UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Investigation into the productivity of single- and mixed-species, second-growth stands of western hemlock… Collins, D. Bradley 2001

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

Item Metadata

Download

Media
831-ubc_2001-0015.pdf [ 4.57MB ]
Metadata
JSON: 831-1.0099569.json
JSON-LD: 831-1.0099569-ld.json
RDF/XML (Pretty): 831-1.0099569-rdf.xml
RDF/JSON: 831-1.0099569-rdf.json
Turtle: 831-1.0099569-turtle.txt
N-Triples: 831-1.0099569-rdf-ntriples.txt
Original Record: 831-1.0099569-source.json
Full Text
831-1.0099569-fulltext.txt
Citation
831-1.0099569.ris

Full Text

I N V E S T I G A T I O N I N T O T H E P R O D U C T I V I T Y O F S I N G L E - A N D M I X E D - S P E C I E S , S E C O N D - G R O W T H S T A N D S O F W E S T E R N H E M L O C K A N D W E S T E R N R E D C E D A R by D . B R A D L E Y C O L L I N S B . S c , The University of British Columbia, 1997 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L M E N T OF T H E R E Q U I R E M E N T S F O R T H E D E G R E E OF M A S T E R O F S C I E N C E in T H E F A C U L T Y OF G R A D U A T E STUDIES Department of Forest Sciences FACULTY OF FORESTRY We accept this thesis as conforming to the required standard T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A October 2000 © D . Bradley Collins, 2000 In presenting this thesis in partial fulfillment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication o f this thesis for financial gain shall not be allowed without my written permission. (Signature) The University of British Columbia Vancouver, Canada Date ABSTRACT In order to evaluate i f mixed-species stands of western hemlock (Tsuga heterophylla (Raf.) Sarg.) and western redcedar (Thujaplicata Donn ex D . Don) were more productive than single-species stands o f either species, I: (1) reviewed the theory of positive plant interactions and integrated it with the silvical characteristics of hemlock and redcedar to determine i f positive interactions are feasible among these species, (2) located even aged, unmanaged, second growth, single- and mixed-species stands of hemlock and redcedar on intermediate (fresh and nutrient medium) sites, within the perhumid mesofhermal climate of southern coastal British Columbia, (3) compared the productivity of each stand type using relative and absolute yield comparisons, and (4) evaluated the mixed-species stands to determine i f increasing productivity was associated with physical separation of the species, as suggested by the findings in (1). Theory on positive plant interactions logically divides the mechanisms of these interactions into two components: competitive reduction, and facilitation. When applied to existing mixed-species stands of hemlock and redcedar, the mechanisms of competitive reduction and facilitation suggest that hemlock and redcedar may experience positive interactions through: (1) vertical canopy separation and a random spatial pattern of trees, and (2) preferential uptake of different nitrogen forms. The relative yield of hemlock and redcedar was lower in mixture, as compared to single-species stands, due to the effects of competition with each other and non-study species. Mean annual increment and wood volume production increased with increasing presence of hemlock. Basal area increased with increasing presence of redcedar, and the redcedar stand type was also the densest. Redcedar trees in the redcedar stand type were similar in height, diameter, and mean annual increment, relative to those in the hemlock-redcedar stand type. Hemlock trees in the hemlock stand type were taller, had larger diameters, and had higher mean annual increments, than those in the hemlock-redcedar stand type. These differences in hemlock and redcedar i i growth in each stand type are thought to be responsible for the differences in mean productivity detected among stand types. In the mixed-species stands, increasing productivity was associated with increasing vertical separation of hemlock and redcedar canopies, where the area of hemlock canopy overtopping that of redcedar, and crown depths of both species, were maximized. Increased productivity was not associated with increasing levels of randomness in stand spatial tree patterns, but was associated with decreasing stand density, and a tendency towards regularity in the stand spatial tree patterns. Maximum productivity in fully stocked hemlock-redcedar mixtures can be attained with low density stands in which hemlock does not experience an establishment lag. Given the choice between single- and mixed-species stands of hemlock and redcedar, hemlock stands should be established i f maximum wood production is the management objective. If maintaining a mixture, or improving site nutrient status, is the management objective, then hemlock-redcedar stands should be established. Maximum wood production in a hemlock-redcedar stand w i l l be achieved where stands are fully stocked, but established at low numbers of stems per hectare. These recommendations are for intermediate (blueberry) sites in the Submontane Very Wet Maritime Coastal Western Hemlock ( C W H v m l ) variant, for stands less than 80 years old. i i i TABLE OF CONTENTS A B S T R A C T i i T A B L E O F C O N T E N T S iv L I S T O F T A B L E S v i L I S T O F F I G U R E S v i i i A C K N O W L E D G E M E N T S x 1. G E N E R A L I N T R O D U C T I O N 1 2 . S T U D Y A R E A A N D S T A N D S 4 3 . M A T E R I A L S A N D M E T H O D S 7 3.1 D A T A C O L L E C T I O N 7 3 .2 P R O D U C T I V I T Y A N A L Y S E S 7 3.2.1 Relative Y i e l d 7 3 .2 .2 Y i e l d 8 3.3 S T R U C T U R A L A N A L Y S E S 1 2 3.3.1 Canopy stratification 13 3 .3 .2 Spatial randomness 1 5 Refined nearest neighbour analysis 1 6 Combined count and distance analysis 18 4 . P O S I T I V E P L A N T I N T E R A C T I O N S - A P P L Y I N G C U R R E N T T H E O R I E S T O M L X E D - S P E C I E S S T A N D S O F H E M L O C K A N D R E D C E D A R 2 0 A B S T R A C T 2 0 4.1 I N T R O D U C T I O N 2 1 4 . 2 C O M P E T I T I V E R E D U C T I O N 2 2 4.2 .1 Reduced Crown Competition 2 2 Physiological Separation 2 2 Spatial Separation 2 3 Temporal Separation 2 4 Incidence of Reduced Crown Competition in Hemlock-Redcedar Stands 2 5 4 . 2 . 2 Reduced Root Competition 2 7 Physiological Separation 2 7 Spatial Separation 2 7 Temporal Separation 2 8 Incidence of Reduced Root Competition in Hemlock-Redcedar Stands 2 9 4 .3 F A C I L I T A T I O N 3 0 4 .3 .1 Resource Modification 3 1 Light and Temperature 3 1 Soil Moisture 3 2 Soi l Nutrients 3 2 Incidence of Resource Modification in Hemlock-Redcedar Stands 3 4 4 . 3 . 2 Substrate Modification 3 6 iv Aeration/Soil Oxygenation 3 6 Substrate Supply and Stabilization 3 6 Incidence o f Substrate Modification in Hemlock-Redcedar Stands 3 7 4 . 3 . 3 Mortality Agent Attenuation 3 7 Competitor Control 3 7 Herbivory Reduction 3 8 Amelioration of Environmental Extremes 3 8 Incidence of Mortality Agent Attenuation in Hemlock-Redcedar Stands 3 9 4 . 3 . 4 Additional Effects 4 1 Pollination and Propagule Concentration 41 Incidence of Additional Facilitating Effects in Hemlock-Redcedar Stands 4 2 4 . 4 T H E D Y N A M I C S A N D I M P L I C A T I O N S O F P O S I T I V E I N T E R A C T I O N S A M O N G P L A N T S 4 2 4 .5 M E T H O D S O F C O M P A R I N G P R O D U C T I V I T Y OF S I N G L E - A N D M I X E D - S P E C I E S S T A N D S 4 4 4 . 6 S U M M A R Y A N D C O N C L U S I O N S 4 7 5. T H E P R O D U C T I V I T Y OF S I N G L E - A N D M L X E D - S P E C I E S S T A N D S OF H E M L O C K A N D R E D C E D A R 4 8 A B S T R A C T 4 8 5.1 I N T R O D U C T I O N 4 8 5.2 R E S U L T S 4 9 5.2.1 Productivity Comparisons - Relative Y i e l d 5 2 5 .2 .2 Productivity Comparisons - Absolute Y i e l d 5 3 5.3 D I S C U S S I O N 5 7 5.3.1 Productivity Comparisons - Relative Y i e l d 5 7 5 .3 .2 Productivity Comparisons - Absolute Y i e l d 5 9 5.4 C O N C L U S I O N S 6 1 6. A R E I N C R E A S I N G C A N O P Y S E P A R A T I O N A N D S P A T I A L R A N D O M N E S S A S S O C I A T E D W I T H I N C R E A S E D P R O D U C T I V I T Y I N M L X E D - S P E C I E S S T A N D S OF H E M L O C K A N D R E D C E D A R ? 6 3 A B S T R A C T 6 3 6.1 I N T R O D U C T I O N 6 3 6 .2 R E S U L T S 6 4 6.2.1 Canopy stratification 6 4 6 .2 .2 Spatial randomness 6 6 Refined nearest neighbour analysis 6 6 Combined count and distance analysis 6 8 6.3 D I S C U S S I O N 7 1 6.3.1 Canopy stratification 7 1 6 . 3 . 2 Spatial randomness 7 3 Refined nearest neighbour analysis 7 3 Combined count and distance analysis 7 4 6.4 C O N C L U S I O N S 7 8 7. G E N E R A L C O N C L U S I O N S 8 0 8. L I T E R A T U R E C I T E D 8 2 v LIST OF TABLES Table 2.1 Latitude, longitude, and elevation of study areas in southern coastal British Columbia 4 Table 2.2 Percentage of basal area contribution to each stand, stratified according to stand type 5 Table 2.3 Selected characteristics (means) of the study stands stratified according to study area and stand types ( H - western hemlock, H C - hemlock-redcedar, C - redcedar) 6 Table 4.1 The mechanisms of positive plant interactions (adapted from: Callaway 1995; M a n and Lieffers 1999) 21 Table 5.1 Selected characteristics of study stands, stratified according to stand type; where: x is the stand type mean, s - is the standard error of the stand type mean, and mixed-species stand values separated by a slash represent those for each of redcedar / hemlock 51 Table 5.2 Relative yield ( R Y ) and total relative yield ( R Y T ) values comparing volume of hemlock and redcedar stands with stands of hemlock-redcedar mixtures, with and without the volume contributions of non-study species 52 Table 5.3 Mean basal areas, standard errors (in parentheses), and percent contribution of hemlock, redcedar, and non-study species to total stand basal area, stratified according to stand type 53 Table 5.4 Analysis of variance summary for mean annual increment (m3ha" 'yr" 1), where: hemlock, hemlock-redcedar, and redcedar are stand types; Capilano, Malco lm Knapp, and Mission are locations 54 Table 5.5 Least squares means and standard errors (in parentheses) of mean annual increment (MAI) , volume, and basal area ( B A ) of 18 study stands, stratified according to density covariate and stand type. Values in the same row with the same superscript are not significantly different (or = 0.05) 55 Table 5.6 Individual mean tree characteristics, and standard errors of the means (in parentheses), o f each species in single- and mixed-species stands. Redcedar tree values in the same row with the same lowercase superscript are not significantly different, and hemlock tree values in the same row with the