil. The concentration of roots near the surface may be explained by many factors, including biological activity brought about by increases in organic substrate, superior soil structure, aeration, and/or temperature occurring close to the surface. Table 32 details selected whole soil nutrient contents (sums of the three soil layers) and treatments before, after one season, and the difference in nutrient content (kg/ha) between the measurement periods. Figure 12 presents the change in whole soil total N , one season after site preparation treatments. Of note were relatively small decreases in whole soil total N (109 kg/ha) following broadcast burning, even less than one-season losses occurring from the control treatment (207 kgN/ha) (Table 32). Burned pile microsites resulted in relatively large declines in the whole soil total N (617 kg/ha) relative to other microsites. Greatest changes in total N from the PB burned-pile microsite occurred from the mineral soil layer (-421 kgN/ha) possibly suggesting greater solubility or mineralization of total N . Short-term increases in whole soil total N content after site preparation were observed in the DT between-trench and berm microsites (393 and 145 kg/ha, respectively). Most of the increases in total N were related to changes occurring in the 0-20 mineral layer. There was an apparent increase in fine fraction bulk density in the DT between-trench microsite between measurement periods (possibly as a result of very slight compaction by skidders during disc trenching activity). This increase in bulk density likely increased nutrient content of mineral layers despite similar N concentrations between 1992 and 1993. The increase in whole 67 89 2. = o co IS u OI s i r £ 3 a u . . * e to u <2 o § 3 (O CD O 0) CA o 3 C CD 3 0) u> ro • CO (O CJ . N N _ I KI "* • * 9 ' CO CO ™ vl M 3 BBSS s S I S ? S3-» 8 O c 3 c_ (D 3 C < Si 2 _ -. 75 x o O O is o £ CO 3 o u o CD D) u O 3 (ft CD •a CD •o 0) . - » M — u i » u s -» 8 o co z co cn a K s s i j 0) S C+ co co W s S § o * IS < £? ^ N _ S * J -n 3 - • N (O u u c " S 3 u o -> u U CO CO vl K s i z 3-M M M M O CO N M CO IN* CO c N 01 7 W 18 * o c 3 < CD :+ " o> vi £ 2 -n » ^ CD 3 O 2. CA er CD o CD C/>_ 3/ CD ^ CD •o 0) 0) W l~¥ H 0) C £ CD (A) ro o c 3 c_ 0) «•+ <" CD «-+ O 9L W o — * o o 3 CO CO ro 3 c r i -- i 5" 3 r-f W cr < 5T < CD • J , « in ; • -» m -» 10 — . - » , CO OI vl -> C N « » 3 K vi , -> CO ft cn ^ c a « a 3 _, r\> N _ «= co oi rl 10 co 3 * to -• S is ft 5 co -» . N) N) fo u ai co co c co -> S co ft s °> co S S ro CO _ : ft ft „ -> o S o -> ™ cn o 5 N vl ft 2 . ro co M ft ft S -* ° C o -> R cn o 5 N vl ft s CO CO CO -> vl O .. ., ro ro co « K K N « c - „ N O « -J ro S « w u 5 H _, -» N _ ~ co co 2 ft 0 0 ^ z * vl O * ., ro ro co K M co c K ro cn ^ co fo co s CD CD O 3 CD Q> CA C CD 3 CD •o CD O a. • t i l ' ^ ft "n £ 0> « ft -n | - « 2 S 2 1 J N * o s y * J s s y s ° i r v» 4^ ^ -* n -? » K s 5 S 8 3 . - » M — "4, S » co 3 w 2 o co s 8 — VI (.1 * CO vl s cn 3 cn cn S S ft g " N ™ 2 J i K 3 S 8 8 8 S i n * a s i s j i i «i -; i i E _. vi oi •£ TI 3 » o ™ o 6 I " -1* M S j 2 c o> ft » - S S ft CO CO vl 2 5 3 8. -> Is) OI OI 0 -> 01 M ., —1 ro co OI ft K vl 00 c CO N S CO p S N 5 TH N 5 s 0 vl Tl . 01 -* 2 . -» ro vi ft rr en co o vi * o co — M - * co co e» K vi ft £ vi g is ro 5 CO oi i* 200 '5s U - -z -100 -"5 o -200 -cu o -300 -* -400 --500 -a JS O -600 --700 -° S « ! o - 0 CD a> • a> I -Q ' 0); Q) CO co a) a> a) -a •o CD Q) J3 Figure 12. Change in whole soil total N (kg/ha) one season after site preparation treatments. soil total N associated with the DT berm microsite over the measurement period (Table 32) may be explained by increased incorporation of organic substrates into the soil medium along with increased microbial biomass and activity. Changes in whole soil exchangeable K (Table 32, Figure 13) on a kg/ha basis following site preparation are less dramatic than for total N but even small changes in K can have long-term consequences on soil productivity (Goulding and Stevens 1988). Although little research is available regarding potassium deficiencies in soils of the interior of B.C. foliar and soil analyses would suggest that exchangeable or available K concentrations are often low, and soil material likely contains less exchangeable K than the tree biomass which is largely removed at harvest (Goulding and Stevens 1988). Greatest losses of whole soil exchangeable K after one growing season at the nutrient sampling locations occurred in the DT hinge (52 kgK7ha), berm (46 kgFC/ha), and trench (68 kgFv/ha) microsites, respectively (Figure 13). This comparison assumes that there is no nutrient contribution of adjacent forest floor/mineral materials to the sampling point. 69 Figure 13. Change in whole soil exchangeable K (kg/ha) one season after site preparation. Greatest increase in whole soil exchangeable K (Figure 13) followed pile burning in windrows (69 kgK/ha) and this net increase was largely as a result of accumulation of exchangeable K in lower mineral horizons following burning, despite decreases in K content of decayed wood and forest floor layers over the same period. Broadcast burning at this site resulted in little change in whole soil exchangeable K (-3 kg/ha), likely as a result of a low intensity broadcast burning treatment with minimal duff consumption or ash deposit on the surface relative to the pile burning treatment. 3.4 Changes in Soil Moisture Soil moisture is of critical importance in the survival and early establishment of conifer plantations. Site moisture supply is influenced by soil water storage properties (e.g. texture, coarse fragment content, porosity), precipitation and atmospheric and vegetative demands for water. Site preparation methods are often prescribed to help limit unfavourable environmental stresses such as water excess or shortage in the 70 rooting zone. For example disc trenching and mounding form microsites which can help reduce excess water in the rooting zone (Macadam 1989), while small scalps (Page-Dumroese et al. 1986) and broadcast burning (Haase 1986; Page-Dumroese et al. 1986) can help conserve water mainly by controlling competing vegetation and associated water losses by transpiration. Conversely, large scalps (exposed mineral soil) can lead to high surface evaporation rates and droughty soils in summer months (Fleming 1993), which could lead to early seedling mortality. In the ESSFmc and SBSmc subzones, precipitation inputs are believed sufficient during the growing season (Banner et al. 1993) to avoid major periods of surface soil drought, particularly where forest floor remains essentially intact on the surface. However, in particularly dry years, or, if the planting microsite has been sufficiently modified during site preparation to cause a distinctive change in the soil water regime, water shortage may be a concern to newly planted seedlings. This study monitored several microsites for soil moisture changes over two growing seasons to examine which microsites possessed growing season moisture supplies associated with good seedling growth. Most plants, including conifers, show greatest productivity under conditions where moisture content of the soil is plentiful or near field capacity. Field capacity is generally believed to exist at a soil tension of between 0.1 bars (10 kPa) and 0.3 bars (30 kPa) (Russell 1988) or the soil tension approximately 24 to 48 hours after a fully wetted soil has been allowed to drain. Field capacity is achieved after most of the larger macropores have emptied of water and micropores or capillary pores are still filled with water for plant use. At the dry-limiting end, many plants, including conifers, generally show reduced vegetative growth near the permanent wilting point (near 15 bars) (Hillel 1980). The soil tension at wilting point has been shown to be somewhat dependant on the species of concern, soil texture (Dosskey and Ballard 1980) and pore size 71 distribution of a given soil. Others have found reduced root growth, root respiration and photosynthesis of conifer seedlings to occur at soil tensions greater than 2-3 bars (Lopushinsky and Klock 1974; Havranek and Benecke 1978). Havranek and Benecke (1978) found that at low soil moisture content (>1.5 bar tension), pine (Pinus cembra) seedlings had a higher net photosynthesis/transpiration ratio relative to spruce (Picea abies) suggesting that pine used limited soil water more slowly and economically than spruce. Others have shown reduced nutrient uptake to occur in only slightly dry soils (soil tensions of 0.3-2.0 bars), depending on the species of concern, nutrient content within plant foliage and other stresses occurring on the foliage at the time of moisture stress (Plaut 1973). Soil water potential data were collected at weekly intervals in several locations per microsite type in 1993 and 1994, the first and second year after site preparation. The 1993 field season had an unusually high rainfall (Table 33), and consequently, soils remained moist to wet for the majority of the growing season for all microsites and treatments (Figure 14). Excessive soil moisture, including water ponding and poor soil aeration, owing to high water tables, may have been more of a concern to seedling health in many of the planting microsites of this relatively fine-textured site, including the DT trench, PB track-impression, PB between-piles-duff-intact, and other slightly compacted soils where skidder traffic had occurred. During the 1993 growing season (late May to late August), water tables remained within 20 cm of the mineral/organic interface until mid-July in all but some of the DT berm microsites. The DT berm was the first microsite to become freely drained to the 20-cm depth. However, this slight drying trend occurred in only one of the three DT treatment units in the beginning of August (Figure 14). Others workers have found that excessive moisture and periodic flooding of varied duration have resulted in root mortality and plant moisture stress. Levan and 72 Riha (1985) studied the response of root systems of nursery transplants of four conifer species (white pine, red pine, white spruce and black spruce) to flooding in a growth Figure 14. Change in soil tension for four well drained treatments/microsites over the 1993 growing season. chamber. They found flooding longer than 1 day killed all flooded root tips in both white and black spruce, but lateral roots of these two species were entirely replaced 6-22 days after soils were drained in 3-, 5-, and 7-day flooding treatments. Transpiration rates were depressed from the first day of flooding and dropped to 50-60% of the control rate in the 5- and 7-day flooding treatments; however, transpiration rates of all species recovered rapidly soon after soil drainage improved. Owing to more typical precipitation during the 1994 growing season relative to the wetter 1993 season (Table 33), greater drying trends in mineral soil also occurred between microsites types and treatments during 1994 (Figure 15). As a result, the 1994 soil moisture data set was primarily used for discussion of soil moisture differences between microsites formed by site preparation treatments. Figures 15 includes average values of soil moisture tension from representative well drained plots between May 19 73 Table 33. Comparison of monthly precipitation totals over three years at the Regan Creek study site. Total Precipitation by Month/Year Month (mm) 1992 1993 1994 May 15-30 n/a 62.9 19.0 June 23.0 108.5 77.5 July 3.7 138.6 30.0 August 18.3 33.0 45.0 Sept. 92.9 12.9 31.1 Oct. 1-15 38.8 19.3 n/a and Sept. 22, 1994 and include DT3, PB3, and NT2. Broadcast and burned pile microsites are not included in the figure for reasons of clarity but both treatments have drying trends similar to the PB duff-intact microsite shown in Figures 14 and 15. A n A N O V A (Table 34) was used to assess treatment/microsite effect on soil moisture meter readings over a 14-week period from June 23-Sept. 22 during the 1994 growing season. Soils were moist to wet before to June 23 measurement period, thus this date was selected as the starting point where microsite effects on soil drying began appearing. The greater frequency of frost events after mid-September was used to identify the end of the active growing season, and, in addition, rainfall increased steadily into October, as did soil moisture content. The 13 microsites (treatments) tested for differences in soil moisture included BB, DT berm, DT hinge, DT trench and DT between-trenches (duff intact), PB burned-pile, PB between-piles (duff intact), PB under-excavator-track (duff intact), PB under-track-mineral-exposed, PB under-track-74 Figure 15. Change in soil tension for four selected treatments/microsites over the 1994 growing season for well drained treatment units. Table 34. A N O V A used for determining effect of microsite on soil moisture over a 14-week period in the 1994 growing season. Source of Variation df SS MS F prob>F Treatment 3 3052 1017 6.88 0.0132 Plot(treatment) 8 Microsite (plot x treatment) 27 8257 305.8 2.68 0.0025 Samples(micro x plot treatment) 180 Total 218 intermixed-material, PB scrape-mineral-exposed, PB scrape-intermixed-material and NT or untreated. Several microsite types had significant effects on soil moisture over the 1994 growing season (Table 34, Figure 15). 75 Table 35 presents a summary of frequency of soil moisture tension for three classes of soil dryness, 0.3-1.0 bars (moist to slightly dry), > 2 bars (slightly to moderately dry), and > 15 bars (very dry) over a 14-week period from June 23 - Sept. 22, 1994. Data in Table 35 are presented as percentage of observations falling in each of the above moisture classes over the 1994 growing season. Average % values were calculated by taking the sum of observations in each of the three soil moisture classes over the 14-week period / the total number of possible observations over that same 14-week period in each plot. A n example calculation is given in Table 35. The reason for expressing numbers as a proportion was a result of varying numbers of weekly soil moisture readings collected for various microsites within a given treatment unit depending on complexity and availability of microsites for sampling. Sample size varied from n=5 for each of the PB microsite types, n=6 for each DT microsite type, and n=7 for each of the NT and B B plots. It appears from the 1994 data (Figure 15) that once spring rains had subsided, removing the slash (e.g. PB duff-intact, broadcast burn, burned pile) from the surface results in a slow but steady drying trend down to the 10-cm mineral soil depth. This drying trend is likely a result of increase surface evaporation and decreased shading of soil layers by woody debris (Spittlehouse and Childs 1990). Where treatments resulted in forest floor removal and exposure of mineral soil, including DT berm and DT hinge (Figure 16), PB mineral-soil-exposed track impressions, and PB scrape-intermixed and mineral-exposed microsites (Figure 17), this accelerated soil drying to an even greater extent. Disc trench berm and hinge, and PB scalps (mineral and intermixed to surface) microsites had commonly dried to soil tensions of > 2 bars by June 23, and these microsites remained drier for longer duration over the growing season as compared to other microsites under comparison (Table 35). In contrast, the microsites untreated, PB burned-pile, and PB under-excavator-track (duff intact) did not start to show a drying 76 Table 35. Average soil moisture tension between June 23 - Sept. 22, 1994 as a proportion of total possible observations in each of three soil moisture classes. Treatment » J I S ^ . « . « « B S * ^ 1 S o i l Tension (bars) 0.3-1.0 >2.0 >15 BB BB 0 .69 2 0.17 0.03 DT BM 0.38 0.42 0.12 DT HG 0.43 0.34 0.07 DT TR 0.73 0.08 0.00 DT INB 0.70 0.12 0.00 NT UN 0.82 0.02 0.00 PB BP 0.71 0.04 0.00 PB SI 0.57 0.24 0.04 PB S M 0.48 0.26 0.03 PB Tl 0.76 0.10 0.00 PB T M 0.64 0.14 0.00 PB TDUFF 0.81 0.07 0.00 PB INB 0.69 0.13 0.00 1Microsite type codes include: BB:broadcast burn; DT BM: disc trench berm; DT HG: disc trench hinge; DT TR: disc trench trench; DT INB: disc trench between-trenches, forest floor intact; NT: no treatment; PB BP: pile-and-burn burned-pile; PB SI: pile-and-burn scrape-intermixed-material; PB SM: pile-and-burn-scrape-mineral-exposed; PB TI: pile-and- burn-under-track-intermixed-material; PB TM: pile-and-burn-under-track-mineral-soil- exposed; PB TDUFF: pile-and-burn-under-excavator-track with forest floor intact; PB INB: pile-and-burn-between-pile-forest-floor-intact 2 Mean proportion (%) = mean total number of observations over 14-week period for 3 plots (per treatment)/total number of possible observations per microsite and treatment (e.g. % observations in 0.3-1.0 bars soil tension for treatment BB: 0.69 = 73/98 (BB1) + 77/98 (BB2) + 53/98 (BB3)/ 3. Total number of possible observations in the 14-week period depends on the treatment: for DT n= 84; for BB and NT, n= 98 and for PB, n= 70. 