Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Factors limiting early conifer growth in salal-dominated cutovers on northern Vancouver Island, British… Messier, Christian 1991

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

Item Metadata

Download

Media
831-UBC_1991_A1 M47.pdf [ 10.51MB ]
Metadata
JSON: 831-1.0100580.json
JSON-LD: 831-1.0100580-ld.json
RDF/XML (Pretty): 831-1.0100580-rdf.xml
RDF/JSON: 831-1.0100580-rdf.json
Turtle: 831-1.0100580-turtle.txt
N-Triples: 831-1.0100580-rdf-ntriples.txt
Original Record: 831-1.0100580-source.json
Full Text
831-1.0100580-fulltext.txt
Citation
831-1.0100580.ris

Full Text

FACTORS LIMITING EARLY CONIFER GROWTH IN SALAL-DOMINATED CUTOVERS ON NORTHERN VANCOUVER ISLAND, BRITISH COLUMBIA by CHRISTIAN MESSIER B.Sc.F., Universite Laval, 1984 M.Sc.F., Universite Laval, 1986 A THESIS SUBMITTED IN PARTIAL F U L F I L M E N T OF T H E REQUIREMENTS FOR T H E D E G R E E OF DOCTOR OF PHILOSOPHY in T H E F A C U L T Y OF GRADUATE STUDIES DEPARTMENT OF FOREST SCIENCES We accept this thesis as conforming to the required standard T H E UNIVERSITY OF BRITISH COLUMBIA FEBRUARY 1991 (c) Christian Messier, 1991 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 Date _ ^ DE-6 (2/88) Je dedie cette these d ma petite fille, Emilie, nee Ie 15 mars 1991. i i ABSTRACT Nutritional stress has been reported in planted and naturally-regenerated conifers growing in association with an ericaceous species, salal (Gaultheria  shallon Pursh), in cutovers previously occupied by old-growth western redcedar (Thuja plicata Donn) and western hemlock (Tsuga heterophylla (Raf.) Sarg.) (CH sites) on northern Vancouver Island. No such stress was apparent in cutovers previously occupied by natural, second growth western hemlock and amabilis fir (Abies amabilis (Dougl.) Forbes) stands (HA sites) that developed following windthrow in 1907. The CH ecosystem type occupies as much as 100,000 ha in coastal British Columbia. In the spring of 1987, a series of field and pot experiments was initiated to investigate some of the ecological processes affecting the early growth of conifers on recently logged and burned 2- to 10-year-old CH and 2- to 4-year-old HA sites. The overall objective of the research was to quantify some of the possible factors limiting early conifer growth on northern Vancouver Island. The research encompassed studies of: (1) below- and above-ground non-crop vegetation recovery, forest floor nutrient availability and soil microenvironmental modification following clear-cutting and burning; (2) competition for nutrients by the non-crop vegetation; (3) interference by salal of the mycorrhizal development on conifer seedling roots; (4) conifer seedling growth under several different experimental conditions, and; (5) relationship between microsite factors and western redcedar seedling growth within clear-cut and burned CH sites. Salal was the main non-crop species found on the CH sites. It reestablished itself rapidly, both above- and below-ground, following clear-cutting and burning on this type of site. The total above-ground vegetation biomass quadrupled from 1372 kg ha"1 on the 2-year-old CH sites to 5574 kg ha - 1 on the 8-year-old CH i i i sites, whereas the total below-ground biomass increased six times from 1908 kg ha _ l on the 2-year-old C H sites to 11415 kg ha'l on the 8-year-old C H sites. Similar amounts of total above-ground non-crop biomass were found on HA cutovers for the first 4 years, but the non-crop vegetation was composed of half salal half fireweed (Epilobium angustifolium L.). The regrowth of the non-crop vegetation immobilized annually 9 and 0.9 kg ha'l of N and P, respectively, on the C H sites during the first eight years. This was estimated to represent potentially between 30 and 45% of the available N on these sites. A model of the development of live fine-root, leaf, stem and rhizome biomass of salal over a 60 year period is proposed based on the result of this study and of other studies. This model suggests that the net immobilization of nutrients in salal biomass will cease between 10 and 20 years after clear-cutting and burning as salal stops expanding its biomass. This model implies that the nutritional stress caused by salal should decline after 10 to 20 years at which time conifer nutrition and growth should slowly improve, especially as salal is eliminated by light competition as the conifer canopy begins to close. Several biotic and abiotic factors were studied in relation to their possible effects on the early growth of western hemlock, Sitka spruce and western redcedar. The growth of these three conifer species was compared between 2- to 4-year-old C H cutovers (2+B CH), 8- to 10-year-old C H cutovers (8+B CH), and 2- to 4-year-old H A cutovers (2+B HA) with and without the influence of the non-crop vegetation. No moisture deficit was measured in the field throughout the year on all sites. Both the growth of the conifers and the availability of N (+36%) and P (+25%) were increased by the removal of the non-crop vegetation; however, no difference in cellulose decomposition and soil moisture, and only very small i v difference in soil temperature was measured. The better conifer growth on 2+B H A > 2+B C H > 8+B C H without the influence of the non-crop vegetation was associated with an overall better forest floor nutrient status; however, no difference in soil moisture and p H , and only small difference in soil temperature were measured. Western redcedar and western hemlock were the best growing species on C H and H A sites, respectively. Both western hemlock and Sitka spruce (Picea sitchensis) were very responsive to the different site conditions ( C H vs H A sites) and to planting treatments that increased or decreased conventional measures of nutrient availability caused by the different treatments, whereas western redcedar was not. The presence of salal was found to have no effect on the total percent mycorrhizae found on the roots of the three conifer species studied three years after planting. Both field and pot experiments yielded comparable results. Slow-release fertilizer, at the time of planting, increased growth only for the first two years after application. Western redcedar growth was significantly greater in depressions than on flats and mounds, but this difference was not related to any major differences in the forest floor variables measured between the three microtopographic positions. These results indicate that the nutritional stress and poor growth reported in conifers, especially in Sitka spruce, on C H cutovers on northern Vancouver Island can be explained by a combination of (1) inherently low forest floor fertility in cutovers originating from the old-growth C H forests, (2) salal competition for scarce nutrients and their immobilization in salal biomass, and (3) declining site fertility caused by the termination of the flush of nutrients that occurs in the immediate post-logging and burning period on C H sites. V TABLE OF CONTENTS Page Abstract. . ii Table of contents v List of tables viii List of figures x Acknowledgements xiii CHAPTER ONE 1 Problem definition and objectives 1 The problem 1 Hypotheses concerning poor tree growth on CH sites 5 Study objectives and research approach 8 CHAPTER TWO 13 Study area and ecosystem description 13 CHAPTER THREE . 19 Above- and below-ground vegetation recovery following clear-cutting and slashburning on CH sites 19 Introduction 19 Materials and methods 23 Research sites 23 Below-ground biomass 24 Above-ground biomass, percentage cover and leaf area 26 Statistical analyses 26 Results and discussion 28 Above-ground biomass, percentage cover and leaf area 28 Below-ground biomass and distribution 32 Growth allocation strategy during the period of vegetation recovery following clear-cutting and burning 38 Ecological and silvicultural implications 41 Summary 48 CHAPTER FOUR 49 Growth stress in Picea sitchensis plantations on the CH sites 49 Introduction 49 Materials and methods 51 Research sites 51 Field seedling bioassays 51 v i Pot seedling bioassays 54 Nutrient content in non-crop vegetation biomass 55 - Forest floor properties and microenvironment 56 Results 60 Field seedling bioassays 60 Nutrient content of non-crop vegetation 64 Forest floor properties and microenvironment 67 Pot seedling bioassays 69 Discussion 74 Summary 79 CHAPTER FIVE 81 Effects of competing vegetation, slow-release fertilizer, and different site conditions on the early growth of western hemlock, western redcedar, and Sitka spruce seedlings on CH and HA sites 81 Introduction 81 Materials and methods 83 Field seedling bioassays 83 Pot seedling bioassays 87 Forest floor properties 88 Non-crop vegetation 88 Results 90 Field seedling bioassay 92 Shoot and root biomass 100 Mycorrhizal colonization 100 Pot seedling bioassay 102 Discussion 104 Conifer seedling growth in contrasting site conditions 104 2+B CH vs 8+B CH sites 104 2+B HA vs 2+B CH sites 107 Seedling growth and competing vegetation I l l Slow-release fertilizer 113 Summary 116 CHAPTER SIX 118 Relationships between microsite factors and western redcedar growth on CH sites 118 Introduction 118 Materials and methods 120 v i i Study area 120 Microsite selection 120 Western redcedar growth 120 Forest floor factors 121 Fireweed and salal vegetation 121 Statistical analyses 122 Results and Discussion 123 Relationships between microtopography and forest floor nutrient status 123 Relationships between western redcedar growth and microsite factors 123 Relationships between fireweed and forest floor nutrient status 129 Summary 133 CHAPTER SEVEN 135 General discussion and concluding remarks 135 Conditions prior to disturbance 135 Regrowth of salal and fireweed following clear-cutting and burning 137 Possible factors limiting early conifer growth on CH cutovers 139 1. Competition by salal 139 2. Inhibition of mycorrhizal infection by salal 141 3. Short-term flush of nutrients on CH sites 142 4. Ha vs CH site conditions 142 5. Microtopography on CH cutovers 143 6. Conifer species . . 144 Long-term effects of salal 145 Additional research needs 147 Limitations of this research 147 Literature cited 149 v i i i L I S T O F T A B L E S Table 1. Experimental treatments and abiotic and biotic factors investigated in this thesis 11 Table 2. Typical soil profile description of the CH ecosystem 16 Table 3. Typical soil profile description of the HA ecosystem 18 Table 4. Mean above-ground biomass (kg ha"l) and leaf area index (m^  m"2) of the vegetation on 2-, 4- and 8-year-old intensively-studied CH site ages 30 Table 5. Vertical and diameter-class distribution of below-ground biomass of the Gaultheria-Vaccinium and Epilobium-Cornus species groups on the 2-, 4- and 8-year-old intensively-studied CH site ages 35 Table 6. ANOVA summary table showing variance ratios (F), P-values and error mean-square for Sitka spruce height increments in 1987, 1988, 1989 between the two planting treatments and two site ages 62 Table 7. Comparison of percent mycorrhizal colonization on Sitka spruce fine-roots between the vegetation-not-removed and vegetation-removed planting treatments on the 2-year-old CH sites 65 Table 8. Comparison of some forest floor properties between the 3 and 9 years after clear-cutting and burning CH site ages 68 Table 9. Comparison of soil temperatures at depths of 3, 10 and 25 cm between the 2 and 8 years after clear-cutting and burning CH site ages 70 Table 10. Comparison of planted conifers, main non-crop species, forest floor properties, and soil temperature between three site-age combinations 91 Table 11. ANOVA summary table showing variance ratios (F), P-values, and error-mean-square for total height and diameter increments of western hemlock, western redcedar ix and Sitka spruce over three growing seasons between the three planting treatments and three site-age combinations . . . 94 Table 12. Comparison of the shoot and root dry mass (g) and shoot/root ratios of 3-year-old western hemlock, western redcedar and Sitka spruce seedlings for the vegetation-not-removed treatment between the 2+B CH and HA sites 101 Table 13. Comparison of different forest floor properties between the three microtopographic positions: mounds, flats and depressions. 124 Table 14. Pearson correlation matrix for western redcedar height and diameter increment, salal abundance, fireweed abundance and vigour, and forest floor variables.. . 127 X L I S T O F F I G U R E S Figure lr View of the HA and CH forest ecosystems before and 10 years after clear-cutting and burning 2 Figure 2. View of a 16-year-old Sitka spruce plantation established on CH sites 3 Figure 3. Planted 9-year-old western redcedar growing on clear-cut and burned CH sites 5 Figure 4. Flow-chart showing the contributions of the different chapters of this thesis to the overall research effort being understaken by a team of researchers (i.e. SCHIRP) to understand the factors limiting early conifer growth on CH cutovers on Northern Vancouver Island 10 Figure 5. Location of the study area on northern Vancouver Island, British Columbia 14 Figure 6. Percent cover of salal on the 2-, 4-, and 8-year-old intensively-studied cutovers and on 11 other cutovers representative of the population of cutovers found in the CH ecosystem near the research sites 29 Figure 7. Changes in the percentage cover of the vegetation during the first 10 years following clear-cutting and burning based on 3 years of data from 2 different site ages: 2-to 4- and 8- to 10-years post-logging and burning site ages 33 Figure 8. Comparison of the vertical distribution of (A) fine-root (0-2 mm) and (B) new rhizome (>2 mm) biomass of Gaultheria-Vaccinium and Epilobium-Cornus species groups within the upper 45 cm of the forest floor of 8-year-old CH site age 36 Figure 9. Above- and below-ground biomass for the (A) Gaultheria- Vaccinium and (B) Epilobium-Cornus species groups on 2-, 4-, and 8-year-old intensively-studied CH site ages 39 x i Figure 10. Hypothetical development of live fine-root, leaf, stem and rhizome relative abundance of salal over a 60 year period following the clear-cutting and burning of old-growth forests of western redcedar and western hemlock (CH) on northern Vancouver Island 45 Figure 11. Sitka spruce annual height and diameter increments on the 2 and 8 years after clear-cutting and burning CH site ages using two treatments: vegetation-not-removed and vegetation-removed treatments 61 Figure 12. Comparison of the height and diameter increments of Sitka spruce after three growing seasons between the vegetation-not-removed and vegetation-removed planting treatments 63 Figure 13. Total amount of N and P contained in above-ground (shoot) and below-ground (root + rhizome) biomass of non-crop vegetation on the 2, 4, and 8 years after clear-cutting and burning CH sites 66 Figure 14. Comparison of height and diameter increments over two growing seasons of Sitka spruce seedlings growing in pots containing forest floor materials taken from the 3 and 9 years after clear-cutting and burning CH site ages between the different spruce-salal combinations 71 Figure 15. Comparison of height and diameter increments over two growing seasons of Sitka spruce seedlings growing in pots between the forest floor materials taken from the 1, 3 and 9 years after clear-cutting and burning CH site ages and two different forest floor depths 73 Figure 16. Vegetation-removed treatment applied to western redcedar growing on the 2 years after clear-cutting and burning CH sites 85 Figure 17. Comparison of total height (A to C) and diameter (D to F) increments of western hemlock, western redcedar and Sitka spruce seedlings after three growing x i i seasons between the three planting treatments and three site-age combinations 93 Figure 18. Comparison of height and diameter increments over three growing seasons between the three conifer species and three site-age combinations 95 Figure 19. Comparison of the annual height increment of western hemlock, western redcedar and Sitka spruce seedlings for the first three years after planting on 8+B CH, 2+B CH, and 2+B HA sites between the vegetation-not-removed, vegetation-removed, and fertilized planting treatments 97 Figure 20. Symptoms of nutrient deficiency in Sitka spruce seedlings growing on the 2+B CH site appearing on the third year after planting for the fertilized treatment 99 Figure 21. Comparison of the 1988 and 1989 annual height and diameter increments of Sitka spruce and western redcedar seedlings growing in pots containing forest floor materials taken from the 2+B HA, 2+B CH, and 8+B CH sites 103 Figure 22. Comparison of height and diameter increments of western redcedar seedlings in 1989 (three years after planting) between mounds, flats and depressions 125 Figure 23. Percentage cover and average height of fireweed between mounds, flats and depressions 130 Figure 24. Relationships between the experimental treatments and biotic and abiotic factors investigated in this thesis 140 x i i i A C K N O W L E D G E M E N T S I am grateful to my supervisor, Dr. J.P. Kimmins, for his foresight, his editorial contribution, his continual encouragement and his never ending energy. Helpful advice and support was obtained from the other members of my supervisory committee: Drs. Ballard, Klinka, and Bunnell. This project would not have been possible without the dedication of several people. I wish to thank the following individuals: K.J. Mackenzie, E. Morton, S. Williams, T. Honer, C. Trethewey, R. Oran, A. Quinde, H. Granander, J. Glaubitz, R. Keenan, L. Ruddick, G. Glover, and P. Warnes. Thanks are also due to A. Ruth for the mycorrhizal work and M. Tsze for the laboratory work. Finally, I would like to thank B. Dumont, M. Watkinson, S. Joyce and P. Bavis for their helpful discussion and continual support and encouragement during the "Port McNeill" work. Western Forest Products Ltd. kindly provided lodging facilities and a multitude of other services. This research was supported through a Forest Resource Development Agreement (FRDA) contract 2.31, a National Sciences and Engineering Research Council of Canada scholarship and a B.C. Science Council GREAT scholarship. Un merci tres special a mes parents pour leur continuel encouragement. Finally, a very special "MERCI" to Lana, for her love, patience, understanding, and positive attitude throughout this dissertation. 1 CHAPTER ONE PROBLEM DEFINITION AND OBJECTIVES THE PROBLEM In the windward, submontane, wetter variant of the wet subzone (CWHbyjjj) of the Coastal Western Hemlock biogeoclimatic zone on northern Vancouver Island (Green et al. 1984), two very different kinds of forest ecosystems occur side by side on what is believed to be a single forest ecosystem association (Lewis 1982). The first of these two forest ecosystems, an old-growth western redcedar-western hemlock ecosystem (hereafter referred to as the CH) occurs extensively on the east and west side of northern Vancouver Island, and is characterized by open stands of western redcedar (Thuja plicata Donn) and western hemlock (Tsuga  heterophylla (Raf.) Sarg.) (Figure 1). Although individual trees are periodically windthrown, the CH ecosystem does not appear to have experienced catastrophic windthrow for a long period of time (probably more than 1000 years). Following clear-cutting and burning, tree growth on CH sites is often good for a few years but soon becomes poor. The ericaceous shrub salal (Gaultheria shallon Pursh), which forms a well developed understory in the old-growth forest, completely dominates the site within 10 years and persists due to generally incomplete canopy closure as the stand develops. This forest ecosystem is believed to account for as much as 100,000 ha of coastal forests in British Columbia (Weetman et al. 1990). A series of fertilization experiments on CH cutovers, done by Germain (1985) and Weetman et al. (1989a,b), identified inadequate availability of N and P as a major limitation on tree growth. 2 The second forest ecosystem, the disturbed western hemlock-amabilis fir (Abies amabilis (Dougl.) Forbes) ecosystem (hereafter referred to as the HA), occurs on sites that have apparently been subjected to periodic massive disturbance by windthrow (Lewis, 1982). The HA ecosystem, which is characterized by very dense closed stands of western hemlock and amabilis fir with a sparse understory (Figure 1), occurs as scattered patches in a matrix of the CH ecosystem. This ecosystem occupies as much as 30% of the area on northern Vancouver Island, but is much less common in many other coastal locations where the CH ecosystem occurs. Tree growth on HA sites is rapid following clear-cutting and burning, and salal is much less abundant than on CH sites, and is soon shaded out by dense regeneration. Forest ecosystems intermediate between the Figure 1. View of the HA and CH forest ecosystems (upper and lower half, respectively) before and 10 years after clear-cutting and burning. Note the difference in tree growth after clear-cutting between the two forest ecosystems. 3 HA and CH can be found, which, according to Lewis (1982), suggests the potential for the HA ecosystem to develop into the CH ecosystem given enough time without windthrow or other catastrophic disturbance. Slow growth and severe signs of nutritional stress in planted and naturally regenerated conifers on salal-dominated CH sites following clear-cutting and/or clear-cutting and slash-burning has raised questions concerning the factors that may be limiting tree growth in these young stands. One such question relates to approximately 1500 ha of Sitka spruce (Picea sitchensis (Bong.) Carr.) plantations (Figure 2) established by Western Forest Product (WFP) between 1966 and 1976 Figure 2. View of a 16-year-old Sitka spruce plantation established on CH sites. Note the extremely slow height growth for the last 5 to 8 years. 4 on clear-cut CH sites on T.F.L. 25 near Port McNeill, northern Vancouver Island. These plantations initially grew well, but are now experiencing severe growth stagnation and nutritional stress (Weetman et al. 1990). Barker et al. (1987) showed that the growth started to decline 5 to 8 years following planting. Fertilization with N and P alleviates stagnation symptoms, but for only a few years (Germain, 1985). The early research on the CH growth stagnation problem revealed that both seedlings and saplings of western redcedar growing on clear-cut and burned CH sites perform better than those of Sitka spruce and western hemlock. Weetman et al. (1989a,b) showed that the annual height increment of 10 to 16-year-old natural western redcedar (33 cm/year) was greater than that of planted 16-year-old Sitka spruce (12 cm/year) or natural 10- to 16-year-old western hemlock (13 cm/year) during 1985, 1986 and 1987. The same studies also showed that the complete removal of salal from 1984 to 1987 by manual and herbicidal (Garlon) means significantly (P<0.05) improved the annual height increment of the natural western redcedar (50 cm/year), but not of Sitka spruce (16.5 cm/year) and western hemlock (20 cm/year). Fertilization with 250 kg N ha"l as ammonium nitrate or urea and 100 kg P ha-^ as triple superphosphate had a significant positive effect on the annual height increment of western redcedar (47 cm/year), Sitka spruce (36 cm/year), and western hemlock (52 cm/year) for the first three years following fertilization. As a result of these studies, western redcedar has been the preferred species on clear-cut and burned CH sites in the last 10 years, and has achieved an annual height growth of 25 to 30 cm per year (Figure 3). Germain (1985) suggested that several repeated fertilizations may be necessary in order to get sufficient canopy closure to suppress the salal understory. It is not known, 5 Figure 3. Planted 9-year-old western redcedar growing on clear-cut and burned C H sites. Note that the average height increment is approximately 30 cm per year. however, what coniferous overstory canopy closure and resultant light environment is required to shade out salal. No such decline has been reported for conifers growing on H A sites (Weetman et al. 1990). H Y P O T H E S E S C O N C E R N I N G P O O R T R E E G R O W T H O N C H S I T E S Several hypotheses have been advanced to explain the poor tree growth on C H sites. One set of hypotheses deals with differences in tree growth between the 6 CH and HA forest ecosystems prior to harvesting, whereas the second set of hypotheses deals with the poor tree growth on CH sites following clear-cutting and slash-burning. 1. The windthrow, site, and western redcedar hypotheses The striking differences in tree growth between adjacent CH and HA sites both before and after harvesting have been hypothesized to be associated with periodic windthrow that has improved the physical and chemical properties of the soils in the more wind-prone HA ecosystem (Lewis, 1982). Germain (1985) and C. Prescott (pers. comm.) found better forest floor nutrient status in HA in contrast to CH ecosystems. Repeated windthrows are also believed to be responsible for the improved productivity on some sites in Coastal Alaska (Bormann, 1989; Ugolini et aL 1990). More recently, however, two alternative site-related hypotheses have been suggested to explain the differences in tree growth between CH and HA sites. The first alternative of these states that because the CH forest floor is dominated by rotten western redcedar, it has a low mineralization potential and immobilizes N in the decomposer community, creating a low rate of nutrient cycling and nutrient availability. The second alternative hypothesis states that CH and HA sites are on different landscape units; that the HA sites are better drained and more exposed to wind than the poorly drained and less exposed CH sites. These different hypotheses are currently being investigated by researchers in the SCHIRP (Salal Cedar Hemlock Integrated Research Program) project. 7 2. The salal and flush of nutrients hypotheses The growth decline and stagnation reported in planted conifers in cutovers has been correlated over time with the re-establishment and complete occupation of the site by salal and with a decrease in forest floor nutrient availability (Germain, 1985; Weetman et al., 1990). Below-ground antagonistic interferences by salal (competition for nutrients and/or allelopathy) have been suggested as an explanation for the poor growth of conifers (Germain, 1985; Weetman et al. 1990) but neither of these mechanisms has been quantified. Because stagnated conifers are growing well above the salal cover, competition for light has been ruled out as a possible antagonistic factor. The allelopathy hypothesis was suggested following the observation that the stagnation symptoms of the Sitka spruce on salal-dominated sites on northern Vancouver Island are similar to those exhibited by Sitka spruce growing on sites dominated by heather (Calluna vulgaris) in Scotland (Weetman et al. 1990). Field, laboratory and greenhouse studies have demonstrated the potential allelopathic effects of heather (Handley, 1963; Robinson, 1972; Read, 1984). Other ericaceous species have been implicated in nutritional stress in conifer plantations in various parts of the world (de Montigny and Weetman, 1990; Mallik, 1987; Robinson, 1972; Handley, 1963; Rose etal. 1983). As noted above, it has also been suggested that salal interferes with western redcedar and western hemlock growth, but L. Husted (pers. comm.) failed to find any evidence of salal toxicity on western redcedar seedlings grown in pots. 8 STUDY OBJECTIVES AND RESEARCH APPROACH In spite of the research done up to 1987, there was insufficient evidence to strongly support or reject any of these hypotheses. This thesis attempted to test the hypothesis that the reported decline in conifer growth a few years following clear-cutting and burning on CH sites is the results of both competition by salal for nutrients, and reduced availability of these nutrients following an initial flush of nutrients. Particular attention was paid to the below-ground factors that limit early growth of conifers in the salal-dominated CH had four main specific objectives: 1. To quantify changes in the above- and below-ground biomass and nutrient content of the non-crop vegetation in a chronosequence of CH sites varying in age from 2 to 10 years after clear-cutting and burning. 2. To separate, using field and pot experiments, the effects of (1) salal interference from the effects of (2) the flush of nutrients occuring on CH sites following clear-cutting and burning on the early growth of Sitka spruce seedlings. 3. To characterize, using field and pot experiments, the effects of (a) non-crop vegetation, (b) slow-release fertilizer, (c) time since clear-cutting and burning, and (d) two contrasting forest ecosystems (CH versus HA) on the growth of Sitka spruce, western hemlock and western redcedar seedlings. A brief investigation was also undertaken of the degree of mycorrhization of the fine roots of the tree seedlings. 9 4. To determine, using western redcedar and fireweed as bio indicators, the microsite conditions associated with poorer and better growth of conifer seedlings on 4-year-old clear-cut and burned CH sites. Because Weetman et al. (1990), Weetman et al. (1989a,b), and Germain, (1985) had already described the nutritional status of conifer seedlings on CH and HA sites, the focus of the thesis was to study the possible causes contributing to the observed temporal and spatial variations in seedling nutrition and growth. Figure 4 illustrates the relationships of the studies reported in this thesis to the broader program of research that is being conducted by researchers in the SCHIRP project to understand the factors limiting early conifer growth on CH cutovers in Northern Vancouver Island. Several biotic and abiotic factors were studied in relation to their possible effects on the early growth of Sitka spruce, western hemlock and western redcedar. Table 1 indicates the different biotic and abiotic factors and experimental manipulations that were investigated. To address the overall objective of this thesis, a series of field and pot experiments was carried out near Port McNeill on northern Vancouver Island. A chronosequence of CH cutovers was selected to cover the first 10 years following clear-cutting and burning. These CH cutovers were selected to look specifically at the effects of time since clear-cutting and burning on the development of the non-crop vegetation, on site nutrient status and microenvironment, and on conifer seedling growth. In addition, 2-year-old clear-cut and burned HA cutovers were selected to investigate the differences in site nutrient status, microenvironment, non-crop vegetation and conifer seedling growth between CH and HA ecosystems soon after harvesting. HA forest Clear-cutting and slash-burning Microsite factors on CH cutovers A i i i • Invasion by Fireweed and salal >i8turbance v — 7 Short-term flush of nutrients (2) ( ^ C h « p t e f ~ 4 ^ ) I 1 I , Clear-cutting and 1 slash-burning —> 1 1 1 haptar 3 Vigorous re growth of salal @ CH forest Utilization of organic compounds by salal ^ J ( ^ C h a p t e r ~ 6 ^ ) Growth of conifers in cutovers Competition • Allelopathy - - Root membrane - - Mycorrhizae C ^ C h a p U r * 4 * 6 L - Protein complexing Figure 4. Flow-chart showing the contributions of the different Chapters of this thesis to the overall research effort being undertaken by a team of researchers (SCHIRP) to understand the factors limiting early conifer growth on CH cutovers on Northern Vancouver Island. Researchers working on northern Vancouver Island: 1 Keenan and Kimmins, Prescott and McDonald; 2 Weetman et al. (1990) and Germain (1985); 3 Weetman et al. (1990); 4 Preston and Weetman, Chanway and Turkington; 5 de Montigny and Lowe, McDonald, Berch and Ghoping; 6 Berch and Ghoping. 