same uppercase superscript are not significantly different ( a = 0.05) 57 v i Table 6.1 Least squares means and standard errors (in parentheses) of mean annual increment per tree, stratified according to hemlock-redcedar stand. Values in the same row with the same superscript are not significantly different (a =0.05) 64 Table 6.2. Percentage of basal area contribution ( % B A ) , quadratic mean diameter, mean tree height, mean height to live crown, and crown overlap ratio (COR), of hemlock and redcedar in each of the hemlock-redcedar stands. Stands are arranged in order of decreasing productivity. Standard errors of means appear in parentheses. Tree height values in the same row with the same lowercase superscript are not significantly different, and height to live crown values in the same row with the same uppercase superscript are not significantly different (a = 0.05) 65 Table 6.3. Percentage of basal area contribution ( % B A ) , quadratic mean diameter, mean tree height, and mean height to live crown, of hemlock and redcedar in the most (M2) and least (M3) productive hemlock-redcedar stands. Standard errors of means appear in parentheses. Tree height values in the same row with the same lowercase superscript are not significantly different, and height to live crown values in the same row with the same uppercase superscript are not significantly different (a = 0.05) 66 v i i LIST OF FIGURES Figure 3.1 Schematic diagram illustrating parameters used in calculating the stand ratio of crown overlap, where: ' A ' represents the mean crown depth (from the top of the tree to the closest of: 66% of top height, or height of live crown) of hemlock, ' B ' represents the mean crown depth of redcedar, ' C represents the mean, maximum canopy depth of the stand, and ' D ' represents the mean depth of crown overlap between hemlock and redcedar 14 Figure 4.1 (From Kelty 1992) Results from hypothetical replacement series experiment, expressed as relative and absolute yield. Dashed lines represent expected yield i f intra- and inter-specific interactions were equivalent. Solid lines represent experimental yields 45 Figure 4.2 (From Jolliffe 1997) Example plots in which the same area is planted in a (a) 2:1 mixture of two species, (b) 2:1 ratio of pure sub-plots, (c) pure stand of species 1, and (d) pure stand of species 2 45 Figure 6.1. Output of refined nearest neighbour analysis for the most (M2) and least (M3) productive hemlock-redcedar stands, in terms of mean "2 1 1 annual increment (m ha" yr") . The horizontal line with Y intercept = 0 represents the expected nearest neighbour function for a Poisson forest [G{d) = \-eXni<\ 67 Figure 6.2 Ripley's Ll2 (d) results for composite (stand - A ) , and individual (redcedar - B ; hemlock - C), spatial point patterns for the most productive hemlock-redcedar stand (M2). The horizontal line with Y intercept = 0 represents the expected value [L 12(d) = 0] o f a random spatial point pattern, and the dotted lines represent the upper and lower pointwise 95% confidence envelope boundaries 69 Figure 6.3 Ripley's Ln (d) results for composite (stand - A ) , and individual (redcedar - B ; hemlock - C) , spatial point patterns for the least productive hemlock-redcedar stand (M3). The horizontal line with Y intercept = 0 represents the expected value [L 12(d) = 0] o f a random spatial point pattern, and the dotted lines represent the upper and lower pointwise 95% confidence envelope boundaries 70 v i i i Figure 6.4 Top down view of the most productive hemlock-redcedar stand M 2 . Each circle represents an individual stem, stratified by colour according to species, and the diameter of each circle is based on the diameter at breast height of each tree 75 Figure 6.5 Top down view of the least productive hemlock-redcedar stand M 3 . Each circle represents an individual stem, stratified by colour according to species, and the diameter of each circle is based on the diameter at breast height of each tree 77 ix ACKNOWLEDGEMENTS It has been a long road. Without the patience, guidance, and support of my supervisor, Dr. Karel Kl inka , I would doubtless have become lost in some remote wasteland of academia. Karel has been a taskmaster, motivator, tutor, and friend, and I am better for it. I would also like to thank Dr. Peter Marshall for helping me with critical strategic planning, and providing crucial navigational assistance through some intimidating analytical landscapes. It was Dr. John Barker who reminded me to think about what I was writing, which, frighteningly, was a revelation to me at the time. Thanks. There are a great many other professors who have provided invaluable advice, but I could not possibly list them all within the confines of this page. Y o u know who you are, and I thank you. I also extend my gratitude to the Research Branch of the B . C . Ministry of Forests, and the National Science and Engineering Research Council , for their generous funding of this project. Incredibly, those who helped me collect my data and write my dissertation are still friends. To: Dave Affleck with his insatiable appetite for plot reconnaissance and sometimes rapier wit, Dave Brisco and Joseph Bartel who collected data in deplorable environmental conditions with stoic cheerfulness, Christine Chourmouzis who never hesitated to provide graphical expertise, and Pal Varga who repaid incessant pestering with critical hardware, software, and editorial assistance. If it weren't for you all , I 'd probably be downtown, piloting an espresso machine through a frenetic ocean of yuppies, while bitterly begrudging my parents' generation for the economic good fortune of their demographic. However, it is to my family that I am most indebted. M y wife Sue has taught me how to love and to live, and has renewed my faith in believing that the whole is more than the sum of its parts. Her tolerance, support, and love have been boundless. M y baby to come has cast the world in a new light, a light in which inspiration comes easier and the mundane isn't quite so unpalatable. Thanks little one, I ' l l try to live up to your example. Lastly, I dedicate this work to my late mother, Audrey Nazar. Thanks M o m , for showing me how to plant seeds that were dreams, so that one day I could walk through the forest we grew... B . C . x 1. GENERAL INTRODUCTION British Columbia (BC) forest policy recommends that harvested areas be regenerated with a mixture of tree species, rather than with a single species, wherever a mixture is suited to the site (Anonymous 1995). This policy is based on the assumptions that stand productivity, reliability, or biodiversity w i l l all be enhanced in mixed-species stands. Analogous assumptions have been made in most European countries, including, in Germany, the assumption that mixed-species stands w i l l also enhance aesthetic values (Burschel et al. 1992). Tree species composition has impacts on all stand characteristics, including productivity, structure, and nutrient cycling, and the B C Ministry of Forests has implemented tree species selection guidelines recognizing these impacts (Kl inka 1977; Green et al. 1984; K l inka and Feller 1984; Green and Kl inka 1994). However, the knowledge justifying the policy of regenerating a mixture of tree species, and the practice of choosing an appropriate mixture, is at best incomplete, and at worst anecdotal. Current policy is to mimic naturally occurring mixtures, however, the de facto assumption that mixtures are generally beneficial is yet to be demonstrated. Studies determining how effectively different mixtures meet specific value objectives are required to guide regeneration policy. Traditionally, the population dynamics, distribution rates and patterns, and productivity of plants have been attributed mostly to competition (Connell 1983; Fowler 1986; Aarssen & Epp 1990). A large body of complex, experimental evidence exists supporting the role and mechanisms of competition. This is the basis, either implicitly or explicitly, o f most general conceptual ecological models (Callaway 1995). However, an increasing number of studies are providing evidence to support alternate theories, in which positive plant species interactions also play a major role in the reproduction, distribution, diversity, and productivity of plant communities (Hunter & Aarssen 1988; DeAngelis et al. 1986; Grace and Tilman 1990; Wang 1 1997). Positive plant interactions are meaningfully divided into two components (Vandermeer 1989; Kel ty 1992; Callaway 1995): 1. competitive reduction through structural and physiological differences in above and below ground structures, (this mechanism is also known as niche separation), and 2. facilitation through any positive effect on the growing environment of one plant species by another species. Those mixtures of plant species in which competition is reduced, or one species facilitates the growth of the other, are said to have high ecological combining ability (Harper 1977; Kelty 1992), increasing the efficiency of plants to utilize a limited resource pool (Kelty 1992). A mixed stand of plant species with high ecological combining ability should be more productive than single-species stands of either component species. Mixtures may also increase pest and disease resistance, increase biodiversity values, and reduce the risk of crop failure. However, silvicultural practice generally favours high-value monocultures, and mixed-species stands are usually only established where the associated benefits are deemed necessary to reduce risk to the species of high commercial value (Kelty 1992). In recent decades interest in the productivity of mixed-species stands has intensified. Some naturally occurring mixtures have been investigated to determine whether potential productivity advantages can be gained by growing mixed-species stands. Mixtures have been found to be both more or less productive than their component species in monocultures, depending on both the species and site in question (Kelty 1992; Binkley et al. 1992; Burschel et al. 1992; Callaway 1995; DeBel l et al. 1997). The results of these studies suggest that no a priori assumptions should be made about the productivity advantages or disadvantages of a particular species mixture. 2 A complex balance between competition and positive interactions exists between plant species occurring in mixtures. This balance may be affected by the life stage of the species involved, environmental harshness, and the degree to which resources required for growth are limited (Callaway 1995). It is necessary to study the productivity of each mixture relative to a particular site type and life stage of both species. The objectives of this study, with respect to naturally established, unmanaged stands of hemlock, redcedar, and their mixtures, are: (1) to review the mechanisms of positive plant interactions and their potential to occur in these mixtures, and (2) to compare the productivity of these three stand types, using relative yield and absolute yield, and (3) to determine i f increased productivity in these mixtures is associated with competitive reduction, through the vertical separation of hemlock and redcedar canopies, or increasing spatial randomness in tree spatial patterns. 