77 trend (i.e. lower than 0.3-1.0 bar) moisture for another full month, or until the end of July. These microsites were considerably moister over the 14-week measurement period than DT berm' and hinge microsites. The increase in drying trend for the DT berm and hinge and P B scalp microsite was likely the result of increased surface area of mineral soil exposed for evaporation associated with all three microsites, and improved drainage associated with the DT berm and hinge microsites. Capillary water movement from moist to wet underlying mineral layers would also be less in intermixed or inverted microsites (Hillel 1980) relative to microsites where surface soil layers have a more uniform and continuous 40 j ^ <n 35 • p i l l c = .o . p m£= Broadcast DT berm DT hinge DT trench DT No burn between treatment trenches Figure 16. Total number of weekly observations (all replicates combined) at several levels of soil tension between disc trench and other microsites from June 23-Sept. 22, 1994. pore network. Microsites showing an intermediate number of slightly to moderately dry soil tensions (>2 bars) during the growing season included broadcast burn, PB track mineral-exposed and PB between-pile-forest-floor-intact (Table 35). 78 > 15 bars 5-15 bars • 3.0-5.0 bars • 2.0-3.0 bars H 1.0-2.0 bars CD 5 o .2 CD E » 3 C (0 .o o 45 40 35 30 25 20 15 10 | P B BURNED PILE P B SCRAPE INTERMIXED P B SCRAPE MINERAL P B TRACK INTERMIXED P B TRACK MINERAL P B TRACK DUFF INTACT P B DUFF INTACT Figure 17. Total number of weekly observations (all replicates combined) at several levels of soil tension between pile-and-burn and other microsites from June23-Sept. 22, 1994. The berm and hinge microsites had the greatest number of observations of very dry soil condition reaching >15 bars tension (11 and 7% of the total observations over the 1994 growing season, respectively). The DT berm microsite was drier for a greater proportion of time (Table 35) over the 1994 growing season relative to all other microsites. Despite the trend of drier microsites associated with berms, and, to a lesser extent, hinge microsites, it is the belief that these dry spells are insufficient in number or duration to inhibit seedling growth by the second growing season. While water ponded in the DT trench and all PB excavator track impressions for most of the early growing season following snow melt (i.e. May to late June), soils of these microsites started to show drying trends (>2 bars tension) later in the summer by August. As determined by Levan and Riha (1985) root mortality can occur soon after sustained flooding in the rooting zone and this would likely be a concern for the 79 seedlings planted in the DT trench and PB under-excavator-track microsites. It is likely that the disc trenching treatment would help in draining soils in this fine-textured site by helping to lower water tables, especially during early spring periods. Planting seedlings on the high hinge or berm microsite, in such a moist climate, should help keep seedling roots well above high spring water tables. The untreated (NT), PB track-forest-floor-intact and PB track-intermixed-material were moist for the longest duration during the 1994 growing season (Table 36, Figure 17). These microsites had a significantly greater number of measurement periods with lower soil tension (between 0.3-1.0 bars) and moist soil conditions relative to the DT berm microsite (Table 36). Moist, but well drained soil conditions are believed ideal for seedling growth (given adequate temperature and other environmental conditions) (Russell 1988, Orlander et al. 1990) while at the same time possibly lessening the frequency of frost damage to certain microsites (e.g. berm) limited by temperature extremes. Under high water contents, soils generally have a greater soil heat capacity and take longer to warm up, a factor to be discussed in the next section. Very dry soil moisture conditions were relatively rare during the first 3 years following harvesting/site preparation in this study and are not expected to limit growth in this relatively moist, high elevation climatic zone. Soil moisture deficits negatively affecting seedling growth have been shown to be of much greater concern in certain microsites (i.e. mounds or scalps) of drier climates as in the southern interior of B.C. (Bassman 1989; Fleming 1993). 3.5 Temperature Both air and soil temperature were measured in the 1994 growing seasons as described in Section 2.5.6.2.2. As no replication between treatment units was possible, no statistical inferences are attempted for temperature data, but descriptive statistics are presented and discussed when attempting to detect seasonal trends. 80 Soil temperature has been determined to be a very important factor influencing seedling growth, particularly in high elevation biogeoclimatic zones in the northern interior of B.C. where the growing season is very short and limited by late snow melt. Most workers working on seedling physiology studies with white and/or Engelmann spruce believe that root and shoot growth increases with soil temperatures up to 20-25 °C (Dobbs and McMinn 1977; Heninger and White 1974). In addition, Heninger and White (1974) found optimal mycorrhiza development in white spruce occurred at 19 °C. Lopushinsky and Max (1990) found that root growth in high elevation sites of lodgepole pine and Douglas-fir at a soil temperature of 10 °C was only 10% and 17%, respectively, of root growth at 14.5 °C. Cold soil temperatures (5-9 °C) have been shown to decrease water uptake and movement through root systems of both spruce and pine as a result of increased viscosity of water, decreased permeability of root membranes and increased root resistance (Kaufmann 1975, Running and Reid 1980, Lopushinsky and Kaufmann 1984). Pine species appear to have a slightly higher minimum threshold temperature and Orlander and Due (1986) found water uptake of pine to be reduced by only 30% at 7°C. In cold, northern latitudes, various site preparation methods are employed to improve warming of soils. In Sweden, several site preparation methods are commonly used (Orlander 1990) with the main objective to increase soil temperature. These methods include scarification, burning, mounding and disc trenching. A l l methods used help reduce the insulative capacity of thick forest floor layers and/or vegetation layers (Coates et al. 1991) and may increase the absorption of solar radiation at the soil surface by exposing mineral material (Bassman 1989). Disc trenching and mounding type treatments can provide raised planting sites which drain better, and therefore, have a lower heat capacity. The sloping sides and larger surface area of exposed mineral material have a different radiation balance than a flat surface, absorbing more 81 solar energy during the day and losing more at night with the end result being higher mean soil temperature during the growing season (Stathers and Spittlehouse 1990). At this study site, cold snow melt runoff persists often until mid-late May, and, with radiation frosts common by the end of August, seedlings are exposed to a very short growing season. Frost damage can be of concern at any time of the growing season on high elevation harvested sites similar to the study area. Frost damage to flushing shoots most commonly occurs in the spring months when air temperatures drop to -3 to -8 0 C (Orlander et al. 1990), although species tolerance, duration of freezing temperatures and light intensity following a frost can all influence the damage which ultimately occurs. Burning treatments can reduce forest floor depth (and subsequent insulative capacity of that layer), and reduce slash and vegetation cover which shade the surface. In addition burning decreases the albedo of the ground surface (Stathers and Spittlehouse 1990), thus increasing the amount of incoming solar radiation absorbed by the surface. This can greatly increase the surface temperature by 10 °C during the daytime on hot, sunny days of summer. Improved drainage of soils through site preparation may be required particularly on moist sites to reduce the heat capacity before any improvement in soil warming can be expected (Stathers and Spittlehouse 1990). 