11 Table 1. Experimental treatments and abiotic and biotic factors investigated in this thesis. Experimental treatments Abiotic and biotic factors measured 1) Time after logging on CH cutovers 1) Soil moisture 2) Site conditions (CH vs HA cutovers) 2) microtopography 3) Removal of non-crop vegetation 3) Non-crop vegetation 4) Fertilization at planting 4) Available soil nutrients 5 ) Decomposition of cellulose 6) Soil temperature 7) Microbial activity 8) Mycorrhizal status 9) Growth of Sitka spruce, western hemlock and western redcedar 12 A description of the study area and of the two main forest ecosystems is presented in Chapter 2. Chapter 3 reports on the above- and below-ground recovery of the non-crop vegetation on the CH chronosequence. Chapter 4 investigates some possible factors contributing to the nutritional stress reported in Picea sitchensis plantations in CH cutovers. In Chapter 5, the effects of non-crop vegetation, slow-release fertilizer, and contrasting CH and HA site conditions on the early growth of western hemlock, western redcedar and Sitka spruce are studied. Chapter 6 reports relations between different microsite factors and the growth of western redcedar and the vigour and abundance of fireweed on 4-year-old CH cutovers. Finally, a synthesis of the cumulative results, some conclusions, additional research needs, and some limitations of the research are presented in Chapter 7. 13 CHAPTER TWO STUDY AREA AND ECOSYSTEM DESCRIPTION The study area is located in Block 4 of T.F.L. (Tree Farm Licence) 25 between Port McNeill and Port Hardy in the submontane variant of the CWHvm biogeoclimatic subzone (Green et al. 1984) on northern Vancouver Island, B.C., Canada (50° 607127° 35') (Figure 5), and has an elevation of 100 m. The study area receives approximately 1700 mm of rain annually, with 65% of the precipitation occurring between October and February. Although the summer months experience less rainfall than the winter months, rainfall during the growing season is thought to be sufficient to prevent any soil moisture deficit (Lewis, 1982). The number of hours of sunshine varies from an average high of 6.4 h/day in July to an average low of 1.5 h/day in December; these low values reflect the frequent occurrence of fog in the summer and frontal clouds in the winter. Mean daily temperature ranges from a low of 3.0°C in January/February to a high of 13.7°C in July/August. All weather data were obtained from the Port Hardy Airport weather station located within 15 km of the study area at an elevation of 50 m. These data represent an average for the last 36 years. All the research sites are in what Lewis (1982) called the Thuja plicata -Tsuga heterophylla - Abies amabilis - Gaultheria shallon - Rhytidiadelphus loreus, "salal-moss" ecosystem association (SI). This ecosystem is the climatic climax association for the CWHyj^ variant and covers approximately 60% of Block 4 of T.F.L. 25. The area is characterized as having a gently undulating topography (Suquash lowlands) which rarely exceeds 300 m in elevation. The surface material consists of deep (> 1 m in many places) unconsolidated fluvioglacial sediments. This surface material is underlain by Cretaceous sedimentary rocks. Lewis (1982) further divided this SI ecosystem association into two phases: (1) the undisturbed 14 0 50 100 R e s e a r c h a rea kilometres ' W a s h i n g t o n Figure 5. Location of the study area on northern Vancouver Island, British Columbia old-growth western redcedar/western hemlock phase (CH ecosystem), and (2) the disturbed western hemlock/ amabilis fir phase (HA ecosystem). These two forest ecosystem types are found on sites with similar parent material and underlying deposit. According to Lewis (1982), the CH ecosystem is the climatic climax community and consists of open western redcedar/western hemlock stands. This 15 ecosystem has not been catastrophically disturbed for several thousand years. The open canopy allows light to penetrate the tree cover which promotes the growth of a dense understory of salal and Vaccinium parvifolium, Howell and V. alaskaense, Smith. Only sparse ferns (Blechnum spicant (L.)) and mosses (Hylocomium  splendens (Hedw.) B.S.C. and Rhytidiadelphus loreus (Hedw.) Warnst.) are found under the Gaultheria-Vaccinium cover (Germain, 1985). This forest ecosystem has a thick (20-60 cm, but mostly > 45 cm), compacted humus layer rich in decaying wood (Lignohumimor: Klinka et al. 1981) overlying a moderately-well to somewhat imperfectly drained Ferro-Humic Podzol. A description of a typical soil profile is given in Table 2. Western hemlock and western redcedar germinants are found in abundance, mainly on rotten logs, but the stand structure suggests that few of these survive more than a few years. Following clear-cutting or clear-cutting and burning, the CH ecosystem is quickly reinvaded by salal from rhizomes that are present in the old-growth forest. Natural regeneration of trees following disturbance is slow and sparse and consists mainly of western redcedar and western hemlock seedlings. Following clear-cutting, this ecosystem is easily recognizable by the presence of large western red cedar stumps (2 to 3 m in diameter). The CH ecosystem is equivalent to CWHbj^) as classified by Green et al, (1984). The HA ecosystem occurs on sites that are subjected to periodic catastrophic windthrow, and is characterized by closed stands of western hemlock and amabilis fir. Salal is virtually absent from the understory. Only some sparse Vaccinium alaskaense, and V. parvifolium, herbs (Blechnum spicant L., Polystichum munitum (Kaulf.) Presl and Tiarella trifoliata L.) and mosses Hylocomium splendens (Hedw.) B.S.C.) are present in the understory (Germain, 1985). This forest ecosystem is frequently situated on upper slopes, and has a relatively thinner (10-40 cm) friable humus layer (Humimor: Klinka et al. 1981) 16 Table 2. Typical soil profile description of the CH ecosystem (adapted from Germain 1985). Horizon Depth (cm) Description LF 60- 55 Variegated mixture of coniferous litter and salal Utter and mosses; loose consistency; lots of fine-roots and mycorrhizal hyphae; abrupt wavy boundary to, H 55- 0 Reddish black (5R 2.5/1 m), dark reddish brown (5YR 3/2 d); massive; greasy; abundant roots of all sizes; abrupt wavy boundary to, Ae 0- 4 Grey (5YR 6/1 m) to brown (7.5YR 4/2 m);sandy loam; medium subangular blocky, friable; few fine and medium roots; clear, broken boundary to, Bhf 4- 19 Red (2YR 4/6 m), yellowish brown (10YR 5/8 d); sandy loam; weak medium subangular blocky; firm when moist; non-sticky and slightly plastic, wet; few fine-roots, few medium and coarse roots; abrupt wavy boundary to, Bf 19- 34 Yellowish red (5YR 5/6 m), yellowish brown (10YR 5/8 d); sandy loam; medium subangular blocky, firm when moist; non-sticky and non-plastic, wet; very few fine-roots, few medium and coarse roots; abrupt wavy boundary to, Bfgj 34- 55 Yellowish brown (10YR 5/8 m), yellowish brown (10YR 5/8 d); gravelly sandy loam;weak, medium and coarse subangular blocky, extremely firm when moist; non-sticky and non-plastic; no roots; common faint mottles, seepage water present; abrupt wavy boundary to, BCc 55+ Olive grey (5Y 5/2 d), strongly cemented to indurated gravelly sandy loam; no roots. 17 overlying a moderately-well drained Ferro-Humic Podzol. A description of a typical soil profile is given in Table 3. Western hemlock and amabilis fir germinants are found in abundance on the forest floor and on rotten logs. Western redcedar either does not germinate or germinants do not become established under the HA ecosystem canopy. Following clear-cutting or clear-cutting and burning, this ecosystem is rapidly invaded by a dense cover of fireweed (Epilobium  angustifolium L.). Natural regeneration after disturbance is rapid and dense, consisting mainly of western hemlock. Transitions between these two forest ecosystems on the landscape are often abrupt and are not always related to obvious topographic features. The HA ecosystem is equivalent to CWHb]^) as classified by Green et al. (1984). Table 3. Typical soil profile description of the HA ecosystem (adapted from Germain, 1985). Horizon Depth (cm) Description LF 25- 22 Variegated mixture of coniferous litter and mosses; loose consistency; lots of fine-roots and mycorrhizal hyphae; abrupt wavy boundary to, H 22- 0 Black (2.5YR 2.5/0 m), dark reddish brown (5YR 3/2 d); highly decomposed organic matter; granular, slightly greasy in lower horizon; abundant root of all sizes; abrupt, wavy boundary to, Aeu 0- 3 Grey (5YR 6/1 m) to brown (7.5YR 4/2 m); sandy loam; medium subangular blocky; friable; few fine roots; clear, broken boundary to, Bhfu 3- 18 Reddish brown (2.5YR 4/4 m), strong brown (7.5YR 5/8 d); sandy loam; weak medium subangular blocky; friable when moist; non-sticky and slightly plastic, wet: few to abundant fine- and medium-roots; abrupt wavy boundary to, Bf 18- 40 Strong brown (7.5YR 5/6 m), yellowish brown (10YR 5/8 5/6 d); gravelly sandy loam; weak medium subangular blocky; friable when moist; non-sticky and non-plastic, wet; few fine-roots; some medium to large roots; clear wavy boundary to, Bfgj 40- 60 Yellowish brown (10YR 5/8 m), brownish yellow (10YR 6/8 d); gravelly sandy loam, weak, medium and coarse subangular blocky, firm when moist; non-sticky and non-plastic, wet; very few roots; few faint mottles; abrupt wavy boundary to, BCc 60+ Olive grey (5Y 5/2 d), strongly cemented to indurated gravelly sandy loam; no roots. 19 CHAPTER THREE ABOVE- AND BELOW-GROUND VEGETATION RECOVERY FOLLOWING CLEAR-CUTTING AND SLASHBURNING ON CH SITES INTRODUCTION Regrowth of minor vegetation is a very important factor in the recovery of forest ecosystems from severe disturbance (Likens et al. 1970). However, abundant regrowth of competing vegetation can interfere with the early establishment of conifer trees (Stewart et al. 1984), and thereby slow down the process of secondary succession. The net positive or negative impact of the competing vegetation depends largely on the rapidity with which it establishes itself following disturbance. This varies according to the intensity of the disturbance (Dyrness, 1973; Halpern, 1988), the fertility of the site (Boring et al. 1981; Gholz et al. 1985; Hamilton and Yearsley, 1988), the type of understory vegetation present in the forest prior to disturbance (Halpern, 1988), and the amount, availability and establishment success of off-site seeds (Hamilton and Yearsley, 1988). In some cases, the recovery of the vegetation is extremely rapid. Marks (1974) estimated the time needed for pin cherry (Prunus pensylvanica L.) to fully occupy both the above- and below-ground environment of a site following clear-cutting in a hardwood forest in northeastern U.S. to be approximately four years. Raich (1980), working in a tropical forest in Costa Rica, reported fine-root biomass to be fully recovered to pre-disturbance levels after one year following clear-cutting. Recent studies indicate that maximum fine-root biomass in temperate conifer forests occurs during the first 10 to 15 years following clear-cutting. Vogt et al. (1987), comparing the fine-root biomass of 11- to 150-year-old Douglas fir 20 (Pseudotsuga menziesii) stands in coastal Washington (USA), found the highest amount to be in 11- and 12-year-old plantations dominated by competing vegetation of mainly salal. Similarly, Yin et al. (1989) found a greater amount of fine-roots in a 6-year-old clear-cut dominated by ferns than in an adjacent undisturbed northern red oak (Quercus rubra L.) forest. The potential for competing vegetation to occupy the below-ground environment quickly and abundantly can confer erosion protection on steep sites, and help to retain nutrients on the site by absorbing the nutrients in the soil that are available for leaching, but may also retard forest re-establishment. Vegetation management may be required if occupancy of the below-ground environment by competing species prevents or reduces the growth of the associated conifer species. Although an understanding of the recovery of competing vegetation following forest removal has very important implications for forest regeneration, very few studies have examined the below-ground recovery during the first 10 years especially in conifer forests. Marks (1974) studied the early above- and below-ground recovery of pin cherry in a northern hardwood forest ecosystem using a chronosequence of 1-, 4-, 6- and 14-year-old sites, but did not describe the depth of root sampling, the size of the roots sampled, and whether the roots were alive or dead. Berish (1982) sampled root biomass in a chronosequence of 1-, 8-, and 70-year-old sites in a tropical forest in Costa Rica, but she did not distinguish between dead and live roots. Raich (1980) reported on the fine-root biomass one year after clear-cutting in a tropical forest in Costa Rica, and Yin et al. (1989) compared the live fine-root biomass between a 6-year-old clear-cut and an undisturbed hardwood forest in Wisconsin. Other studies of the early recovery of both the above- and below-ground biomass include Boring and Swank (1984), Uhl and Jordan (1984), and Auclair (1985), but none of these studies provided detailed information regarding the below-ground biomass. 21 Several thousand hectares of poorly growing coniferous plantations on northern Vancouver Island, British Columbia, are completely dominated by an ericaceous species, salal (Gaultheria shallon Pursh). It has been hypothesized that salal is at least partially responsible for the poor conifer growth either by competing directly for nutrients (Germain, 1985), or by inhibiting their availability to trees (Germain, 1985; Weetman et al. 1989a). On these sites, salal reestablishes itself quickly following disturbance, mainly by resprouting from old rhizomes already present in the undisturbed old-growth forest of western redcedar (Thuja plicata Donn) and western hemlock (Tsuga heterophylla Raf. Sarg.). Salal is an erect to nearly prostate ericaceous shrub growing up to 2.5 m tall. It has an extensive, shallow root system with spreading rhizome-like structures. Salal occurs all along the Pacific coast of British Columbia and U.S. in lowland coniferous forest. Its distribution suggests that it requires a humid to perhumid coastal climate with mild temperatures, little snow, and unfrozen soils in winter. A few studies have documented the above-ground biomass (Sabhasri, 1961; Stanek, et al. 1979; Vales, 1986) and below-ground biomass of salal (Sabhasri, 1961; Vogt et al. 1987) growing in a forest understory, but none has investigated both the above- and below-ground recovery of salal and associated species on recently disturbed sites. There is a paucity of data on plant succession in coastal British Columbia, and the functional role of different species in early succession is even less well known. The objective of the study reported in this chapter was to document the patterns of above- and below-ground biomass recovery of the competing vegetation in an age sequence of salal-dominated plantations from 2 to 10 years following clear-cutting and burning on CH sites. A greater understanding of the successional patterns (both above-and below-ground) during early ecosystem 22 recovery is necessary to (1) understand the poor conifer plantation performance on the CH sites, and (2) prescribe effective silvicultural treatments to enhance forest regeneration and subsequent forest growth. The information gathered in this chapter will be used in chapters 4 and 5. 23 MATERIALS AND METHODS RESEARCH SITES Field studies started in the summer of 1987 on an age sequence of mesic clear-cut CH sites planted with western redcedar. Two ages of sites were chosen for this study: 2- and 8-years post-logging and burning. No suitable sites of intermediate age were found, so the 2 years after clear-cutting and burning sites were remeasured in the summer of 1989 to provide data for four years after clear-cutting and burning. The remeasurement data were collected in different areas of the sites, so the 2- and 4-year-old observations were independent of each other within the sites. For each site age, two different cutovers (8 to 12 ha in size and 2 to 5 km apart) were selected. All four cutovers were selected based on their homogeneity and similarity: similar slope position, aspect, surface material, forest floor thickness, soil characteristics, severity of burn (as determined from Western Forest Products files and from talking to the foresters in charge of the burning), and tree species composition prior to disturbance as determined by the stumps. This kind of age sequence (or chronosequence) research assumes that all the cutovers had similar ecological attributes prior to disturbance, that they received a similar degree of disturbance, and that they share a similar post-disturbance stand history (Cole and Van Miegroet, 1989). This assumption is believed to be satisfied by the sites used in this study because of the great care taken in the selection of the study sites and the characteristic uniformity of the CH ecosystem in the area of northern Vancouver Island where the study was conducted. Four pits were dug on each of several potential cutover research sites, and the soil horizons described. Only cutovers that had very similar soil profile characteristics were selected for the study. The four sites that were selected after this reconnaissance were considered to be very representative of the SI CH 24 ecosystem described by Lewis (1982). However, to further test the assumption that they were representative of the population of cutovers in the area, and that the temporal patterns revealed by the intensively-studied cutovers were mainly a function of time since disturbance, a visual survey was conducted of % cover of salal in 11 other rather similar cutovers (i.e. all sites had a thick forest floor, non-crop vegetation dominated by salal, stumps of western redcedar, moderately-well drained sites, and similar aspect, slope and elevation) representative of the SI CH ecosystem, and varying in ages from 1 to 15 years after clear-cutting and burning. The age sequence approach has been used by several researchers (Wallace and Freedman, 1981; Covington, 1981; Martin, 1985) BELOW-GROUND BIOMASS The vertical distribution and biomass of live salal, Vaccinium spp, fireweed, and bunchberry (Cornus canadensis L.) roots were measured 2,4, and 8 years after clear-cutting and burning on the CH ecosystem by taking 12 root cores (7.4 cm in diameter) at the beginning of June 1987 on each of 2- and 8-years post-clear-cutting and burning sites, and at the beginning of June 1989 on the sites that had been 2 years old in 1987, and were 4 years old at the time of sampling. Six cores were taken along two parallel line transects situated in the middle of each of the two cutovers for each site age. One core sample was taken at every 15 m along each transect for a total of three samples per transect. Stumps, logs and holes in the forest floor were avoided. The cores were stored at 3°C until processing. In the June 1989 sample on the 4-year-old sites, the root cores were taken in the center of the 12 l-m^ quadrats that were subsequently used at the end of July 1989 to assess the above-ground biomass (see below). Pre-sampling indicated that very few roots were present below 45 cm. Therefore, the cores were taken from depths 0 to 15 cm, 15 to 30 cm, and 30 to 45 cm starting at the top of the forest floor, 25 providing a total of 108 root-core samples (3 site-age combinations x 3 depths x 12 cores). Because the depth of the forest floor was greater than 45 cm in most places, most cores were composed almost entirely of organic matter. The non-root organic matter contained in each core was carefully washed through a 2-mm sieve, and the roots and rhizomes were separated and sorted by hand into different sizes (0-1 mm, 1-2 mm, 2-5 mm, and >5 mm) and species groups. Kurz (1989), using the same technique, recovered only a very small amount of live fine-roots in a 0.5-mm sieve placed under the 2-mm sieve. Therefore, only the 2-mm sieve was used in this study. Gaultheria shallon and Vaccinium spp were combined in a single group, and Cornus canadensis and Epilobium angustifolium in another group, due to the difficulty in distinguishing between fine-roots of individual species within these two groups. Distinction of fine-roots between the two species groups was possible due to the differences in root morphology, resilience when bent, and colour. Roots were visually separated into live and dead categories: living roots were resilient, slightly translucent and light brown (Gaultheria-Vaccinium) or translucent and white (Epilobium-Cornus); dead roots fragmented easily, were dull and were darker in colour. Only the recently produced salal rhizomes were measured in order to eliminate the old rhizomes produced prior to harvesting. Recently produced rhizomes were found to be those less than 8 mm in diameter with a white stele and a dense cover of short hairs on the rhizome. Some of those recently produced rhizomes may have been produced prior to harvesting, however. After sorting, the roots were dried at 70°C for 24 hrs and weighed to determine the oven-dry biomass per hectare. Each 645 cm^ core required an average of 20 person-hours of work. 26 ABOVE-GROUND BIOMASS, PERCENTAGE COVER AND LEAF AREA Above-ground biomass, percentage cover and leaf area were assessed on the same sites as the below-ground biomass (Le, 2, 4 and 8 years after clear-cutting and burning). Twelve 1-m^  quadrats were clipped at the end of July along two parallel line transects situated in the middle of each cutover to assess the above-ground biomass. One quadrat was clipped at every 7.5 m along each line transect for a total of six quadrats per line transect. The biomass collected from each quadrat was separated by species and then further divided into leaf and stem+fruit components. The biomass was dried at 70°C for 48 hrs and weighed to determine the oven-dry weight for each species and plant component. The percentage cover of each species was assessed visually at the beginning of April and at the end of July in 1987, 1988 and 1989 in twelve permanent 1-m^  quadrats within each cutover; these quadrats were not clipped. Total leaf area (one-side measurement) for each species was determined by calculating a conversion factor between oven-dry leaf weight and leaf area; correlation coefficients between leaf area and oven-dry leaf weight were all > 0.88 based on a subsample for each species. Leaf area was measured using grid paper. STATISTICAL ANALYSES The study did not use any particular design, but consisted of sampling two cutovers for each site age selected. For the analysis of variance, the cutovers were nested within site ages (3 site ages x 2 cutovers/site age) The Tukey HSD multiple comparison test was used to compare the treatment means. Log- or square-root-transformed values were used when the variances were not homogeneous based on Bartlett's test. Both transformed and untransformed data were checked for homogeneity of variances and normality of 27 distribution. Only untransformed means are presented, but the statistics of some of the means were performed on transformed means. 28 RESULTS AND DISCUSSION ABOVE-GROUND BIOMASS, PERCENTAGE COVER AND LEAF AREA Figure 6 shows the percent cover of salal in 6 intensively-studied cutovers plus 11 other cutovers representative of the population of cutovers in the CH ecosystem. This figure illustrates that the 6 intensively-studied cutovers are representative of the population of cutovers found in the region. There was no statistical difference (at P<0.10) between the two cutovers within each site age for most above-ground variables measured. Only the leaf biomass of salal was statistically (P=0.035) different between the two cutovers for the 2-year-old site. Therefore, the above-ground biomass data from the two cutovers within each site age were combined. Above-ground biomass of individual species in 2-, 4- and 8-year-old sites are given in Table 4. Two years after clear-cutting and burning, salal and fire weed accounted for 77% and 17% of the total above-ground biomass, respectively. The few clumps of Vaccinium spp. and conifers present represented only 2% to 3% of the total above-ground biomass. Bunchberry and other species were extremely sparse at this age, making up less than 1% of the total above-ground biomass. From 2 to 4 years after clear-cutting and burning, salal, Vaccinium spp, bunchberry and conifers more than doubled in biomass, whereas fireweed increased only slightly (Table 4). The increase was significant (P< 0.05) only for salal. Salal leaf biomass increased dramatically during this two year period, going from 596 kg ha"* to 2462 kg ha"*, a 4-fold increase. By four years after clear-cutting and burning, salal had increased its dominance of the site and represented 87% of the total above-ground vegetation biomass. Hendrickson (1988) and Boring 29 % s a l a l c o v e r 1 0 0 8 0 6 0 4 0 20 x Other cutovers A Studied cutovers x x X A A X X X _ J I l _ A X A X X A X A X J I I L_ 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Y e a r s a f t e r c l e a r - c u t t i n g a n d b u r n i n g Figure 6. Percent cover of salal on the 2-, 4- and 8-year-old intensively-studied cutovers and on 11 other cutovers representative of the population of cutovers found in the CH ecosystem near the research sites. 30 TABLE 4. Mean above-ground biomass (kg ha"1) and leaf area index (m^ m"^ ) of the vegetation on 2-, 4- and 8-year-old intensively-studied CH site ages. Numbers in rows followed by the same lower case letter are not significantly different (P> 0.05) between sites. Values in parentheses represent one standard error of the mean. n=24 for each cell. Plant species and components Biomass (kg ha"^ ) and leaf area (m2 specified site ages m"2) at 2yr 4yr 8yr A. Biomass Gaultheria shallon Leaf Stem+Fruit Total 596a (141) 462a (129) 1058a (267) 2462b (366) 1164b (185) 3626b (519) 2219b (246) 1858c (227) 4078b (454) E. angustifolium Leaf Stem+Fruit Total 126a (56) 110a (52) 236a (107) 131a (26) 162a (35) 293a (58) 66a (26) 103a (44) 169a (70) Vaccinium spp. Leaf Stem+Fruit Total 10a (6) 18a (15) 28a (22) 14a (11) 40a (32) 54a (43) 85a (39) 290b (121) 375b (148) Cornus canadensis Total la (1) 26a (14) 406b (79) Conifers Total 43a (16) 163ab (78) 446b (159) Other species Total 6a (2) 6a (5) 101a (52) Grand total Leaf Stem+Fruit Total 757 615 1372 2705 1464 4169 2846 2728 5574 B. Leaf Area Index 0.67 2.31 2.53 31 et al. (1981) also found that sprouting species dominated their sites rapidly following clear-cutting. Between 4- and 8-year-old sites, Vaccinium spp., bunchberry, conifers and other species all increased their above-ground biomass substantially, whereas salal increased only slightly, and fireweed decreased slightly (Table 4). Of all these changes between 4- and 8-year-old sites, only the increases in Vaccinium  spp. and bunchberry biomass were significant (P<0.05). The increase in salal biomass was entirely the result of an increase in the stem and fruit component; no increase in leaf biomass was detected. On the 8-year-old CH sites, the above-ground dominance of salal was reduced to 73% of the total above-ground biomass. The range in above-ground biomass values among individual quadrats was substantial in this study, as indicated by the high error of the means (Table 4). Weetman et al. (1990), working on similar sites nearby, found smaller salal above-ground biomass values on 2- and 4-year-old CH sites (500 and 2000 kg ha'^ , respectively), and slightly higher values on an 8-year-old site (7600 kg ha"^ ). The total above-ground biomass values found on our sites at 2, 4 and 8 years were substantially lower than those reported by Marks (1974) for a northern hardwood forest ecosystem (17,573 and 32,556 kg ha'^ at 4 and 6 years after clear-cutting, respectively), by Boring and Swank (1984) for a southern Appalachian hardwood forest ecosystem ( 21,993 kg ha"l at 4 years after clear-cutting), by Hendrickson (1988) for a northern mixed conifer-hardwood forest ecosystem (7,878 kg ha'l at 4 years after clear-cutting), by Uhl and Jordan (1984) for an Amazonian forest ecosystem (28,880 kg ha"^  at 4 years after clear-cutting), and by Outcalt and White (1981) for an Abies balsamea-Betula papyrifera forest ecosystem in northern Minnesota (4,181 and 2,604 kg ha'l at 2 years after clear-cutting for whole-tree and tree-length logged and burned sites, respectively). They are higher, 32 however, than the values reported by Auclair (1985) for a boreal forest ecosystem in northern Quebec (1,369 and 3,015 kg ha-^ at 4 and 7 years after wildfire, respectively), and similar to those reported by Gholz et al. (1985) for a Douglas fir forest ecosystem in Oregon (1,633 kg ha"l at 2 years after clear-cutting). Leaf area index (LAI) increased rapidly between the 2- and 4-year-old-sites, but increased only slightly between the 4 and 8-year-old CH sites (Table 4). Figure 7 shows the changes in percentage cover of the main species found on the initially 2- and 8-year-old sites measured when they were 2, 3 and 4 years old, and 8, 9 and 10 years old, respectively. Salal was always the dominant species on both site ages, and reached about 50% cover four years after clear-cutting and burning. Bunchberry was the second most abundant species on the 8-year-old site age, and reached its maximum cover of about 30% nine years after clear-cutting and burning. The planted and naturally-regenerated conifer species increased slowly in cover over time, reaching 23% by year ten on the 8-year-old sites. The small change in percentage cover of salal between 4 and 10 years suggests that it had reached its maximum cover by 4 years. Fireweed reached its peak 2 to 4 years after clear-cutting and burning, and then declined slowly. This pattern is very similar to that reported by Halpern (1989) on two clear-cut and burned Douglas-fir sites in Oregon. Weetman et al. (1990) estimated the percentage cover of salal at 55% on a similar 8-year-old CH site. The high percentage cover of salal found at 4 years on the initially younger sites (58%) indicates that salal can reach these high cover levels very rapidly. BELOW-GROUND BIOMASS AND DISTRIBUTION There was no statistical difference (at P<0.1) between the two cutovers within each site age for all of the below-ground variables measured. Therefore, the 33 Figure 7. Changes in the percentage cover of the vegetation during the first 10 years following clear-cutting and burning based on 3 years of data from 2 different site ages: 2- to 4- and 8- to 10-years post-logging and burning site ages. Gaul= Gaultheria  shallon; Ep= Epilobium angustifolium; Corn= Cornus canadensis; Conif=conifers; and Oth=other species. 