3 2. STUDY AREA AND STANDS This study was conducted in southern British Columbia, in the foothills of the Coastal Mountains. The areas under study were within the Submontane Very Wet Maritime Coastal Western Hemlock ( C W H v m l ) variant, characterized by a perhumid, cool mesothermal climate (Klinka et al. 1991). Candidate stands were located in three areas: Capilano River (Capilano), University of British Columbia Malco lm Knapp Research Forest (Malcolm Knapp), and Miss ion Tree Farm No . 26 (Mission) (Table 2.1). Table 2.1 Latitude, longitude, and elevation of study areas in southern coastal British Columbia. Study area Latitude Longitude Elevation (m) Capilano 4 9 ^ 2 8 ' N 123" 08' W 308 Malco lm Knapp 4 9 ° 1 7 ' N 122° 36' W 362 Mission 4 9 ° 2 1 ' N 1 2 2 ° 2 1 ' W 371 The selection criteria required study stands to be as similar as possible in history, site quality, and structure (i.e., age, species composition, stocking, density, spatial (horizontal and vertical) tree distribution). Within each area (location), 3 stands were selected in each of 3 stand types: 3 hemlock, 3 redcedar, and 3 mixtures (with approximately equal proportions of hemlock and redcedar), for a total of 27 study stands. Final assignment of each stand to one of the 3 stand types was based on basal area. Those stands with greater than 66% basal area contribution of either species were assigned to the hemlock or redcedar stand types; those stands with less than 16% difference in basal area contributions by hemlock and redcedar were assigned to the hemlock-redcedar stand type. Those stands with more than 15% basal area contribution by non-4 study species were excluded from the study. These criteria resulted in the selection of 18 stands for the study: 4 redcedar stands, 7 hemlock stands, and 7 hemlock-redcedar stands (Table 2.2). Table 2.2 Percentage of basal area contribution to each stand, stratified according to stand type. Stand type Stand Identification Redcedar Basal area contribution (%) Hemlock Non- study species Redcedar C I 75 15 10 C2 73 26 1 C3 69 27 4 C4 67 33 0 Hemlock- M l 55 39 6 Redcedar M 2 52 40 8 M 3 52 43 5 M 4 48 52 0 M 5 42 58 0 M 6 40 51 9 M 7 39 52 9 Hemlock H I 16 73 11 H2 12 73 15 H3 10 76 14 H4 6 90 4 H5 4 94 2 H6 2 95 3 H7 0 95 5 A l l second growth study stands were naturally regenerated after clearcutting and slashburning of old-growth stands, were unmanaged, fully stocked (Hamilton and Christie 1971), and even-aged (where <20 years at breast height age difference existed between all stands' site trees). Study stands were located on the most commonly occurring, intermediate (zonal) H w B a -Blueberry (Vaccinium ovalifolium Sm.) sites (Green and Kl inka 1994); selection of study stands with fresh and medium soil moisture and nutrient regimes kept the edaphic conditions of all sample plots within a narrow range. Study stands were free of disease and insect problems, and represented a late stem exclusion stage of stand development (Oliver and Larson 1996). This development stage allowed sufficient time for the overstory to influence forest floor nutrient properties, but preceded the understory reinitiation stage and the influence this understory vegetation may have on forest floor nutrient properties. Forest floor depths ranged from 2 to 23 cm, and potential rooting depths ranged from 50 to over 100 cm; soils were typically mid-slope, coarse-skeletal, Humo-Ferric or Ferro-Humic Podzols (Soil Classification Working Group 1998), with textures ranging from sandy to loamy, derived from glacial t i l l and underlain by granitic rocks, and coarse fragment contents ranging from 10 to 70%. Humus forms were generally thick (6 to 14 cm) Mors in hemlock stands, and Mormoders in redcedar stands and hemlock-redcedar stands. Although the study stands varied somewhat in the individual physical properties (Table 5.1), such as topographic and soil characters, because of the compensating effects of environmental factors, they were considered to be ecologically-equivalent, i.e., to provide the same growing conditions and to have the same vegetation potential (Cajander 1926; Bakuziz 1969; Major 1977). Table 2.3 Selected characteristics (means) of the study stands stratified according to study area and stand types ( H - western hemlock, H C - hemlock-redcedar, C -redcedar). Study area Capilano Malcom Knapp Mission Stand H H C C H H C C H H C C Number of stands 3 3 3 3 3 3 3 3 3 Slope gradient (%) 25 20 35 28 32 34 37 33 26 Forest floor (cm) 17 19 9 13 14 5 7 10 4 Content of coarse fragments (%) 25 32 40 20 25 30 30 33 30 Rooting depth (cm) 35 25 33 45 72 65 70 58 53 Cover of understory vegetation (%) 21 27 28 62 35 44 36 2 3 Forest floor cover (%) 65 62 65 72 71 69 65 55 53 Decaying wood cover (%) 22 22 13 25 23 20 30 28 35 Coarse fragment cover (%) 13 11 20 5 6 11 5 13 10 Mineral soil cover (%) 0 1 2 0 0 0 1 4 2 6 3. MATERIALS AND METHODS 3.1 Data Collection Within each study stand a 30 x 30 m (0.09 ha) area, uniform in structural, topographic, and soil characteristics, was established. The understory vegetation and environment of each study stand was described. The vegetation description included strata coverage of the tree, shrub, herb and moss layers, and coverage of component species within each stratum, expressed as percent ground cover. The environmental description included elevation, slope gradient, aspect, slope position, microtopography, parent materials, surface substrate, humus form, potential rooting depth, soil texture and structure, coarse fragment content, thickness and sequence of soil horizons, and soil identification. The corners of each study stand were marked with painted stakes, and a 5 m x 5 m grid was established within the stand. The grid perimeter and grid intersection points were marked with labeled flagging pins. The species, height, height to live crown, diameter at breast height (dbh), and position relative to the grid (within 0.1 m), of all live and dead trees taller than 1.3 m were recorded. The ages at breast height (1.3 m) of the three dominant trees of each leading species per stand were determined through increment coring techniques. 3.2 Productivity Analyses 3.2.1 Relative Yield The relative yields of the mixed- and single-species stands were compared using the 3 1 1 mean annual increment (m ha" yr") of each stand, based on the sum of individual live tree volumes (inside bark, gross volume from stump to tree top) divided by the mean age at breast height of three dominant trees of each leading species per stand. Relative yield comparisons 7 were made with and without the volume contributions of non-study (inclusion) species. For mixtures of species X and Y , relative yield ( R Y ) and total relative yield ( R Y T ) comparisons were calculated: R Y x = Yie ldx in mixture / Yie ldx in monoculture [3.1] R Y T = R Y x + R Y y [3.2] If both species use resources in mixture identically to that in monoculture, the R Y of each should be 0.5 and the R Y T should be 1.0. Where competition is reduced or species are facilitated, R Y T values greater than 1.0 are expected. Where competitive interactions are dominant among the species, R Y T values of less than 1.0 are expected. 3.2.2 Yield Productivity was compared among stand types using mean annual increment. These comparisons were augmented with comparisons using stand volume and basal area. Gross tree volume equations were obtained from the B C Ministry of Forests (Anonymous 1976), in accordance to the appropriate age and forest inventory zone. The gross volume equations utilized were: logic VHemiock = - 4.418820 + 1.8677801ogio(dbh) + 1.0998901og10(Ht) [3.3] logio VRedcedar = - 4.139118 + 1.7167701og10 (dbh) + 1.00474201ogi 0(Ht) [3.4] logio Vpacific silver fir 4.266202 + 1.7829601og10 (dbh) + 1.1038201og10 (Ht) [3.5] logio Voougias-fr = - 4.319071 + 1.8138201og10 (dbh) + 1.0424201og10 (Ht) [3.6] where dbh is measured in centimeters, and Ht represents tree height measured in meters. Individual tree volumes were back-transformed from log volumes before summing the stand total for further analysis. Stand volume totals included non-study tree species for which 8 volume equations were available, namely Pacific silver fir (Abies amabilis (Dougl. ex Loud.) Dougl. ex J. Forbes) and Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco var. menziesii) (equations [3.5] and [3.6]). These inclusions contributed an average basal area of 5.6% overall, with inclusions in the hemlock stand type at 7.7%, the hemlock-redcedar stand type at 5.3%, and the redcedar stand type at 3.8%. Hardwood species in the shrub layers (<10 m height) were not included in stand volume calculations. Productivity comparisons among stand types (hemlock, hemlock-redcedar, and redcedar) were performed with analysis of covariance ( A N C O V A ) models, according to the general equation: Y„ =ii. +t, + yx {Xin -X.x) + r2 (Xij2 -X.2) + ...+ ev [3.7] where YtJ is the value of the productivity measure of the jth stand in the z'th stand type, /u. is the overall mean of that measure, r , are the fixed stand type effects, yx is the regression coefficient for the relation between Y and the first model covariate Xi]X, XijX represents the value of the first model covariate for they'th stand in the z'th stand type, X..\ is the overall mean of the first model covariate, and £ are independent. Additional model covariates were added according to the specific analysis, as detailed in the remainder of this section. Productivity measures from each stand type were assessed for normality with probability plots and subjected to variance homogeneity tests via Bartlett's test (Bartlett 1937) ( a = 0.05). Tests for outliers were conducted with studentized residual analysis (Neter et al. 1996). Interactions were not detected between stand types and covariates in A N C O V A models, indicating stand type response variables were close to parallel. Volume was the only productivity measure with non-homogenous variance among stand types, and model coefficients were fitted via iteratively re-weighted least squares procedures until stable coefficients were attained (Neter et al. 1996). 