3.5.1 Air temperature Potentially damaging frost events occurred on June 12 (-2.9 °C) and 20 (-4.2 °C), 1993 measured 20 cm above the ground surface soon after planting (June 10). Figure 18 details soil and air temperatures associated with the more severe frost event on June 20, 1993. The minimum air temperatures used in Figure 18 represent the average of 82 Figure 18. Minimum air (20 cm and 1 m) and soil (10 cm) temperatures associated with a frost event occurring soon after planting. the values recorded from PB, DT, and BB plots. Visually, several spruce seedlings showed some negative effects to these early frost events, including purpling of needle tips and browning of needles. No effects of frost were observed on pine seedlings. During frost events, air temperatures 1 m above the surface were always slightly warmer (0.5-1.5°C) than recorded at the 20-cm height, suggesting greater longwave radiative heat loss near the surface on this calm, clear night. Treatments having intact forest floor at the surface were least affected by the surface frost events in terms of changes in minimum 10-cm soil temperature (Figure 18). Figure 19 shows minimum air temperatures recorded 20-cm and 1-m above the ground surface during a single frost event early in the second growing season (1994). Approximately double the number of ground frosts were recorded at the 20-cm height 83 relative to 1-m height. Site preparation played a major role in minimizing the number of frosts events close to the ground surface (Figure 19). Frequency of frost events for the 20-cm air temperature between May 16 - Sept. 13 in the 1994 growing season increased in the order broadcast burn (16 events) < pile-and-burn-duff-intact (23 events) < untreated treatment (37 events). Only the untreated plot (20-cm) had air temperatures cold enough (<-4°C) to potentially cause needle injury to flushing seedlings (Orlander et al. 1990). Fowler and Helvey (1981) studied air temperatures at 10- and 50-cm above the ground surface after following several residue treatments and found that broadcast and pile burning treatments increased daily accumulations of heat into air whereas clearing (i.e. PB forest-floor-intact) treatments were equal to untreated plots. They also found that broadcast burn treatments had consistently warmer air temperatures close to the surface at sunrise, similar to the results found in this study. Although not presented, maximum air temperatures over the growing season follow the trend of 10-cm soil temperatures as discussed in the next section. 3.5.2 Soil temperature Several studies have found cold soil temperatures to be one of the most limiting features to early seedling growth in northern climates (Bassman 1991; Macadam 1991; Lopushinsky et al. 1992; Delucia 1986; Hellmers et al. 1970). This study wil l use the threshold soil temperature in the rooting zone of 8 ° C to represent the temperature at which stomatal conductance and photosynthesis in some conifer species has been shown to decline (Delucia 1986). Others have found that very slow rates of root growth to occur at rooting zone soil temperatures in the range of 2-10° C [Heninger and White (1974); Lopushinsky and Max (1990)]. Soil temperatures below 5°C are generally believed to be associated with dormant season for most conifers with minimal root growth occurring (Stathers and Spittlehouse 1990). Growing season daily maximum and minimum soil temperatures at 10-cm depth 84 Figure 19. Minimum air temperatures during the 1994 growing season. 85 below the surface are presented in Figure 20 for several selected treatments during the 1994 growing season. For reasons of clarity, the PB burned-pile and PB duff-intact microsites are not included in the Figure 20 but soil temperatures of these microsites reacted similar seasonally to those of the broadcast burn and DT between-trench-duff-intact microsites, respectively. The rate of soil warming in the spring wil l depend partially on the amount of cold soil water and snow present in/on soils during the spring runoff period. Non site prepared soils were always wet or saturated for several weeks after snow melt at this site owing to the insulating capacity of the slash/duff materials. Site preparation treatments at this site were noticed to have a strong influence on soil temperature. On clear warm days after June 20, maximum soil temperatures in the DT berm microsite were 12-15°C warmer than untreated soils. Highest soil temperatures were generally associated with well drained soil material (e.g. berms) having a high surface area of mineral material to conduct solar radiation more efficiently. Conversely, these microsites are subject to greater radiative cooling at night as indicated by daily minimum temperatures. Treatments with insulating forest floor layers present at the surface including DT between-trench-duff-intact, PB between-piles-duff-intact, PB burned-pile, untreated, and broadcast burn had much lower daily maximums relative to the berm and hinge microsites. Maximum soil temperature (10-cm depth) rarely exceeded 15 °C in microsites having a forest floor layer present and such low temperatures are believed associated with greater soil moisture and volumetric heat capacity present in these microsites. In addition, microsites with intact forest floors have lower thermal diffusivity and would remain colder for longer periods, despite warming air temperatures, relative to mineral-exposed microsites. Relatively high soil maximum temperatures occurred for the PB-under track microsite relative to the control treatment during the 1994 growing season and is believed as result of increased thermal conductivity of duff and mineral layers following compaction of mineral and forest floor materials. 86 Figure 20. Daily average soil (10-cm depth) minimum and maximum temperature over the 1994 growing season 87 Figure 21 depicts diurnal trends in 10-cm soil temperature during a typical warm, sunny day in mid-summer (July 22, 1994). Coldest daytime temperatures generally occur between 4:00 and 6:00 in the morning for all microsites or treatments. Superior soil warming occurs in the disc trench berm, hinge, and to a lesser extent, trench microsites, relative to other treatments where forest floor remained fully or partially intact at the surface (Figure 21)- Soil temperatures in the disc trench berm — — DT berm DT hinge -©— DT trench — DT duff inset a— Broadcast bum Untreated PB buned pile PB track o — PB duff intact Figure 21. Hourly average soil temperatures at the 10-cm depth for a 24-hour period during a warm day in July, 1994. and hinge microsites are in the range believed associated with optimum root growth (18-25 °C) for over half of the day, while untreated soils warmed little above 13 °C. The sum of daily hours of soil temperature greater than 8 degrees during the growing season for various treatments allows a comparison of the number of hours above a threshold temperature at which root growth and function should increase. Figure 22 compares cumulative total hours greater than 8 degrees in the 10-cm soil o o o o o o o o o o o o O O O O O O O O O O O C N J ' T C O O O O C M T C O O O C D C M i - t - T - T - i - C\J CM Time of day on July 22 88 layer for various treatments during a 2-week period early in the spring when soil temperatures are often limiting to root growth. This figure shows how all site preparation treatments have a distinct advantage over the untreated plots in terms of warming soils up to a level believed adequate for proper root growth (>8°C). Treatments which promote early spring soil drainage (i.e. DT berm, and hinge), appear to warm quicker, as do treatments which may have undergone some degree of soil compaction (e.g. DT duff intact and PB track). Conversely, treatments which contain greater amounts of moisture for longer periods or have greater slash/duff o m CD TJ CO (ft — $ O * E > u '5 •o 0) > E o 240 220 200 180 160 140 120 100 80 60 40 Untreated Broadcast DT-berm DT-hinge DT-trench DT-duff PB-bumed PB-lrack PB-duff bum intact pile intact Figure 22. Cumulative daily hours greater than 8 degrees at the 10-cm soil depth during an early 2-week spring period (May 17-31) in 1994. cover (e.g. untreated) would limit the rate of snow melt and evaporation, decrease thermal diffusivity (for wet soils) and increase the volumetric heat capacity. A l l these factors would lead to soils which would warm up slowly in the spring. Less obvious treatment differences to soil warming occur later in the growing season (e.g. late July to early August) as the soil dries out more uniformly across all microsites. 