34 below-ground data from the two cutovers within each site age. were combined. The biomass and distribution of the below-ground components of the Gaultheria- Vaccinium and Epilobium-Cornus species groups for each site age are shown in Table 5. The total below-ground biomass on 2-, 4- and 8-year-old CH sites averaged 1908, 4110, and 11,415 kg ha'^ , respectively. The total biomass of fine-roots (0-2 mm) at these ages was 1621, 2119, and 5311 kg ha"*, respectively. The Gaultheria-Vaccinium group (of which salal was by far the dominant species) comprised 95%, 90%, and 75% of the total live fine-root biomass on the 2-, 4-, and 8-year-old sites, respectively. The proportion of live fine-roots to total live roots was greatest (85%) on the 2-year-old CH sites and smallest (47%) on the 8-year-old CH sites. The amount of fine-roots of shrubs is known to fluctuate widely throughout the year (Yin et al. 1989; Kurz, 1989), however, and the proportion of fine-roots to total roots probably varies from one month to another. Although the above-ground vegetation was very sporadic on the 2-year-old sites, live fine-roots were found in all core samples at all depths. Between year 2 and year 4 on the initially 2-year-old sites, fine-roots increased only marginally, whereas rhizomes (roots >2 mm) increased from 169 kg ha"^  to 1542 kg ha'l. Most of the Gaultheria-Vaccinium live fine-roots were found in the 1 mm diameter category (Table 5). The amount of live fine-roots found in the 8-year-old sites (5311 kg ha"*) is similar to that found by Vogt et al. (1987) in a low productivity open 11-year-old Douglas-fir stand dominated by a dense understory of salal (5780 kg ha"*). It is, however, significantly less than the 12,910 kg ha"* found in a 8-year-old tropical clear-cut in Costa Rica (Berish, 1982), and the 9800 kg ha"* found in a 6-year-old northern hardwood clear-cut in Wisconsin (Yin et al. 1989). Figure 8 compares the distribution of Gaultheria-Vaccinium and Epilobium-Cornus fine-roots and rhizomes between the three sampling depths for 35 Tab le 5. V e r t i c a l a n d d iameter -c lass d i s t r i bu t i on of be low-ground b iomass of the G a u l t h e r i a - V a c c i n i u m a n d E p i l o b i u m - C o r n u s species groups on the 2-, 4- a n d 8-year-old in tens ive ly -s tud ied C H si te ages. V a l u e s i n paren theses represent one s t a n d a r d er ror of the m e a n . T w e l v e root cores were t a k e n on each s i te for each dep th . Biomass (kg/ha) Gaultheria-Vaccinium Epilobium-Cornus Site Depth (cm) 0-lmm l-2mm 2-5mm >5mm 0-lmm l-2mm 2-5mm >5n 0-15 805 48 89 39 39 32 48 0 (149) (37) (62) (30) (9) (23) (48) (0) 15-30 395 32 41 0 5 0 0 70 (135) (29) (38) (0) (2) (0) (0) (68) 30-45 248 13 0 0 4 0 0 0 (60) (9) (0) (0) (2) (0) (0) (0) Total 1448 93 130 39 48 32 48 70 0-15 843 102 (121) (50) 15-30 659 105 (173) (40) 30-45 253 57 (57) (25) Total 1755 264 264 950 55 (206) (610) (28) 139 189 27 (103) (193) (14) 000 000 9 (00) (00) (3) 403 1139 91 9 50 28 (8) (50) (30) 0 25 330 (0) (26) (316) 0 16 000 (0) (16) (00) 9 91 358 0-15 2189 309 1007 2666 625 373 14 0 (238) (85) (369) (485) (209) (138) (10) (0) 15-30 748 68 202 1318 125 22 2 0 (197) (24) (134) (658) (39) (21) (2) (0) 30-45 557 120 275 620 100 75 00 0 (77) (42) (227) (635) (35) (51) (0) (0) Total 3494 497 1484 4604 850 470 16 0 36 100 80 60 40 20 F i n e - r o o t Percentage of live root biomass (A) Gaultheria-Vaccinium  Y/A Epilobium-Cornus • 0-15 cm 15-30 cm 30-45 cm R h i z o m e Percentage of live root biomass 0-15 cm 15-30 cm 30-45 cm Forest floor depths Figure 8. Comparison of the vertical distribution of (A) fine-root (0-2 mm) and (B) new rhizome (> 2 mm) biomass of Gaultheria-Vaccinium and Epilobium-Cornus species groups within the upper 45 cm of the forest floor of the 8-year-old CH site age. The distribution is expressed as the % of the total live fine-roots or total live rhizomes in the upper 45 cm found at each of the three depths. 37 the 8-year-old CH site age. The Gaultheria-Vaccinium species group concentrated 57% of its fine-roots in the top 15 cm, whereas the proportion was 74% for the Epilobium-Cornus species group. Many studies have found the roots of herbs and shrubs to be concentrated in the upper soil horizons (Kummerow et al. 1977; Ruark and Bockheim, 1987; Aerts et al. 1989; Yin et al. 1989), and especially in the forest floor (Kimmins and Hawkes, 1978; Persson, 1980). Persson (1980) found that 89% and 98% of the live fine-roots of Calluna vulgaris and Vaccinium vitis- idaea, respectively, were in the upper 10 cm in a 15- to 20-year-old Scots pine (Pinus sylvestris) stand in Sweden. Aerts et al. (1989), working in a wet heathland in the Netherlands, reported that 95% of the roots of Calluna vulgaris were in the upper 20 cm of the soil. Most of the studies cited above were done on sites with relatively thin forest floor layers (<15 cm), whereas the forest floor on my study sites was more than 45 cm in most places. Kimmins and Hawkes (1978) reported that the shallow rooting habit of the vegetation on their study site could have been explained by the low nutrient status of the mineral soil underlying the forest floor. The thick forest floor found on my study sites could explain why I found somewhat less of a concentration of fine-roots in the upper few centimeters of the forest floor compared to other studies dealing with similar species but thinner forest floor layers. The observed degree of concentration of roots in the top 15 cm of the forest floor in this study correlated well with a better nutrient status in the 0-8 cm layer compared to the 8-25 layer (Chapter 4; Table 8), and with the higher soil temperature measured during the growing season at 3 and 10 cm as compared to 25 cm (Chapter 4; Table 9). Moisture was not a major factor on my study sites because of the frequent rainfall and heavy fog present throughout the growing season. 38 GROWTH ALLOCATION STRATEGY DURING THE PERIOD OF VEGETATION RECOVERY FOLLOWING CLEAR-CUTTING AND BURNING Figure 9 compares the pattern of above- and below-ground biomass accumulation of the two main species groups (Gaultheria-Vaccinium and Epilobium-Cornus) following the clear-cutting and burning of old-growth western redcedar and western hemlock forests (CH). During the first two years, few new rhizomes were produced by the Gaultheria-Vaccinium species group, the new growth being concentrated in the fine-roots and leaves. The strategy of these species for the first two years thus appears to be to produce leaves and fine-roots with which to exploit the newly unoccupied above- and below-ground environment created by clear-cutting and burning before producing new rhizomes. Once the Gaultheria-Vaccinium species group has produced enough leaves and roots from its old rhizomes to fully occupy its immediate territory, new rhizomes and shoots are produced. Between 2 and 4 years, the amount of rhizomes had increased by 9 times (Table 5), the leaves by more than 4 times (Table 4), and the fine-roots by only approximately 25% (Table 5). Between the 4- and 8-year-old sites, however, the increase in Gaultheria-Vaccinium species group biomass came mainly from fine-roots (a 2 times increase) and rhizomes (a 4 times increase). The large increase in Epilobium-Cornus leaf and fine-root biomass between the 4- and 8-year-old sites was entirely due to bunchberry (Table 4). Moreover, the large increase in the 1-2 mm Epilobium-Cornus fine-root category from the 4- to 8-year-old sites came mainly from small bunchberry rhizomes. Most of bunchberry rhizomes were less than 2 mm in diameter, and were therefore sorted as fine-roots instead of as rhizomes. It is important to remember that the comparisons between 4 and 8 years are based on different sites, whereas the comparisons between 2 and 4 years are 39 GAULTHERIA-VACCINIllM Live biomass (1000 kg/ha) (A) Leaf '• Stem+fruit —*— Fine-root ~ e — New rhizome / ,1 K 1 1 0 2 4 8 10 EPILOBIUM-CORNUS Live biomass (1000 kg/ha) - (B) — Leaf —>— Stem+fruit —*- Fine-root —s— New rhizome i ,1 1 1 1 0 2 4 8 10 Years after clear-cutting and burning Figure 9. Above- and below-ground biomass for the (A) Gaultheria-Vaccinium and (B) Epilobium-Cornus species groups on 2-, 4- and 8-year-old intensively-studied CH site ages. Note that the scale of the y-axis for A is different from that of B. Note that most of what is reported as fine-roots for the Epilobium-Cornus species group on the 8-year-old CH site age is in fact mainly small Cornus rhizomes. 40 based on the same sites, but sampled at a two year interval. It is felt that all the sites were similar enough to warrant the conclusion that the difference between 4 and 8 years was mainly due to time, and not due to any major difference in stand history and/or ecological attributes of the different sites. Furthermore, the non-statistical difference (P>0.05) found between the two cutovers within each site age for most biomass components suggests that our site selection criteria were successful in allowing the selection of ecologically similar sites. The pattern of above- and below-ground recovery found at the Port McNeill study sites differed somewhat from those reported for other regions. In northeastern (Marks, 1974) and southeastern (Boring and Swank, 1984) USA hardwood forests, the below-ground:above-ground biomass ratio has been shown to decrease gradually following the revegetation of the sites. Uhl and Jordan (1984) showed that the roots in a tropical forest in Amazonia remained a fairly constant percentage of total plant biomass throughout the first five years of succession following clear-cutting. In contrast, we found that the below-ground: above-ground biomass ratio increased from 1.4 in the 2-year-old sites to 2.5 in the 8-year-old sites. However, the low ratios found in years 2 and 4 would have been greater if we had included in the calculations the old rhizomes produced prior to disturbance, but which were still present following the disturbance. Sabhasri (1961) found a below-ground:above-ground ratio of 2.4 for salal growing under a 120-year-old Douglas fir stand in Washington. He reported that most of the above-ground biomass of salal was in the stem, but did not make any distinction between fine-roots and rhizomes for the below-ground biomass component. The high below-ground:above-ground biomass ratios found in this study resemble those found on other sites dominated by shrub and perennial herb communities (e.g. Whittaker and Marks, 1975; Yin et al. 1989). 41 In 1989, the below-ground biomass on the 4-year-old CH sites was assessed by taking a root core in the center of the 12 l-rn^ quadrats that had been also used to assess the above-ground biomass. A correlation analysis was done between the different components of the above- and below-ground biomass to see if any useful relationships existed. Although both above- and below-ground biomass components varied greatly between the 12 quadrats, no statistically significant (at P< 0.1) correlation (0.18 > r > -0.38) was found between any of the above- and below-ground biomass components. It appears that the amount of above-ground vegetation does not indicate how much below-ground vegetation is present on a particular microsite in these post-harvest communities. The rhizomatous nature of the main species, salal, may explain this lack of correlation; it reflects the comparative independence of local shoot and root production in such species. Salal may produce its shoots in areas that have appropriate above-ground resources (e.g. space, light), but which are not necessarily good for the production of fine-roots (e.g. old decaying logs), and vice versa. The lack of correlation may also be due to the small number of samples used, however. ECOLOGICAL AND SILVICULTURAL IMPLICATIONS One the most striking features of recently clear-cut and burned sites on northern Vancouver Island is the strong above- and below-ground dominance by salal. Hendrickson (1988) stated that the rapid occupancy of clear-cut areas by aspen and maple sprouts near the Petawawa Research Forest in Ontario, Canada, probably prevented the growth of non-woody species on his study sites. On our study sites, however, salal does not occupy a significant proportion of both the above- and below-ground environment until 3 to 4 years after clear-cutting and burning. This delay should give enough time for any potential invaders (e.g. fireweed, red alder (Alnus rubra) and salmonberry (Rubus spectabilis)) to get 42 established. However, of all the potential invaders, only fireweed colonized a significant fraction of the site (around 12% at 4 years). Fireweed appears to prefer small depressions on these sites (Chapter 5, Figure 23), although no edaphic explanation for such preference was established. One possible explanation of the failure of many early successional species to get established on my study sites is that the thick, wet and decay resistant forest floor of mainly decaying wood was not disturbed enough by clear-cutting and burning to allow for the rapid invasion and growth of early serai species such as fireweed, salmonberry and red alder1, all of which require a certain amount of disturbance to get established (Kimmins, 1987). Salal is able to persit and dominate these nutrient poor sites because of its ability to survive clear-cutting and burning (by vigorously resprouting from rhizomes and thereby quickly occupying the site), and to resist invasion by other species. Resistance to invasion may be achieved by making nutrients unavailable and/or by inhibiting the germination and growth of other species by direct or indirect allelochemical mechanisms (Kimmins, 1987). Read (1984) discusses the possible toxicity syndrome created by Calluna (an ericaceous shrub like salal) in Scotland. He proposes that Calluna acidifies the soil to a pH of 3 to 4 and produces toxic compounds that form organic complexes with nitrogen which slow down decomposition of the humus. It has been proposed that a similar phenomenon may be occuring on the salal-dominated cutovers on northern Vancouver Island (Weetman et al. 1990). Moreover I estimate, in Chapter 5, that 30 to 45% of the potentially available N may be tied up annually in living tissue of salal on these sites during the first 8 years. Such immobilization of nutrients on these Red alder is, however, present on exposed mineral soil on road sides. 43 nutrient-poor sites may be sufficient to prevent the establishment of more nutrient-demanding early serai species. The dominance of salal on nutrient-poor CH sites supports Tilman's theory (Tilman, 1990) that plants with a low R (the level of resource below which the population is unable to maintain itself) should dominate. Similarly, Berendse and Elberse (1990) suggested that the ericaceous species Erica is more successful in nutrient-poor environments because it is more economical with the nutrients that it has acquired. Connell and Slatyer (1977) described three models of species replacement during secondary succession: the facilitation model, the tolerance model, and the inhibition model. Based on the results of this study and of other studies done in the same area, it would appear that succession on these salal-dominated sites more or less follows the tolerance model in which slower growing, more tolerant western redcedar and western hemlock seedlings invade and grow in the presence of salal right after disturbance, but are adversely affected by the presence of salal. Eventually, however, the conifer trees will grow above salal and shade it out (Messier et al. 1989). Because it can take as many as 40 years for the conifers to shade out salal (Messier et al. 1989), and because salal has been shown to interfere effectively with the growth of conifer trees (Chapters 4 and 5), there is some evidence that salal may also exhibit some of the attributes of the inhibition model. As reported in this Chapter, salal produces large amounts of very fine-roots (< 1 mm) that can effectively preempt site resources whereas conifer fine roots tend to be largely in the 2-5 mm size class. Depletion of nutrients (or water) by plants with similar total root biomass has been shown to be greater for species with thinner roots (Goldberg, 1990). 44 It has been suggested that salal inhibits the growth of other species by direct or indirect allelochemical mechanisms (Germain, 1985; Weetman et al. 1989a). However, salal may also play an important role in nutrient conservation in the study area (cf. Prunus pennsylvanica: Marks 1974), since no other species seems able to quickly colonize these sites following the removal of the forest canopy. The negative consequences of nutrient competition over the first 10 to 20 years post-disturbance may turn into a positive long-term effect by maintaining nutrients on site in salal biomass. The hypothetical development of salal biomass in the understory of a young conifer stand over a 60 year period is presented in Figure 10. This model was built based on the results from this study and of a few other studies (Messier and Kimmins, 1991; Messier et al. 1989; Vogt et al. 1987; Vales, 1986). Based on this model, the development of the live fine-root, leaf, stem and rhizome biomass of salal over a 60 year period involves three different stages in salal development: (1) a rapid increase in salal live fine-root, leaf, stem and rhizome biomass during the first 8 to 15 years followed by (2), a rapid decline in salal live fine-root biomass and a more gradual decrease in leaf and stem biomass following tree canopy closure and (3), the virtually complete elimination of salal as the overstory tree canopy becomes very dense (Figure 10). No data are available for salal rhizome biomass between 10 and 60 years after clear-cutting and burning, but it is believed to continue to increase slowly for a few years. The first stage in this post-disturbance salal development (0-15 years) is based on data from this Chapter. The second stage (15-45 years) is based on above-ground biomass data from Vales (1986), fine-root data from Vogt et al. (1987), and above- and below-ground biomass data from a pot experiment conducted at Port McNeill (Messier and Kimmins, 1991). Vogt etal.'s (1987) study 45 Salal relative abundance Leaf Fine-root New-rhizome Stem «_ Stage 1 _ J „ Stage 2 I Stage 3 —r-15 - 1 -30 45 Old-growth forest 60 Years after clear-cutting and burning Figure 10. Hypothetical development of live fine-root, leaf, stem and rhizome relative abundance of salal over a 60 year period following the clear-cutting and burning of old-growth forests of western redcedar and western hemlock (CH) on northern Vancouver Island. Based on the results of this study and of Messier and Kimmins (1991), Messier et al. (1989); Vogt et al. (1987), and Vales, (1986). There are no data for rhizome biomass in the 10 to 60 year period. See text for information about how these curves were obtained. 46 showed that the understory angiosperm fine-root biomass (predominantly of salal) on low-productivity Douglas-fir stands in western Washington state decreased from between 5500 and 7000 kg/ha at 11 and 12 years in relatively open stands to between 400 and 700 kg/ha at 33 and 69 years in relatively closed stands, yet the percent cover of the understory salal (25 to 50%) remained unchanged. Their study indicates that salal shifts a major portion of its carbon allocation from fine-root to leaf tissues as it experiences lower light availability caused by the closing of the overstory canopy. Similar results were obtained in a pot experiment in which salal was grown under various light intensities (Messier and Kimmins, 1991). Such a shift in carbon allocation has been observed in many studies (e.g. Chapin et al. 1987). Vales (1986) examined the relation between salal above-ground biomass and both overstory tree cover and the resultant level of solar radiation reaching the salal in twelve 28- to 60-year-old western hemlock-Douglas-fir stands on Vancouver Island. His data showed that salal maintains a more or less constant amount of above-ground biomass under tree cover varying from approximately 35% to 80%. The third stage (45-60 years) is based on data obtained from Messier et al. (1989) and from Vales (1986) who found the above-ground biomass of salal to be drastically reduced under tree canopy covers greater than 80%. The hypothetical curves presented in Figure 10 provide a basis for predicting post-logging and burning patterns of salal above- and below-ground biomass accumulation on similar sites in the CWHbvm biogeoclimatic variant in coastal British Columbia. The model suggests that the net storage of nutrients in live fine-root, leaf, and stem of salal will cease between 10 and 20 years after clear-cutting and burning as salal stops expanding its biomass in these components. Between 20 and 45 years, salal biomass will decline, especially the fine-root biomass, so that below-ground competition by salal is almost certainly 47 less than that suggested by the magnitude of the above-ground biomass. After approximately 45 years the salal understory almost completely disappears as the tree canopy reaches 80% cover or more. No data is available on salal rhizome biomass after 8 years on similar sites. However, data reported by Sabhasri (1961) from an open 120-year-old Douglas-fir stand of site quality V in Washington state where salal is very abundant in the understory suggest that the rhizome biomass could continue to increase beyond the amount reported at 8 years in this study. The scenario presented here is plausible on the nutrient poor CH sites only if a fairly dense stand (> 4000 trees/ha) is maintained (Messier et al. 1989). On nutrient rich sites, however, this scenario may be achieved at somewhat lower tree densities, or at similar densities but at an earlier age. Following the clear-cutting and burning of old-growth western redcedar and western hemlock forests, Gaultheria shallon, Epilobium angustifolium, Vaccinium spp., Cornus canadensis and a variety of other minor species colonized the study sites. However, by eight years after forest removal, all of the vascular plant species that were present in the old-growth forest were represented in the cutovers, suggesting that clear-cutting and burning of these ecosystems does not result in any loss of vascular plant species diversity. 48 SUMMARY The above- and below-ground vegetation recovery was assessed 2, 4 and 8 years after clear-cutting and burning on an age sequence of CH sites dominated by salal (Gaultheria shallon, Pursh) on northern Vancouver Island, British Columbia. The total above-ground vegetation biomass quadrupled from 1372 kg ha"* on the 2-year-old sites to 5574 kg ha"* on the 8-year-old sites. These are low values for post-harvesting above-ground biomass when compared to many other forest ecosystems. Salal was the dominant species on these sites, representing 77%, 87% and 73% of the total above-ground biomass on the 2-, 4- and 8-year-old sites, respectively. Leaf area index increased from 0.67 to 2.31 m 2 m"2 between the 2- and 4-year-old sites, but was only 2.53 m 2 m"2 on the 8-year-old sites. The total below-ground biomass increased six times from 1908 kg ha"* on the 2-year-old-sites to 11415 kg ha"* on the 8-year-old sites. The upper 15 cm of the forest floor in the 8-year-old CH site was found to contain 56% and 74% of the live fine-roots and 64% and 49% of the new rhizomes of the Gaultheria-Vaccinium and Epilobium-Cornus species groups, respectively. No correlation was found between the amount of salal found above-ground and below-ground at the microsite level. The post-disturbance dominance of salal on these sites following clear-cutting and burning appears to be due to: (1) its ability to survive these treatments and rapidly occupy both above-ground and below-ground environments from rhizomes present prior to disturbance, and (2) to resist invasion by other species by preempting resources (nutrients in this case). The projected development of the live fine-root, leaf, stem and rhizome biomass of salal over a period of 60 years presented in this study suggests that the net storage of nutrients by salal will cease between 10 and 20 years after clear-cutting and burning. 49 C H A P T E R F O U R G R O W T H S T R E S S IN P I C E A SITCHENSIS P L A N T A T I O N S O N T H E C H SITES I N T R O D U C T I O N The implementation of successful regeneration programs requires, among other things, the identification of the external factors that limit plantation establishment and growth (Hobbs, 1984). Factors in the environment that may limit plant growth can be divided between (1) those that can be affected by other plants, and (2) those that are more or less intrinsic to the site. Non-crop vegetation often interferes with the growth of conifers directly through resource competition (i.e. light, moisture and nutrients) and chemical inhibition (i.e. allelopathy), or indirectly by modifying the environment in a way that is detrimental to the conifers (Weidenhamer et al. 1989). Numerous examples for each of these different possible types of interference exists in forestry: competition for water (Price et aL 1986; Flint and Childs, 1987; Petersen et al. 1988), nutrients (Carter et al. 1984; Cole and Newton, 1986; Elliot and White, 1987), and light (Brand, 1986; Brand and Janas, 1988); modification of the environment (Damman, 1971; Read, 1984); and direct allelopathic interferences (Del Moral and Cates, 1971; Rose et al. 1983; Hanson and Dixon, 1987; Mallik, 1987; Cote" and Thibault, 1988). All of these effects have the potential to interfere with the normal ability to obtain resources by conifers. Site conditions following forest harvesting are often found to be too dry or too wet, too hot or too cold, prone to late spring or early fall frost, or too poor in nutrients to sustain the normal growth of a particular species. However, clear-50 cutting and burning practices are known to increase temporarily resource availability (Binkley, 1984; Krause and Ramlal, 1986; David, 1987), and this often allows crop trees to grow very well for the first few years following forest removal (Martin, 1985). In some cases, both the termination of the increase in resource availability (flush of nutrients) due to clear-cutting and burning and the full occupancy of the above- and below-ground environment by non-crop vegetation can hinder the growth of crop trees. Three hypotheses to explain the growth stress observed in the Sitka spruce plantations on northern Vancouver Island (Chapter 1) were tested in this chapter: (1) that salal competition for scarce nutrients, especially N, can provide a partial explanation for the observed growth stress; (2) that the presence of salal reduces mycorrhizal infection and/or modify the types of mycorrhizae found on Sitka spruce seedling roots; and (3) that forest floor nutrient availability on these CH sites declines after 5 to 8 years following clear-cutting and burning, and that this further contribute to the observed growth stress. A series of pot and field experiments was carried out to test these hypotheses. 51 MATERIALS AND METHODS RESEARCH SITES Field studies were initiated in the summer of 1987 on the same age . sequence of mesic clear-cut CH sites as for Chapter three. FIELD SEEDLING BIOASSAYS Nursery grown "plug type" 1-0 seedlings of Sitka spruce were planted in April 1987 on each of two cutovers for each of two site ages (2 and 8 years after clear-cutting and burning) and left to grow for three growing seasons (i.e. until the sites were 4 and 10 years old, respectively) for a total of 144 seedlings. The seedlings were between 15 and 25 cm in height at the time of planting with an average below- and above-ground biomass of 0.71 g and 2.04 g, respectively. The study was a 2 x 2 nested-factorial experiment with 36 seedlings per treatment. The two main factors examined were: (i) two site ages (2 and 8 year old) and (ii) two planting treatments: in treatment 1, seedlings were planted without additional treatment (Vegetation-not-removed): In treatment 2, seedlings were planted in the middle of 200 cm diameter patches from which all above-ground vegetation was continuously removed by clipping, and from which the below-ground competition from adjacent vegetation was eliminated by periodically cutting to a depth of 40 cm in the forest floor around the patches (vegetation-removed). The vegetation surrounding both planting treatments was top pruned to eliminate any potential competition for light. This was done to emulate the usual field conditions on CH sites in which conifer saplings are growing above a dense understory of non-crop vegetation. 52 This experiment was designed to isolate the effects of the flush of nutrients ("assart effect") from the effects of salal interference on the growth of Sitka spruce in natural conditions. The comparison between vegetation-not-removed and vegetation-removed planting treatments looks at the effects of salal interference upon Sitka spruce growth, and the comparison between Sitka spruce planted in the vegetation-removed treatment between the two site ages looks at the effect of time since clear-cutting and burning. The height and basal stem diameter of each seedling were evaluated just after planting in 1987 and at the end of each growing season in 1987, 1988 and 1989, except for the diameter which was remeasured only at the end of the last two growing seasons. Orthogonal and non-orthogonal contrasts were used to compare the treatment means. For the contrasts that were non-orthogonal, I used the FT error instead of the H error (Sokal and Rohlf, 1981). The FT error was obtained from: H^l-d-H)1^. where k= degrees of freedom Log- or square-root transformed values were used when the variances were not homogeneous, using Bartlett's test. Both transformed and untransformed data were checked for homogeneity of variances and normality of distribution. Only untransformed means are presented, but the statistics of some of the means were performed on transformed means. The degree of mycorrhizal colonization was evaluated on a subsample of 8 seedlings from each planting treatment, on the initially 2-year-old CH sites when the sites were 4 years old, to look at the effects of salal upon Sitka spruce percent 53 mycorrhizal colonization. The fine-roots, which were obtained by carefully excavating as much as possible of the root system were kept at 3°C until analyzed. The root system of each seedling was cut in 5 cm sections and thoroughly mixed before taking 3 subsamples for analyses. A t-test was carried out to compare the frequency of each of the mycorrhizal types between the two planting treatments. The mycorrhizal status of the seedlings was not quantified at planting, but many of the root tips were observed to be mycorrhizal. Preparation of the roots for the determination of the mycorrhizal colonization was achieved using the following method: (1) soaking and heating (80-90°C) for 3 hrs in 10% KOH, (2) bleaching in a mixture of water, 30% H 2 0 2 and ammonia for 2 hrs, (3) acidifying in 85% lactic acid for 20 minutes, and (4) staining in a mixture of 85% lactic acid, glycosol, water and trypan blue. Each root sample was divided into 3 subsamples and the mycorrhizal status was assessed using a dissecting microscope. The mycorrhizal status was assessed by calculating the percentage of root tips colonized by each one of 5 different types: 1. Cenococcum geophilum (Cg): Tips with black hyphae forming Hartig net, hyphae mostly 4-5 um wide, mantle when present with radiate pattern. 2. Brown (B): Tips with brown hyphae forming a Hartig net, hyphae mostly 2 um wide, mantle with jigsaw pattern. 3. Thelophora terrestris (Tt): Tips without brown hyphae, stele of the root obscured, Hartig net present, hyploid cystidia few to abundant, 80-130 um long x 3 um wide at clamped base, with radial taper to 1.5 um at the tip. 4. Others (O): all others. 