9 Despite controlling as much as possible for tree age, site quality, and stand density during the selection of study stands, some variation in these factors could not be avoided. Since tree age, site quality, and stand density all affect productivity, variation in these factors was addressed by adding these covariates to analysis of covariance models. Adding model covariates for age, site, and density provided estimates of productivity in which the means were adjusted to reflect trends in stand productivity for equivalent age, site, and density. Within each stand, tree age was measured as mean age at breast height (Table 2.3), based on three dominant trees of each leading species per stand. Dominant trees were selected as those trees, free from damage or disease, of largest dbh in the stand. Site was measured as site index (Table 2.3), which, in the hemlock-redcedar stand type, represents a modified site index for each stand. The site indices of both hemlock and redcedar, and the proportion of contribution of both hemlock and redcedar to stand basal area, were determined for each stand. Each species' site index was then weighted by the proportion of basal area contribution of that species to the stand total, and then summed to give a pooled site index for each stand. Density was measured as number of stems per hectare (SPH), quadratic mean diameter (QDIA), and Curtis' relative density index (1982) (RD): QDIA = ^dbhf [3.8] n RD = — ^ [3.9] Quadratic mean diameter reflects the diameter of trees of average basal area ( B A ) in the stand, and thus contains information on basal area and stems per hectare in the model. Curtis' (1982) 10 relative density index provides a density index independent of stage of stand development, as basal area is divided by the diameter of tree of average basal area in the stand. In addition to age, site quality, and stand density, productivity differences among Capilano, Malco lm Knapp, and Mission could also affect productivity comparisons among stand types. To determine i f productivity comparisons among stand types required correction for location, a two-factor analysis of variance was performed testing stand types, locations, and their interaction. Although mean annual increment was not found to differ significantly among locations, and no significant stand type by location interaction was found, location was inserted as an additional factor in analysis of covariance models in order to reduce model variability. Mean annual increment (MAI) has an intrinsic age correction factor, therefore the analysis of covariance model [3.7] only required the addition of site index (Xyi), density (X^), and location {Xys) covariates. Volume ( V O L ) and basal area ( B A ) are not age corrected, and the analysis of covariance model [3.7] for each productivity measure included the addition of site index (Xyi), age at breast height (Xyi), density (X^), and location (X^) covariates. Each of S P H , Q D I A , and R D was inserted separately in the analysis of covariance for each productivity measure, generating three sets of estimates of stand type means for each model. Least squares means were used for final comparisons of productivity among stand types. Means were compared with multiple, pair-wise t-tests, wherein probability levels were adjusted for the number of pairs examined according to the Bonferroni multiple comparison procedure (Neter et al. 1996). To help explain differences or similarities in productivity among stand types, the individual characteristics of both redcedar and hemlock trees were compared in the single-species stands versus the mixed-species stands. The mean values of dbh, tree height, height to live crown, and mean annual increment of redcedar trees in redcedar stands were compared to 11 those of the redcedar trees in the hemlock-redcedar stands. The same comparison was made among hemlock trees in hemlock and hemlock-redcedar stands. Variances of mean tree characteristics were subjected to normality and variance homogeneity tests, via probability plots and Bartlett's test (Bartlett 1937) ( a = 0.05). Where variances were found to be non-homogenous, differences in mean tree characteristics were assessed with the Wilcoxon Rank Sum test (Venables and Ripley 1999). Where variances were found to be homogenous, differences in mean tree characteristics were determined through analyses of variance. Differences in cumulative distribution functions of each tree characteristic were assessed with the Kolmogorov - Smirnov goodness of fit test (Kaluzny et al. 1998; Venables and Ripley 1999). Alpha levels were held at 0.05 in all cases, unless otherwise stated. 3.3 Structural Analyses Each o f the hemlock-redcedar stands was evaluated to determine i f increasing stand productivity was associated with increasing vertical canopy separation or spatial randomness in tree patterns. The productivity of each hemlock-redcedar stand was assessed by comparing the average mean annual increment of all live hemlock and redcedar trees in each stand. Within each stand, the mean annual increments were calculated for each tree. A n analysis of covariance was performed to determine whether significant differences in mean annual increment per tree existed among the mixed-species stands. Site index (Xyi) and nearest neighbour distance (Xyi) covariates were used to correct for variability in site and density in the analysis of covariance model [3.7]. Least squares means were used for comparisons of mean annual increment per tree among hemlock-redcedar stands. Means were compared with multiple, pair-wise t-tests, wherein probability levels were adjusted for the number of pairs examined according to the Bonferroni 12 multiple comparison procedure (Neter et al. 1996). Stands were ranked according to their means, in descending order of productivity. Canopy stratification, and refined nearest neighbour and combined count and distance (Upton and Fingleton 1985), analyses were conducted on all hemlock-redcedar stands. In the least and most productive stands, the results of each analysis were contrasted and compared. 3.3.1 Canopy stratification Canopy stratification in mixed-species stands is generally the manifestation of the component species having different shade tolerances (Kelty 1992). Stratification can also occur where two species have established at the same time, but have different height growth rates/patterns, or where species establish at different times due to succession (Man and Lieffers 1999). When a canopy is stratified between two species, each species' crowns should occupy different niches within the canopy. These vertical niches can be distinguished from each other by the differences in tree height, and the amount of crown overlap between the two species. The height differences between hemlock and redcedar in each mixture were broken into two subcomponents for analysis: (1) the differences in mean tree height of each species in each stand, and (2) the differences in the cumulative height distribution functions of each species in each stand. Variances of mean tree characteristics were subjected to normality and variance homogeneity tests, via probability plots and Bartlett's test (Bartlett 1937). Where variances were found to be heterogenous, differences in mean tree characteristics were assessed with the Wilcoxon Rank Sum test (Venables and Ripley 1999). Where variances were not found to be heterogenous, differences in mean tree characteristics were determined through analyses of variance. A visual reference of height structure was constructed via height frequency distributions for both hemlock and redcedar in each stand. The cumulative height distribution 13 functions for hemlock and redcedar in each stand were tested for similarity with a Kolmogorov -Smirnov goodness of fit test (Kaluzny et al. 1998; Venables and Ripley 1999). To complete the assessment of canopy stratification, a crown overlap ratio (COR) was calculated for hemlock and redcedar in each mixture: COR = ^; where : D = min (HHw ;HCJ- max (HcHw ; HcCw ) L * J C = max (H H w ;HCw)-min(HcHw ; HcCw) where min and max indicate the minimum and maximum values occurring between the adjacent bracketed terms, HHw and HCw are the average heights of hemlock and redcedar in each mixture, and HcHw and HcCw are the average heights of competitive live crown for hemlock and redcedar in each mixture (Figure 3.1). Figure 3.1 Schematic diagram illustrating parameters used in calculating the stand ratio of crown overlap, where: ' A ' represents the mean crown depth (from the top of the tree to the closest of: 66% of top height, or height of live crown) of hemlock, ' B ' represents the mean crown depth of redcedar, ' C represents the mean, maximum canopy depth of the stand, and ' D ' represents the mean depth of crown overlap between hemlock and redcedar. 14 Competitive live crown for each tree was defined as 66% of the top height, or the height to live-crown, whichever was closer to the top of the tree. Defining the zone of maximum competition as 66% of tree height is in agreement with the range of values (50% - 75%) determined in other studies (Biging and Dobbertin 1995; Latham et al. 1998). 3.3.2 Spatial randomness Reduced competition for light occurs where different species capture light at different intensities and at different positions in the canopy. Canopy stratification, as described in the previous section, is a two-dimensional characteristic and is not sufficient, in and of itself, to prove above ground competitive reduction in mixtures. In addition to the vertical component, the horizontal arrangement of species on the ground surface is of vital importance to maximize light capture. In addition to canopy stratification, the spatial tree pattern of each species must be considered when evaluating aboveground species interactions. A spatial tree pattern is said to be regular when the average distance between trees is greater than would be observed when those trees are arranged in a completely random orientation. Alternately, a clustered spatial pattern is characterized by an average inter-tree distance which is less than would be observed in a completely random pattern. When analyzing the spatial patterns of trees, regularity suggests competition, while clustering suggests facilitation (Upton and Fingleton 1985). Where facilitation does not occur, but competition is reduced, a random pattern intermediate between clustering and regularity is expected. Competitive reduction is maximized with a fine-grained intermixing of species (i.e., trees must be adjacent to trees of a different species) on a two-dimensional plane (Kelty 1992). This type of intermixing is most closely approximated by spatial randomness in the patterns of each species, and the pattern of both species combined, where each species' pattern is independent from that of the other species. A mixture of species exhibiting complete spatial randomness