89 3.6 Tree Seedling Response Site preparation is commonly practiced to improve seedling survival and early growth response over untreated site conditions. As mentioned in previous chapters, many microclimate and nutritional factors are believed to be improved following site preparation. In particular, low soil temperatures and moisture stress (excessive and limited) are believed most important to survival and early establishment of seedlings in high elevation forests of the interior of British Columbia (Bassman 1989; Macadam 1991; McMinn 1982; Goldstein et al. 1985; Coates et al. 1991) and forest types under similar climatic conditions in Scandinavia (Orlander et al. 1990). Visual observations of seedlings for two growing seasons after planting would suggests that both of these factors were likely important to early growth performance in this study and that microsite alteration through site preparation may have influenced response. With the exception of the disc trench hinge microsite, microsites types were planted as they were encountered on the 2.5-m grid pattern. Sample size was generally inadequate to allow for statistical comparison of seedling response within microsites types of a given treatment (i.e. n<5) thus treatment effects are presented and compared. 3.6.1 Seedling condition over the first two years Pine and spruce seedlings planted in June 1993 were assessed for condition in October, 1993 and again, prior to growth initiation in May of 1994 to determine the influence of winter weather on seedling condition, mortality and other snow/ice effects on stem form. Results of seedling condition and stem form after the first growing season and winter are presented in Table 36 for pine and Table 37 for spruce. By the end of the first growing season (Oct., 1993) 35-67% of the pine seedlings and 16-45% of spruce seedlings were in good condition. Untreated plots had the greatest number of seedlings of both species in good condition after the first season. The first winter had a substantial effect on pine and spruce health and only 14-29% of 90 Table 36. SeecQing condition and damage codes for pine^ after one growing season and the following spring after one winter. Plot Good Good Fair Poor Dead Dead Chlorosis Snake 2 '93 '94 '94 '94 '93 '94 '94 '94 PB1 n/a 3 40 6 n/a 0 38 9 PB2 22 8 28 13 0 0 40 16 PB3 40 9 33 5 0 2 45 7 NT1 40 10 35 3 0 1 32 7 NT2 29 8 32 6 3 3 36 7 NT 3 30 4 42 3 0 0 42 9 BB1 35 9 31 5 3 4 32 16 BB2 30 17 23 4 4 5 23 1 BB3 39 13 29 4 2 3 22 3 DTI 16 10 35 3 0 1 27 6 DT2 30 25 21 3 0 0 19 14 DT3 40 7 35 6 0 1 40 8 1 Numbers represent number of seecllings representing a condition class prior to fill planting May 30, 1994 or at the end of the previous growing year (October, 1993). 2 Snake refers to "S-shaped" seedlings likely caused as a result of ice/snow creep. Table 37. Seedling condition and damage codes for spruce 1 after one growing season and the following spring after one winter. Plot Good Good Fair Poor Dead Dead Chlorosis Snake 2 '93 '94 '94 '94 '93 '94 '94 '94 PB1 n/a 4 42 3 n/a 0 27 21 PB2 8 4 38 4 1 3 41 20 PB3 8 11 34 3 1 1 41 16 NT1 22 3 45 1 0 0 38 19 NT2 27 6 40 3 0 0 37 22 NT3 17 3 37 7 2 2 37 12 BB1 17 12 32 4 1 1 33 20 BB2 13 13 33 1 1 2 31 7 BB3 19 4 41 2 1 2 37 3 DTI 48 12 34 3 0 0 22 11 DT2 n/a 10 34 5 n/a 0 31 17 DT3 25 26 23 0 0 0 31 21 Numbers represent number of seedlings representing a condition class prior to fill planting in May 30, 1994 or at the end of the previous growing year (October, 1993) Snake refers to "S-shaped" seedlings likely caused as a result of ice/snow creep 91 pine and 8-33% of spruce (depending on plot) were in good condition when assessed the following spring. Figure 23 reveals how seedling health of pine and spruce seedlings had dropped between the end of the first growing season (Oct., 1993) and at the start of the second growing season (May, 1994). Seedling health generally improved by the end of the second growing season and between 67-91% pine seedlings and 51-92% spruce seedlings were in good condition, depending on treatment, by September, 1994. The healthiest seedlings of both species were on the disc trench and broadcast burn plots after two growing seasons. The large majority of the spruce seedlings of the PB treatment in good condition were from the burned pile microsite. Overall seedling chlorosis of both species was most prominent in the PB Pine 100-r o o 80-cn _c c CO o 60 -rJling ondi 40--0) CU CO BB DT NT PB • Oct-93 • May-94 • Sep-94 Spruce •O 100 0 01 80 — o CO \J5 or . -60 "0 z s 40 0) 20 co I II I BB DT NT PB • Oct-93 O May-94 • Sep-94 Figure 23. Percentage of seedlings in good condition averaged for each of four treatments and three measurement periods during the first two growing seasons. treatment and least common in the BB treatment after the first winter. Snow press and ice-creep was a likely cause of "S-shaped" seedlings or "snaky" stem form after the 92 first winter. Spruce seedlings were 1.5 to 2 times more likely to sustain snaky stem form after one winter than pine (Figure 24). Survival after the first growing season was very high for both pine and spruce seedlings irrespective of treatment, ranging from 94-100%. The first winter resulted in killing an additional 1-2% of seedlings in most treatments. The survival results at this site would suggest that initial seedling survival is not perhaps as large an issue as factors influencing initial growth performance of seedlings. Figure 24. Percentage of seedlings having snaky stem form after the first winter for each treatment. 3.6.2 Height and root collar diameter growth Table 38 and 39 detail one-way A N O V A effects of treatment on pine and spruce seedling growth, respectively, over the first two growing seasons. The first growing season is generally seen as important to seedling establishment and survival and most energy goes into establishing a healthy root system followed by increased root, diameter and shoot extension in successive years. As 93 Table 38. A N O V A for treatment effects on pine seedling growth over the first two years. Total 1 Total Height Height Caliper height height increment increment '94 !93 '94 193 '94 Source df Pr>F Treatment 3 0.1154 0.0473 0.0777 0.0557 0.0910 Plot(treatment) 8 Er ror 2 576 Total 587 1 Based on sample size of n= 49 seedlings for both seasons 2 Plot nested within treatments was used as an error term for treatment effect Table 39. A N O V A for treatment effects on spruce seedling growth over the first two years. Total 1 Total Height Height Caliper height height increment increment '94 ^3 '94 !93 '94 Source df Pr>F Treatment 3 0.2073 0.0286 0.1807 0.0045 0.0271 Plot (treatment) 8 Error 2 396 Total 407 1 Based on sample size of n= 49 seedlings in 1993 and n=34 seedlings in 1994. 2 Plot nested in treatment was used as an error term. Error term df presented is for 1994 data. Error term df for 1993 data is same as for pine where df= 576. a result, the influence of the soil medium (including fertilizers) supplied within the plug confounds influence of treatments in the first year after planting. The PB treatment would appear to have conditions which are least favourable to first-year seedling growth relative to the other treatments studied. Relatively poor mean height increment and caliper growth associated with the PB treatment may have been caused by soil compaction during site preparation and slower soil warming (particularly in spring 94 periods) as a large majority of forest floor materials were left largely intact following piling. First-year height increment growth was greatest for both pine and spruce in the better drained plots (i.e. DT3, BB2, NT1, PB3), overall, where water tables dropped quicker and soils warmed up sooner in the growing season. Seedlings generally experienced very slow height growth over the first two years and treatment did not have a significant effect on first-year total and incremental seedling height growth of either species (Table 38 and 39). Although not significant, first-year average total height of spruce in the PB treatment (all microsites combined) was slightly less than other treatments. Height increment means (cm/yr) for seedlings in each treatment unit and averages for each treatment overall after two growing seasons are found in Table 40. By the end of the second season of growth, both pine and spruce seedlings appear to be starting to show some preference (statistically significant) to treatment. Treatment had a significant effect (p=0.0473) on pine (Table 38) and spruce (p=0.0286) (Table 39) total height by the end of the second growing season. The BB treatment contained significantly taller pine seedlings relative to the PB treatment (all microsites combined) after two growing seasons. Figure 25 compares height increment means by treatment between first and second season growth for pine. Mean height increment growth of pine