54 5. Non-mycorrhizae (NM): Tips without brown hyphae, stele of root clearly visible, Hartig net absent. POT SEEDLING BIOASSAYS In March 1988, two pot experiments were initiated in an open area near the research sites. The first experiment was designed to evaluate the effect of different densities of salal on Sitka spruce seedling growth and mycorrhizal infection to help evaluate the results obtained from the field studies. Nursery-grown Sitka spruce seedlings (1-0 plug type) and salal were planted at different densities (4:6, 4:0, 3:2, and 2:4 spruce-salal plants, respectively) in pots 20x40x20 cm in size. These densities were used to cover a wide range of salal and spruce combinations. There were 4 ,3 and 2 seedlings per pot per treatment, for a total of 24 pots. Two types of growth media were used: the upper 8 cm of the forest floor from a 3-year-old and a 9-year-old clear-cut and burned CH site. The experiment was a 2x4 factorial using a completely randomized design. Salal plants were established as 10 cm long rhizomes with at least two healthy buds. In the second experiment, the soil fertility along an age sequence of CH sites was assessed in the absence of salal using Sitka spruce seedlings as a bioassay. This was done to confirm the results from the field studies of seedling growth when salal interference was eliminated, and the temporal patterns of forest floor characteristics reported in this Chapter. Sitka spruce 1-0 plug seedlings were established in pots filled with forest floor material taken from three CH site ages (1,3 and 9 years after clear-cutting and burning) and from two depths (0 to 8 cm and 8 to 20 cm). There were 4 seedlings per pot per treatment 55 for a total of 18 pots. The experiment was a 2x3 factorial using a completely randomized design. All rhizomes and most fine and medium roots were removed from the forest floor material which was then mixed within each site age and depth to reduce variability prior to filling two pots per treatment. All pots were maintained in full sunlight, and the forest floor near field capacity by periodic watering. At the end of each of the 1988 and 1989 growing seasons, the height and basal diameter increments of the Sitka spruce were measured, and pot averages calculated. The degree of mycorrhizal colonization of the Sitka spruce was determined for the 4:0 and 4:6 spruce-salal combinations of the first experiment for the 3-year-old CH site age at the end of the 1989 growing season. NUTRIENT CONTENT IN NON-CROP VEGETATION BIOMASS The N and P concentrations for each species and biomass component were either measured in the laboratory and/or obtained from the literature (Klinka, 1976; Sabhasri, 1961; Weetman et al. 1990). The N and P concentrations were measured by digesting 0.2 g (oven-dry weight) of plant material overnight in a mixture of potassium sulfate, sulfuric acid and selenium in a block digester. The digest solutions were then analyzed for N and P using a Technicon AutoAnalyser II (Technicon Instrument Corp., Tarrytown, N.Y.). The nutrient concentrations of salal roots measured in the laboratory agreed with those reported by Sabhasri (1961). The above- and below-ground biomass data of the competing vegetation for the 2, 4 and 8 years after clear-cutting and burning CH site ages measured in Chapter three were used to calculate the amount of N and P immobilized in the vegetation. This information was used to calculate the net immobilization of N 56 and P in the non-crop vegetation biomass during the first 8 years following clear-cutting and burning. FOREST FLOOR PROPERTIES AND MICROENVIRONMENT Eighteen forest floor cores were taken in 1988 from depths of 0-8 and 8-20 cm on each of the two cutovers that were 2 and 8 years old at the start of the study. These sites were 3 and 9 years old at the time of sampling in 1989. The fresh samples were stored at 3°C for less than one week prior to being passed through a 2-mm sieve before analysis. A subsample was oven-dried at 70°C for 24 hrs to determine the moisture content. All the results are reported on an oven-dry basis. Forest floor pH was determined in distilled water with a glass electrode using a soil-water ratio of 1:4 (gram:mL). Total N and P were measured by digesting 0.2 g (oven-dry mass) of forest floor material overnight with a mixture of potassium sulfate, sulfuric acid and selenium in a block digester. Extractable N was determined by extracting and shaking 5 g (fresh mass) of forest floor material with 100 mL 2 M KC1 solution for one hour. Extractable P was determined by extracting and shaking 5 g (fresh mass) of forest floor material with 100 mL 0.01 M HCL solution for five minutes. Mineralizable N was determined by extracting 5 g (fresh mass) of forest floor material with 100 mL 2M KC1 solution following anaerobic incubation in 25 ml of distilled water at 30°C for 7 days. The digest solutions were then analyzed for N and P using a Technicon AutoAnalyser II (Technicon Instrument Corp., Tarrytown, N.Y.). Soil organic matter was determined by loss on ignition (24 hrs at 500°C), and carbon content calculated by dividing the organic matter content by 1.723 (Armson, 1979). 57 Microbial activity was assessed in the laboratory by measuring the amount of CO2 being released over 48 hrs. Twelve samples of forest floor material were used. Most of the live root and the undecomposed woody debris was removed from these samples, and eighty grams (fresh mass) of the residual forest floor per sample were added to 500 mL containers (canning jars, Le Parfait). A crucible containing twenty mL of 0.1 M NaOH was placed in each 500 mL container. The containers were tightly sealed and incubated in the dark for 48 hrs at room temperature (18 to 21°C). The amount of CO2 trapped in the 0.1 M NaOH was determined by precipitating the carbonate with 0.3 mL of 50% BaCl2 and then titrating the excess alkali with 0.1 M HCL The experiment was a 2x2 nested-factorial using a completely randomized design. The two main factors were: (i) two site ages (3 and 9 years after clear-cutting and burning) and (ii) two soil depths nested within the cutovers (two cutovers per site age). Orthogonal contrasts were used to compare the treatments means. No transformation were necessary. In early May 1988, twenty-four ion-exchange resin bags of mixed cation (21 g of 68% moisture Amberlite IRC-50 CP. RCOO-H-) and anion (29 g of 65% moisture Amberlite IRC-45 CP. RNH3+OH-) exchange resin enclosed in stocking bags, and twenty-four cellulose discs (4.25 cm diameter Whatman #1) enclosed in one-mm nylon mesh bags, were buried for four months at depths of 10 and 25 cm in the forest floor. These were used to compare the relative levels of forest floor ammonium, nitrate and phosphate availability and relative decomposition rate in the seedling's root environment between the two site ages at the two forest floor depths. Fox and VanCleve (1983) have shown that the Jenny's index of decomposition rate, k, was well correlated with annual cellulose (filter-paper) decomposition differences among 16 Alaskan taiga forest stands. The resin mixture bags had cation and anion exchange capacities of approximately 33 mmolc each. To prevent microbial growth on the resin, approximately 4% of the 58 exchange capacity of the resin was loaded with mercuric chloride (HgC^). Laboratory and field studies have shown a good correlation between the ion-exchange resin bag method and many conventional methods of assessing nutrient availability (Binkley and Matson, 1983; Binkley et al. 1986; Lajtha, 1988). Moreover, it is believed that in situ ion-exchange resin bags may behave similarly to plant roots in terms of ion uptake (Gibson, 1986), and be sensitive to micro-environmental conditions that influence nitrogen availability (Binkley and Matson, 1983). After two months, the resin was removed from the bags, air-dried, shaken with 200 mL 1 M KC1 for 1 hour, and the extract analyzed for NH4+, NO3" , and phosphate-P as described above. The cellulose discs were cleaned in distilled water, dried at 70°C for 24 hrs, and their loss of mass determined. The experiment was a 2x2x2 nested-factorial. The three main factors were as follows: (i) two site ages (3 and 9 years old); (ii) two forest floor depths nested within the cutovers (two cutovers per site age); and (iii) two planting treatments (vegetation-not-removed and vegetation-removed) nested within the cutovers. Orthogonal contrasts were used to compare the treatments means. No transformation were necessary. Eighteen soil temperature (using dial soil thermometers) and ten soil moisture (using quick draw soil tensiometers) measurements were made intermittently at depths of 10 and 25 cm in 1987 and at depths of 3, 10, and 25 cm in 1988 on each of the two site ages and for each planting treatment. Measurements were made between 11:00 and 13:00 pm twice every month from May to September of both 1987 and 1988. Occasionally, soil temperature was measured simultaneously on 2 or 3 different site ages. Soil temperature and moisture were obtained for a total of 20 dates. 59 A l l these forest floor and microenvironmental measurements were designed to characterize the changes in forest floor characteristics with time since clear-cutting and burning and the changes induced by the removal of the non-crop vegetation. 60 R E S U L T S FIELD SEEDLING BIOASSSAYS Figure 11 compares the annual height and diameter increments of Sitka spruce seedlings over the first three growing seasons after planting for the 2 and 8 years after clear-cutting and burning C H site ages and for the two planting treatments. From the analyses of variance of height increments in 1987, 1988 and 1989 (Table 6), the following general statements can be made: (1) the growth was significantly greater for the 2 years than for the 8 years after clear-cutting and burning C H site ages; (2) there were no statistical differences (at P<0.1) between the two cutovers within each site age; and (3) the only significant (at P<0.05) interaction between site ages and planting treatments was for the annual height increment in 1988. The results of the analyses of variance done on diameter increments were almost identical to those done on height increments, and therefore are not reported here. Vegetation removal (planting treatment 2) increased growth during the second and third post-treatment years on the 2 years after clear-cutting and burning C H site age, but only during the third post-treatment year on the 8 years after clear-cutting and burning C H site age (Figure 11). Figure 12 compares the height and diameter increments of Sitka spruce seedlings over three growing seasons between the two site ages and planting treatments investigated in order to separate the effects of the site ages from the effects of salal interference. The growth differences between the two site ages in graph A, where the effects of the non-crop vegetation was not removed, shows the site age and interference effect. The growth differences between the two site ages in graph B where the effects of the non-crop vegetation was removed shows the 61 Figure 11. Sitka spruce annual height and diameter increments on 2 (filled symbols and continuous lines) and 8 (open symbols and dotted lines) years after clear-cutting and burning CH site ages using two treatments: vegetation-not-removed (-0 • -), and vegetation-removed (-A A .) treatments. The annual diameter growth shown for each of the first 2 years after planting was calculated as the average value over the first 2 years. Vertical bars are one standard error. Table 6. ANOVA summary table showing variance ratios (F), P-values and error mean-square for Sitka spruce height increments in 1987, 1988, and 1989 between the two planting treatments and two site ages. The data for the height increments of 1987, 88 and 89 were transformed using the square root transformation. Ht increment in 1987 Ht increment in 1988 Ht increment in 1989 Source Df F-ratio P Df F-ratio P Df F-ratio P Cutl (2+B CH) 1 0.16 0.688 1 2.03 0.156 1 1.81 0.181 Cut2 (8+B CH) 1 2.15 0.146 1 0.05 0.829 1 0.03 0.866 Site age (S) 1 6.11 0.015 1 162.40 0.000 1 139.89 0.000 Treatment (T) 1 0.14 0.714 1 18.29 0.000 1 50.44 0.000 SxT 1 1.35 0.247 1 10.91 0.001 1 0.18 0.674 CutlxT 1 0.07 0.788 1 3.89 0.051 1 3.27 0.061 Cut2xT 1 0.20 0.653 1 0.06 0.805 1 0.03 0.873 Error mean-square 126 2.38 126 5.18 126 6.83 63 • 4 — Interference effects —• (A) Vegetation-not-removed H e i g h t Increment (cm) Diameter increment 8 0 6 0 4 0 2 0 0 • * - S i t e a g e a n d i n t e r f e r e n c e e f f e c t s - * 2 * B C H 8 » B C H - | 3 0 2 5 8 0 - 2 0 6 0 16 4 0 10 6 2 0 - 0 0 S i t e a g e a I H e i g h t increment \/A Diameter Increment (B) Vegetation-removed K r e m e n t (cm) Diameter increment (mm) 2 * B C H 8 * B C H 84te a g e a I H e i g h t Increment CZ2 Diameter Increment Figure 12. Comparison of the height and diameter increments of Sitka spruce after three growing seasons between the vegetation-not-removed and vegetation-removed planting treatments. The growth differences between the two site ages in graph A where the effects of the non-crop vegetation was not removed shows the combined site age and interference effect. The growth differences between the two site ages in graph B, where the effects of the non-crop vegetation was removed, shows the site age effect. The growth differences for the same site age between the two planting treatments (i.e. between graphs A and B) shows the interference effect. The vertical bars are one standard error of the mean. 30 to 36 seedlings were sampled per treatment. 64 site age effect. The growth differences for the same site age between the two planting treatments (graphs A and B) shows the interference effects. The effects of interference by salal reduced Sitka spruce height and diameter growth by almost 2 times on both site ages, whereas the effects of the site ages reduced height increments by 4 times and diameter increments by 3 times (Figure 12). These results shows that the effects of the site age on the growth of Sitka spruce is proportionally more important than the effects of interference by salal. Table 7 compares the percent mycorrhizal colonization of Sitka spruce fine-roots between the vegetation-not-removed and vegetation-removed treatments on the 2 years after clear-cutting and burning CH site age when the seedlings were 3-year-old. Five types of mycorrhizal association were identified. No statistical difference (P=0.856) in total percent mycorrhizal infection was found between the two planting treatment. For both planting treatments, the percent mycorrhizal infection on spruce roots was greater than 99%. There was, however, statistically more (P=0.002) and less (P=0.02) percent colonization by Cenococcum geophilum and Telephora terrestris, respectively, on the vegetation-not-removed than the vegetation-removed treatments. NUTRIENT CONTENT OF NON-CROP VEGETATION Figure 13 shows the increase in the amount of N and P within the non-crop vegetation across the chronosequence of CH site ages. On average, approximately 9 and 0.8 kg ha"* of N and P, respectively, were tied up in the accumulating biomass of competing vegetation annually over the 2 to 8 years after clear-cutting and burning period. These quantities do not include nutrients in undecomposed 65 Table 7. Comparison of percent mycorrhizal colonization on Sitka spruce fine-roots between the vegetation-not-removed and vegetation-removed planting treatments on the 2-year-old CH sites. Values in parentheses are one standard error of the mean. Numbers in columns followed by the same lower case letter are not significantly different (P>0.05) between site types. NM: Non-mycorrhizal; CG: Cenococcum geophillum; B: Brown; TT: Thelophora terrestris. Eight seedlings were sampled for each treatment. Planting treatments Percent mycorrhizal colonization for specified mycorrhizal types (average per seedling) NM CG B TT Other Total Vegetation-not-removed 0.7a (0.2) 12.3b (2.9) 7.0a (2.9) 68a (6.1) 12a (4.9) 99.3a (0.2) Vegetation-removed 0.5a 0.8a 0.5a 88.3b 9.9a 99.5a (0.2) (0.2) (0.2) (2.5) (2.2) (0.2) 66 120 100 80 60 40 20 0 N (kg/ha) P (kg/ha) R o o t & r h i z o m e S h o o t N 2 2 3 R o o t & rhizomeEESS S h o o t P 2 4 8 Y e a r s a f t e r c l e a r - c u t t i n g a n d b u r n i n g Figure 13. Total amount of N and P contained in above-ground (shoot) and below-ground (root + rhizome) biomass of non-crop vegetation on the 2, 4 and 8 years after clear-cutting and burning CH sites. 67 (not quantified). The N/P ratio of the non-crop vegetation of mainly salal was approximately 11 on all sites (Figure 13). FOREST FLOOR PROPERTIES AND MICROENVIRONMENT Table 8 compares the forest floor properties between the 3 and 9 years after clear-cutting and burning CH site ages. Because no statistical difference (at P<0.1) was found between the cutovers within each site age for most forest floor variables, the forest floor data from the two cutovers within each site age were combined. The statistical P-values for the comparison between the two site ages and two depths for each of the forest floor variables are shown in Table 8. With the exception of resin NH4+, pH, and CO2 evolution, all forest floor variables indicated greater fertility on the 3 than on the 9 years after clear-cutting and burning CH site age. Similarly, with the exception of extractable NH4+ and all resin variables, all forest floor variables indicated a decline in fertility between the upper and lower sampling depths. There was an unexplainable discrepancy between the results from the extractable and mineralizable NH4+ and the resin-NH4+. Further work is required to explain this discrepancy. Ion-exchange resin bags and cellulose discs were used to compare nutrient availability and relative decomposition rate between the vegetation-not-removed and vegetation-removed planting treatments. No significant (at P<0.1) difference was found in the relative decomposition rate of the cellulose discs between the vegetation-not-removed and removed treatments, but significantly more (P<0.05) resin-P (25% more) and NH4+ (36% more) were found for the vegetation-removed treatment. These increases in P and NH4+ availabilities were obtained even though the roots of the non-crop vegetation had been partly replaced by the increase in roots of the better growing conifer seedlings. Since ion-exchange resin bags are reported to behave similarly to plant roots in terms of ion uptake 68 Table 8. Comparison of some of forest floor properties between the 3 and 9 years after clear-cutting and burning CH site ages. Values in parentheses are one standard error of the mean. 3 yrs 9 yrs ANOVA (P-value) Forest floor depths Factors 0-8 cm 8-25 cm 0-8 cm 8-25 cm Sites Depth pH 4.47 (0.10) 4.03 (0.07) 4.33 (0.04) 3.91 (0.03) 0.156 0.000 C/N ratio 45.2 (1.7) 60.7 (2.5) 58.3 (5.4) 68.5 (3.4) 0.024 0.000 COo evolution 0.56 (mg C02/g/24 hr) (0.05) 0.38 (0.04) 0.59 (0.04) 0.33 (0.02) 0.874 0.000 Total N (%) 1.25 (0.05) 0.95 (0.03) 1.10 (0.06) 0.84 (0.03) 0.010 0.000 Total P (%) 0.068 (0.017) 0.045 (0.013) 0.063 (0.017) 0.042 (0.012) 0.082 0.000 Extractable NH4+ (ppm) 0.14 (0.01) 0.13 (0.02) 0.05 (0.01) 0.05 (0.01) 0.000 0.560 Mineralizable NH4+ (ppm) 0.37 (0.03) 0.24 (0.02) 0.28 (0.02) 0.17 (0.01) 0.006 0.000 Extractable P (ppm) 0.220 (0.056) 0.043 (0.016) 0.005 (0.000) 0.001 (0.000) 0.000 0.000 8 cm 20 cm 8 cm 20 cm Sites Depths Resin N H 4 + (mg g_1) 0.472 (0.015) 0.469 (0.011) 0.580 (0.035) 0.632 (0.027) 0.000 0.463 Resin N O 3 " (mg g"1) 0.047 (0.002) 0.047 (0.001) 0.045 (0.002) 0.042 (0.002) 0.657 0.743 Resin P (mg g_1) 0.160 (0.030) 0.165 (0.027) 0.053 (0.011) 0.025 (0.006) 0.000 0.132 Cellulose decomposition (% lost) 25.9 (4.0) 19.7 (3.8) 27.3 (4.7) 19.7 (4.3) 0.823 0.010 69 (Gibson, 1986), the nutrients that they capture is greatly influenced by the amount of roots present nearby. No soil moisture deficit was observed throughout the summers of 1987 and 1988 on any site age and for any planting treatment. All soil matric water potential values were above -0.021 MPa. Furthermore, no difference in soil matric water potential or gravimetric soil moisture was measured between site ages. Soil temperature was generally a few degrees (0.5 to 2.5°C) higher on the 2 than on the 8 years after clear-cutting and burning CH site ages in 1987 but not in 1988, and on the vegetation-removed as compared to the vegetation-not-removed treatments (Table 9). The differences were statistically significant (P<0.01) mainly at the 3 and 10 cm depths. POT SEEDLING BIOASSAYS The combined Sitka spruce height and diameter growth data for site ages 3 and 9 for the spruce-salal 4-0 combination had statistically lower values (P<0.01) than the comparable data for the 3-2 or 2-4 combinations (Figure 14). It was observed that on average one Sitka spruce seedling produced four to six times more biomass (above- and below-ground) over the two growing seasons than one salal plant. Therefore, the Sitka spruce seedlings from the 3-2 and 2-4 spruce-salal combinations suffered from less potential competition (combined intraspecific and interspecific) than the Sitka spruce from the 4-0 spruce-salal combination based on biomass data alone. On the 3-year-old site, the 4-6 combination had lower growth values than the 4-0 combination, whereas there was no statistical difference between the two combinations on the 9-year-old site. No statistically significant differences (at P<0.05) in height and diameter growth and total mycorrhizal infection (both in terms of total percent colonization and proportion of 70 Table 9 Comparison of soil temperatures at depths of 3,10, and 25 cm between the 2 and 8 years after clear-cutting and burning CH site ages. VNR=Vegetation-not-removed treatment; VR= Vegetation-removed treatment. Values in parentheses are one standard error of the mean. Soil temperature at different soil depths for the 2 different site ages Sampling date Planting & sky conditions treatment 3 cm 10 cm 25 cm 3 cm 10 cm 25 cm 2 yrs 8 yrs July 12/87 VNR 17.6 15.4 15.5 13.8 (sunny day) (0.3) (0.2) (0.6) (0.4) VR ~ ~ ~ ~ — — May 23/87 VNR 12.3 8.5 9.8 8.1 (cloudy day) (0.5) (0.4) (0.5) (0.6) VR — — — — — ~ 3 yrs 9 yrs July 20/88 VNR 19.4 14.9 19.1 14.8 (sunny day) (0.6) (0.5) (0.7) (0.5) VR 20.2 15.2 — 20.6 15.0 (0.5) (0.5) (0.5) (0.4) June 25/88 VNR 18.8 15.6 11.8 18.2 15.5 13.4 (Partly cloudy) (2.0) (0.8) (0.5) (0.9) (0.6) (0.3) VR 21.0 15.7 12.0 23.0 17.6 12.1 (1.8) (0.6) (0.4) (0.9) (1.1) (0.5) August 13/88 VNR 21.6 17.1 14.9 19.3 17.2 13.7 (sunny day) (0.9) (0.4) (0.3) (0.5) (0.4) (0.2) VR 71 3 years site age 50 40 30 20 10 0 Height increment (cm) Diameter Increment (mm Height EZ2 Diameter 4-6 4-0 3-2 Spruce-salal combination 2-4 9 year site age 50 40 30 20 10 Height increment (cm) Diameter increment (mm Height V/A Diameter 4-6 4-0 3-2 Spruce-salal combination 2-4 Figure 14. Comparison of height and diameter increments over two growing seasons of Sitka spruce seedlings growing in pots containing forest floor materials taken from the 3 and 9 years after clear-cutting and burning CH site ages between the different spruce-salal combinations. Each bar represents an average for 3 pots. 7 2 colonization by different types of mycorrhizal fungi) was found between the spruce-salal 4-6 and 4-0 combinations. For both spruce-salal combinations, the total percent mycorrhizal infection on spruce roots was greater than 96%. In the second pot experiment, statistically (P<0.01) greater height and diameter growth were generally found in the growth medium from the first year after clear-cutting and burning CH site age than from the 3 and 9 year site ages for both soil depths (Figure 15). The 3 year site age produced significantly greater height growth than the 9 year site age for both the 0-8 cm and 8-20 cm depths, whereas this was true for diameter growth only for the 0-8 cm depth. Statistically lower growth (P<0.01) was also obtained from the 8-20 cm depth substratum than the 0-8 cm depth substratum for all site ages. These results confirm those obtained in the field (Table 8; Figures 11 and 12). 73 Figure 15. Comparison of height and diameter increments over two growing seasons of Sitka spruce seedlings growing in pots between the forest floor materials taken from the 1, 3, and 9 years after clear-cutting and burning CH site ages and two different forest floor depths. Each bar represents an average for 3 pots, each pot containing 4 seedlings. 74 DISCUSSION The results of the field seedling bioassay indicated that the growth of Sitka spruce seedlings was improved markedly by the removal of the competing vegetation of mainly salal. Moreover, the needles of the seedlings where the competing vegetation had been removed were greener than the needles where the competing vegetation had not been removed, suggesting increased nutrient uptake. Weetman et al., (1989a,b) showed that Sitka spruce growing on such CH sites are deficient in both N and P, and that the application of fertilizer alleviated the nutritional stress. My results confirm that salal interferes with the availability of an adequate supply of N and P to Sitka spruce, and that this occurs early after planting. No moisture deficit was found on these CH site ages throughout the year, and therefore competition for water was ruled out as a possible factor. Moreover, the soil temperature in the first 10 cm of the forest floor in both the vegetation-not-removed and removed planting treatments did not differ much (Table 9), and was within the range for optimal root growth reported by Coutts and Philipson (1987). Many studies done by Ingestad and co-workers (Ingestad, 1973, 1979, 1981; Ingestad and Lund, 1986) have reported the optimum growth for many plant species (e.g. Pinus silvestris, Picea abies, Alnus incana, Betula verrucosa,  Vaccinium vitis idaea, Vaccinium myrtillus) to occur when the N/P ratio in plant tissues lies between 5 and 8. The high N/P ratio (i.e. approximately 11) found in the non-crop vegetation of mainly salal growing on these sites (Figure 13) strongly suggests that plants were suffering of phosphorus deficiencies. Similar high N/P ratios have been found in salal leaves from adjacent sites (Weetman et al. 1990). 75 The amount of N available annually for plant uptake on these poor sites following clear-cutting and burning has been estimated by Weetman et al. (1990) to be between 20 and 30 kg ha"*. This figure is in the lower part of the range of annual N uptake requirements for conifer stands (6.5 to 88 kg N ha'l yr~l) reported by Cole and Rapp (1981). From Figure 13, it was calculated that approximately 9 kg ha'l of N yr"^  is tied up in living tissue of the competing vegetation; this constitutes between 30 and 45% of Weetman et al.'s estimate of the potentially available N on the site. If Weetman et al.'s estimate is correct, this immobilization of N in salal biomass can therefore potentially explain much of the differences in growth between the two planting treatments reported in Figure 12. It is believed that the development of competing vegetation biomass will continue for some time beyond 8 years before it reaches its maximum development (Chapter 3; Figure 10). Sabhasri (1961), for example, estimated the salal biomass under an open 120-yr-old Pseudotsuga menziesii stand of site quality V in Washington state at 35,000 kg, almost twice the maximum amount found in this study. No data on total salal biomass are available for older CH stands or sites that are similar to the 2- to 10-year-old CH sites reported here. Sitka spruce growth was significantly lower on the 8 than on the 2 years after clear-cutting and burning CH sites for both planting treatments (Figures 11 and 12). The second pot experiment also showed that in the absence of competing vegetation the growth of Sitka spruce was lower on forest floor from 9-year-old CH site age than from either 1- or 3-year-old CH site ages (Figure 15). Table 8 shows a decline in the availability of N and P nutrients in the forest floor from 2 to 8 years after burning on these CH site ages. Germain (1985), working on similar sites nearby, reported the post-disturbance "assart" flush of nutrients to last less than 5 years, based on a chronosequence study. This initial increase in nutrient availability following forest harvesting followed by a decline a few years later is a 76 well known phenomenon (Covington, 1981; Krause and Ramlal, 1986; David, 1987). Following the early flush of nutrients, the remaining forest floor is composed largely of rather decay-resistant material which releases nutrients slowly. This decline in nutrient availability in the forest floor measured at 8 years after clear-cutting and burning has been shown to be assocatied with a smaller height and diameter growth of Sitka spruce in comparison to what was measured at 2 years after clear-cutting and burning. Moreover, this occurs both with and without the presence of salal. This decline in forest floor nutrient availability with time imposes nutritional stress on Sitka spruce growing on 8-year-old CH cutovers and older, in addition to that caused by salal competition (Figure 12). This helps explaining the decline in Sitka spruce growth 5 years after planting reported by Barker et al. (1987). Mycorrhizae are known to enhance nutrient uptake in trees (Perry et al., 1987). Walker (1987) reported that such enhancement has been shown experimentally in Sitka spruce for phosphate, nitrate, ammonium, potassium and simple organic nitrogenous compounds. A few studies have reported evidence of an allelopathic effect of shrubby species on mycorrhizal colonization of tree species (Cote" and Thibault, 1988; Robinson, 1972; Handley, 1963). More particularly, Handley (1963) reported that poor growth of Sitka spruce on sites covered by Calluna vulgaris in Scotland was associated with a lack of mycorrhizal development. Laboratory studies showed that extracts from mycorrhizal Calluna roots inhibited the growth of mycorrhizal fungi known to be associated with Sitka spruce (Robinson, 1972). Based on these latter two studies and on field observations on northern Vancouver Island, it has been hypothesized (Weetman et sd., 1989a,b; Germain, 1985) that the release of chemicals by salal into the root environment of Sitka spruce could inhibit the normal mycorrhizal infection of the spruce roots, and consequently the normal uptake of nutrient ions. 77 Neither the field seedling bioassay nor the pot experiment, where Sitka spruce seedlings were growing with and without salal, showed any difference in the total percent mycorrhizal infection. Therefore, the large differences in Sitka spruce growth between the vegetation-not-removed and vegetation-removed treatments found in the field on the 2-year-old sites were not due to a lack of mycorrhizal infection. Furthermore, in comparison to the salal-free 4-0 spruce-salal combination, the Sitka spruce growing in the 3-2 and 2-4 spruce-salal combinations in the first pot experiment (Figure 14) did not show any sign of drastic growth reduction due to the presence of salal. The intraspecific interference by Sitka spruce appeared to be more significant than the interspecific interference by salal, and this difference could be explained by the differences in total biomass measured between the different combination of pots. Finally, a greenhouse bioassay in another study failed to find evidence that the presence of salal directly affected Sitka spruce's ability to take up nutrient ions (McDonald, 1989). These results indicate that the onset of nutritional stress and reduced growth reported in Sitka spruce plantations 5 years after planting on CH sites on northern Vancouver Island can be explained by a combination of (1) salal competition for scarce nutrients and their subsequent immobilization in salal biomass, and (2) declining availability of nutrients in the forest floor following the flush of nutrients associated with clear-cutting and burning. The hypothesis regarding the inhibition by salal of the mycorrhizal infection of Sitka spruce roots was not verified, but neither was it completely rejected. Even though the roots of Sitka spruce were fully mycorrhizal, it is possible that the mycorrhizae are not fully functional in the presence of salal and a difference in the relative importance of two mycorrhizal syrhbionts was noted (Table 7). On-going research is testing 78 the hypothesis of salal's allelopathic effect on mycorrhizal function in the conifer seedlings (S. Berch, pers. comm.). It is interesting to note that N and P deficiencies in young plantations of Sitka spruce in Scotland have been found on sites where Calluna is absent, infrequent or has been killed (Dickson and Savill, 1974; Mcintosh, 1980; Mcintosh, 1983) suggesting that there is a strong site component in the phenomenon. Sitka spruce is considered a nutrient demanding species (Germain, 1985), especially prior to canopy closure (Miller and Miller, 1987), and there are apparently insufficient nutrients on the salal-dominated CH sites on Vancouver Island and on the Scottish sites to sustain rapid growth of this species. The advent of canopy closure conditions resulting in (1) a greater proportion of the annual demand for nutrients being satisfied by internal cycling and (2) a stabilization or reduction in salal biomass, especially fine-root biomass, which reduces competition for nutrients, may reduce the nutritional stress observed in currently stagnated 8- to 14-year-old Sitka spruce plantations (Mcintosh, 1983; Miller and Miller, 1987). This hypothesis is now under investigation by other researchers. 