is 15 characterized by: (1) no interactions between the species' individual spatial point patterns, and (2) a constant intensity (number per unit area) of each species over a bounded region (Diggle 1983; Kaluzny etal. 1998). A random distribution of trees in a stand can be likened to a random spatial point pattern. The probability observing r points (trees) per unit area in a stand of area (A) is defined by the formula: P r = ^ ~ (r = 0 ,1 ,2 . . . ) [3.11] where X is the intensity of points (trees) in a unit area. Such a random distribution of trees in a stand is thus referred to as a Poisson process or Poisson forest (Keuls et al. 1963; Upton and Fingleton 1985). Thus in a mixture of two species, each species may individually be arranged according to a Poisson process. However, the combined, bivariate spatial point pattern of both species may or may not appear random. It is therefore important to evaluate the bivariate pattern for any interrelation between the individual spatial point patterns of each species (Upton and Fingleton 1985). The interrelation between each species' spatial point patterns can reveal attraction (clustering), independence, or repulsion (regularity) (Upton and Fingleton 1985), supporting arguments for facilitation, competitive reduction, and competition, respectively. Refined nearest neighbour analysis Within each hemlock-redcedar stand the spatial point patterns of all stems (live and dead) were subjected to refined nearest neighbour analysis (Roder 1975; Diggle 1979; Upton and Fingleton 1985), examining the complete distribution function of nearest neighbour distances as a function of distance. If trees are arranged according to a Poisson process and the mean intensity of stems per unit area is X, then the area of a circle centered on tree i, with radius (dt) 16 extending to the nearest neighbouring tree, is a random variable defined by the probability density function: g(d) = Ikndfi**** [3.12] The cumulative distribution function based on equation [3.12] describes the probability that a nearest neighbour is within distance (d) o f a specified tree (/) in a circular region of area (Moeur 1993): P{dt <d) = G(d) = \-eXnd'1, d>0 [3.13] The corresponding test statistic (Diggle 1979) was also computed for each study stand: dt=sapG(d)-G(d) [3.14] which represents the largest absolute deviation between the theoretical, random distribution G(d) and the estimated distribution G(d). Bias due to edge effects was avoided by calculating the estimated distribution according to the procedure outlined by Upton and Fingleton (1985), which considers only those trees that are closer to their nearest neighbouring trees than they are to any plot boundary: * #(b.>d>d.) #(bi > d) where # indicates the number of trees for which the argument within the adjacent brackets is true, and bt is the distance from tree i to its nearest plot boundary. Plots of G(d) - G(d) over d were made for each stand, providing initial evidence of pattern tendencies. Clustering is 17 indicated where G(d)>G(d), independence is indicated where G(d)=G(d) , and regularity is indicated where G(d) <G(d). Combined count and distance analysis The interrelation between both species' spatial point patterns within each stand was evaluated using Ripley's (1976, 1977) KiJ(d)and LtJ (d)functions. Ripley's functions focus on the inter-tree (point to point) distances on a plane, as do the refined nearest neighbour analyses. However, nearest neighbour methods ignore the majority of spatial relations between points in a plane (Clark and Evans 1954). The Ripley functions consider not only nearest neighbour distances, but all inter-tree distances on the plane. This allows greater insight into the univariate and bivariate spatial point patterns under investigation, at multiple scales (Upton and Fingleton 1985; Moeur 1993; Chen and Bradshaw 1999). If trees within a stand are arranged according to a Poisson distribution, and the spatial point pattern is bivariate (there are two tree species within the stand), then that bivariate pattern can be defined by a set of functions whose estimator is calculated from the combined distributions of distances from type 1 trees to type 2 trees, and vice versa (Lotwick and Silverman 1982; Moeur 1993): v fj\ n2k\2(d) + nxK\\{d) Kn(d) = ; [3.16] nx + n2 where: K n(d) = A ^ ^ ^ Z , md K v(d) = A ^ ^ - ^ - , for i * j i=\ j=\ nx n2 i=x j=x nx n2 where ni and ri2 are the number of type 1 and type 2, for all pairs of trees with dtj <d,Ais the total area, and w,y is an edge effect correction factor (Ripley 1977; Moeur 1993). If trees of type 1 and type 2 are independent, then KX2(d)-nd2. Besag [in the discussion of Ripley (1977)] 18 suggested applying a square root transformation to Kn (d), which linearizes the function, stabilizes its variance, and has an expected value of zero under the Poisson assumption (Moeur 1993; Chen and Bradshaw 1999). The linearized, Ln (d) function was computed as: Ln(d) = [3.17] The significance of the deviations of Kl2 (d) and LX2 (d), from Kn (d) and L n (d) respectively, were determined by comparing the Ripley functions against confidence envelopes generated via Monte Carlo procedures. Monte Carlo methods allowed approximate tests for randomness at any point d along the distribution in question, and allowed confidence envelopes to be generated for statistics whose underlying distributions were not necessarily known (Upton and Fingleton 1985; Moeur 1993). The confidence envelopes (a = 0.05) for each study stand were generated according to the procedures detailed by Moeur (1993), based on the method detailed by Besag and Diggle (1977). Confidence envelopes for each mixture, and each of hemlock and redcedar within each mixture, were derived from 100 simulations of random, spatial point patterns. Simulations were based on the same numbers of stems as present in the actual study stand. A l l spatial analyses were performed with the S-Plus 2000 statistical software package (Mathsoft Inc. 1999) spatial analysis module (Kaluzny et al. 1998). 19 4. POSITIVE PLANT INTERACTIONS - APPLYING CURRENT THEORIES TO MIXED-SPECIES STANDS OF H E M L O C K AND REDCEDAR Abstract Theory on positive plant interactions logically divides the mechanisms of these interactions into two components: competitive reduction, and facilitation. Although naturally occurring mixtures of hemlock and redcedar are very common in the coastal and southern interior wet belt of British Columbia (BC) , the productivity and growth dynamics of these mixtures have been little studied. Theory on the mechanisms of competitive reduction suggest that hemlock and/or redcedar may benefit in mixtures through: (1) below ground physiological separation via the preferential uptake of different forms of nitrogen, and (2) above ground spatial and temporal separation of canopies via differences in growth habit, spatial arrangement, and longevity. Theory on the mechanisms of facilitation suggests that hemlock and redcedar may benefit in mixtures through: (1) substrate stabilization where redcedar enhances the windfirmness of hemlock, and (2) herbivory reduction where each species shields the other from their respective, common defoliators. When applied to the mixed-species study stands, facilitation among hemlock and redcedar is unlikely due to site and stand selection criteria, but these species may experience competitive reduction through: (1) vertical canopy separation and a random spatial pattern of trees, and (2) preferential uptake of different nitrogen forms. 20 4.1 Introduction The study of positive plant interactions has intensified in recent years, providing mounting evidence that these interactions play a major role in the reproduction, distribution, diversity and productivity of plant communities (DeAngelis et al. 1986; Hunter & Aarssen 1988; Grace and Tilman 1990; Wang 1997). The mechanisms of positive plant interactions are logically divided into those of competitive reduction, and facilitation (Vandermeer 1989; Kelty 1992; Callaway 1995), and each of which can be subdivided into several components (Table Table 4.1 The mechanisms of positive plant interactions (adapted from: Callaway 1995; M a n and Lieffers 1999). 4.1). Competitive Reduction Facilitation 1 Reduction of crown competition - physiological separation - spatial separation - temporal separation 1 Resource modification - light and temperature - soil moisture - soil nutrients 2 Reduction of root competition - physiological separation - spatial separation - temporal separation 2 Substrate modification - aeration / soil oxygenation - substrate supply and stabilization 3 Mortality agent attenuation - competitor control - herbivory reduction - amelioration of environmental extremes 4 Additional effects - pollination - propagule concentration Hemlock and redcedar commonly occur in mixtures in the coastal and southern interior wet belt regions of British Columbia, and each species has been the subject of studies investigating 21 various aspects of growth and yield. However, how these species grow together in mixture has been the subject of little study. Applying the theory of positive plant interactions to hemlock-redcedar mixtures may provide useful clues to the growth potential of these mixtures. Thus the objectives of this chapter are to: (1) review current theories on the mechanisms of positive plant interactions, and (2) hypothesize the degree to which these mechanisms may occur generally in hemlock-redcedar mixtures, and specifically in the hemlock-redcedar study stands. 4.2 Competitive Reduction Competitive reduction requires a resource for which competition is reduced to be limiting production (Kelty 1992). When competition is reduced between two species in a mixture relative to what each species would experience in a single-species stand, it can be reduced both above- and belowground. Aboveground competitive reduction is characterized by competition for light, the resource of primary limitation. Belowground competitive reduction is characterized by competition for soil moisture and / or nutrients. These belowground resources may be limiting on a site- and season-specific basis. 