was greater in the DT, NT and PB treatments after the first year and growth in the BB treatment was initially quite slow. By the end of the second growing season, treatment effect on pine height increment growth was marginally significant (p=0.0557) and most of these effects were accounted for by differences between the BB and PB treatments (Table 40). Average root collar diameter of pine and spruce seedlings after two growing seasons are summarized in Table 41. Treatment effects on diameter (caliper) of PI seedlings (Tables 38 and 41) were not significant (p=0.0910) after the second growing season, despite relatively large caliper differences between seedlings from the B B 95 Table 40. Average height increment (cm/yr) of seedlings in the second year after site preparation1. Pine Plot n Mean Std dev Std err Var CV Min Max BB 117 10.5 4.7 0.8 21.7 44.5 1.1 21.7 BB1 372 10.2 4.8 0.8 23.5 47.3 0.3 24.0 BB2 43 11.0 4.5 0.7 19.9 40.7 0.7 21.6 BB3 37 10.2 4.7 0.8 21.6 45.5 2.2 19.5 DT 132 7.9 4.1 0.6 18.4 52.3 0.6 17.3 DTI 44 6.4 3.4 0.5 11.8 54.0 0.0 12.3 DT2 46 11.7 5.9 0.9 34.7 50.3 2.3 27.2 DT3 42 5.6 2.9 0.5 8.7 52.5 -0.5 12.4 NT 132 6.1 3.3 0.5 11.1 54.7 1.0 15.8 NT1 44 7.2 3.9 0.6 14.8 53.6 1.8 20.0 NT2 43 6.3 3.2 0.5 10.3 50.5 1.5 14.5 NT3 45 4.7 2.8 0.4 8.1 60.0 -0.4 13.0 PB 129 5.3 3.3 0.5 11.2 64.4 0.0 17.4 PB1 45 4.4 2.9 0.4 8.6 66.4 0.6 18.4 PB2 48 4.7 3.6 0.5 13.2 76.7 -0.2 18.4 PB3 36 6.8 3.4 0.6 11.8 50.1 -0.5 15.5 Spruce Plot n Mean Std dev Std err Var CV Min Max BB 100 5.5 3.7 0.7 14.1 68.0 0.2 14.1 BB1 33 5.9 4.4 0.8 19.3 74.6 0.0 14.0 BB2 33 6.2 3.6 0.6 12.6 57.7 1.0 16.3 BB3 34 4.5 3.2 0.6 10.4 71.6 -0.3 12.0 DT 102 5.5 3.5 0.6 12.2 64.1 -1.0 17.4 DTI 34 6.3 3.2 0.6 10.4 51.0 0.5 16.5 DT2 34 4.7 3.7 0.6 13.6 78.0 -2.1 24.5 DT3 34 5.6 3.5 0.6 12.5 63.4 -1.3 11.3 NT 100 3.3 2.1 0.3 5.1 74.5 -0.1 9.4 NT1 33 2.2 1.8 0.3 3.1 80.3 -0.4 8.3 NT2 33 6.3 3.2 0.5 10.3 50.5 1.5 14.5 NT3 34 1.5 1.4 0.2 2.0 92.6 -1.3 5.5 PB 96 3.0 2.5 0.5 6.4 93.5 -0.9 8.8 PB1 33 2.6 2.1 0.4 4.4 82.5 -1.0 7.0 PB2 32 1.7 2.2 0.4 4.7 131.0 -2.3 7.5 PB3 31 4.8 3.2 0.6 10.2 66.9 0.5 12.0 Height increment means are averaged over all microsite types in each treatment 2 Different number of seedlings per plot as a result of removal of dead top/bud seedlings from data analysis 96 i e .£ 4 • 1 st year 2nd year broadcast burn pile and treatment Figure 25. Mean height increment growth of pine over the first two growing seasons. (Similar letters associated with each mean value are not significantly different using Tukey's Studentized Range test, p=0.05) treatment (7.8 mm) and untreated plots (5.5 mm). Spruce seedlings, similar to pine, didn't show significant growth response to treatments until the second growing season (Tables 39 and 41). Disc trench and broadcast burn height increments were significantly greater than untreated and PB treatments (Figure 26). Treatment had a significant Table 41. Average root collar diameter by treatment for pine and spruce after two growing seasons. T R E A T M E N T Pine M E A N (mm) STDERR (mm) Spruce M E A N (mm) STDERR (mm) Broadcast burn 7.8 0.23 8.0a 1 0.27 Disc trench 6.0 0.26 6.9ab 0.23 No treatment 5.5 0.17 5.4c 0.15 Pile and burn 6.1 0.21 6.3ab 0.18 iMeans with the same letter are not significantly different using Tukey's Studentized Range test, p=0.05 97 8 -r b u r n t r e a t m e n t b u r n Figure 26. Height increment growth of spruce over the first two growing seasons. (Similar letters associated with each mean value are not significantly different using Tukey's Studentized Range test, p=0.05) effect on second season spruce root collar diameter (Table 41). Spruce seedlings from broadcast burn plots had significantly larger mean calipers (8.5 mm) when compared to seedlings from untreated plots (5.4 mm) after the second growing season. 4 SUMMARY AND MANAGEMENT IMPLICATIONS Superior root collar diameter and height growth of both species on the broadcast burn plots are likely a consequence of improved nutrient availability in soils and some gains in soil and air warming in the first two years after burning. The results of this study agree with the work of others (Lopushinsky et al. 1992) who found improved growth of lodgepole pine and Douglas-fir in burned treatments relative to piling, cleared and untreated sites in a high elevation site in Washington. They attributed improved growth on burned sites to be related to adequate soil temperature for root growth, sufficient nutrient availability, minimal soil compaction and animal damage to 98 seedlings relative to other treatments. The disc trench hinge position also provided a good growing environment for seedlings as reflected in growth response after the second season. Several growth limiting factors were shown to have been reduced for seedlings grown on the DT hinge versus the NT and PB treatments, and include (1) improved soil warming in the rooting zone for longer duration, particularly early in the growing season, and, (2) improved nutrient status as a result of organic matter incorporation into the rooting medium of the hinge and berm microsites, and, (3) improved soil drainage, aeration, and significantly lower 0-20 cm fine fraction bulk density, presumably contributing to an improved environment for biological activity, root establishment and overall soil productivity. One main factor likely limiting growth performance of seedlings at this relatively fine textured site was excessive soil moisture (owing to minimum cover of transpiring vegetation cover for the first two years after harvesting) and high water tables, particularly in the first half of the growing season, despite the site being classified as moderately-well to well drained prior to logging. Water tables persisted within 20 cm of the surface often to mid-June in microsites where forest floor remained intact. This in turn increased the volumetric heat capacity of the soil and slowed soil warming until the moisture content of the soil dropped. Treatments which improved drainage (i.e. DT hinge and berm) resulted in warmer growing season soil temperatures and overall improved seedling health and growth. Others (von der Gonna 1989; McMinn 1982) have found that spot scalping treatments on wet and moist sites resulted in poorer height, diameter, and root growth of white spruce than in untreated sites but that the reverse was true on fresh or well drained sites. Macadam (1991) studied Sx seedling growth in inverted and mineral mounds from a well drained (and wetter) site in the SBSmc zone and found that elevated microsites had warmer and better drained rooting environments particularly in May-early June and ultimately produced the largest and 99 healthiest seedhngs. Slow first-year growth response for both pine and spruce seedlings in the DT hinge microsite may have been the result of periodic moisture stress affecting poorly established root system in newly planted seedlings. This problem would likely be even more evident on looser berm microsites with a larger percentage of organic material incorporated into the microsite, or similarly in raised microsites such as mounds observed by other workers (Macadam 1991; Bassman 1989; Orlander et al. 1990). Results of this study agree with others in that seedlings in raised microsites generally recover and outperform other treatments within 2-3 years in fresh to wet sites as extensive root systems thrive in the relatively porous environment. The berm and hinge microsites had 0-20 cm bulk density (effective rooting zone in first two growing seasons) significantly less than burned, pile-and-burn, or untreated treatments one growing season after treatment. Bulk density of the 0-20 cm layer of berm and hinge microsites decreased 50 and 24%, respectively relative to pre-treatment densities. Despite good seedling growth and condition of both pine and spruce in the PB burned-pile microsite, early treatment performance in the pile-and-burn treatment overall (all microsites included) was generally poorer after 2 growing seasons relative to other treatments studied. A 19% increase in bulk density of near surface (2-7 cm) mineral soil layers under excavator tracks (10% total surface area of the PB treatment) resulted in prolonged water ponding on the surface of tracks in spring months and generally poor seedling growth and condition. In this study, the trench microsite of the disc trench treatment (17% of total treated area) had 2-7 cm bulk density values 19% greater than no treatment plots. The denser soils associated with the trench microsite were also prone to water ponding for extended periods in early spring and the few seedlings which were planted in this microsite showed poor growth and condition relative to other disc trench microsites. Growth performance of pine and spruce seedhngs was poorest in untreated units 100 where slash was left intact. The insulating capacity of slash and forest floor layers limited soil drying and resulted in cold soils (<8 deg C) for longer periods than other treatments studied. This study revealed that the berm (and to a lesser extent, hinge) microsite minimized many of the factors believed limiting to early seedling growth in this subzone. The observation of soil moisture, temperature and nutrient status would suggest that the berm microsite of the disc trench treatment should be considered for operational planting (in addition to the more traditional hinge position), particularly in high elevation, moist subzones containing soils of medium to fine texture. 