79 SUMMARY The effects of salal and of a reduction in forest floor nutrient availability on Sitka spruce early growth was investigated on an age sequence of CH sites up to 10 years after clear-cutting and burning. No soil moisture deficit was measured in the field throughout the year. Moreover, soil temperature in the forest floor was found to be within the range for optimal root growth for Sitka spruce on both site ages. The regrowth of above- and below-ground non-crop vegetation immobilized annually about 9 ka ha"l for at least the first 8 years following clear-cutting and burning. This was estimated to represent between 30 and 45% of the available N on these sites. The removal of the non-crop vegetation around individual Sitka spruce seedlings increased the availability of NH4+ by 36% and phosphate-P by 25%, did not affect cellulose decomposition and soil moisture (i.e. water tension), and slightly increased soil temperature. Therefore, the immobilization of nitrogen and of other nutrients in expanding salal biomass and other non-crop species could provide an explanation for much of the large differences in Sitka spruce growth found between the vegetation-not-removed and vegetation-removed planting treatments. Moreover, forest floor analyses and field and pot seedling bioassays demonstrated a reduction in the forest floor nutrient availability in the period 8 to 10 years after clear-cutting and burning in comparison to the period 2 to 4 years. This was shown to contribute to the early growth stress experienced by Sitka spruce on these CH sites in addition to the effects of salal. Sitka spruce fine-roots growing with and without salal had very high percent mycorrhizal colonization in both field and pot experiments. Furthermore, no sign of a dramatic growth reduction in Sitka spruce seedlings due to the presence of salal was found in a pot experiment. 80 These results indicate that the onset of nutritional stress and reduced growth reported in the Sitka spruce plantations 5 years after planting on northern Vancouver Island can be explained by a combination of (1) salal competition for scarce nutrients and their subsequent immobilization in salal biomass, and (2) declining forest floor nutrient availability caused by the termination of the flush of available nutrients that occurs in the immediate after clear-cutting and burning period. Although the results do not confirm that all of the observed nutritional stress can be attributed to these two factors alone, and do not provide a basis for rejecting alternative hypotheses, they demonstrate that these two factors could account for much of the observed growth problem. 81 C H A P T E R F I V E E F F E C T S O F C O M P E T I N G V E G E T A T I O N , S L O W - R E L E A S E F E R T I L I Z E R A N D D I F F E R E N T SITE CONDITIONS O N T H E E A R L Y G R O W T H O F W E S T E R N H E M L O C K , W E S T E R N R E D C E D A R A N D SITKA S P R U C E S E E D L I N G S O N C H A N D H A SITES I N T R O D U C T I O N In Chapter 4,1 discussed the potential causes for the nutritional stress reported in Sitka spruce seedlings and saplings growing on CH cutovers. Similar nutritional stress can be found in western hemlock, and to a lesser extent in western redcedar. However, conifer plantations of these same conifer species established on adjacent HA cutovers are growing much better, and do not show any sign of nutritional stress. Although it is well recognized that the growth of planted conifer seedlings varies between the CH and HA ecosystems (Germain, 1985; Barker et al. 1987; Weetman et al. 1990), no proper comparative experiment has been done to quantify the growth performance of these different conifer species between these two forest ecosystems. Moreover, very little information is available about the factors associated with these growth differences. The main objective of the study reported in this chapter was to compare the growth of three conifer species under different experimental conditions in order to determine which factors are associated with the very different conifer seedling growth found between CH and HA cutovers. A secondary objective was to assess slow-release N-P-K fertilization at the time of planting as a way to stimulate rapid stand establishment on CH sites. To achieve these objectives, the early growth of western hemlock, western redcedar and Sitka spruce seedlings were compared on CH sites that were 2 and 8 years post-clear-cutting and burning (hereafter 82 referred to as 2+B CH and 8+B CH, respectively) and on HA sites that were 2 years after clear-cutting and burning (hereafter referred to as 2+B HA). On each of these three site-age combinations, three planting treatments were applied to the seedlings: (1) no removal of non-crop vegetation, (2) removal of all non-crop vegetation and (3) application of slow-release fertilizer at the time of planting without any removal of non-crop vegetation. The effects of the different types of sites and of salal on the growth of three conifer species were investigated both in the field and in pots. 83 MATERIALS AND METHODS FIELD SEEDLING BIOASSAYS Field studies were initiated in the summer of 1987 on the same age sequence of mesic clear-cut and burned CH sites as for the experiments described in Chapters 3 and 4. In addition, similar experiments were installed on two cutovers of the HA ecosystem that had been clear-cut and burned 2 years previously. These two HA cutovers were selected based on their homogeneity, their similarity, and their representativeness of the HA cutovers in the study area. They were situated within 300 m of the two 2-year-old CH cutovers; both had similar slope, aspect and burning history. In this study, I focuss on the comparison between the 2+B CH and the 8+B CH site age combinations, and between the 2+B CH and the 2+B HA site age combinations in order to investigate the effects of (1) time since clear-cutting and burning in the CH ecosystem, (2) differences in the amounts and types of non-crop vegetation, and (3) differences between forest ecosystems (CH vs HA). Nursery-grown "plug type" 1-0 seedlings of each of western redcedar, Sitka spruce, and western hemlock were planted in April 1987 on two cutovers for each of the three site-age combinations near Port McNeill, for a total of 810 seedlings. The seedlings were between 15 and 25 cm in height at the time of planting. The oven-dry biomass of the below- and above-ground components were respectively 0.71 g and 2.04 g for Sitka spruce, 0.68 g and 1.57 g for western hemlock, and 0.88 g and 2.11 g for western redcedar. These seedlings were left to grow for three growing seasons until the sites were 4,10 and 4 years after clear-cutting and burning, respectively. HA sites that were eight years after clear-cutting and burning were not investigated because by that age such sites are densely revegetated with 3 to 5 84 m tall western hemlock which made the establishment of my experiments impossible. The experiment was a 3x3x3 nested-factorial in which two cutovers were nested within each type of site. The three main factors investigated were: (i) 3 coniferous species (western redcedar, Sitka spruce and western hemlock), (ii) 3 site-age combinations (CH sites 2 and 8 years after clear-cutting and burning, and HA sites 2 years after clear-cutting and burning, and (iii) 3 planting treatments: (a) seedlings planted without any additional treatment except for the elimination of light competition (vegetation-not-removed treatment); (b) seedlings planted in the middle of 200 cm diameter patches from which all above-ground vegetation was continuously removed by clipping and from which below-ground competition from adjacent vegetation was eliminated by periodically cutting around the patches to a depth of 40 cm (vegetation-removal treatment: Figure 16); and (c) seedlings planted and supplied with 40 g of a 14-14-14 NPK slow release fertilizer (Osmocote, Sierra Chemical Company, California: nutrients released over a 3 month period at 25°C) placed at a depth of 10 cm in four holes equally spaced at a distance of 10 cm from the seedlings (fertilizer treatment). The vegetation surrounding the seedlings in all planting treatments was continually pruned sufficiently to eliminate any competition for light. This was done to emulate the natural field conditions in which conifer saplings are growing above a dense understory of non-crop vegetation. The height and basal diameter of each seedling were measured just after planting in the spring of 1987, and at the end of the 1987, 1988 and 1989 growing seasons. Orthogonal and non-orthogonal contrasts were used to compare the 85 Figure 16. Vegetation-removed treatment applied to western redcedar growing on the 2 years after clear-cutting and burning CH sites. Note that all the vegetation has been removed from an area one meter in radius all around the seedling. treatment means. For the contrasts that were non-orthogonal, I used the H' error instead of the H error (Sokal and Rohlf, 1981). The H' error was obtained by the following equation: H^l-Q-H) 1 ^. where k= degrees of freedom Log- or square-root-transformed values were used when the variances were not homogeneous using Bartlett's test. Both transformed and untransformed data 86 were checked for homogeneity of variances and normality of distribution. Only untransformed means are presented, but the statistics of some of the means were performed on transformed means. The effects of the vegetation-not-removed and vegetation-removed treatment on percent mycorrhizal colonization of all three conifer species were examined in 1989 using seedlings from the 2+B CH sites. A t-test was carried out to compare the frequency of each mycorrhizal type on each conifer species between the two planting treatmnets. Collection and preparation of the roots for the determination of the mycorrhizal infection was done as described in Chapter 4. Each root sample was divided into 3 subsamples and the % colonization by mycorrhizal fungi was assessed by observing the stained root system under a dissecting microscope. Western hemlock and Sitka spruce mycorrhizal status was assessed by calculating the percentage of root tips colonized by each one of the 5 different types as described in Chapter four. Western redcedar mycorrhizal status was assessed by calculating the proportion of root length colonized by VA mycorrhizae using the gridline intersect method (Kormanik and McGraw, 1982). The above- and below-ground biomass of 6 randomly chosen seedlings for each of the three conifer species growing in the vegetation-not-removed treatment was compared in 1989 between the 2+B CH and HA sites. The experiment was a 3x2 factorial in which three conifer species were compared on the two site-age combinations. The above-ground biomass was obtained by clipping each seedling at the forest floor level, whereas the below-ground biomass was obtained by manually removing the growth medium (mainly organic matter) surrounding the 87 roots of each seedling. It took on average 2.5 hours for 3 people to excavate one seedling. POT SEEDLING BIOASSAYS In March 1988 a pot experiment was initiated in an open area near the research sites on northern Vancouver Island. The experiment investigated the growth of Sitka spruce and western redcedar in forest floor material taken from the 0-8 cm depth of the 2+B HA, 2+B CH and 8+B CH sites. Western hemlock was not studies here in order to reduce the number of pots required. It was designed to provide a more controlled environment to evaluate the effects of the 3 site-age combinations on the growth of conifer seedlings growing without non-crop vegetation. The experiment was a 3x2 factorial (3 site-age combinations of site x 2 conifer species) using a completely randomized design. There were 4 seedlings per pot and 3 pots per treatment, for a total of 18 pots. Orthogonal contrasts were used to compare the treatment means within each growing season. No transformation were necessary. The growth medium for each site was thoroughly mixed and all rhizomes and most fine and medium roots were removed prior to filling the pots. All pots were maintained in full sunlight, and the soil was held near field capacity by watering. This was done to standardize as much as possible the environmental conditions in order to identify substrata effects on seedling growth. At the end of each growing season, the height and diameter increments of the seedlings were measured, and pot averages calculated. 88 FOREST FLOOR PROPERTIES The forest floor properties of the 0-8 cm and 8-20 cm depths for the 2+B CH and 8+B CH sites were described in Chapter 4. Similar forest floor analyses were performed on forest floor material from the 2+B HA sites (see Chapter 4, p. 53-55, for a description of the methodology). A one-way analysis of variance was carried out to compare the different forest floor factors measured between the three site-age combinations. The Tukey HSD multiple comparison test was used to compare the treatment means. Log- or square-root-transformed values were used when the variances were not homogeneous using Bartlett's test. Both transformed and untransformed data were checked for homogeneity of variances and normality of distribution. Only untransformed means are presented, but the statistics of some of the means were performed on transformed means. NON-CROP VEGETATION The non-crop vegetation data for the 2+B CH and 8+B CH sites are presented in Chapter 3. The above-ground biomass of the non-crop vegetation on the 2+B HA sites was estimated in 1987 and 1989 at the end of July on each cutover. One quadrat was clipped at every 7.5 m along each of two parallel line transects situated in the middle of each cutover for a total of six quadrat per line transect. The above-ground biomass collected from each quadrat was separated by species and then further divided into leaf and stem+fruit components. These biomass components were dried at 70°C for 48 hrs and weighed. No below-ground biomass assessment was made for this type of site. The percentage cover of each species was visually assessed at the end of July on 12 randomly located 1 m 2 quadrats on each cutover. An one-way analysis of variance was carried out to compare each of the non-crop vegetation parameters measured between the three 89 site-age combinations. The Tukey HSD multiple comparison test was used to compare the treatment means. Log- or square-root-transformed values were used when the variances were not homogeneous using Bartlett's test. Both transformed and untransformed data were checked for homogeneity of variances and normality of distribution. Only untransformed means are presented, but the statistics of some of the means were performed on transformed means. 90 RESULTS There were considerable diffrenc^ s in the amount of non-crop vegetation and in the forest floor properties between the three site-age combinations investigated in this experiment (Table 10). In 1989, both of the 2+B CH and 8+B CH sites were dominated by salal, whereas the 2+B HA sites had both fireweed and salal in almost equal amounts. Fireweed biomass did not change much from 2 to 4 years after clear-cutting and burning in the HA sites, whereas during that period salal biomass increased by more than 6 times, from 300 to 1950 kg ha-1. All three site-age combinations had similar total amounts of above-ground non-crop vegetation biomass in 1989, but the 8+B CH site had significantly (P<0.01) more below-ground biomass than the 2+B CH sites. No below-ground biomass data were available for the 2+B HA sites. Lack of resources prevented these measurements from being made. The 2+B CH sites had a significantly (P<0.05) higher total N, available P, mineralizable N, and Resin-P in the first 8 cm of the forest floor than the 8+B CH sites, whereas the 2+B HA sites had a significantly (P<0.05) thinner forest floor, smaller total N, slightly higher soil temperature, higher rate of cellulose decomposition, and higher amounts of Resin-NH^ than the 2+B CH sites. Similar differences were found for the 8-20 cm forest floor depth. Only small differences in soil temperature were found among the three site-age combinations at depths of 3,10 and 25 cm between 11:00 and 13:00 pm for 10 dates from May to September in 1987 and 1988; Table 10 reports the soil temperature at 10 cm for 4 dates in 1988 only. No difference in soil moisture (both gravimetric and soil tension) was measured at 10 cm between the three site-age combinations throughout the summer of 1987 and 1988. 91 Table 10. Comparison of planted conifers, main non-crop species, forest floor properties, and soil temperature between three site-age combinations. Numbers in rows followed by the same lower case letter are not significantly different (P>0.05) between site-age combinations. Site-age combinations 2+B HA 2+B CH 8+B CH Conifers planted by WFP species trees/ha age in 1989 (yrs) Height in 1989 (m) Non-crop vegetation in 1989 Cover (%) western hemlock 900 3 1.5 western redcedar 900 3 0.7 western redcedar 900 9 2.7 Salal: 30a 60b 65b Bunchberry: 0 7a 30b Fireweed: 45b 10a 8a Shoot biomass (kg/ha) Salal: 1950a 3626b 4078b Bunchberry: 0 26a 406b Fireweed: 1910b 293a 169a Total: 3860a 3945a 4653a Root biomass (kg/ha) Gaultheria-Vaccinium: 3561a 10079b Epilobium-Cornus: — 549a 1336b Forest floor properties as measured in 1988 (0-8 cm depth) Thickness (cm) 10-40 20-70 20-70 pH (0-8 cm) 4.49 a 4.47 a 4.33 a Total N (%) 1.05 a 1.25 b 1.10 a Total P (%) 0.072 a 0.068 a 0.063 a Extractable P (ppm) 0.20 b 0.22 b 0.005 a C/N ratio 52 a 45a 58 a Min. N (ppm) 0.37 b 0.37 b 0.28 a Cellulose dec. (%) 65.0 b 25.9 a 27.3 a Resin-NH "^1" (mg/g) 0.873 b 0.472 a 0.580 a Resin-No3 (mg/g) 0.043 a 0.047 a 0.047 a Resin-P (mg/g) 0.124 b 0.160 b 0.053 a Soil Temperature (at 10 cm) Sept. 15/88 (sunny day) 16.4 b 14.4 a — July 20/88 (sunny day) 20.5 a 19.4 a 19.1a June 25/88 (cloudy day) 16.2 b 15.6 a 15.5 a May 23/88 (cloudy day) 11.4 b 12.3 b 10.1a 92 FIELD SEEDLING BIOASSAY The biotic and abiotic differences reported in Table 10 between the three site-age combinations were associated with large differences in early growth of western hemlock, western redcedar and Sitka spruce in the field (Figure 17). Large, statistically significant (P<0.001) differences were found between the three sites, the three planting treatments and the three species (Table 11). In general, both total height and diameter increments were found to be significantly (P<0.001) greater on the 2+B HA than the 2+B CH, and on the 2+B CH than on the 8+B CH sites. Both vegetation-removed and fertilized treatments significantly (P<0.001) increased total height and diameter increments over the vegetation-not-removed treatment. The increase was particularly striking for western hemlock on the 2+B CH and 8+B CH sites (Figure 17). Western redcedar total height and diameter increments were significantly (P<0.001) lower than those of Sitka spruce and western hemlock. Sitka spruce diameter increment was significantly (P<0.001) greater than that of western hemlock, whereas western hemlock height increment was significantly (P=0.002) greater than that of Sitka spruce. No significant (at P<0.05) difference was obtained between cutovers within each site-age combination, except between the cutovers in the 8+B CH sites in which case a small significant difference was found for total height (Table 11). All of the interactions between site-age combinations, species and planting treatments were significant at P<0.01, but only the interaction between site-age combinations and species was important (Le^  contributed more than 5% to the total sum of squares of all the factors and interactions). This latter interaction indicated that western redcedar was a lot less affected by the differences found between the three site-age combinations regardless of the planting treatments than Sitka spruce and western hemlock (Figure 18). This figure also illustrates WESTERN HEMLOCK Height increment after 3 seasons (cm) 160 100 8+B CH 2+B CH Site types 2+B HA WESTERN HEMLOCK Diameter Increment after 3 seasons (mm) 8+B CH 2+B CH Site types 2+B HA WESTERN REDCEDAR Height Increment after 3 seasons (cm) 160 100 8+B CH 2+B CH Site types 2+B HA WESTERN REDCEDAR Diameter Increment after 3 seasons (mm) 8+B CH 2+B CH Site types 2+B HA SITKA SPRUCE Height Increment after 3 seasons (cm) 150 100 50 Veg.-not-removed Veg.-removed Fertilized aJLl 8+B CH 2+B CH Site types 2+B HA 30 25 20 16 10 SITKA SPRUCE Diameter Increment after 3 seasons (mm) HB Veg.-not-removed INN Veg.-removed \Z2 Fertilized r-d 8+B CH 2+B CH Site types 2+B HA Figure 17. Comparison of total height (A to C) and diameter (D to F) increments of western hemlock, western redcedar and Sitka spruce seedlings after three growing seasons between the three planting treatments and three site-age combinations. The vertical bars are one standard error of the mean. n=30 to 36. Table 11. ANOVA summary table showing variance ratios (F), P-values and error mean-square for total height and diameter increments of western hemlock, western redcedar and Sitka spruce over three growing seasons between the three planting treatments and three site-age combinations .The data for the total height and diameter increments were transformed using the square-root transformation. Total height Total diameter Source Df F-ratio P Df F-ratio P Cut 1 (2+B HA) 1 1.34 0.248 1 0.56 0.453 Cut 2 (2+B CH) 1 0.02 0.895 1 2.93 0.088 Cut 3 (8+B CH) 1 4.63 0.032 1 2.66 0.103 Site-age (S) 2 600.30 0.000 2 635.92 0.000 2+B HA vs 2+B CH 0.000 0.000 2+B CH vs 8+B CH 0.000 0.000 Treatment (T) 2 144.23 0.000 2 228.08 0.000 Fert. vs Veg.-not-removed 0.000 0.000 Veg.-removed vs Veg.-not-removed 0.000 0.000 Species (Sp) 2 26.36 0.000 2 42.73 0.000 WRC vs SS + WH 0.000 0.000 SS vs WH 0.002 0.000 SxT 4 6.85 0.000 4 4.03 0.003 SpxT 4 5.98 0.000 4 5.05 0.001 SxSp 4 67.57 0.000 4 48.49 0.000 SxSpxT 8 7.45 0.000 8 7.50 0.000 CutlxT 2 0.48 0.615 2 1.03 0.360 Cut2xT 2 4.34 0.013 2 7.50 0.000 CutlxSp 2 1.40 0.247 2 0.17 0.849 Cut2xSp 2 3.93 0.019 2 4.72 0.009 Cut3xT 2 2.64 0.073 2 2.59 0.077 Cut3xSp 2 1.96 0.141 2 2.16 0.116 CutlxTxSp 4 1.488 0.204 4 1.55 0.187 Cut2xTxSp 4 1.58 0.178 4 0.97 0.422 Cut3xTxSp 4 0.67 0.613 4 0.19 0.945 Error mean-square 867 8.70 870 1.53 95 Diameter increment over 3 seasons (mm) - • - western redcedar —*— western hemlock -- ° Sitka spruce -• 1 2+B HA 2+B CH 8+B CH Sites F i g u r e 18. C o m p a r i s o n of he igh t a n d d iame te r i nc remen ts over th ree g r o w i n g seasons be tween the three coni fer species a n d th ree s i te-age comb ina t ions . E a c h va lue for each species a n d type of s i te represen ts the average for the three p l a n t i n g t rea tmen ts . T h i s f igure i l l us t ra tes the i n te rac t i on be tween si te-age comb ina t ions a n d species s h o w n i n T a b l e 11. 96 that both Sitka spruce and western hemlock show a very similar pattern of height and diameter increments between the three different site-age combinations. The other statistically significant, but less important, interactions indicated that (1) western redcedar total height and diameter increments were less affected by the removal of the non-crop vegetation than were western hemlock and Sitka spruce (Species x Treatments), and (2) the differences in western redcedar total height and diameter increments between planting treatments and site-age combinations differed slightly from those of western hemlock and Sitka spruce (Site-age combinations x Treatments x Species). The interactions between cutovers and the main treatments were either statistically non-significant (P>0.05) or, if significant, not important (Le, small sum of squares). The general pattern of growth differences obtained between site-age combinations, planting treatments and species were very similar for both total height and diameter increments (Figure 17), and therefore the same general conclusion can be drawn from either of these two growth variables. Figure 19 compares the annual height increment of western hemlock, western redcedar and Sitka spruce between the three planting treatments within each of the three site-age combinations. One year after planting, the height increment was significantly (P<0.001) greater for the fertilized treatment than for the vegetation-not-removed and vegetation-removed treatments for all three species on all site-age combinations; no significant (at P<0.01) difference was found between these two later treatments. One year after planting, there were also some statistically (P<0.01) significant differences between conifer species within and between site-age combinations. Two years after planting, the height increment for the fertilized treatment was still generally significantly (P<0.001) 97 Western Hemlock Western Redcedar Sitka Spruce 1 2 a 1 * 1 1 1 3 Years after planting Figure 19. Comparison of the annual height increment of western hemlock, western redcedar and Sitka spruce seedlings for the first three years after planting on 8+B CH, 2+B CH and 2+B HA sites between the vegetation-not-removed (— O —), vegetation-removed ( » " C i " . i ) , and fertilized (-- A --) planting treatments. 98 greater than the vegetation-not-removed and removed treatments. Also, the height increment for the vegetation-removed treatment was significantly (P<0.001) greater than the vegetation-not-removed treatment for all species on all site-age combinations, except for Sitka spruce on the 2+B HA and 8+B CH sites, and for western redcedar on the 2+B CH site. Three years after planting, the height increment for the vegetation-removed treatment had surpassed both the vegetation-not-removed and fertilized treatments for all species on all site-age combinations, except for western redcedar on the 2+B CH and 2+B HA sites. By that time, the height increment for the fertilized treatment on both 2+B CH and 8+B CH sites was reduced below the level obtained the second year, except for western redcedar on the 2+B CH sites in which case there was no change. Associated with this reduced growth in the third year were symptoms of nutrient deficiency (yellowing of all foliage, reduction in size of new foliage, and reduced leader growth) (Figure 20). The slow-release fertilizer treatment had only a small effect in improving the growth of conifer seedlings on the 2+B HA sites, except for western redcedar where the effect was greater than on the 2+B CH sites (Figures 17 and 19); no visual symptoms of nutrient deficiency were apparent on this type of site in the third year. The percent mortality of planted conifer seedlings three years after planting was less than 2% for all three species on all site-age combinations and for all planting treatments, except for western hemlock and Sitka spruce on the vegetation-not-removed treatment on the 8+B CH sites where it was 8 and 10%, respectively. However, an inspection made at the end of the fourth growing season after planting (i.e. 1990) revealed that the percentage of dead western hemlock and Sitka spruce on the 8+B CH sites for the vegetation-not-removed treatment had increased to approximately 70%. In comparison, the percentage mortality of 100 SHOOT AND ROOT BIOMASS Root and shoot dry weight and shoot/root ratios of seedlings excavated at the end of the third growing season for the vegetation-not-removed treatment on the 2+B CH and HA sites are shown in Table 11. On the 2+B CH sites, Sitka spruce had the highest shoot and root dry weight and lowest shoot/root ratio of all three species; western hemlock had the highest and lowest values, respectively on the 2+B HA sites. Western redcedar had a significantly (P<0.001) lower shoot and root dry weight and a higher shoot/root ratio than western hemlock and Sitka spruce on both sites 2+B CH and HA sites. The three conifer species combined had a significantly (P<0.001) higher root and shoot dry weight and shoot/root ratio on the 2+B HA than on the 2+B CH sites. Western hemlock and Sitka spruce seedlings produced long lateral roots (up to 2.5 m long compared with mean seedling total height of 80 cm) in the top 5 cm of the forest floor of the 2+B CH site, whereas western redcedar produced shorter roots (up to 0.8 m long compared with mean seedling total height of 85 cm), but these were more evenly distributed in the forest floor than the roots of the hemlock and spruce. MYCORRHIZAL COLONIZATION Five types of mycorrhizal fungi were recognized on western hemlock and Sitka spruce seedlings growing on the 2+B CH sites sampled in 1989. No significant (at P<0.05) difference in percent mycorrhizal colonization was obtained for western hemlock between the two planting treatments for any of the mycorrhizal types recognized. However, significantly more (P=0.005) and less (P=0.02) percent mycorrhizal infection was found on Sitka spruce roots for the types Cenococcum geophilum and Telephora terrestris, respectively, on the vegetation-not-removed compared to the vegetation-removed treatments (Table 7). 99 Figure 20. Symptoms of nutrient deficiency in Sitka spruce seeldings growing on the 2+B C H sites appearing in the third year after planting for the fertilized treatment. Note the yellow foliage, stunted new needles and reduced leader growth. planted seedlings for the other planting treatments had increased only very slightly. Clearly, the 8+B C H sites differ drastically in seedling survival to the fourth post-planting year as well as in growth of these two species. The cause of mortality was not determined, but it is thought to be related to the severe nutritional stress experienced on the 8+B C H site with no control of competition by non-crop vegetation. Table 12. Comparison of the shoot and root dry mass (g) and shoot/root ratios of 3-year-old western hemlock, western redcedar and Sitka spruce seedlings for the vegetation-not-removed treatment between the 2+B CH and HA sites. Values in parentheses are one standard error of the mean. Types of Component Western Western Sitka site hemlock redcedar spruce 2+B CH Root 15.1 (4.2) 6.6 (1.9) 31.2 (8.2) Shoot 40.2 (12.7) 22.5 (6.9) 68.5 (21.2) Shoot/root 2.54 (0.14) 3.52 (0.36) 2.04 (0.15) 2+B HA Root 66.1 (17.2) 9.6 (1.1) 58.1 (8.7) Shoot 211.1 (61) 37.7 (5.6) 189.1(19.7) Shoot/Root 3.26 (0.35) 3.96 (0.46) 3.45 (0.33) Note: Six seedlings per species per site were sampled. 102 For both planting treatments and both conifer species, the total percent mycorrhizal colonization was greater than 90%. No significant (P=0.874) difference in percent mycorrhizal colonization was found on western redcedar growing in the 2+B CH sites in 1989 between the vegetation-not-removed compared to the vegetation-removed treatments. For both planting treatments the total percent mycorrhizal colonization was greater than 95%. POT SEEDLING BIOASSAY Figure 21 shows the annual height and diameter increments for 1988 and 1989 of Sitka spruce and western redcedar seedlings growing in pots with forest floor materials taken from the 2+B HA, 2+B CH and 8+B CH sites. The height and diameter increments of Sitka spruce for the first growing season (i.e. 1988) were significantly (P<0.01) greater than those of western redcedar growing in forest floor taken from both 2+B CH and 8+B CH sites, whereas the height increment was significantly (P<0.01) smaller in spruce than redcedar seedlings in forest floor taken from the 2+B HA sites. For the second growing season (i.e. 1989), however, the height and diameter increments of western redcedar were significantly (P<0.01) greater than those of Sitka spruce for all three site-age combinations. The height and diameter increments of Sitka spruce were significantly (P<0.01) greater on the 2+B HA than on the 2+B CH site material and on the 2+B CH than on the 8+B CH site material for both growing seasons. For western redcedar, the height and diameter increments were also significantly (P<0.01) greater on the 2+B HA than on the 2+B CH site material, but they were significantly (P<0.01) smaller on the 2+B CH than on the 8+B CH sites material for the 1989 growing season. 103 Annual height increment (cm) 88 89 2+B HA 88 89 2+B CH 88 89 8+B CH Annual diameter increment (mm) L^l Sitka spruce V/A western redcedar 88 89 2+B HA 88 89 2+B CH 88 89 8+B CH Sites Figure 21. Comparison of the 1988 and 1989 annual height and diameter increments of Sitka spruce and western redcedar seedlings growing in pots containing forest floor materials taken from the 2+B HA, 2+B CH and 8+B CH sites. The vertical bars are one standard error of the mean. Each bar represents an average for 3 pots, each pot containing 4 seedlings. 104 DISCUSSION CONIFER SEEDLING GROWTH IN CONTRASTING SITE CONDITIONS Figures 17 and 19 clearly illustrate that the three conifer species performed very differently among the three different site-age combinations, both with and without non-crop vegetation. In general, growth trends recorded in the pot experiments agreed with those in the field; growth of Sitka spruce was greatest on the 2+B HA and poorest on the 8+B CH, whereas growth of western redcedar did not show such a pattern. Munson and Timmer (1989) also found a close correspondence between pot and field results for black spruce (Picea mariana) during the first growing season. 