4.2.1 Reduced Crown Competition Plants compete above ground primarily for one resource: light. Only a single leaf can occupy a specific area to intercept light, and the amount of space that leaves can fill to intercept light is finite. Competition in this limited space can be minimized i f species can separate themselves to unique niches in the canopy layer. This separation can occur physiologically, physically, or temporally (Table 4.1). Physiological Separation Competition for light has resulted in physiological adaptations among plants, conferring a gradient of light use efficiencies among species. Some plant species have lower compensation 22 points, the point where respiration requirements are met by photosynthesis, than others (Boardman 1977). The lower compensation points of 'shade-tolerant' species result in increased light use efficiency, allowing increased carbon fixation per unit of light (Evens et al. 1988). However, shade-tolerant species also have lower maximum photosynthetic rates and photosynthetic light saturation points than shade-intolerant species (Boardman 1977). Differences in compensation point, saturation point, and light use efficiency allow species of different shade tolerances to reduce their competition for light. Shade-tolerant species can exploit the space in lower canopy strata by tolerating low light conditions, resulting in less direct competition in the main canopy. The combination of a shade-intolerant overstory with a shade-tolerant understory would result in less light competition per canopy stratum than a single-species stand of either component species of equivalent density (Kelty 1992). Spatial Separation The physiological adaptations associated with shade tolerance are complemented with physical adaptations of the crown. Shade-tolerant species w i l l allocate resources among the stem, leaves, and branches, and orient their leaves and branches, differently than shade-intolerant species (Kelty 1992; M a n and Lieffers 1999). Shade-tolerant species allocate more resources to lateral growth than height growth, and have lower juvenile height growth rates than shade-intolerant species (Kelty 1992). Given these growth allocation differences, mixed-species stands of a shade-tolerant and -intolerant species would result in the species occupying different canopy strata. This layered structure w i l l increase stand vertical canopy depth, and increase light use efficiency through a greater volume of canopy space. Shade-tolerant species are able to support greater leaf areas than shade-intolerant species, because their increased efficiency at lower light levels allows more foliage to survive in deeper crown layers, increasing light capture (Assman 1970; Kelty 1992). Shade-adapted foliage is generally oriented along horizontal planes to maximize light interception, enhancing capture of 23 diffuse light filtered through higher canopy strata. This physical orientation augments the physiological characteristics of shade tolerant species to capture diffuse light under the canopy of a shade intolerant species. Competition for light is reduced most efficiently in a fine-grained intermixing (i.e., trees must be adjacent to trees of a different species) of shade-tolerant and -intolerant tree species (Kelty 1992). Temporal Separation Plant species can reduce their competition for light i f they can separate the times when they use the resource. This separation can occur seasonally, through phenological separation, or over longer time scales, through differences in growth strategy and longevity. Evergreen trees are able to photosynthesize both earlier and later in the growing season than deciduous species. Evergreen species exploit the temporal windows in which deciduous species must overcome bud dormancy and flush new foliage, and then sequester nutrients and shed their foliage. This phenological separation results in a longer effective growing season for the more shade-tolerant conifer species that underlie shade-intolerant deciduous species (Man and Lieffers 1999). Shade intolerance is usually indicative of an early serai species' successional role. Early serai species quickly occupy and capture disturbed sites, and have rapid height growth rates in their juvenile life-stage (Oliver and Larson 1996). Late serai species usually establish themselves after a site has been occupied by species of earlier successional roles, and have the physiological and physical attributes characteristic of shade tolerance. Late serai species usually also have increased longevity relative to early serai species, assuming dominance in late serai stages of stand development (Oliver and Larson 1996; M a n and Lieffers 1999). 24 Incidence of Reduced Crown Competition in Hemlock-Redcedar Stands Aboveground competitive reduction in hemlock-redcedar stands may occur due to physical canopy separation. Competition for light is reduced most efficiently in a mixture of species with the highest degree of niche separation for the light resource, typically that of a shade-tolerant and -intolerant species. Hemlock and redcedar are both shade-tolerant species, although there have been arguments that hemlock is slightly more shade-tolerant than redcedar (Minore 1979, 1990). More recently, redcedar has been found to be a very shade-tolerant species, as much so as hemlock or Pacific silver fir (Abies amabilis (Dougl. E x Loud.) Dougl. ex J. Forbes) (Wang et al. 1994). However, hemlock has a different height growth pattern than redcedar, overtopping redcedar by age 15 (Gashwiler 1970; Curran and Dunsworth 1988). Therefore, even-aged hemlock-redcedar stands in the earlier stages of stand development are likely to be mixtures in which the crowns o f each species are vertically separated. This may reduce competition for light between hemlock and redcedar relative to what either species would experience in equally dense single-species stands. Wherever vertical canopy separation occurs in the hemlock-redcedar study stands, it may be augmented by the relative performance of hemlock and redcedar with respect to climate. On intermediate sites the mean site index of redcedar has been observed to decrease by 5 m with progression in elevation from the Dry Maritime Coastal Western Hemlock (CWHdm) subzone to the Very Wet Maritime C W H (CWHvm) subzone (Green and Kl inka 1994; Kayahara et al. 1997). Given these findings, the location of all study stands within the C W H v m subzone would give dominant hemlock a 5 m height growth advantage over redcedar, enhancing canopy stratification in even-aged, hemlock-redcedar mixtures. However, in addition to redcedar's site index decreasing with progression from the C W H d m to the C W H v m zone, hemlock's site index has also been observed to decrease by the same amount (Site Productivity Working Group 1997), resulting in no net change in site index differential between the two species. Both studies 25 suggest the site index of dominant hemlock, on intermediate sites in the C W H v m , is 4 m greater than that of redcedar. This supports the assertion that these species w i l l have vertically separated canopies in the mixed-species study stands. Despite the species' similar shade tolerances, competition among hemlock and redcedar may be reduced in these mixtures. The degree to which this stratification is combined with a random spatial pattern of trees w i l l be assessed in Chapter 6, but a random pattern is assumed to further reduce competition for light among hemlock and redcedar in these stands. Redcedar shoots have a longer growth period than any associated conifer species (Minore 1990), suggesting a longer photosynthetically active period than its associates. This longer photosynthetically active period could give redcedar a longer growing season relative to hemlock, but the relative difference in growth periods between the two species may be small enough to make redcedar's advantage negligible. Considering that hemlock and redcedar are both very shade-tolerant species (Minore 1990; Packee 1990), suggesting similar specific leaf area capacities (Kayahara et al. 1997), phenological separation of these species is unlikely. Both hemlock and redcedar are able to occupy different successional roles. Both species can act as early serai species, but their shade tolerance, longevity, and regeneration characteristics allow them to act as climax species. However, the lifespan for hemlock is 400 to 500 years (Packee 1990), whereas lifespan for redcedar is 800 to over 1000 years (Minore 1990). This difference in longevity between hemlock and redcedar suggests that temporal separation between the species' canopies occurs on a time scale of centuries. Hemlock w i l l overtop redcedar, assuming no establishment lag, early canopy dominance, and optimal growth rates. Redcedar has the physiological mechanisms to persist beneath and within the hemlock canopy for the first 300-500 years. Dominant and co-dominant hemlock w i l l then be replaced by redcedar, which w i l l assume dominance and co-dominance for the next 3-5 centuries. Thus, by intermixing hemlock, which dominates light resources relatively earlier, with redcedar, which 26 dominates light resources relatively later, the light competition each species experiences over 800-1000 years would be less than each would experience i f grown in similar density single-species stands. This type of temporal stratification is impossible to assess in the study stands, given the brief amount of time since establishment. 4.2.2 Reduced Root Competition Soil moisture and nutrients are the belowground resources that may limit plant growth, and thus are the resources for which competition may be reduced. The extents to which these resources are limiting are site and species specific, and depend on such factors as climate, topography, soils, and the species in question. Potential mechanisms for belowground competitive reduction can be grouped similarly to those for aboveground competition: physiologically, spatially, and temporally (Table 4.1). Physiological Separation Species growing in mixture can reduce competition for nutrients by preferentially using different forms of those nutrients through physiological adaptations (Kelty 1992). Nitrogen, often the most often limiting nutrient in north temperate forests, is available to plants in both nitrate and ammonium forms. Certain species take up different proportions of these nitrogen forms (Waring and Schlesinger 1985). If a mixture is made up of two species that are adapted to uptake nitrogen of different forms, those species are physiologically separated. In such a physiologically separated mixture, the nitrogen competition each species experiences would be less than each would experience in a single-species stand of similar density. Spatial Separation Water and nutrients can also be partitioned between competing species, i f those species occupy different rooting depths, and/or have different rooting structures (Man and Lieffers 27 1999). Species with deeper rooting habits have access to moisture and nutrients that shallow-rooted species do not have. A mixture of species with similar rooting habit w i l l pack all roots into the same strata, resulting in intense competition. M i x i n g shallow- and deep-rooting species w i l l distribute the same root mass over a greater depth, and thus into a larger volume of soil. A