101 5 LITERATURE CITED Ballard, T .M. 1985. Spruce nutrition problems in the Central Interior and their relationship with site preparation. In Interior spruce seedling performance; state of the art. Proceedings of the Northern Silviculture Committee Workshop, February 5-6, 1985, Prince George, B.C. Ballard, T .M. and B.C. Hawkes. 1989. Effects of burning and mechanical site preparation on growth and nutrition of planted white spruce. For. Can., Pac. For. Cent., Victoria, B.C. BC-X-309. 19p. Banner, A., W. MacKenzie, S. Haeussler, S. Thompson, J . Pojar, and R. Trowbridge. 1993. A field guide to site identification and interpretation for the Prince Rupert Forest Region. B.C. Min . For., Smithers, B.C. Land Manage. Handb. 26. 503p. Bassman, J .H . 1989. Influence of two site preparation treatments on ecophysiology of planted Picea engelmannii x glauca seedlings. Can. J . For. Res. 19:1359-1370. Blackwell, B.A. 1989. Some ecological effects of operations used to convert densely stocked lodgepole pine stands into young pine plantations in west central British Columbia. M.Sc. thesis. University of British Columbia, Vancouver, B.C. 288p. Bosworth, B, and D. Studer. 1990. Comparison of tree height growth on broadcast burned, bulldozer-piled, and non-prepared sites 15 to 25 years after clearcut logging. Paper presented at the Symp. on Management and Productivity of Western-Montane Forest Soils, Boise, ID, Apr. 10-12, 1990. Brockley, R.P. and D. Yole. 1985. Growth response of lodgepole pine to operational fertilizer application near Burns Lake, B.C. Min . For. Res. Rep. RR85005-HQ. 21p. Cafferata, P. 1992. Soil compaction research. In Forest soils and riparian zone management: The contributions of Dr. Henry A. Froehlich to forestry. Nov. 17, 1992, Ore. State Univ., Corvallis, Ore. p. 8-22 Carr, W. W. 1988. Nutritional and soil compaction aspects of establishing forest cover on winter landings in the Fort St. James area. Can. For. Serv., B.C. Min . For., Victoria, B.C. F R D A Rep. 47. 6p. Coates, K.D. , W.H. Emmingham, and S.R. Radoseviche. 1991. Conifer seedling success and microclimate at different levels of herb and shrub cover in a Rhododendron-Vaccinium-Menziesia community of south central British Columbia. Can. J.For. Res. 21:858-866. Cochran ,P .H. and T. Brock. 1985. Soil compaction and initial height growth of planted ponderosa pine. U.S.D.A. For. Serv., Portland, Ore. Res. Note PNW-434. 4p. 102 Curran, M . and S. Thompson. 1991. Measuring soil disturbance following timber harvesting. B . C . Min . For., Nelson, B .C . Land Manage. Handb. Field Guide Insert 5. 25p. Debell, D.S. and C.W. Ralston. 1970. Release of nitrogen by burning light forest fuels. Soil Sci. Soc. Am. Proc. 34:936-938. Delucia, E . H . 1986. Effect of low root temperature on net photosynthesis, stomatal conductance, and carbohydrate concentration in Engelmann spruce seedlings. Tree physiology 2: 143-154. Dobbs, R . C . and R .G . McMinn. 1977. Effects of scalping on soil temperature and growth of white spruce seedlings. In Energy, water, and the physical environment of the soil: Sixth B . C . Soil Science Workshop. B .C . Min . Agric. Victoria, B . C . p.66-73. Donnelly , J. B. Shane, H .W. Yawney. 1991. Harvesting causes only minor changes in physical properties of an upland Vermont soil. J. For. 8:33-36. Dosskey, M . G . and T . M . Ballard. 1980. Resistance to water uptake by Douglas-fir seedlings in soils of different texture. Can J. For Res. 10: 530-534. Feller, M . C . 1982. The ecological effects of slashburning with particular reference to British Columbia: a literature review. B .C . Min. For., Land Manage. Rep. No. 13. 60p. Fleming, R . L . 1993. Effects of site preparation in interior plateau clearcuts on the soil water regime and the water relations of conifer seedlings. Ph.D. thesis. University of British Columbia. Vancouver, B .C . 191p. Fowler, W.B . and J.D. Helvey. 1981. Soil and air temperature and biomass after residue treatment. USDA For. Serv. , Pac. Northwest For. and Range Exp. Stn., Portland, Ore. Res. Note PNW-383. 8p. Froehlich, H . A . 1979. Soil compaction from logging equipment: effects on growth of young ponderosa pine. J. Soil Water Cons. 34:276-278. Goldstein, G . H . , L . B . Grabaker and T . M . Hinckley. 1985. Water relations of white spruce (Picea glauca (Moench) Voss) at tree line in north central Alaska. Can. J. For. Res. 15: 1080-1087. Goulding, K . W. T. and P. A . Stevens. 1988. Potassium reserves in a forested, acid upland soil and the effect on them of clear-felling versus whole tree harvesting. Soil Use and Mgt. 4:45-51. Green, R . N . , R . L . Trowbridge, K . Klinka. 1993. Towards a taxonomic classification of humus forms. For. Sci. Monogr. 29:1-48. 103 Haase, S.M. 1986. Effect of prescribed burning on soil moisture and germination of Southwestern ponderosa pine seed on basaltic soils. U.S.D.A. For. Serv. Rocky Mount. Exp. Sta. RM-462. 6p. Harvey A.E . , M.F . Jurgensen, M . J . Larsen and R.T. Graham. 1987. Decayed organic materials and soil quality in the Inland Northwest: A management opportunity. U.S.D.A. For. Serv., Int. Res. Sta., Ogden, Utah. Gen. Tech Rep. INT-225. 15p. Havranek, W . M . and U . Benecke. 1978. The influence of soil moisture on water potential, transpiration and photosynthesis of conifer seedlings. Plant and soil 49:91-103. Heilman, P. 1981. Root penetration of Douglas-fir seedlings into compacted soil. For. Sci. 27:660-666. Hellmers, H . , M . K . Genthe, F. Ronco. 1970. Temperature affects growth and development of Engelmann spruce. For. Sci. 16:447-452. Heninger, R.L. and D.P. White. 1974. Tree seedling growth at different soil temperatures. For. Sci. 20:363-367. Hillel, D. 1982. Introduction to soil physics. Acedemic Press, Orlando, Fla. 288p. Johnson, C.E., A . H . Johnson, T.G. Huntington, T.G. Siccama. 1991. Whole tree clear-cutting effects on soil horizons and organic-matter pools. Soil Sci. Soc. Am. J . 55:497-502. Kaufmann,M.R. 1975. Leaf water stress in Engelmann spruce - influence of the root and shoot environments. Plant Physiol. 56:841-844. Klinka, K., R .N. Green, R.L. Trowbridge andL.E. Lowe. 1981. Taxonomic classification of humus forms in ecosystems of British Columbia. B.C. Min . For., Victoria, B.C. Land Manage. Rep. 8. 54p. FLrajina, V J . 1969. Ecology of Forest trees in British Columbia. Ecol. West. N . Ameri. 2: 1-146. Levan, M . and S.J. Riha. 1985. Response of root systems of northern conifer transplants to flooding. Can. J . For. Res. 16: 42-46. Lewis, T. and W.W. Carr. 1993. Hazard assessment keys for evaluating site sensitivity to soil-degrading processes - Interior sites. B.C. Min . For., Victoria, B.C. Land Manage. Handb. 8. 16p. Lieffers, V . J . and R.L. Rothwell. 1986. Effects of depth of water table and substrate temperature on root and top growth of Vicea mariana and Larix laricina seedlings . Can. J . For. Res. 16:1201-1206. 104 Lindeburgh, S.B. 1990. Effects of prescribed fire on site productivity: A literature review. B.C. Min . For., Victoria, B.C. Land Manage. Rep. 66. 15p. Lopushinsky, W. and M.R. Kaufmann. 1984. Effects of cold soil on water relations and spring growth of Douglas-fir seedlings. For. Sci. 30:628-634. Lopushinsky, W, and G.O. FQock. 1974. Transpiration of conifer seedlings in relation to soil water potential. For.Sci. 20:181-186. Lopushinsky, W. and T.A. Max. 1990. Effect of soil temperature on root and shoot growth and on budburst timing in conifer seedling transplants. New Forests 4: 107-124. Lopushinsky, W, D. Zabowski and T.D. Anderson. 