2+B CH vs 8+B CH sites Large growth differences were found in all three conifer species between the 2+B CH and 8+B CH sites. These large differences relate well to the differences in forest floor nutrient status and below-ground non-crop vegetation biomass (Table 10) found between the two site-age combinations. These two factors probably explain most of the differences in conifer growth between these two CH site-ages (see Chapter 4). However, the high percent mortality of planted western hemlock and Sitka spruce seedlings observed on the vegetation-not-removed treatment on the 8+B CH sites during the third and fourth years after planting is more difficult to explain. Conifer seedlings are not thought to have suffered from a lack of light or water on any of the study sites. Allelopathic interference by salal may have been involved, although no evidence for such interference has yet been demonstrated (McDonald, 1989; L. Husted, pers. comm.; Chapter 4). 105 The mycorrhizal status of planted conifers was not assessed on the 8+B CH sites. Parke et al. (1984) reported reduced ectomycorrhizal colonization for Douglas-fir and ponderosa pine (Pinus ponderosa Laws.) seedlings grown in soil from 10- to 20-year-old clear-cuts compared to undisturbed forests. Moreover, Perry et al. (1987) mentioned that cutovers which are invaded by plants that do not form ectomycorrhizae may have low site inoculum potential of this type of mycorrhizal-forming fungus. Vegetation on the 8+B CH sites consisted mostly of western redcedar (which forms vesicular arbuscular mycorrhizae: VAM), salal (which forms ericoid and possibly ectomycorrhizae; Largent et al. 1980) and fireweed (which can form VAM). It is therefore possible that western hemlock and Sitka spruce seedlings planted on the 8+B CH sites suffered from an inadequate mycorrhizal colonization due to a lack of ectomycorrhizae inoculum on the site by the time the seedlings were planted. This hypothesis needs to be tested. However, low colonization frequency could also reflect extreme nutritional stress. Seedlings that are excessively stressed do not form abundant mycorrhizae (Hacskaylo, 1972; Kimmins, 1987). The better height growth measured for western redcedar compared to western hemlock and Sitka spruce on both CH site-age combinations for the vegetation-not-removed treatment agrees with the observations of Curran and Dunsworth (1988) for the same three species growing on similar nutrient-poor sites on the west side of Vancouver Island. Several possible mechanisms may explain the relative good growth of western redcedar on CH sites. Firstly, the slightly better height growth of western redcedar on the 2+B CH sites was achieved with 2 to 3 times less root and shoot biomass than western hemlock and Sitka spruce (Table 12). This suggests that either western redcedar seedlings needed to take up less nutrients to achieve a similar height growth than Sitka 106 spruce and western hemlock or that their root systems are more efficient. The former was suggested by Miller and Miller (1987) to explain the differences in early growth between Sitka spruce, Corsican pine, and Scots pine on nutrient-poor sites in Britain. They found that Corsican pine and Scots pine seedlings needed to produce less foliage to achieve a certain growth than Sitka spruce seedlings early in the life of a plantation which required less nutrient uptake. Secondly, the deeper rooting habit of western redcedar fine-roots observed in this study may allow this species to obtain nutrients not available to the more shallow rooting Sitka spruce and western hemlock seedlings. Thirdly, western redcedar is colonized by vesicular-arbuscular mycorrhizae that may provide this species with an advantage on these nutrient-poor sites which are characterized by very high accumulations of organic matter in the forest floorl. Finally, Ryan et al. (1986) reported western redcedar to be more tolerant of nutrient solutions of low pH which contain relatively high levels of Al ions than western hemlock and particularly Sitka spruce. This greater tolerance of western redcedar to acidic conditions might also confer a growth advantage on redcedar growing on the acidic CH cutovers. This latter possibility is questionable, however, since both western hemlock and Sitka spruce grew very well on HA cutovers which have pH values that are similar to those on CH cutovers (Table 10). The greater ability of western redcedar to achieve satisfactory height growth on these nutrient-poor salal-dominated CH sites in comparison to the spruce and hemlock is probably the result of a complex of these factors. Current research is addressing this possibility (Berch, S., pers. comm.) 107 Frieberg (1989), using stem analysis, studied the pattern of growth of 25- to 30-year-old western hemlock and western redcedar saplings growing in a naturally regenerated stand on a nearby CH site. He found the height and diameter growth of the 15 tallest western redcedar (38 cm and 0.7 mm per year, respectively) to be greater than those of western hemlock (28 cm and 0.3 mm per year, respectively). His findings suggest that the initial height growth advantage measured for western redcedar on the CH ecosystem in this study will continue for at least 25 years. Omule (1988) also reported an average annual height growth of 40 cm for 25-year-old western redcedar saplings growing on ecologically similar sites (CWHb-]^ )) on the western side of southern Vancouver Island. 2+B HA vs 2+B CH sites Both western hemlock and Sitka spruce grew substantially better on the 2+B HA sites than on the CH sites for both the vegetation-not-removed and the vegetation-removed planting treatments. The greater rate of cellulose decomposition and amount of resin-NH4+ found on the 2+B HA than on the 2+B CH sites suggest that the 2+B HA site had a better overall forest floor nutrient status than the 2+B CH site (Table 10). The slight difference in soil temperature measured between these 2 site-age combinations in 1987 and 1988 appears to be insufficient to affect conifer seedling growth significantly. There was no difference in total above-ground biomass of the non-crop vegetation between the two site-age combinations 4 years after clear-cutting and burning, but the composition of the non-crop vegetation differed markedly: the 2+B CH sites were completely dominated by salal during the first 4 years, whereas the 2+B HA sites were dominated by fireweed during the first two years 108 and fireweed and salal during the third and fourth year. The below-ground biomass of the non-crop vegetation may also have differed between the two site-age combinations, but because it was not measured on the 2+B HA sites, no comparison is possible. The removal of the non-crop vegetation had a greater effect on the height and diameter increments of Sitka spruce and western hemlock on the 2+B CH than on the 2+B HA sites. Because of this, the differences in conifer seedling growth between the two site-age combinations were lower for the vegetation-removed than vegetation-not-removed planting treatments. This smaller between-site differences for the vegetation-removed treatment may indicate that the availability of nutrients was more limiting on the 2+B CH sites than 2+B HA sites. Such an explanation is substantiated by the smaller growth improvement obtained by the application of fertilizer at the time of planting for both western hemlock and Sitka spruce onthe 2+B HA sites compared to the 2+B CH sites. Therefore, part of the differences in western hemlock and Sitka spruce growth between the 2+B CH and HA sites can be explained by differences in the overall forest floor nutrient status. The very different nature of the non-crop vegetation found on the 2+B HA and CH sites could also be of importance in explaining the large differences in early conifer growth measured between these two site-age combinations. Fireweed is an annual plant that produces a litter that decomposes rapidly (Taylor et al. 1991), whereas salal is a perennial plant that produces leaves that decompose relatively slowly (Feller et al. 1982; DeCatanzaro and Kimmins, 1985). Moreover, salal keeps its leaves for more than 6 years when growing under canopy (Bunnell, F., pers. comm.). The phenological and functional differences between these two non-crop species may contribute somewhat to the differences in forest floor nutrient availability measured between the two site-age combinations in the first few years following clear-cutting and burning. It was observed that a greater level 109 of micro and mesofaunal activity was associated with the higher decomposition of the cellulose discs on the 2+B HA sites. The yearly input of easily decomposable leaves of fireweed on the 2+B HA sites may supply an excellent source of carbohydrates and nutrients for the microfauna and flora. In contrast, very little input of fresh leaves occurred on the 2+B CH sites during the first four years, and very little sign of micro and mesofaunal activity was observed on the cellulose discs on this site-age combination. The differences in micro and mesofaunal activity observed between these 2 site-age combinations may be important in explaining the large differences in the rate of cellulose decomposition obtained in this study (Table 10), and consequently in forest floor nutrient availability. Flanagan and Van Cleve (1983) and Berg and Staaf (1987) discussed the possible beneficial effects of yearly inputs of litter that decomposes rapidly on decomposition and nutrient cycling in boreal forests. The inability of western redcedar to substantially increase its growth on the nutrient richer 2+B HA sites in the field experiment further suggests that on these acidic, ammonium-dominated sites, western redcedar seedlings do not respond very well to an increase in nutrient availability (see Chapter 6, p. 126-127, for further discussion on this subject). Omule (1988) also found very little difference in height growth between 25-year-old western redcedar growing in nutrient-poor sites (CWHb^g)) as compared to those growing on nutrient-medium sites (CWHb|(4p on the west coast of southern Vancouver Island. These 2 types of sites are ecologically similar to the CH and HA sites, respectively. Both of these experiments indicate that silvicultural treatments to increase N availability on such sites will tend to favour western hemlock and Sitka spruce over western redcedar. 110 The higher shoot-to-root ratios found on the 2+B HA compared to the 2+B CH sites in the field experiment support the idea that seedlings growing on nutrient-poor sites allocate proportionally more photosynthates to their roots. Moreover, conifer seedlings growing on nutrient-poor sites have been found to produce thinner fine-roots (Nambiar and Zed, 1980; McKay, 1988), which, for a similar amount of biomass, further increases their ability to take up scarce nutrients. This latter adaptation was not investigated in this study. The growth differences found between western redcedar and western hemlock growing on 2+B CH and HA sites may help explain the current composition and structure of the mature forest. In the old-growth CH ecosystem, western redcedar dominates, but western hemlock makes up a significant proportion of the stand, whereas in the mature HA ecosystem western hemlock and amabilis fir completely dominate, and very little western redcedar is found. The poorer growth of western redcedar found on the HA ecosystem compared to that of western hemlock may explain the absence of western redcedar in the HA ecosystem. Following the opening of the HA forest canopy through natural disturbance, such as windthrow or human-made disturbance such as clear-cutting, a dense stand of western hemlock with some amabilis fir establishes rapidly. Some western redcedar seedlings can be found in the first few years following such disturbance, but due to their slower initial growth (demonstrated in this study) and the very high density of western hemlock seedlings (as many as 200,000 trees per hectare), it is plausible that most of the redcedar seedlings are eventually shaded out by the dominating overstory of western hemlock and amabilis fir. I l l SEEDLING GROWTH AND COMPETING VEGETATION All three site-age combinations investigated had a substantial amount of non-crop vegetation surrounding the conifer seedlings. On both CH site ages, salal was the dominant non-crop vegetation, whereas on the HA sites both salal and fireweed were abundant. The removal of the non-crop vegetation on the two CH site ages resulted in a large increase in N (and possibly P) uptake by western hemlock and Sitka spruce seedlings, as evidenced by the increase in growth and by the dark green coloring of the foliage compared to the vegetation-not-removed treatment. Very similar results were obtained for Douglas-fir seedlings growing in salal-dominated ecosystems in southern Vancouver Island (Green, 1990), and for Sitka spruce seedlings growing in Calluna-dominated sites in Scotland (Dickson and Savill 1974). Weetman et al. (1989a,b) also reported an increase in growth following the complete removal of salal in 8- to 15- year-old western hemlock, Sitka spruce and western redcedar trees in CH ecosystems close to my study sites. Western hemlock and Sitka spruce responded noticeably more to the removal of competing vegetation than did western redcedar on both 2+B CH and 8+B CH site ages. This is in contrast to Weetman et al.'s (1989a,b) results obtained on similar CH sites nearby in which the improvement in height growth after 3 years following treatment was proportionally greater for western redcedar (60%) than for Sitka spruce (40%) or western hemlock (20%). The release of Sitka spruce and western hemlock occured mainly during the third year after the removal of salal, however, indicating that they may have been under greater stress than western redcedar. Squire (1977) and Neary et al. (1990) also found below-ground competition by non-crop vegetation (grasses in these two cases) to be a major factor restricting the early growth of conifer seedlings. 112 The better growth of conifer seedlings on CH sites where the non-crop vegetation was removed can probably be attributed to the increase in nutrient availability found in the forest floor in the vegetation-removed treatment (see Chapter 4, p. 64). Competition for light was not a factor in this study because all seedlings were growing in full sunlight conditions. Moreover, competition for water was non-existent due to the regular rainfall and fog which occurred throughout the growing season. The slight increase in soil temperature at 3 and 10 cm caused by the removal of the vegetation reported in Chapter 4 (Table 9) appears to be insufficient to have affected conifer seedling growth significantly. Competition for nutrients, although often mentioned as a possible factor affecting conifer growth, has rarely been carefully investigated. However, two recent studies (Eissenstat and Mitchell, 1983; Neary et al., 1990) have isolated competition for nutrients as the main factors affecting conifer growth. The removal of the competing vegetation did not have any marked effect on the mycorrhizal colonization of the three conifer species on the 2+B CH sites; they all had very high percent mycorrhizal colonization on both the vegetation-not-removed and removed treatments. No difference in mycorrhizal colonization was reported between weeded and unweeded 7-year-old black spruce (Picea mariana) saplings in a raspberry (Rubus spp)-dominated site in Quebec (Cote- and Thibault, 1988). Both height and diameter responded in the same way to the removal of the non-crop vegetation, but diameter was slightly more responsive than height to a reduction in interfering vegetation. This difference in responsiveness between height and diameter has been noted by others (e.g. Zutter et al., 1986). The conifer seedlings growing on the 2+B HA sites did not respond as much to the removal of the non-crop vegetation as on the 2+B CH sites. This lack of growth response indicates that either (1) little below-ground competition for 113 nutrients was occurring or (2) the non-crop vegetation did not deplete the nutrient resource to the level by which competition could seriously affect the growth of the conifer seedlings on the HA sites. The latter explanation is substantiated by the very small conifer growth response obtained by the application of fertilizer at planting on this sites. SLOW-RELEASE FERTILIZER The very substantial increase in growth following N-P-K fertilisation at the end of the first year after planting in the fertilizer treatment indicated that Sitka spruce, western hemlock, and to a lesser extent western redcedar were experiencing nutrient deficiencies on both CH and HA sites, but especially on the former. Weetman et al. (1989a,b) concluded that western hemlock, western redcedar and Sitka spruce saplings growing on similar CH sites nearby were suffering from a lack of both N and P. The good growth response obtained with western hemlock in this study and in the studies of Carlson (1981) and Arnott and Burdett (1988) differs from the often very erratic results obtained from conventional fertilisation (Radwan and Shumway, 1983). The lack of response to fertilisation in western hemlock has been attributed to some form of nitrogen "shock", and Weetman et al. (1989b) suggested that slow-release fertilizer should be used instead. Our results and those of Carlson (1981) and Arnott and Burdett (1988) agree with such a suggestion for the first few years after planting. The slow-release fertilizer treatment increased the growth of seedlings for the first two years after planting. However, growth was dramatically reduced in the third year on both 2+B CH and 8+B CH sites, especially for Sitka spruce and western hemlock. The growth reduction was still present at the end of the fourth growing season (based on visual observation only). This reduction in growth of 114 fertilized seedlings a few years after planting has been reported by other researchers (Arnott and Burdett, 1988; Brockley, 1988; Green, 1990). Four possible causes have been hypothesized to explain such post-fertilization growth reduction: (1) that conifer seedlings take up most of the extra nutrients during the first year, and that by the third year the effect has completely disappeared (Weetman, G.F., pers. comm.); (2) that non-crop vegetation is enhanced by fertilisation which increases competition near the seedling (Brockley, 1988; Tiarks and Haywood, 1986); (3) that conifer seedlings allocate proportionally more photosynthate to their shoots than their roots during the first and second years, so that by the third year they need to reallocate most of their photosynthate to their roots; and (4) that fertilization at the time of planting affects mycorrhizal colonization on the seedlings (Alexander and Fairley, 1983; Shaw et al. 1987). Arnott and Burdett (1988) found that fertilisation at time of planting had little effect on the shoot-root allometry of western hemlock seedlings. Coutts and Philipson (1976) and Philipson and Coutts (1977) reported that localized application of fertilizer stimulated root growth of Sitka spruce, but only within the area of application. In contrast, Carlson (1981) and Carlson and Preisig (1981) reported that the application of slow-release fertilizer in one particular spot near the seedling stimulated root growth in all zones surrounding the seedling, and not just the zone in which fertilizer was applied. No differences in the shoot:root ratio were found between fertilized and unfertilized western hemlock (Carlson, 1981) and Douglas-fir (Carlson and Preising, 1981) seedlings. Alexander and Fairley (1983) reported a smaller percent mycorrhizal colonization in Sitka spruce seedlings from fertilized (34%) than unfertilized (71%) seedlings. Based on the results obtained so far in this study and other studies (reviewed by Brockley 1988), it is doubtful that a small one-time application of 115 slow-release fertilizer at the time of planting could be used to significantly shorten the time required for the stand to attain canopy closure conditions and shade out competing vegetation on CH sites. In this study, only 5 kg of N/ha was used however, and it is possible that a greater amount of slow-release fertilizer with a longer release period would give better results. 116 SUMMARY The effects of non-crop vegetation and of slow-release fertilizer treatment applied at time of planting on western hemlock, western redcedar, and Sitka spruce growth were assessed on CH sites that were 2 and 8 years after clear-cutting and burning, and on HA sites that were 2 years after clear-cutting and burning on northern Vancouver Island. The main objective of this study was to compare the early growth of these three conifer species under different experimental conditions to determine which factors are associated with the very different conifer growth found between CH and HA cutovers. A secondary objective was to investigate the efficacy of slow-release fertilizer at the time of planting in improving early conifer growth. Both field and pot experiments were used to achieve these objectives. Western hemlock and Sitka spruce responded more than did western redcedar to all treatments that affected nutrient availability, and to site differences that were also related to site fertility. This response indicates that silvicultural treatments applied early following planting to increase N and P availability on these sites will favour western hemlock and Sitka spruce over western redcedar. The better conifer growth on the 2+B CH sites than on the 8+B CH sites , and on the 2+B HA sites than on the 2+B CH sites were associated with better forest floor nutrient status. Similarly, the increase in conifer growth due to the removal of the non-crop vegetation was associated with an increase in nutrient availability. Very little difference in soil moisture and temperature were measured between the three different site-age combinations and between the vegetation-not-removed and vegetation-removed treatments. All three conifer species had very high percent mycorrhizal colonization on the 2+B CH sites three years after planting, and except for two symbionts in Sitka spruce, this was not altered by the removal of the non-crop vegetation. Western redcedar achieved 117 a better early height growth for the vegetation-not-removed treatment than western hemlock and Sitka spruce on CH sites, and this was related to (1) its ability to achieve a certain height increment with less biomass production, and therefore with less total nutrient uptake, and (2) its deeper rooting habit. Slow-release fertilizer treatment applied at the time of planting increased above-ground conifer growth only for the first two years after planting. During the third and fourth years, a sharp decline in above-ground growth was observed for both Sitka spruce and western hemlock. 118 C H A P T E R SIX R E L A T I O N S H I P S B E T W E E N M I C R O S I T E F A C T O R S A N D W E S T E R N R E D C E D A R G R O W T H O N C H SITES I N T R O D U C T I O N The forest floor and surface soil horizons of a forest or a clear-cut are often highly variable, both in microelevation and micro-scale chemical properties (Beatty and Stone, 1986). This small-scale heterogeneity over a few square meters has often been ignored in site research, which has focussed on larger-scale (i.e. hectare and 100 m scales) gradients in the environment (Beatty, 1984). Micro-scale heterogeneity of the forest soil has been attributed to tree falls (Beatty and Stone, 1986), decaying wood (Quesnel and Lavkulich, 1981; Sidle and Shaw, 1983), variations in overstory canopy species composition (Turner and Franz, 1986; Beatty, 1984; Zinke, 1962; Boerner and Koslowsky, 1989), and distance from individual trees (Lodhi, 1977). Improved growth for individual trees has also been associated with more fertile microsites (Husted, 1982; Adams, 1974). On the northern part of Vancouver Island in coastal British Columbia, early height growth of western redcedar (Thuja plicata Donn) on clear-cuts originating from old-growth western redcedar and western hemlock (Tsuga heterophylla (Raf.) Sarg.) forests (CH), is generally poor, and highly variable (Le. 10 to 40 cm height growth per year). Although some of this variability may be explained by variation in stock quality, seedling genotype, and the handling and planting of seedlings, it is hypothesized that small-scale differences in site quality are also important. Genetic variability in western redcedar is reported to be small (Bainer and Dunsworth, 1988). Patchy distribution of fireweed (Epilobium angustifolium 119 L.) on these sites could also indicate differences in site quality at the microsite level. The objective of this study was to relate western redcedar seedling growth and fireweed distribution and height to microtopography, to some measures of forest floor nutrient status, and to salal (Gaultheria shallon Pursh) abundance on CH sites 4-year-old logged and burned CH sites. A better understanding of the microsite factors that influence the growth of western redcedar seedling may help to identify the main factor(s) that limit early conifer seedling growth on these sites. The following three hypotheses were addressed: (i) depressions have a better forest floor nutrient status than mounds; (ii) the growth of western redcedar seedlings is related to the nutrient status of the forest floor immediately surrounding its roots; (iii) the abundance and height of fireweed is related to the forest floor nutrient status at the microsite level. Hypothesis one is based on personal observations that western redcedar growth and fireweed abundance and height are greater in depressions than on mounds. Hypotheses two tests the idea that better growth of individual trees can be related to some measures of forest floor nutrient availability measured near individual trees (Husted, 1982; Adams, 1974). Finally, hypothesis three arises from the reports by Tamm (1956) and Ruwet (1980), among others, that the abundance and total height of fireweed are good indicators of the supply of nutrients in the soil, especially nitrogen. All hypotheses were tested in the field using microsite plots centered on 3-year-old western redcedar seedlings. 120 MATERIALS AND METHODS STUDY AREA The study was conducted in two different cutovers which had been clear-cut in 1985 and slashburned in 1986 within the CH ecosystem (same cutovers as the 2 years after clear-cutting and burning CH cutovers described in Chapter three). The harvesting technique (cable logging) minimized site disturbance, so that the micro topographic pattern was not altered significantly by the operation. The two cutovers were both well drained and situated within 1.5 km of each others. MICROSITE SELECTION Western redcedar seedlings, planted 2 years previously as part of the vegetation-not-removed treatment in Chapter 5, were randomly selected in the spring of 1989 on each of the two cutovers as the center of 20 microsite plots per cutover for a total of 40 microsite plots. The microsite plots were then classified into 3 microtopographic positions: mounds, flats, and depressions. Mounds and depressions were defined as being at least 0.5 m above and below the mean ground level, respectively. Plots at mean ground level were classified as flats. The vertical distance between mounds and depressions varied between 1.0 and 1.8 m. WESTERN REDCEDAR GROWTH The total height and root collar diameter of each seedling were measured at the beginning (April) and at the end (September) of the 1987, 1988, and 1989 growing seasons. 121 FOREST FLOOR FACTORS Four forest floor cores (7.4 cm in diameter) were collected to a depth of 15 cm from equally-spaced sampling spots within 20 cm of each seedling at the end of September 1989. These were bulked to form one forest floor sample per seedling. Forest floor pH was determined in water from fresh samples passed through a 2-mm sieve. A subsample was oven-dried at 70°C for 24 hrs to determine the moisture content. All the results are reported on an oven-dry basis. Forest floor pH, total N and P, and organic matter content were measured using standard soil analyses (see Chapter 4 for a detailed description of the methodology). In early May 1989, two ion-exchange resin bags of mixed cation (21 g of 68% moisture Amberlite IRC-50 CP. RCOO-H-) and anion (29 g of 65% moisture Amberlite IRC-45 CP. RNH3+OH-) exchange resin enclosed in two different stocking bags and two cellulose discs (4.25 cm diameter Whatman #1) enclosed in two different one-mm nylon mesh bags were buried for 2 months at 15 cm in the forest floor 20 cm on each side of each seedling to estimate the relative nutrient availability and decomposition rate, respectively, in the seedling's root environment. A description of the ion-exchange resin bag and cellulose disc methods is given in Chapter 4. In October 1989, the forest floor surrounding each seedling was excavated and the rooting substrata was described in terms of presence and absence of mineral soil and decaying wood. FIREWEED AND SALAL VEGETATION The percentage cover of fireweed and salal in 1-m^ plots centered on the seedling was estimated visually in July 1989. The total height was measured for each fireweed stem within each plot at the end of August 1989. 122 STATISTICAL ANALYSES One-way analysis of variance was used to compare redcedar seedling growth, fireweed abundance and vigour, forest floor variables, and salal abundance between the 3 different microtopographic positions. Tukey's HSD multiple comparison test was used to compare the treatments means. The data were checked for both homogeneity of variances and normality of distribution. No transformation were required. Multiple correlation analyses (using the Pearson correlation matrix) were used to estimate the degree of association between any pair of variables. A probability test for each correlation coefficient was also performed (Wilkinson, 1988). 