larger volume of soil is likely to have greater moisture and nutrient pools available for growth. Thus, root stratification not only pulls the roots of different species apart, reducing direct competition, but also may provide the mixture with more moisture and nutrients due to increased soil volume, providing these resources exist in the lower soil layer. These beneficial effects of root stratification are likely to be enhanced by a fine-grained distribution of trees of the component species (Kelty 1992). Temporal Separation Species that separate their individual growth into different time frames, reduce belowground and abovegroiind competition simultaneously. Evergreen species exploit the temporal windows in which deciduous species are not photosynthesizing. Deciduous species must overcome bud dormancy and flush new foliage in spring, and then sequester nutrients and shed their foliage in autumn. These windows of opportunity allow evergreen species to uptake soil moisture and nutrients with less competition from deciduous species than would be encountered in the middle of the growing season (Kelty 1992). Species with similar phenologies w i l l compete directly for moisture and nutrients over the same time frame. Species with different phenologies reduce the overlap of direct competition, and spread moisture and nutrient demands over a longer portion of the growing season. This w i l l result in less nutrient demand per unit time in a mixture of two species than either species would experience in single-species stands of the same density. The beneficial effects of phenological separation may be enhanced by canopy stratification in mixtures of shade-tolerant and -intolerant species (Kelty 1989; 1992). 28 The physiological differences between early and late serai species may be associated with different soil moisture and nutrient demands. Early serai species generally have high evapotranspiration and nutrient cycling rates, a relatively short lifespan, and establish themselves at the onset of stand development. Typically, late serai species have relatively lower evapotranspiration and nutrient cycling rates, and relatively longer lifespans. Each site has a finite resource pool, capable of supporting a maximum density of a species. If that species is early serai, full utilization of the limiting resource is likely to be rapid and of short duration. Alternately, a late serai species is likely to take longer to maximize utilization of the limiting resource, and maintain the utilization for a longer time. Species of similar successional roles w i l l experience high levels of competition due to similar growth patterns and longevities. If early and late serai species are mixed, resources w i l l be fully utilized more rapidly and for a longer duration than could be realized in monocultures of each component species. Early and late serai species reduce competition by maximizing the demand for limiting resources at different times. Early serai species are commonly deciduous and shade-intolerant, while late serai species are typically evergreen and shade-tolerant. The associated differences in physiology, structure, and phenology may further reduce competition in mixtures of species with different successional roles. Incidence of Reduced Root Competition in Hemlock-Redcedar Stands Redcedar is known to uptake high concentrations of nitrate relative to hemlock (Pharis et al. 1964; Krajina et al. 1973). Hemlock tolerates high soil concentrations of ammonium better than redcedar, does not have high biomass nitrate concentrations, and tends not to nitrify ammonium, even in conditions with sufficient ammonium and phosphorus (Turner and Franz 1985). These findings suggest that redcedar preferentially uptakes nitrate, while hemlock preferentially uptakes ammonium. This physiological separation reduces competition for 29 nitrogen, commonly the most limiting nutrient for growth in the Pacific Northwest (Weetman 1962; Heilman 1979; Pritchet and Fisher 1987), among hemlock and redcedar. These findings suggest that this mechanism of competitive reduction may be acting in the hemlock-redcedar study stands. Hemlock is a shallow-rooted species, with a concentration of fine roots very near the surface (Packee 1990). Redcedar is a deeper-rooted species that has an extensive root system, with a poorly defined taproots and a dense root network of fine roots near the surface (Minore 1979, 1990). In two stands of redcedar, western white pine (Pinus monticola Dougl. ex D . Don), western larch (Larix occidentalis Nutt.), and Douglas fir in northern Idaho, redcedar contributed only 17% of basal area, but accounted for 82% of total root length contribution (Leaphart and Grismere 1974). However, differences in rooting habit between hemlock and redcedar may be site specific, as no significant differences between the species' root area, length or depth could be found when grown on substrates with similar physical characteristics (Eis 1973, 1987). Since redcedar and hemlock are shade-tolerant species, and both retain considerable amounts of foliage through winter, phenological separation of below ground competition is unlikely. Differences in the lifespan of the species indicate that competition may be reduced due to successional separation. Although this mechanism of competitive reduction may occur in hemlock-redcedar mixtures of the appropriate age, the age of the study stands dictates that this mechanism does not contribute to positive plant interactions in these stands. 4.3 Facilitation Facilitation in mixed-species stands implies that: (1) one species requires the other to establish or grow, or (2) a synergy exists between the component species. In either case, facilitation in a mixture results in greater productivity than could be realized by each component species in monoculture. Callaway (1995) provides a comprehensive review on positive 30 interactions among plants, and his categorization of facilitative mechanisms has been adapted to provide the framework for the following discussion. 4.3.1 Resource Modification Through its establishment and growth, a plant interacts constantly with the environment. Part of this interaction involves some modification of the environment by the plant. Resource modification by a benefactor species may: (1) facilitate growth of a beneficiary species, where the environment may not have been initially suitable for the beneficiary, or (2) may improve the growth of a beneficiary beyond that which it could achieve in monoculture (Callaway 1995). Light and Temperature The presence of an overstory canopy w i l l impact the amount and quality of light available to plants in lower strata. Plants perform better as the amount of available light increases. However, shade can have many beneficial effects on plants. Shade reduces water loss, photoinhibition during stomatal closure, and temperature extremes, even though it also reduces the energy available for photosynthesis (Callaway 1995). Those plants adapted to the low energy environment in the shade, especially those in arid climates, may be facilitated by the ancillary benefits associated with shade. Trees on high-elevation sites modify the microclimate around them, often to the benefit of understory species in close proximity. Larger stems absorb short- and longwave electromagnetic radiation, and then re-radiate this energy in their immediate vicinity (Oke 1987). This phenomenon creates snow craters, in which understory species are facilitated due to a longer effective growing season (Brooke et al. 1970). 31 Soil Moisture Plants species commonly compete for soil moisture, however there are ways in which one species can improve moisture conditions for another. Soil improvement through organic matter contributions and hydraulic lift, are mechanisms in which benefactor plants may improve soil moisture conditions for beneficiaries. Those benefactor plants able to establish on mineral soil, and their organic inputs to that soil, would facilitate species favoring organic seedbeds. Organic matter accumulations are known to improve soil structure, and increase porosity, improving the soil's moisture holding capacity over time (Brady and W e i l 1996). Plants with a deep rooting habit create water potential at depth, hydraulically lifting tapped water through their root systems into shallower soil layers, improving water relations for shallow-rooted species (Caldwell and Richards 1989; Dawson 1993). Trees on high-elevation sites tend to grow in irregularly grouped cohorts, and so modify soil moisture conditions by creating 'tree islands' or 'ribbon forests'. The modified microclimate around tree stems creates snow craters (Brooke et al. 1970). Drifting of snow in the lee of these trees then creates store of soil moisture enhancing the growth of the species established in the craters (Billings 1969). This sequence allows the gradual establishment of tree species in a ribbon of ameliorated conditions in the lee of the benefactor tree. Soil Nutrients Benefactor species may improve soil nutrient availability for other plants through soil organic matter contributions, canopy drip, and symbiotic organisms. Some of the nutrients that plants extract from the soil are incorporated into leaf tissues for maintenance and growth. Varying amounts of these leaf-stored nutrients are sequestered in the plant at senescence, but the remaining nutrients are returned to the forest floor in litter. Soil nutrient capital can be increased, given appropriate amounts and composition of litter. Sub-canopy nutrient enrichment, often with subsequent species changes and increased productivity, has been widely reported, especially in 32 savanna habitats (Callaway 1995). Hardwood-softwood mixtures have exhibited increased nutrient cycling rates, relative to softwood stands, due to the relatively higher nutrient content and lower levels of compounds which inhibit decomposition in hardwood litter (Gordon 1983; Kelty and Cameron 1995). Leaves can have concentrations of minerals and nutrients on their surfaces, from the leaves themselves or from airborne particles intercepted by the leaves. Precipitation washes these minerals and nutrients into solution, and returns them to the soil as canopy drip and stem flow. Nutrient-rich canopy drip is more common among deciduous hardwoods. For example, bigleaf maple (Acer macrophyllum Pursh) leaves contain high concentrations o f potassium, calcium, and other nutrients (Tarrant et al. 1951), and the species improves the nutrient status of the sites on which it grows (Minore and Zasada 1990). Plant symbioses with nitrogen-fixing organisms often result in soil enrichment, as these species shed leaves with high nitrogen concentrations. These symbioses act as natural