1992. Early survival and height growth of Douglas-fir and lodgepole pine seedlings and variations in site factors following treatment of logging slash residues. U.S.D.A. For. Serv., Pacific Northwest Sta., Portland, Ore. Res. Pap. PNW-451. 22p. Luttmerding, H.A., D.A. Demarchi, E.C. Lea, D.V. Meidinger, and T. Void. 1990. Describing ecosystems in the field. 2nd ed. B.C. Min . Environ, and Min . For., Victoria, B.C. Min . Environ. Manual 11. Macadam, A . M . 1987. Effects of slash burning on fuels and soil chemical properties in the Sub-boreal Spruce Zone of central British Columbia. Can J . For. Res. 17:1577-1584. . 1989. Effects of prescribed fire on forest soils - A training manual. Ministry of Forests, Research Section, Prince Rupert Forest Region, Smithers, B.C. . Unpubl. Rep. 15p. . 1991. Effects of microsite alteration on soil climate, nitrogen mineralization, and establishment of Picea glauca x engelmannii seedlings in the Sub-boreal Spruce Zone of west-central British Columbia. M.Sc. thesis. O.S. Univ, Corvallis, Ore. Melton, L and S. Childs. 1989. Low soil moisture can increase seedling frost damage. FIR Rpt. 10(4):2-4. Medford, Ore. 2p. Minore, D., C.E. Smith, and R.F. Woollard. 1969. Effects of high soil density on seedling root growth of seven northwestern tree species. U.S.D.A. For. Serv., NW For. Range Exp. Sta., Res. Note PNW-112. 6p. Minore, D. 1986. Effects of site preparation on seedling growth: a preliminary comparison of broadcast burning and pile burning. U.S.D.A. For. Serv., NW For. Range Exp. Sta., Res. Note PNW-452. lOp. Morris L .A. and R.F. Lowery. 1988. Influence of site preparation on soil conditions affecting stand establishment and tree growth. S. J . Appl. For. 12:170-173. 105 Mroz, G.D., M.F . Jurgensen, A . H . Harvey and M . J . Larsen. 1980. Effects of fire on nitrogen in forest floor horizons. Soil Sci. Soc. Am. J . 44:395-400. McMinn, R.G. 1982. Ecology of site preparation to improve performance of planted white spruce in northern latitudes. Paper presented at Third Annual Workshop, International Committee on Regeneration of North Latitude Forest Lands, IUFRO W.P.S 1.05-08, Prince George, B.C. Aug. 30-Sept. 1, 1981. Orlander, G. and K. Due. 1986. Water relations of seedlings of Scots pine grown in peat as a function of soil water potential and soil temperature. Studia For. Suedica No. 175. Orlander, G.P. Gemmel, and J . Hunt. 1990. Site preparation: A Swedish overview. B.C. Min . For., Research Branch, Victoria, B.C. FRDA Rep. 105. 61p. Page-Dumroese, D.S., M.F . Jurgensen, R.T. Graham, A .E . Harvey. 1986. Soil physical properties of raised planting beds in a Northern Idaho Forest. U.S.D.A. For. Serv. Ogden, UT, Res. Pap. INT-360. 6p. Plaut, Z. 1973. The effect of soil moisture tension and nitrogen supply on nitrate reduction and accumulation in wheat seedlings. Plant and Soil 38:81-94. Running, S.W. and C P . Reid. 1980. Soil temperature influences on root resistances ofPinus contorta seedlings. Plant Physiol. 65: 635-640. Russell, E . J . 1988. Soil conditions and plant growth. Ed. A. Wild. Dept. Soil Sci. Univ. Reading, John Wiley and Sons, New York. 588p. SAS Institute Inc. 1988. SAS/STAT user's guide, release 6.03 ed. Cary, NC. SAS Institute Inc. 1028p. Stathers, R.L. and D.L. Spittlehouse. 1990. Forest soil temperature manual. For. Can. and B.C. Min . For., Victoria, B.C. FRDA Report 130. 47p. Taylor, S.W. and M.C. Feller. 1987. Initial effects of slashburning on the nutrient status of Sub-boreal Spruce Zone ecosystems. In papers presented at the Fire Management Symposium, April 8-9, 1987, Prince George, B.C. Central Interior Fire Protection Committee, Smithers, B.C. Trowbridge, R., B. Hawkes, A. Macadam, J . Parminter. 1987. Field handbook for prescribed fire assessments in British Columbia: logging slash fuels. B.C. Min . For., Victoria, B.C. Land Manage. Hand. 11. 63p. von der Gonna, M.A. 1989. First year performance and root egress of white spruce and lodgepole pine seedlings in mechanically prepared and untreated planting spots in north central British Columbia. M.Sc. thesis, Univ. of B.C. 130p. 106 Weetman,G.F. 1987. The importance of forest humus manipulation in silviculture practice. Paper presented at B.C. Soil Science Workshop, U.B.C., Vancouver, B.C. Unpubl. Rep. 21 p. Zar, J .H . 1984. Biostatistical analysis. Prentice Hall , Englewood Cliffs, N J . 718pp. Personal Communications Dunbar, J . 1993. Assistant Regional Site Preparation Coordinator, Min . For., Regional Silviculture Section PRFR, Smithers, B.C. Dunbar, M . 1992. Houston Forest Products, Houston, B.C. Kranabetter, M . 1995. Min . For., Forest Sciences Section, PRFR, Smithers, B.C. Pinkerton, J . 1993. Regional Site Preparation Coordinator, Min . For., Regional Silviculture Section, PRFR, Smithers, B.C. Wong, J . 1993. Woodmere Nursery, Telkwa, B.C. 107 '1 s s w o O <N 9 -e ° GO « <N CJ I *-' o d d GO o cn o d cn 00 •5 a , cj T3 I I 1-^ <—4 6 3 6 cd ca o a . (N P< >n l « o l « o I * '3 .3 £ 2 11 Q u 53 13 S t 3 -rt 3 H J cn 0 0 .3 Q W £2 es S Q w d « Da oa J o §3 Q d is 1 l l o cn O K cn 8 8, I a co a, cu J2 CJ o o co o TT •O in o o oa oa <N oa oa oa H Q P Q Q 108 o 5 1 us « © •3 CH a § 8 g. cct o l 5 g I o a, s o 00 CN Si, SiC SiL SiL SiL.SiC 1 SiL _ _ _ H 3 e 3 g compact at 30 ci compact at 30 ci compact at 30 ci compact 38 cm . none none M 1-1 Q P o o • J O CH pa CN m PH q 1-H o PH _1 -i d ^ CH PH 03 H J O a er er er low upp low CJ 9* o Q O CH 73 CJ 1) 4* • ^ <*H a. o I t 13 d O O CH € CO O 2 *H r-v CH W .a «> - O 1H •a g S 2 "i" < 00 ? C3 c^ -C/ Co W "jH 2P~ £ .9 a S ^ co o III 5 2 5 r— CN r-i u-i 109 Appendix 2. Analytical procedures used for Regan Creek soil samples (Ministry of Forests, Forest Sciences Lab, Victoria, B.C.). The laboratory uses two references for methods including: Carter, M.R. 1993. Soil sampling and methods of analysis. Kalra and Maynard. 1991. Forestry Canada Inf. Rep. NOR-X-319. Soil Analysis Exchageable cations and CEC Neutral ammonium acetate extraction of cations followed by ethonal rinse to remove unbound ammonium; sodium chloride (10%) extraction of retained ammonium for CEC. Final determination of cations by ICP and ammonium by Technicon Auto-analyzer. Available phosphorus Bray P I procedure (TJV/Visible spectrophotometer) Total C, N, and S Combustion elemental analysis (Leco method) Mineralizable N 2-week, 30 degree C anaerobic incubation followed by I N K C L extraction of ammonium-N. Ammonium-N analyzed by a Technicon Auto-analyzer. pH p H (deionized water, 1:1 for mineral; 2:1 or 3:1 for organics) and 0.01N Calcium chloride Forest floor total nutrients Totals of Ca, K, Mg, P, B, Fe, Zn, Cu, and Mn. Involves microwave digestion with l g sample treated 10ml of a series of concentrated acids including H N 0 3 , H 2 P 0 4 , A N D H C L followed by analysis of soluble digests by ICP. 110 Appendix 3. Mean forest floor nutrient concentrations in various treatments after harvesting and prior to site preparation (standard error in brackets). Variable Units Broadcast Burn Disc Trench Untreated Pile & Burn Exch Ca (cmolc/kg) 20.9 (1.32) 25.9 (2.08) 19.9 (1.17) 23.5 (1.91) Exch K cmolc/kg) 3.4 (0.11) 3.2 (0.18) 3.9 (0.12) 3.2 (0.11) Exch Mg cmolc/kg) 3.9 (0.22) 4.5 (0.23) 4.1 (0.12) 4.4 (0.16) CEC cmolc/kg) 104 (2.9) 105 (2.0) 97 (2.3) 104 (2.5) Min N (mg/kg) 576 (19.7) 566 (29.3) 609 (22.3) 621 (21.6) Avail P (mg/kg) 117 (7.71b1 138 (8.7)b 178 (7.4)a 130 (5.8)b pH(H 20) 4.29 (0.03) 4.41 (0.07) 4.27 (0.03) 4.43 (0.04) pH(CaCI2) 3.75 (0.04) 3.85 (0.07) 3.72 (0.04) 3.87 (0.05) Total C (cg/g) 49.3 (0.64) 48.4 (0.59) 50.0 (0.69) 49.0 (0.84) Total N (cg/g) 1.25 (0.02) 1.31 (0.03) 1.21 (0.02) 1.30 (0.02) C/N (cg/g) 39.0 (0.53) 37.3 (0.94) 41.7 (0.72) 37.7 (0.48) Total Ca (cg/g) 0.58 (0.053) 0.70 (0.063) 0.53 (0.033) 0.61 (0.040) Total K (cg/g) 0.16 (0.014)b 0.16 (0.01 Dab 0.19 (0.010)a 0.14 (0.007)b Total Mg (cg/g) 0.074 (0.005) 0.090 (0.005) 0.076 (0.004) 0.084 (0.004) Total P (cg/g) 0.13 (0.003)b 0.14 (0.004)ab 0.15 (0.003)ab 0.16 (0.005)a Total S (cg/g) 0.123 (0.005) 0.131(0.005) 0.125 (0.004) 0.128 (0.005) Total Cu (mg/kg) 9.1 (0.39) 10.1 (0.54) 9.5 (0.59) 9.9 (0.52) Total Fe (mg/kg) 2992 (342) 3830 (96.5) 3152 (305) 4183 (254) Total Zn (mg/kg) 49.9 (4.05) 47.9 (4.53) 49.5 (7.97) 48.3 (4.42) Total Mn (mg/kg) 872 (178) 1261 (313) 957 (195) 1221 (313) 1 For different microsite types, means followed by the same letter are not significantly different (Tukey's Studentized Range (HSD) test , P<0.05 level. I l l 

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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0086975/manifest

Comment

Related Items