1 2 3 R E S U L T S A N D D I S C U S S I O N RELATIONSHIPS B E T W E E N MICROTOPOGRAPHY AND FOREST FLOOR NUTRIENT STATUS No significant difference in forest floor pH, cellulose decomposition, total N and P, and resin NH4+, NO3" and Phosphate-P was found between the three microtopographic positions (Table 13). The percentage of organic matter and C:N ratio were significantly (P<0.05) lower in depressions than on flats and mounds. These data indicate that the microtopographic pattern found on my study sites does not induce any major differentiation in the forest floor nutrient status as revealed by the measurements we made. This finding does not support my first hypothesis. This microtopographic pattern appears, based on personal observations made in clearcuts and old-growth forests, to stem from the distribution and accumulation of large pieces of decaying logs on the forest floor; no evidence was observed on my study sites that the major agent responsible for the formation of depressions and mounds was windthrow. A qualitative comparison of the rooting substratum between the three different microtopographic positions indicated mounds to be more often associated with decaying wood than depressions. This is consistent with the observation of higher organic matter and C:N ratio in forest floor (to 15 cm) of mounds compared to depressions(Table 13). RELATIONSHIPS B E T W E E N WESTERN REDCEDAR GROWTH AND MICROSITE FACTORS Western redcedar height and diameter increments in 1989 (the third year after planting) were significantly (P<0.05) higher in depressions than on flats and mounds (Figure 22). There was an upward trend in both growth variables from 124 Table 13. Comparison of different forest floor properties between the three microtopographic positions: mounds, flats and depressions. Values in parentheses are one standard error of the mean. Numbers in rows followed by the same letter are not significantly (P>0.05) different between microtopographic positions. Microtopographic positions Forest floor Variables Mounds Flats Depressions M i n / M a x values Cellulose Dec. (%) 12.8 a (1.9) 14.9 a (2.8) 13.5 a (1.4) 1.2 /52.0 Resin N H 4 + (mg/g of resin) 0.156 a (0.016) 0.149 a (0.010) 0.155 a (0.016) 0.05 /0.25 Resin NO3" (mg/g of resin) 0.003 a (0.001) 0.002 a (0.000) 0.002 a (0.001) 0.000 / 0.006 Resin P (mg/g of resin) 0.140 a (0.023) 0.119 a (0.017) 0.151 a (0.037) 0.02 /0.34 p H 3.93 a (0.08) 4.43 a (0.27) 4.20 a (0.12) 3.61 /4.89 Organic matter (%) 93.2 b (1.8) 88.8 b (3.3) 78.2 a (5.2) 45.3 / 98.3 Total N (%) 0.81a (0.04) 0.86 a (0.06) 0.87 a (0.09) 0.45 /1.45 Total P (%) 0.041 a (0.003) 0.044 a (0.004) 0.057 a (0.013) 0.025 / 0.180 C : N ratio 69.1b (4.3) 64.2 b (4.7) 54.7 a (3.3) 33.0 / 95.6 # of microsite 8 20 12 plots 125 Figure 22. Comparison of height and diameter increments of western redcedar seedlings in 1989 (three years after planting) between mounds, flats and depressions. Vertical bars represent one standard error of the mean. 126 mounds to depressions, the difference being about 2.5 times. Differences in growth increments between the three microtopographic positions became apparent in 1988, 2 years after planting. However, these differences in western redcedar growth are not related to any of my measures of forest floor nutrient status (Table 14). This finding does not support my second hypothesis. Other edaphic or microclimatic factors not investigated in this study are probably involved in explaining the marked differences in the growth of western redcedar seedlings between microtopographic positions reported in Figure 22. Moisture was not measured for this particular study, but, in Chapter 4, it was not found to be a factor limiting conifer growth on these same sites. We observed, however, that western redcedar seedlings growing in fully exposed locations often turn brown for a few years following planting, whereas those partially protected by vegetation and microtopography maintain a green appearance. This browning effect likely affects photosynthesis and ultimately the growth of western redcedar seedlings. We are unaware of any research on this subject, however. Studies examining relationships between tree growth and soil properties immediately surrounding individual trees report contrasting results. Husted (1982) compared the soil properties surrounding well- and poorly-growing Pacific silver fir (Abies amabilis (Dougl.) Forbes) saplings in recently clearcut sites on eastern Vancouver Island, Canada. Forest floor associated with trees showing good growth had significantly higher total N and exchangeable Mg, and a lower C:N ratio compared to forest floor associated with the poorly-growing trees. Similarly, Adams (1974) found a higher soil total N under faster growing 7- to 40-year-old Sitka spruce (Picea sitchensis (Bong.) Carr.) trees compared to slower 127 Table 14. Pearson correlation matrix for western redcedar height and diameter increment, salal abundance, fireweed abundance and vigour, and forest floor variables. 1 2 3 4 5 6 7 8 9 10 11 1. Ht. inc. of cedar 1.00 2. Di. inc. of cedar 0.71c 1.00 3. Salal cover -0.27 -0.31a 1.00 4. Fireweed cover 0.61c 0.41a-0.37a 1.00 5. Fireweed height 0.42b0.19 -0.16 0.57c 1.00 6. Cell. Decomp. 0.24 0.21 -0.39a0.21 0.33a 1.00 7. ResinNH4+ -0.16 -0.17 -0.03 0.01 0.20 -0.27 1.00 8. Org. matter -0.42a-0.15 0.32 -0.19 -0.36 -0.49a 0.06 1.00 9. Total N 0.18 0.28 0.58b 0.04 -0.12 -0.24 -0.29 0.37a 1.00 10. Total P 0.19 0.03 0.22 -0.03 -0.13 -0.19 -0.28 0.16 0.61b 1.00 11. C:N ratio -0.42a-0.40 0.38 -0.18 -0.09 0.06 0.48a 0.27 -0.70b 0.17 1.00 Note: Significance levels: aP=0.05; b P=0.01; c P=0.001 128 growing trees. The differences in soil properties between the faster- and slower-growing trees in these two studies were fairly small, however, and could have been induced by the trees themselves. Shaw et al. (1987) found no difference between the growth of Sitka spruce seedlings planted in rotten wood and those planted in undisturbed forest floor materials in southeast Alaska, although large differences in soil nutrient status were found between the two types of microsites (Sidle and Shaw, 1983). It appears very difficult, based on this study and those mentioned above, to relate the growth of conifers to the forest floor properties immediately surrounding individual trees. Conifers are capable of producing very extensive root and mycorrhizal systems, and may not be completely dependent on the soil properties immediately surrounding them. In my study, most of the western redcedar roots were located within 50 cm of the stem three years after planting, although associated endomycorrhizal hyphae may have extended much further. The interpretation of my results in terms of the nutritional requirements of western redcedar is difficult since there is very little information available on western redcedar nutrition (Weetman et al. 1988). Western redcedar is found on a wide variety of sites in the Pacific Northwest, but grows best on wet, nutrient-rich sites (Krajina et al. 1982). Very good growth of western redcedar has been reported on alluvial sites in Coastal British Columbia (Krajina, 1969; McLennan, D., Pers. Comm.). Krajina (1969) and Krajina et al. (1982) characterized western redcedar as more nutrient-demanding than western hemlock and Sitka spruce. Krajina et al. (1973) found western redcedar to grow better with nitrogen supplied as nitrate than as ammonium. Western redcedar has been found to respond to NPK fertilization in coastal British Columbia (Weetman et al. 1989b). However, 1-to 3-year-old western redcedar seedlings were found to be unresponsive to any 129 treatment that either increase or decrease many measures of forest floor nutrient availability (Chapter 5). Reports concerning the nutritional requirements of western redcedar are contradictory. On one hand, it is considered to be a nutrient demanding species, requiring fertile soil with a well balanced supply of nitrogen, calcium and magnesium (Krajina, 1969). On the other hand, on some nutrient-poor sites its growth may exceed that of other supposedly less nutrient demanding species (Gregory, 1957; Curran and Dunsworth, 1988; Chapter five). One possible explanation is that western redcedar may demonstrate rapid growth rate and high nutrient demand only on sites where nitrate is available (on alluvial sites, for example). On nutrient-poor acidic soils where nitrogen is available only in the ammonium form, its growth and therefore nutrient demand may be decreased. On my study sites, virtually no nitrate was detected (Table 13), and western redcedar was able to grow relatively well with only a small amount of available ammonium. The high level of endomycorrhizal infection (greater than 90%) found on the roots of these western redcedar seedlings (Chapter five) could also explain its ability to grow relatively well on these nutrient poor sites. Further research is required to ascertain the nutritional requirement of western redcedar on edaphically different sites. RELATIONSHIPS BETWEEN FIREWEED AND FOREST FLOOR NUTRIENT STATUS Fireweed was significantly (P<0.05) more abundant and taller in depressions than on flats and mounds (Figure 23). The differences in percentage cover were substantial, going from an average of 5% on mounds to 20% in depressions. These differences in fireweed abundance and height cannot be 130 Percentage (%) Average height (cm) 120 100 80 60 40 20 percentage cover Y//A Average height Mounds Flats Depressions Microtopographic position Figure 23. Percentage cover and average height of fireweed between mounds, flats and depressions. Vertical bars represent one standard error of the mean. 131 explained in terms of the forest floor variables that were measured (Table 14); the correlation between fireweed average height and ammonium availability (resin NH4+) was particularly weak (r=0.2). These findings do not support my third hypothesis. The small differences in percentage organic matter and C:N ratio appear insufficient to explain the marked differences in fireweed abundance and height between depressions and mounds. Snaydon (1962), studying clover (Trifolium repens L.) micro-distribution in a upland pasture in north Wales, also found a greater abundance of clover in small depressions, but was unable to relate it to any specific edaphic factor. These results are surprising since fertilization on similar sites nearby has been found to induce vigorous growth of fireweed (personal observation). These results are consistent, however, with those of van Andel (1976), van Andel et al. (1978), and van Andel and Nelissen (1979) who stated that fireweed can tolerate and grow on a wide variety of soil conditions, and should not be considered a good indicator of nitrogen availability. According to van Andel (1976), its increase in abundance following fertilization or burning could be attributed to its ability to tolerate the newly created conditions rather than indicating a high nutrient requirement. Fireweed produces long rhizomes that accumulate large amount of carbohydrates over the years, and its growth for any particular year may not be dependent on the current forest floor nutrient status immediately surrounding its shoots. Myerscough (1980), reviewing the literature on fireweed, stated that the successful establishment of this species on recently disturbed sites appears to be limited to open moist sites of at least moderate fertility, and little initial vegetative competition. Myerscough and Whitehead (1967) found the establishment of fireweed to be restricted by the nutritional status of their experimental sites. Only a low and patchy establishment of fireweed occurred on my study sites four years after clear-cutting and burning (approximately 12% 132 cover), and this may be due to the low forest floor nutrient availability found on these CH sites compared to HA sites (Chapter 5). Kuiters et al. (1987), working in the laboratory, reported fireweed to be very sensitive to high levels of phenolic acids. It appears possible that the failure of fireweed to completely occupy my study sites (in spite of an abundant source of seeds from adjacent nutrient richer sites) could also be attributed to the presence of high concentrations of phenolic acid in the forest floor, especially on the mounds that consist largely of decaying western redcedar wood. Salal cover varied from 2.5% to 85% within the forty 1-m2 microsite plots. Only a very weak negative correlation was obtained between western redcedar growth and salal cover (Table 14). This finding appears to further indicate that western redcedar is affected little by the presence of salal (Chapter 5). This may be due to the low nutrient requirement of western redcedar on sites where only ammonium is available and/or to the high levels of endomycorrhizal infection found on its roots. However, my findings in Chapter 4 (p.40) suggests that there is no good relationship between the amount of above-ground and below-ground biomass of salal for a particular microsite. Therefore, salal above-ground cover may not be a good indicator of the potential severity of below-ground salal competition on a particular microsite. SUMMARY The growth of western redcedar (Thuja plicata) seedlings were studied in relation to microtopography, to forest floor nutrient status, and to non-crop vegetation on 4-year-old logged and burned CH sites dominated by salal (Gaultheria shallon) on northern Vancouver Island, British Columbia. These relationships were sought to determine some possible factors at the microsite level that influence the growth of western redcedar on recently clear-cut sites, and thereby to contribute to the understanding of the factors that control growth on CH sites in general. No significant difference in my measures of forest floor nutrient status was found between the three microtopographic positions investigated in this study. However, both western redcedar growth and fireweed abundance and height were greater in depressions than on mounds. This suggests that small depressions on these sites constitute good planting microsites for western redcedar. It is speculated that the better western redcedar growth in depressions may be due to protection from direct sunlight. No correlation were found between western redcedar height and diameter increments and my measures of forest floor nutrient status. This lack of correlation may have been due to the selection of western redcedar as the bio-indicator species. In effect, in chapter 5, western redcedar was found to be little affected by any treatments that increase or decrease many measures of forest floor nutrient availability. Sitka spruce or western hemlock might have been better bio-indicators since they have been shown to be more sensitive to changes in forest floor nutrient availability (Chapter 5). 134 Finally, the lack of correlation between fireweed abundance and height and my measures of forest floor nutrient status suggests that fireweed growth is not directly influenced by the availability of nutrients measured four years after logging and burning on CH sites. 135 CHAPTER SEVEN GENERAL DISCUSSION AND CONCLUDING REMARKS Many of the factors limiting early conifer growth in the salal-dominated CH ecosystem, and many of the differences in productivity between CH and HA forest ecosystems following clear-cutting and burning, are the results of the conditions prevailing in the forest prior to harvesting. It is, therefore, important to examine briefly the findings reported in this thesis in relation to what is known of the main ecological differences found between these two forest ecosystems prior to disturbance. CONDITIONS PRIOR TO DISTURBANCE The old-growth CH forest ecosystem is characterized by very large and old western redcedar (many are more than 600-year-old), some smaller, poorly growing western hemlock and Pacific silver fir, a dense understory of salal, and a thick accumulation of organic matter on the forest floor, of which decaying western redcedar forms a substantial proportion. Although individual trees may be windthrown, no major catastrophic disturbance is believed to occur in this forest ecosystem (Lewis, 1982). Little is known about the rate of decay of large western redcedar logs on the study sites, but it is believed to be very slow. The J3-thujaplicin that is found in western redcedar wood is known to inhibit the growth of a wide variety of bacterial species (Trust and Coombs, 1973), and this may contribute to its slow decomposition. As high C/N ratio woody material decomposes, nitrogen may be taken up by the microbial community that is decomposing the woody material, immobilizing some portion of the available nitrogen on the site (Grier, 1978; Covington, 1981; Kimmins, 1987). 136 The open canopy of the old-growth CH forest ecosystem allows a substantial amount of light to penetrate the understory, which promotes the development of a dense understory which consists mainly of salal. Light intensity in gaps averages 25% of full sunlight, but it may reach up to 80% under large gaps (Messier and Kimmins, 1991). Little difference in salal biomass production at light intensities between 30% and 100% full sunlight was found in a pot experiment, in contrast to the large decrease in salal biomass production, especially in fine-roots, that was observed at 5% and 10% of full sunlight (Messier and Kimmins, 1991). It appears, therefore, that the light conditions in medium to large old-growth CH forest gaps (20 to 80% full sunlight) are adequate for salal growth. The ability of salal to dominate the understory of the CH ecosystem may be related to its ericoid-mycorrhizae (Largent et al. 1980), which have been shown to be capable of efficiently utilizing the soil matrix, absorbing organic and inorganic N and P compounds, and resisting soil toxicity (Read, 1984). The dense cover of salal may contribute to the thick accumulation of organic matter on the forest floor by producing a litter that decomposes slowly and incompletely (DeCatanzaro and Kimmins, 1985; Feller et al. 1982) and/or by releasing compounds (e.g. tannins) that reduce the rate of decomposition of the forest floor (de Montigny and Weetman, 1990). Materials rich in lignin and phenolic compounds, such as ericaceous plants, increase the potential for humification (Tate, 1987). The decline in productivity of some types of forest with time after a very prolonged period without catastrophic disturbance is a well documented phenomenon in northern forests (VanCleve and Viereck, 1981). The factors associated with this decline are mainly related to unfavorable conditions for decay of organic matter and the resultant slow rates of nutrient cycling. As time passes, organic matter that is resistant to decomposition tends to accumulate on the forest 137 floor. This leads to a steady increase in immobilization of nutrients in organic matter and an associated decrease in nutrient availability, and ultimately in tree growth. By comparison, the windthrown HA forest ecosystem is characterized by a smaller accumulation of organic matter on the forest floor, of which decaying wood from western hemlock and amabilis fir forms a substantial proportion. The decomposition of the organic matter in this ecosystem, including the decaying wood, appears to be more rapid and/or complete than on the CH forest ecosystem. Very little herbaceous and shrubby understory vegetation is present. The low light intensity measured at the forest floor (2-5% of full sunlight) probably precludes the growth of most plants (Messier and Kimmins, 1991). Consequently, most of the forest floor originates from above- and below-ground litter of western hemlock and amabilis fir. Large amounts of decaying wood from amabilis fir and western hemlock are present, but they appear to decompose within 100 years. The better forest floor nutrient status (Germain, 1985; Prescott, C , pers. comm.), the apparently more rapid decomposition and nutrient cycling, and smaller amount of shrubby understory vegetation in this ecosystem compared to the CH ecosystem appears to create conditions which favour rapid tree growth. REGROWTH OF SALAL AND FIREWEED FOLLOWING CLEAR-CUTTING AND BURNING Following the clear-cutting and burning of the CH forest ecosystem, salal reestablishes itself quickly, mainly by resprouting from rhizomes already present prior to harvesting. Immediately following the disturbance, salal expands its shoots and roots to fully exploit the above- and below-ground environment it 138 already occupies. It then produces new rhizomes to exploit the newly available soil resources created by the disturbance. The result is a large biomass of rhizomes developed by eight years after clear-cutting and burning. In the longer term, this large production of rhizome in the clear-cut (when the resources are plentiful) may allow it to survive for many years even under very dense tree stands. Sabhasri (1961) estimated the below-ground biomass of salal under a 120-year-old Douglas-fir stand at 25,000 kg ha'l. Although not explicitly mentioned in the study, it appears that most of the below-ground biomass was rhizomes. This amount of below-ground biomass is 2.5 times higher than the amount found at 8 years after clear-cutting and burning in this thesis. Few other non-crop species invade clear-cut and burned CH sites. Fireweed invades quickly but exhibits little increase in biomass after two years, and declines slowly in biomass after 4 years. In contrast, bunchberry colonizes the sites more slowly but continues to increase over a longer period. Neither species occupies a very large portion of either the above- or the below-ground environment, however. Following clear-cutting and burning, the HA forest ecosystem is completely covered by fireweed within 2 years. Salal invades HA cutovers by resprouting from the few rhizomes that are already present prior to harvesting and by establishing from seed transported by birds and mammals. In the two intensively-studied HA cutovers in this thesis, fireweed above-ground biomass was 1910 kg ha'l at 2 years, and did not change from 2 to 4 years after clear-cutting and burning, whereas salal increased its above-ground biomass from 300 kg ha"^  to 2000 kg ha'l during that period. An examination of older HA cutovers that were not burned showed that the dense and vigorous cover of western hemlock that 139 characterizes such sites was very successful in shading out both salal and fireweed. Only very tall salal plants (>3 m) were able to survive by etiolation. POSSIBLE FACTORS LIMITING EARLY CONIFER GROWTH ON CH CUTOVERS Figure 24 illustrates the relationships between the experimental treatments and biotic and abiotic factors that were investigated in this thesis. The results of these investigations were used as the basis for concluding about factors that are limiting early conifer growth on CH cutovers on Northern Vancouver Island. 1. Competition by salal It is estimated in this thesis that the regrowth of the above- and below-ground non-crop vegetation immobilizes annually about 9 kg of nitrogen per hectare for at least the first 8 years following clear-cutting and burning on the CH sites (Figure 13). Weetman et al. (1990) estimated (based on mineralizable N using anaerobic incubation at 40°C for 7 days) that, on average, 20 to 30 kg of nitrogen per hectare are available annually for plant uptake on these nutrient-poor clear-cut and burned CH sites. Based on Weetman et al.'s estimation, the regrowth of the non-crop vegetation can potentially immobilize annually between 30 and 45% of the available N on these CH sites in accumulating live biomass, and additional nitrogen is temporarily tied up in slowly decomposing leaf litter. The 140 'Time after logging"" on CH cutovers Site conditions (CH vs HA cutovers) Removal of non-crop vegetation. Soil moisture Microtopography Non-crop vegetation Available soil nutrients Decomposition G r o w t h o f S s , H w a n d C w Soil temperature Microbial activity Fertilization at planting Mycorrhizal status Factors investigated Experimental treatments Significant relationships Figure 24. Relationships between the experimental treatments and biotic and abiotic factors investigated in this thesis. The arrows link the relationships that were found to be significant in this study. Factors or processes in Boxes not linked by arrows were not found in this thesis to contribute significantly to our explanation of the differences in early conifer growth. 141 removal of the non-crop vegetation around individual conifer seedlings on CH cutovers increased the availability of NH4+ by 36% and phosphate-P by 25%, did not affect cellulose decomposition and soil moisture (i.e. water tension), and slightly increased soil temperature. The immobilization of nitrogen and of other nutrients in expanding salal biomass and other non-crop species could provide an explanation for much of the large differences in conifer growth found between the vegetation-not-removed and vegetation-removed planting treatments (Figures 11, 12, 17 and 19) measured in CH cutovers. 2. Inhibition of mycorrhizal infection by salal Weetman et al. (1989a; 1990) suggested that the release of chemicals by salal into the root environment of the conifer seedlings may inhibit the normal mycorrhizal infection of tree roots, and consequently the uptake of nutrient ions. No evidence of a change in total degree of mycorrhizal colonization was found in the field or pot seedling experiments that were conducted with or without the presence of salal. The observed changes in relative frequency of two species of mycorrhizal fungi (Telophora terrestris and Cenococcum geophilum) associated with Sitka spruce's fine-roots between the vegetation-not-removed and vegetation-removed treatments may be interesting, but was not investigated further in this thesis. While the functioning of the mycorrhizae was not investigated in this thesis, a greenhouse biossay in another study failed to find any evidence that the presence of salal reduced the ability of Sitka spruce, western hemlock or western redcedar to take up nutrient ions (McDonald, 1989). Allelopathy is very difficult to prove or disprove, however, and other experiments (L. de Montigny and Ghoping) are in progress to further investigate this possibility. 142 3. Short-term flush of nutrients on CH sites Conifer seedlings planted on CH sites 8 years after clear-cutting and burning grew less than seedlings planted on 2-year-old CH sites with and without the presence of competing vegetation (Figures 11, 12, 17 and 19). Growth of Sitka spruce in pots was also lower in forest floor taken from 9 years after clear-cutting and burning CH sites than 1- and 3-year-old CH sites. Forest floor nutrient availability declined from 3 to 9 years after clear-cutting and burning as indicated by the forest floor factors measured in this study, whereas no difference in soil moisture and only small differences in soil temperature were found (Table 8). Germain (1985), working on similar CH sites nearby, also reported a decrease in nutrient availability 3 to 4 years following clear-cutting and burning. Decline in nutrient availability a few years following the removal of a forest cover has been reported in several studies (Covington, 1981; Binkley, 1984; Martin, 1985; Krause and Ramlal, 1986; David, 1987). This decline in forest floor nutrient availability following clear-cutting and burning was shown to be associated with a growth reduction of these coniferous species in addition to that caused by salal competition (Figure 12). This growth reduction occuring on the 8+B CH sites may also be the results of poor ectomycorrhizal colonization on the roots of western hemlock and Sitka spruce due to a lack of ectomycorrhizae inoculum present on these sites by the time the seedlings were planted. 4. HA vs CH site condition effects Field and pot bioassays showed that western hemlock and Sitka spruce, and to a lesser extent western redcedar, grew better on clear-cut and burned HA than CH sites both with and without competing vegetation. Large differences in the 143 forest floor nutrient status were measured between these two ecosystems, whereas no difference in soil moisture and above-ground biomass of non-crop vegetation, and only very small differences in soil temperature were found (Table 10). Consequently, the differences in conifer seedling growth between these two forest ecosystems following clear-cutting and burning can be explained partly by intrinsic differences in forest floor nutrient status. These differences are probably due to forest floor nutrient status conditions prevailing prior to clear-cutting and burning (Germain, 1985; Prescott, C , pers. comm.). However, the functional differences in the non-crop species invading these 2 forest ecosystems following clear-cutting and burning may also contribute to some of the differences in forest floor nutrient availability found between CH and HA cutovers. 5. Microtopography on CH cutovers Western redcedar and fireweed were used as bio-indicators to determine which factors at the microsite level are associated with poor and good seedling growth on CH cutovers. Small microtopographic depressions were found to be associated with a better growth of western redcedar and a greater vigour and abundance of fireweed, but the results could not be explained in terms of the edaphic factors that were measured. The presence of decaying wood did not appear to be associated with poor western redcedar growth. However, the lack of responsiveness of western redcedar to any of the factors that increase or decrease forest floor nutrient availability (Chapter 5). strongly suggests that Sitka spruce and western hemlock should have been used for this particular sub-study. 144 6. Conifer species Western hemlock and Sitka spruce seedlings responded noticeably more to the removal of competing vegetation of mainly salal than did western redcedar on both 2+B CH and 8+B CH sites. Moreover, western hemlock and Sitka spruce responded more than did western redcedar to all treatments that affected forest floor nutrient availability, and to site differences (i.e. CH vs HA sites) that were related to site fertility. The removal of the competing vegetation of mainly salal did not have any marked effect on the total percent mycorrhizal colonization of the three conifer species growing on the 2+B CH sites; they all had very high percent mycorrhizal colonization (i.e. greater than 90%) on both the vegetation-not-removed and vegetation-removed planting treatments. Western hemlock and Sitka spruce seedlings produced long lateral roots (up to 2.5 m long compared with mean seedling total height of 0.80 m) in the top 5 cm of the forest floor of the 2+B CH sites, whereas western redcedar produced shorter roots (up to 0.8 m long compared with mean seedling total height of 0.85 m), but these were more evenly distributed in the forest floor than the roots of the hemlock and spruce. The deeper rooting habit of western redcedar fine-roots observed in this study may allow this species to obtain nutrients unavailable to the more shallow rooting Sitka spruce and western hemlock seedlings. Western redcedar had a significantly lower shoot and root dry weight and a higher shoot/root ratio than western hemlock and Sitka spruce on both 2+B CH and HA sites. Therefore, the slightly better height growth of western redcedar seedlings for the vegetation-not-removed treatment found on the 2+B CH sites was achieved with 2 to 3 times, less root and shoot biomass than western hemlock and Sitka spruce, which suggests that western redcedar seedlings needed to take up less nutrients to achieve a similar height growth than spruce and hemlock. 145 All of these results indicate that the nutritional stress and poor growth reported in young conifer plantations (especially in Sitka spruce (Germain, 1985; Weetman et al. 1989a,b)) growing on clear-cut and burned CH sites on northern Vancouver Island are the consequences of the combined effects of (1) inherently low forest floor fertility in cutovers originating from old-growth CH forests; (2) salal competition for scarce nutrients and their subsequent immobilization in salal biomass, and (3) reduced forest floor nutrient availability caused by the termination of the flush of available nutrients that occurs in the immediate post-logging and burning period. The low availability of nutrients in the forest floor 5 to 8 years following clear-cutting and burning and the immobilization of large amount of nutrients in non-crop vegetation biomass occuring on CH cutovers does not appear to leave enough nutrients to support rapid tree growth. This is especially true for a nutrient-demanding species such as Sitka spruce (Miller and Miller 1987). These findings lead to the more fundamental question of what are the processes controlling nutrient availability and how these might be manipulated through silviculture treatments to increase conifer seedling performance in the CH forest ecosystem. This thesis results connot rule out an involvement of allelochemicals and a mycorrhizal effect, but demonstrate that the observed growth stress can be accounted for without appeal to these mechanisms. LONG-TERM EFFECTS OF SALAL The long-term effect of salal on conifer growth in the CH ecosystem is not known. However, based on the data obtained so far, it appears possible that with the advent of canopy closure the negative influence of salal will be considerably reduced, if not eliminated. As conifers grow and the young stand develops on clear-cut CH sites, the overstory tree canopy slowly closes and reduces the 146 intensity of light reaching the understory salal. This reduction in light intensity causes changes in the structure, above-ground biomass (Vales, 1986), leaf morphology and biomass distribution of salal (Messier et al. 1989), and in carbon allocation (Messier and Kimmins, 1991). The projected development of the live fine-root, leaf, stem and rhizome biomass of salal over time following the removal of the tree canopy is shown in Figure 10. This figure suggests that the increase in biomass and resulting nutrient immobilization by salal ceases at around 15 years after clear-cutting and burning, and then declines following the gradual closing of the conifer canopy. Turner (1975; in Cole and Rapp, 1981) also showed that as the crown began to close at age 22 year, nutrient immobilization in the understory vegetation declined rapidly. An almost complete elimination of salal has been reported under a very dense (i.e. 20,000 trees/ha) 36-year-old stand of western redcedar and western hemlock that regenerated naturally following the clear-cutting of a CH forest (Messier et al. 1989). The rotation-length impact of early reduction (due to salal) or improvement (due to silvicultural treatments) in tree growth is very difficult to predict unless the observed early tree growth responses are sustained over significant time periods. Short term gains Or losses in tree growth will not necessarily lead to a major difference in harvestable volume at the end of the rotation period. Long-term studies and rotation-length, ecosystem-level management simulation models that can account for the effects of the rotation-length dynamics of nutrients, light competition, and stand dynamics are needed for such an assessment (e.g. FORCYCTE-11: Kimmins, 1988). 147 A D D I T I O N A L R E S E A R C H N E E D S This research has revealed the need for many additional studies, in addition to the studies currently being undertaken, to improve our understanding of the factors limiting early conifer growth following harvesting in salal-dominated ecosystems on northern Vancouver Island: 1. Factors influencing decomposition and nutrient cycling on both CH and HA ecosystems. 2. Salal growth, carbon allocation, biomass, nutrient uptake and net nutrient release under tree canopies of increasing density. 3. Optimum nutritional requirements of western redcedar on these acidic soils. 4. Effects of slow-release fertilizer at time of planting on conifer seedling growth, carbon allocation, and mycorrhizal status. 5. Long-term effects of mechanical site preparation on the decomposition of the organic matter and soil nutrient status in the CH ecosystem. 6. A synthesis of the many process studies that are being conducted to evaluate the long-term behavior of the CH and HA ecosystems. LIMITATIONS O F THIS R E S E A R C H The studies presented in this thesis have several shortcomings that need to be considered in order to fully appreciate the results presented: 148 1. Only one particular type of CH site was investigated in order to reduce between-site variability. This somewhat limits the range of CH sites to which the results can be applied. 2. The time period of most studies reported in this thesis (3 years) may limit the interpretation of some of the results, because many of the factors investigated may require longer periods of time to fully understand their dynamics. A good local example of the potential shortcomings of short term studies is the phenomenon that was the genesis of this thesis: planted Sitka spruce grew well for the first 4 to 6 years on the CH sites and then went into severe growth check. 3. The use of the chronosequence approach (also called space-for-time substitution) to circumvent some of the limitations posed by the long time scales of many phenomena in forest ecosystems requires the acceptance of some basic assumptions (Cole and Miegroet, 1989; Pickett, 1990) that somewhat weaken the confidence in the interpretation of my results. 4. The use of only two replicates per cutover per type of site reduces the confidence that the selected cutovers are representative of the population of cutovers in the region. Three or more cutovers per site type would have provided a better measure of the variability of the factors investigated within each site type. However, there will always be a trade off between the number of replicates and the intensity of study that can be undertaken. Two replicates represented the maximum that could be investigated with the available resources. 149 L I T E R A T U R E C I T E D ADAMS, S.N. 1974. Some relations between forest litter and growth of Sitka spruce on poorly drained soils. J. Appl. Ecol., 11: 761-765. AERTS, R., BERENDSE, F., KLERK, N.M. and BAKKER, C. 1989. Root production and root turnover in two dominant species of wet heathlands. Oecologia, 81: 374-378. ALEXANDER, I.J. and FAIRLEY, R.I. 1983. Effects of N fertilisation on populations of fine roots and mycorrhizas in spruce humus. Plant and Soil, 71: 49-53. van ANDEL. J. 1976. Growth and mineral nutrition of Chamaenerion angustifolium (L.) Scop. (=Epilobium angustifolium L.) on culture solutions. Oecol. Plant., 1: 25-40. van ANDEL, J., BOS, W. and ERNST, W. 1978. An experimental study on two populations of Chamaenerion angustifolium (L.) Scop. (=Epilobium  angustifolium L.) occurring on contrasting soils, with particular reference to the response to bicarbonate. New Phytol., 81: 763-772. van ANDEL, J. and NELISSEN, H. J. M. 1979. Nutritional status of soil and plant species in several clearings in coniferous woods compared to that in two related habitats. Vegetatio, 39: 115-121. ARMSON, K.A. 1979. Forest soils: properties and processes. University of Toronto Press, Toronto. 390 pp. ARNOTT, J.T. and BURDETT, A.N. 1988. Early growth of planted western hemlock in relation to stock type and controlled-release fertilizer application. Can. J. For. Res., 18: 710-717. AUCLAIR, A.N.D. 1985. Postfire regeneration of plant and soil organic pools in a Picea mariana-Cladonia stellaris ecosystem. Can. J. For. Res., 15: 279-291. BAINER, R.C. and DUNSWORTH, B.G. 1988. Provenance test of western redcedar on Vancouver Island. In: Western red cedar - does it have a future? N.J. Smith (Ed.). Conference Proceedings, University of British Columbia, Faculty of Forestry. BARKER, J., WEETMAN, G.F. and FOURNIER, R.M. 1987. Growth and nutrition of Sitka spruce, western hemlock, and western redcedar following fertilization of coastal cedar/hemlock sites. Internal report, Western Forest Products Ltd, Vancouver. BEATTY, S.W. 1984. Influence of microtopography and canopy species on spatial patterns of forest understory plants. Ecology, 65: 1406-1419. BEATTY, S.W. and STONE, E.L. 1986. The variety of soil microsites created by tree falls. Can. J. For. Res., 16: 539-548. 150 BERENDSE, F. and ELBERSE, W.Th. 1990. Competition and nutrient availability in heathland and grassland ecosystems (Chapter 7). In: J.B. Grace and D. Tilman. Perspectives on plant competition. Academic Press, Inc., San Diego. BERG, B. and STAAF, H. 1987. Release of nutrients from decomposing white birch leaves and Scots pine needle Utter. Pedobiologia, 30: 55-63. BERISH, C.W. 1982. Root biomass and surface area in three successional tropical forests. Can. J. For. Res., 12: 699-704. BINKLEY, D. 1984 Does forest removal increase rates of decomposition and nitrogen release? For. Ecol. Manage., 8: 229-233. BINKLEY, D., ABER, J., PASTOR, J. and NADELHOFFER, K. 1986. Nitrogen availability in some Wisconsin forests: comparisons of resin bags and on-site incubations. Biol. Fertil. Soils, 2: 77-82. BINKLEY, D. and MATSON, P. 1983. Ion exchange resin bag method for assessing forest soil nitrogen availability. Soil Sci. Soc. Am. J., 47: 1050-1052. BOERNER, R.E.J. and KOSLOWSKY, S.D. 1989. Microsite variations in soil chemistry and nitrogen mineralization in a beech-maple forest. Soil Biol. Biochem., 21: 795-801. BORING, L.R., MONK, CD. and SWANK, W.T. 1981. Early regeneration of a clear-cut southern Appalachian forest. Ecology, 62: 1244-1254. BORING, L.R. and SWANK, W.T. 1984. The role of black locust (Robinia pseudo- acacia) in forest succession. J. Ecol., 72: 749-766. BORMANN, B.T. 1989. Podzolization and windthrow: natural fluctuations in long-term productivity and implications for management. In: Perry, D.A., Meurisse, R., Thomas, B., Miller, R., Boyle, J., Means, J. and Bowers, R.F. (Eds.). Maintaining the long-term productivity of Pacific northwest forest ecosystems. Timber Press, Portland, Oregon. BRAND, D.G. 1986. Competition-induced changes in the developmental features of planted Douglas-fir in southwestern British Columbia. Can. J. For. Res., 16: 191-196. BRAND, D.G. and JANAS, P.S. 1988. Growth and acclimation of planted white pine and white spruce seedlings in response to environmental conditions. Can. J. For. Res., 18: 320-329. BROCKLEY, R.P. 1988. The effects of fertilization on the early growth of planted seedlings: a problem analysis. FRDA report #011, Ministry of Forests and Lands, Victoria, B.C. CARLSON, W.C 1981. Effects of controlled-release fertilizers on shoot and root development of outplanted western hemlock (Tsuga heterophylla Raf. Sarg.) seedlings. Can. J. For. Res., 11: 752-757. 151 CARLSON, W.C. and PREISIG, CL. 1981. Effects of controlled-release fertilizers on the shoot and root development of Douglas-fir seedlings. Can. J. For. Res., 11: 230-242. CARTER, G.A., MILLER, J.H., DAVIS, D.E. and PATTERSON, R.M. 1984. Effect of vegetative competition on the moisture and nutrient status of loblolly pine. Can. J. For. Res., 14: 1-9. CHAPIN III, E.S., BLOOM, A.J., FIELD, CB. and WARING, R.H. 1987. Plant responses to multiple environmental factors. BioScience, 37: 49-57. COLE, D.W. and RAPP, M. 1981, Elemental cycling in forest ecosystems (P: 341-409). In: Reickle, D.E. (Ed.). Dynamic properties of forest ecosystems. Cambridge University Press. Cambridge. COLE, D.W. and VAN MIEGROET, H. 1989. Chronosequences: a technique to assess ecosystem dynamics. Pp. 5-23. In: W.J. Dyck and CA. Mees (Eds.). Research strategies for long-term site productivity. Proceedings, IEA/BE A3 Workshop, Seattle, WA, August, 1988. IEA/BE A3 Report No. 8. Forest Research Institute, New Zealand, Bulletin 152. COLE, E.C. and NEWTON, M. 1986. Nutrient, moisture, and light relations in 5-year-old Douglas-fir plantations under variable competition. Can. J. For. Res., 16: 727-732. CONNELL, J.H. and SLATYER, R.O. 1977. Mechanisms of succession in natural communities and their role in community stability and organization. Amer. Nat., Ill: 1119-1144. COTE, J.-F. and THIBAULT, J.-R. 1988. Allelopathic potential of raspberry foliar leachates on growth of ectomycorrhizal fungi associated with black spruce. Amer. J. Bot., 75: 966-970. COUTTS, M.P. and PHILIPSON, J.J. 1976. The influence of mineral nutrition of the root development of trees. I. The growth of Sitka spruce with divided root systems. J. Exp. Bot., 27: 1102-1111. COUTTS, M.P. and PHILIPSON, J.J. 1987. Structure and physiology of Sitka spruce roots. In: Henderson, D.M. and Faulkner, R. Sitka spruce. Section B (Biological Sciences). Proc. of the Royal of Edinburgh, 93B: 117-129. COVINGTON, W.W. 1981. Changes in forest floor organic matter and nutrient content following clear cutting in northern hardwoods. Ecology, 62: 41-48. CURRAN, M.P. and DUNSWORTH, B.G. 1988. Coastal western red cedar regeneration: problems and potentials. In: Western red cedar - does it have a future? Smith N.J. (Ed.). Conference Proceedings, University of British Columbia, Faculty of Forestry. DAMMAN, A.W.H. 1971. Effect of vegetation changes on the fertility of a Newfoundland forest site. Ecol. Monog, 41: 243-270. DAVID, CA. 1987. Post-harvest forest floor changes and nitrogen mobilization in an Engelmann spruce subalpine fir forest. PhD thesis, Faculty of Forestry, University of British Columbia, Vancouver, B.C. 152 DeCATANZARO, J.B. and KIMMINS, J.P. 1985. Changes in weight and nutrient composition of litter in three forest ecosystem types on coastal British Columbia. Can. J. Bot., 63: 1046-1056. DEL MORAL, R. and CATES, R.G. 1971. Allelopathic potential of the dominant vegetation of western Washington. Ecology, 52: 1030-1037. DICKSON, D.A. and SAVILL, P.S. 1974. Early growth of Picea sitchensis (Bong.) Carr. on deep oligotrophic peat in northern Ireland. Forestry, 47: 57-88. DWYER, L.M. and MERRIAM, G. 1981. Influence of topographic heterogeneity on deciduous litter decomposition. Oikos, 37: 228-237. DYRNESS, CT. 1973. Early stages of plant succession following logging and burning in the western Cascades of Oregon. Ecology, 54: 57-69. EISSENSTAT, D.M. and MITCHELL, J.E. 1983. Effects of seeding grass and clover on growth potential of Douglas-fir seedlings. For. Sci., 29: 166-179. ELLIOT, K.J. and WHITE, AS. 1987. Competitive effects of various grasses and forbs on ponderosa pine seedlings. For. Sci., 33: 356-366. FELLER, M.C., KIMMINS, J.P. and SCOULLAR, KA. 1982. Forcyte-10: Calibration data and simulation of potential long-term effects of intensive forest management on site productivity, economic performance, and energy benefit cost ratio. In: Ballard, R. and Gessel, S.P (Eds.). IUFRO symposium on forest site and continuous productivity. Seattle, Washington, August 22-28. FLANAGAN, P.W. and VAN CLEVE, K. 1983. Nutrient cycling in relation to decomposition and organic-matter quality in taiga ecosystems. Can. J. For. Res., 13: 795-817. FLINT, L.E. and CHILDS, S.W. 1987. Effect of shading, mulching and vegetation control on Douglas-fir seedling growth and soil water supply. For. Ecol. Manage., 18: 189-203. FOX, J.F. and VANCLEVE, K. 1983. Relationships between cellulose decomposition, Jenny's k, forest-floor nitrogen, and soil temperature in Alaskan taiga forests. Can. J. For. Res., 13: 789-794. FRIEBERG, R.F. 1989. Growth of western hemlock and western redcedar on the SI cedar-hemlock phase of northern Vancouver Island. Technical essay, FOR 348, University of British Columbia, Vancouver. GERMAIN, A. 1985. Fertilization of stagnated Sitka spruce plantations on northern Vancouver Island. M.S. Thesis. Faculty of Forestry, University of British Columbia, Vancouver, B.C., Canada. GHOLZ, H.L., HAWK, G.M., CAMPBELL, A. and CROMACK, Jr., K. 1985. Early vegetation recovery and element cycles on a clear-cut watershed in western Oregon. Can. J. For. Res., 15: 400-409. GIBSON, D.J. 1986. Spatial and temporal heterogeneity in soil nutrient supply measured using in situ ion-exchange resin bags. Plant and Soil, 96: 445-450. 153 GOLBERG, D.E. 1990. Components of resource competition in plant communities (Chapter 3). In: J.B. Grace and D. Tilman. Perspectives on plant competition.Academic Press, Inc., San Diego. GREGORY, R.A. 1957. Some silvicultural characteristics of western redcedar in southeast Alaska. Ecology, 38: 646-649. GREEN, R.N. 1990. Response of Douglas-fir seedlings to salal control. FRDA # 2.17.1. Ministry of Forests, Victoria, B.C., Canada. Unpublished. GREEN, R.N., COURTIN, P.J., KLINKA, K., SLACO, R.J. and RAY, CA. 1984. Site diagnosis, tree species selection, and slashburning guidelines for the Vancouver Forest Region. Land Manag. Handbk. #8. Ministry of Forests, Victoria, B.C., Canada. GRIER, CC. 1978. A Tsuga heterophylla- Picea sitchensis ecosystem of coastal Oregon: decomposition and nutrient balances of fallen logs. Can. J. For. Res., 8: 198-206. HACSKAYLO, E. 1972. Mycorrhizae: the ultimate in reciprocal parasitism. BioSci., 22: 577-583. HALPERN, CB. 1988. Early successional pathways and the resistance and resilience of forest communities. Ecology, 69: 1703-1715. HALPERN, CB. 1989. Early successional patterns of forest species: interactions of life history traits and disturbance. Ecology, 70: 704-720. HAMILTON, E.H. and YEARSLEY, H.K. 1988. Vegetation development after clear-cutting and site preparation in the SBS zone. FRDA report # 018, B.C Ministry of Forests and Lands, Victoria, B.C., Canada. HANDLEY, W.R.C. 1963, Mycorrhizal association and Calluna heathland afforestation. For. Comm. Bull. 36, H.M.S.O., London. HANSON, P.J. and DIXON, R.K. 1987. Allelopathic effects of interrupted fern on northern red oak seedlings: amelioration by Suillus luteus L.: Fr. Plant and Soil, 98: 43-51. HENDRICKSON, O.Q. 1988. Biomass and nutrients in regenerating woody vegetation following whole-tree and conventional harvest in a northern mixed forest. Can. J. For. Res., 18: 1427-1436. HOBBS, S.D. 1984. The influence of species and stock type selection on stand establishment: an ecophysiological perspective, pp. 179-224. In: M.L. Duryea and G.N. Brown (Eds.). Seedling physiology and reforestation success. Martinus Nyhoff/Dr. W. Junk Publishers, London. HUSTED, L.D. 1982. The relationships between foliar and soil chemistry, growth parameters and variable height growth in advance regeneration of amabilis fir. M.Sc. Thesis. University of British Columbia, Faculty of Forestry, Vancouver, B.C., Canada. INGESTAD, T. 1973. Mineral nutrient requirements of Vaccinium vitis idaea and V. myrtillus. Physiol. Plant., 29: 239-246. 154 INGESTAD, T. 1979. Mineral nutrient requirements of Pinus silvestris and Picea  abies seedlings. Phvsiol. Plant., 45: 373-380. INGESTAD, T. 1981. Nutrition and growth of birch and grey alder seedlings in low conductivity solutions and at varied relative rate of nutrient solutions. Physiol. Plant., 52: 454-466. INGESTAD, T. and LUND, A.-B. 1986. Theory and techniques for steady state mineral nutrition and growth of plants. Scand. J. For. Res., 1: 439-453. KIMMINS, J.P. 1987. Forest Ecology. Macmillan Publishing Company, New-York. 531 pp. KIMMINS, J.P. 1988. Community organization: methods of study and prediction of the productivity and yield of forest ecosystems. Can. J. Bot., 66: 2654-2672. KIMMINS, J.P. and HAWKES, B.C. 1978. Distribution and chemistry of fine roots in a white spruce - subalpine fir stand in British Columbia: implications for management. Can. J. For. Res., 8: 265-279. KLINKA, K. 1976, Ecosystem units - their classification, mapping and interpretation in the University of British Columbia Research Forest. PhD thesis, Faculty of Forestry, University of British Columbia, Vancouver, B.C. KLINKA, K., GREEN, R.N., TROWBRIDGE, R.L. and LOWE, L.E. 1981. Taxonomic classification of humus forms in ecosystems of British Columbia. Ministry of Forests, Victoria, B.C., Canada. KORMANIK, P.P. and McGRAW, A.-C. 1982. Quantification of vesicular-arbuscular mycrorrhizae in plant roots. In: Schenck, N.C. (Ed.). Methods and principles of mycorrhizal research. The American Phytopathological Society, St. Paul, Minnesota. KRAJINA, V.J. 1969. Ecology of forest trees in British Columbia. Ecology of Western North America, 2: 1-146. KRAJINA, V.J., KLINKA, K. and WORRALL, J. 1982. Distribution and ecological characteristics of trees and shrubs of British Columbia. Fac. of For., Univ. of British Columbia, Vancouver, B.C, Canada. KRAJINA, V.J., MADOC-JONES, S. and MELLOR, G. 1973. Ammonium and nitrate in the nitrogen economy of some conifers growing in Douglas-fir communities of the pacific north-west of America. Soil Biol. Biochem., 5: 143-147. KRAUSE, H.H. and RAMLAL, D. 1986. In situ nutrient extraction by resin from forested, clear-cut and site-prepared soils. Can. J. Soil Sci., 67: 943-952. 155 KURZ, W.A. 1989. Net primary production, production allocation, and foliage efficiency in second growth Douglas-fir stands with differing site quality. PhD thesis, Faculty of Forestry, University of British Columbia, Vancouver, B.C. LAJTHA, K. 1988. The use of ion-exchange resin bags for measuring nutrient availability in an arid ecosystem. Plant and Soil, 105: 105-111. LARGENT, D.L., SUGIHARA, N. and WISHNER, C. 1980. Occurrence of mycorrhizae on ericaceous and pyrolaceous plants in northern California. Can. J. Bot., 58: 2274-2279. LEWIS, T. 1982. Ecosystems of the Port McNeill block 4 of Tree Farm Licence 25. Int. report. Western Forest Products Ltd., Vancouver, B.C., Canada (unpubl.). LIKENS, G.E., BORMANN, F.H., JOHNSON, N.M., FISHER, D.W. and PIERCE, R.S. 1970. Effects of forest cutting and herbicide treatment on nutrient budgets in the Hubbard Brook watershed-ecosystem. Ecol. Monogr., 40: 23-47. LODHI, M.A.K. 1977. The influence and comparison of individual forest trees on soil properties and possible inhibition of nitrification due to intact vegetation. Amer. J. Bot., 64: 260-264. MALLIK, A.U. 1987. Allelopathic potential of Kalmia angustifolia to black spruce (Picea mariana). For. Ecol. Manage., 20: 43-51. MARKS, P.L. 1974. The role of pin cherry (Prunus pensylvanica L.) in the maintenance of stability in northern hardwood ecosystems. Ecol. Monogr., 44: 73-88. MARTIN, W.L. 1985. Post-clear-cutting forest floor nitrogen dynamics and regeneration response in the Coastal Western Hemlock wet subzone. PhD thesis. Faculty of Forestry, University of British Columbia, Vancouver, B.C. 350 pp. McDONALD, M.A. 1989. Competition for nutrients and chemical interference by salal (Gaultheria shallon Pursh). Vegetation management: an integrated approach - Proceeding of the fourth Annual Vegetation Management Workshop. FRDA report 109. November 1989. McINTOSH, R. 1980. Fertiliser treatment of Sitka spruce in the establishment phase in upland Britain. Scottish For., 35: 3-13. McINTOSH, R. 1983. Nitrogen deficiency in establishment phase Sitka spruce in upland Britain. Scottish For., 37: 185-193. McKAY, H.M. 1988. The influence of pine on the form of Sitka spruce fine roots. J. Exp. Bot., 39: 1263-1266. MESSIER, C , HONER, T.W. and KIMMINS, J.P. 1989. Photosynthetic photon flux density, red:far-red ratio, and minimum light requirement for survival of Gaultheria shallon in western redcedar/western hemlock stands in coastal British Columbia. Can. J. For. Res., 19: 1470-1477. 156 MESSIER, C. and KIMMINS, J.P. 1991. Below-ground factors, limiting conifer seedling growth in recently clear-cut sites dominated by Gaultheria shallon in the CWHb biogeoclimatic subzone on northern Vancouver Island. FRDA report #2.31. Ministry of Forests, Victoria, B.C., Canada (In press). MILLER, H.G. and MILLER, J.D. 1987. Nutritional requirements of Sitka spruce (P: 75-83). In: Hendersen, D.M. and Faulkner,R. (Eds.). Sitka spruce. Proc. of the Royal Society of Edinburgh. Section B (Biological Sciences). de MONTIGNY, L.E. and WEETMAN, G.F. 1990. The effects of ericaceous plants on forest productivity. In: B.D. Titus, M.B. Lavigne, P.F. Newton, W.J. Meades, Editors. The silvics and ecology of boreal forests. 1989 IUFRO working party Sl.05-12 Symp. Proc. Newfoundland, 12-17 Aug. 1989. For. Can. Inf. Rep. N-X-271, pp. 83-90. MUNSON. A.D. and TIMMER, V.R. 1989. Site-specific growth and nutrition of planted Picea mariana in the Ontario Clay Belt. I. Early performance. Can. J. For. Res., 19: 162-170. MYERSCOUGH, P.J. 1980. Biological flora of the British Isles: Epilobium  angustifolium L. J. Ecol., 68: 1047-1074. MYERSCOUGH, P.J. and WHITEHEAD, F.H. 1967. Comparative biology of Tussilago farfara L., Chamaenerion angustifolium (L.) Scop., Epilobium  montanum L., and Epilobium adenocaulon Hausskn. II. Growth and ecology. New. Phytol., 66: 785-823. NAMBIAR, E.K.S. and ZED, P.G. 1980. Influence of weeds on the water potential , nutrient content and growth of young radiata pine. Aust. For. Res., 10: 279-288. NEARY, D.G., ROCKWOOD, D.L., COMERFORD, N.B., SWINDEL, B.F. and COOKSEY, T.E. 1990. Importance of weed control, fertilization, irrigation, and genetics in slash and loblolly pine early growth on poorly drained spodosols. For. Ecol. Manage., 30: 271-281. OMULE, S.A.Y. 1988. Juvenile height growth of western redcedar in four sites on Vancouver Island. In: Western red cedar - does it have a future? N.J. Smith (Ed.). Conference Proceedings, University of British Columbia, Faculty of Forestry. OUTCALT, K.W. and WHITE, E.H. 1981. Understory biomass and nutrients 2 years after timber harvest in northern Minnesota. Can. J. For. Res., 11: 305-308. PARKE, J.L., LINDERMAN, R.G. and TRAPPE, J.M. 1984. Inoculum potential of ectomycorrhizal fungi in forest soils of southwest Oregon and northern California. For. Sci., 30: 300-304. PERRY, D.A., MOLINA, R. and AMARANTHUS, M.P. 1987. Mycorrhizae, mycorrhizospheres, and reforestation: current knowledge and research needs. Can. J. For. Res., 17: 929-940. 157 PERSSON, H. 1980. Spatial distribution of fine-root growth, mortality and decomposition in a young Scots pine stand in Central Sweden. Oikos, 34: 77-87. PETERSEN, T.D., NEWTON, M. and ZEDAKER, S.M. 1988. Influence of Ceanothus velutinus and associated forbs on the water stress and stemwood production of Douglas-fir. For. Sci., 34: 333-343. PHILIPSON, J.J. and COUTTS, M.P. 1977. The influence of mineral nutrition on the root development of trees. II. The effect of specific nutrient elements on the growth of individual root of Sitka spruce. J. Exp. Bot., 28: 864-871. PICKETT, S.T.A. 1990. Space-for-time substitution as an alternative to long-term studies. In: G.E. Likens (Ed.). Long-term studies in ecology: approaches and alternatives. Springer-Verlag, New-York. PRICE, D.T., BLACK, TA. and KELLIHER, F.M. 1986. Effects of salal understory removal on photosynthetic rate and stomatal conductance of young Douglas-fir trees. Can. J. For. Res., 16: 90-97. QUESNEL, H.J. and LAVKULICH, L.M. 1981. Comparison of the chemical properties of forest floors, decaying wood, and fine roots in three ecosystems on Vancouver Island. Can. J. For. Res., 11: 215-217. RADWAN, MA. and SHUMWAY, J.S. 1983. Soil nitrogen, sulfur, and phosphorus in relation to growth response of western hemlock to nitrogen fertilization. For. Sci., 29: 469-477. RAICH, J.W. 1980. Fine roots regrow rapidly after forest felling. Biotropica, 12:231-232. READ, D.J. 1984. Interactions between ericaceous plants and their competitors with special reference to soil toxicity. Aspects of Applied Biology, 5: 195-209. ROBINSON, R.K. 1972. The production by roots of Calluna vulgaris of a factor inhibitory to growth of some mycorrhizal fungi. J. Ecol., 60: 219-224. ROSE, S.L., PERRY, D.A., PILZ, D. and SCHOENEBERGER, M.M. 1983. AUelopathic effects of litter on the growth and colonization of mycorrhizal fungi. J. Chem. Ecol., 9: 1153-1162. RUARK, G.A. and BOCKHEIM, J.G. 1987. Below-ground biomass of 10-, 20- and 32-year-old Populus tremuloides in Wisconsin. Pedobiologia, 30: 207-217. RUWET, A. 1980. L'6tonnante vitality d'Epilobium angustifolium L. au pre" des forges a Mirwart (Province de Luxembourg). Bull. Soc. Roy. Belg., 112: 166-170. RYAN, P.J., GESSEL, S.P. and ZASOSKI, R.J. 1986. Acid tolerance of Pacific northwest conifers in solutions culture. II. effect of varying aluminium concentration at constant pH. Plant and Soil, 96: 250-272. SABHASRI, S. 1961. An ecological study of salal (Gaultheria shallon Pursh). Ph.D. thesis, University of Washington, Seattle, Washington, USA. 158 SHAW, III C.G. JACKSON, R.M. and THOMAS, G.W. 1987. Fertilizer levels and fungal strain influence the development of ectomycorrhizae on Sitka spruce seedlings. New Forests, 1: 215-223. SHAW, III C.G., SIDLE, R.C. and HARRIS, A.S. 1987. Evaluation of planting sites common to a southeast Alaska clear-cut. III. Effects of microsite type and ectomycorrhizal inoculation on growth and survival of Sitka spruce seedlings. Can. J. For. Res., 17: 334-339. SIDLE, R.C. and SHAW, III C.G. 1983. Evaluation of planting sites common to a southeast Alaska clear-cut. I. Nutrient status. Can. J. For. Res., 13: 1-8. SNAYDON, R.W. 1962. Micro-distribution of Trifolium repens L. and its relation to soil factors. J. Ecol., 50: 133-143. SOKAL, R.R. and ROHLF, F.J. 1981. Biometry: the principles and practice of statistics in biological research. 2 n d ed. W.H. Freeman and company. New-York. SQUIRE, R.O. 1977. Interacting effects of grass competition, fertilizing and cultivation of nutrient-absorbing organs: selective exploitation of soil heterogeneity. Plant and Soil, 71: 487-493. STANEK, W., BEDDOWS, D. and STATE, D. 1979. Fertilization and thinning effects on a Douglas-fir ecosystem at Shawnigan lake on Vancouver island: some observations on salal and bracken fern undergrowth. Pacific Forest Research Centre, Victoria, B.C., Canada. STEWART, R.E., GROSS, L.L. and HONKALA, B.H. 1984. Effects of competing vegetation on forest trees: a bibliography with abstracts. U.S. Dep. Agric. For. Serv., Gen. Tech. Rep. WO-43. Washington, D.C. 260 pp. TAMM, CO. 1956. The response of Chamaenerion angustifolium (L.) Scop, to different nitrogen sources in water culture. Physiol. Plant., 9: 331-337. TATE, III, R.L. 1987. Soil organic matter: biological and ecological effects. John Wiley and Sons. New-York. TAYLOR, B.R., PRESCOTT, C.E., PARSONS, W.F.J, and PARKINSON, D. 1991. Substrate control of litter decomposition in four rocky mountain coniferours forests. (Submitted to Can. J. Bot.) HARKS, A.E. and HAYWOOD, J.D. 1986. Pinus radiata L. response to fertilization, herbaceous plant control, and woody plant control. For. Ecol. Manage., 14: 103-112. TILMAN, D. 1990. Mechanisms of plant competition for nutrients: the elements of a predictive theory of competition (Chapter 7). In: J.B. Grace and D. Tilman. Perspectives on plant competition. Aca3emic Press, Inc., San Diego. TRUST, T.J. and COOMBS, R.W. 1973. Antibacterial activity of B-thujaplicin. Can. J. Microbiol., 19: 1341-1346. TURNER, D.P. and FRANZ, E.H. 1986. The influence of canopy dominants on understory vegetation patterns in an old-growth cedar-hemlock forest. Am. Midi. Nat., 116: 387-393. 159 UGOLINI, F.C., BORMANN, B.T. and BOWERS, F.H. 1990. The effect of treethrow on forest soil development in southeast Alaska. (Submitted to Can. J. For. Res.) UHL, C. and JORDAN, CF. 1984. Succession and nutrient dynamics following forest cutting and burning in Amazonia. Ecology, 65: 1476-1490. VALES, D.J. 1986. Functional relationships between salal understory and forest overstory. M.Sc. thesis. University of British Columbia, Faculty of Forestry, Vancouver, B.C., Canada. VAN CLEVE, K. and V1ERECK, L.A. 1981. Forest succession in relation to nutrient cycling in the boreal forest of Alaska (Chapter 13). In: D. C. West; Shugart, H.H.; Botkin, D.B. (Eds.). Forest succession: concepts and application. Springier-Verlag. New-York. VOGT, KA., VOGT, D.J., MOORE, E.E., FATUGA, B.A., REDLIN, M.R. and EDMONDS, R.L. 1987. Conifer and angiosperm fine-root biomass in relation to stand age and site productivity in Douglas-fir forests. J. Ecol., 75: 857-870. WALKER, C. 1987. Sitka spruce mycorrhizas. In: Henderson, D.M. and Faulkner, R. Sitka spruce. Section B (Biological Sciences). Proc. of the Royal of Edinburgh, 93B: 117-129. WALLACE, E.S. and FREEDMAN, B. 1981. Forest floor dynamics in a chronosequence of hardwood stands in central Nova Scotia. Can. J. For. Res., 16: 293-302. WEETMAN, G.F., FOURNIER, R., BARKER, J., SCHNORBUS-PANOZZO, E., and GERMAIN, A. 1989a. Foliar analysis and response of fertilized chlorotic Sitka spruce plantations on salal-dominated cedar-hemlock cutovers on Vancouver Island. Can. J. For. Res., 12: 1501-1511. WEETMAN, G.F., FOURNIER, R., BARKER, J., and SCHNORBUS-PANOZZO, E. 1989b. Foliar analysis and response of fertilized chlorotic western hemlock and western redcedar reproduction on salal-dominated cedar-hemlock cutovers on Vancouver Island. Can. J. For. Res., 12: 1512-1520. WEETMAN, G.F., FOURNIER, R.M., SCHNORBUS-PANOZZO, E. and BARKER, J. 1990. Post-burn nitrogen and phosphorus availability in coastal British Columbia cedar/hemlock forests and the use of fertilization and salal eradication to restore productivity. In: Gessel, S.P. (Ed.), Sustained productivity of forest lands. Proceedings of seventh North American Forest Soils Conference, Friesen Press, Vancouver, B.C., Canada. WEETMAN, G.F., RADWAN, M.A., KUMI, J. and SCHNORBUS, E. 1988. Nutrition and fertilization of western redcedar. In: Western Redcedar -Does It Have a Future? Eds. N.J. Smith, pp 47-60. Conference Proceeding, UBC, Faculty of Forestry, Vancouver, B.C., Canada. WEIDENHAMER. J.D., HARTNETT, D.C. and ROMEO, J.T. 1989. Density-dependent phytotoxicity: distinguishing resource competition and allelopathic interference in plants. J. Appl. Ecol., 26: 613-624. 160 WHITTAKER, R.H. and MARKS, P.L. 1975. Methods of assessing terrestrial productivity. In: H. Lieth and R.H. Whittaker (Eds.), Primary productivity of the biosphere. Springer, New-York, pp. 55-118. WILKINSON, L. 1988. SYSTAT: the system for statistics. Evanston, IL: SYSTAT, Inc. YIN, X., PERRY, J.A. and DIXON, R.K. 1989. Fine-root dynamics and biomass distribution in a Quercus ecosystem following harvesting. For. Ecol. and Manag., 27: 159-177. ZINKE, P.J. 1962. The pattern of influence of individual forest trees on soil properties. Ecology, 43: 130-133. ZUTTER, B.R., GJERSTAD, D.H. and GLOVER, G.R. 1986. Effects of interfering vegetation on biomass, fascicle morphology and leaf area of loblolly pine seedlings. For. Sci., 32: 1016-1031. 

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.831.1-0100580/manifest

Comment

Related Items