fertilizers, extracting and dispersing atmospheric N 2 onto the soil in readily mineralizable forms. Soil nitrogen content has been observed to increase tenfold with the introduction of a nitrogen fixer to a bare soil (Lawrence et al. 1977), and facilitative effects of nitrogen fixers on secondary species have been observed (Walker and Chapin 1986). It is common for particular plant species to be associated with particular mycorrhizal and bacterial communities that enhance nutrient uptake. Beneficiary species can indirectly profit from benefactor symbioses by reaping the benefits of the benefactor's nutrient-rich litter. Beneficiary species can directly profit from the mycorrhizal and bacterial symbioses of a benefactor, where the beneficiary is able to exploit direct symbioses with the fungi and/or bacteria introduced by the benefactor. Such linkages have been observed to allow bi-directional, inter-species transfers of water, nutrients, and photosynthate among trees (Simard et al. 1997). Studies have also been conducted which suggest the possibility of inter- and intra-specific root 33 grafting, resulting in the inter-tree passage of water, nutrients and photosynthate (Bormann and Graham 1959; Bormann 1966). When hydraulic lift occurs in a mixture, 'nutrient pumping' often occurs as a corollary mechanism, whereby deeply-rooted species uptake nutrients and make them available to shallow-rooted species via litterfall and canopy drip (Richards and Caldwell 1987). Incidence of Resource Modification in Hemlock-Redcedar Stands Hemlock and redcedar are similar in shade tolerance, and neither species thrives in conditions of high insolation and temperature (Minore 1990; Packee 1990). Neither species is better suited than the other to act as a shade-providing benefactor where shade protection is required. Therefore, light modification is an unlikely facilitator between these species. Hemlock and redcedar are both able to regenerate on mineral and organic seedbeds (Morris 1970; Parker 1986; Minore 1990; Packee 1990; Feller and Kl inka 1998). Since both species are able to regenerate on mineral soil, and under their own canopy, neither hemlock nor redcedar regeneration is facilitated by organic contributions of the other species. Studies have suggested that redcedar has a deeper rooting habit than hemlock (Minore 1979). However, these differences may be influenced more by differences in rooting substrate than by genetic expression (Eis 1973,1987), eliminating hydraulic lift as a facilitative mechanism among hemlock and redcedar. Since hemlock and redcedar presence decreases with increasing elevation, and neither species is commonly found in the subalpine parkland (Minore 1990; Packee 1990), the mechanisms of facilitation associated with areas of heavy, persistent snowpack would not occur in hemlock-redcedar mixtures. The chemical composition of hemlock litterfall is distinctly different from that of redcedar, resulting in different forest floor and mineral soil nutrient properties beneath each species (Alban 1967; 1969). Forest floors influenced by redcedar litter have higher p H , base 34 saturation, and exchangeable calcium, than forest floors influenced by hemlock litter (Daubenmire 1953; Ovington and Madgewick 1957; Alban 1967, 1969; Turner and Franz 1985; Weetman et al. 1988, Collins et al. in review). Hemlock litterfall acidifies underlying forest floor and mineral soil, promoting the development of fungal populations over bacterial populations (Turner and Franz 1985). The increases in p H , soil microbial communities, and exchangeable base cations in forest floors influenced by redcedar litter, suggest that redcedar may facilitate hemlock by creating improved nutrient conditions. This assumption is uncertain, given that hemlock grows best where nutrients are supplied in small quantities, and tolerates any nutrient deficiency better than other conifers (Kl inka et al. 2000). Additionally, redcedar does lower forest floor N concentrations due to a high nitrogen resorption rate and nitrogen use efficiency (Keenan et al. 1995), and its seedlings have been observed to survive and grow better than most competitors on nitrogen poor sites, promoting further decreases in soil nitrogen content. These findings suggest that redcedar's influence on forest floor nutrient properties is linked to a competitive advantage for its progeny (Tilman 1988; Weetman et al. 1990; Messier 1993; Keenan et al. 1995). However, hemlock also establishes and grows well on nitrogen poor sites, making hemlock uniquely suited, relative to other conifers, to grow on sites where redcedar has depleted the nitrogen capital. Given the uncertainty as to whether or not hemlock growth may be facilitated by redcedar increasing forest floor p H , microbial populations, and base cation concentrations, this type of facilitation cannot be proven. Neither hemlock nor redcedar share symbioses with nitrogen-fixing organisms, however both species do benefit from mycorrhizae. Hemlock is associated with ectomycorrhiza, while redcedar is associated with endomycorrhiza (Rygiewicz et al. 1984a and b). Since hemlock and redcedar profit from different mycorrhizal associations, interspecies linkages and resource exchanges through mycorrhizae are unlikely. 35 4.3.2 Substrate Modification Plants and soils continually affect each other, within a feedback mechanism governed by factors like climate, topography, and disturbance regime. The physical presence of a plant, and its influence on the soil, may be able to modify the substrate enough to facilitate the growth of beneficiary plant species. Aeration/Soil Oxygenation Contribution of organic materials to the soil, from litterfall, ephemeral roots, and root exudates, enhances the suitability of the soil for plant growth. Organic matter contributions decrease soil bulk density, improve soil structure, and increase soil porosity and aeration, resulting in the colonization and growth of new species (Callaway et al. 1991; Joffre and Rambal 1993). Evapotranspiration may also reduce soil moisture levels and increase soil oxygenation near the ground surface, allowing the colonization and growth of beneficiary species (Berendse and Aerts 1984). Substrate Supply and Stabilization Plants can serve as substrate traps and stabilizers in exposed environments, decreasing wind (and water) velocities and thus promoting the deposition of substrate in the lee of the plant (Oke 1987). The shelter and additional substrate deposited may then allow the colonization and growth of plants that could not have done so without the contributions of the benefactor species (Vasek and Lund 1980). Epiphytic plants utilize the stems of live trees as a substrate, and exploit light unavailable to plants in the understory. Many epiphytes appear to be obligatorily dependent on their benefactors for access to light (Callaway 1995). 36 Certain species continue as benefactors after death. Decaying logs provide a favorable substrate for other tree species, allowing seedlings to grow on an elevated platform free from competition for light resources with shrubs and herbs (Harmon and Franklin 1989). Incidence of Substrate Modification in Hemlock-Redcedar Stands Since both hemlock and redcedar regenerate on mineral seedbeds (Minore 1990; Packee 1990), neither species requires the other's inputs of organic matter to modify soil aeration for establishment and growth. In established stands, hemlock and redcedar regenerate on decaying wood (Morris 1970; Parker 1986). Redcedar veglings, from layering, rooting of fallen branches, and branch development on fallen trees, may be more abundant than seedlings in mature stands (Parker 1979). Both species benefit by regenerating on decaying wood, but neither species specifically requires decaying wood of the other for regeneration. 4.3.3 Mortality Agent Attenuation A benefactor species can reduce the likelihood of mortality for a beneficiary by: (1) acting as a buffer between the beneficiary and its direct, vigorous competitors, (2) shielding the beneficiary from herbivory, and (3) ameliorating environmental extremes. Competitor Control Competition among plant species can vary in intensity, depending on the relative competitive strengths and strategies of the species involved. A direct clash of similar competitive strategies w i l l result in the stronger competitor succeeding at the expense of the weaker. A pairing of dissimilar competitive strategies may allow for the growth of both species on a site. These species groupings for which competition is reduced may have an additional, facilitative effect, where the presence of one species inhibits the invasion of a direct, detrimental 37 competitor. In boreal forests, overstory hardwood species are believed to inhibit the growth of highly competitive understory species, allowing the establishment of more shade-tolerant conifers (Man and Lieffers 1999). Herbivory Reduction One plant can directly or indirectly reduce herbivory of another. Benefactor species can provide herbivory protection both chemically and physically (Vandermeer 1989). Palatable species are able to gain chemical protection against herbivory by intermixing, or surrounding themselves, with plants that produce unpalatable secondary compounds (McNaughton 1978; Holmes and Jepson-Innes 1989). The foliage and odours of a benefactor can camouflage beneficiary species, masking the sight and scent of the palatable species to herbivores. Increased complexity and species diversity have been found to decrease herbivore search efficiency, resulting in decreased herbivory. Such visually complex mixtures are known as 'associational plant refuges' (Pfister and Hay 1988). This concept has further been observed with respect to the dilution of host odors by the presence of other, non-host species (May and Ahmad 1983; MacLean 1996). Thorns, spines and prickles are other physical adaptations by which armoured benefactor species can shield palatable plants from attack (Callaway 1995). Amelioration of Environmental Extremes Temperature increases associated with increased light levels can be detrimental to plant growth, and the negative effects are most pronounced in hot, arid environments. Seedlings are particularly susceptible to the temperature regime of the ground surface. The shade provided by a benefactor species can reduce temperature and exposure levels to those tolerated by the beneficiary (Callaway 1995). Blue oak (Quercus douglasii Hook. & Arn.) survivorship of 35% has been observed in the shade, compared to 0% in the open, after one year (Callaway 1992). In 38 certain environments, shading can also result in decreased soil salinity, due to less evaporative soil moisture losses (Callaway 1994). The presence of a plant species can also improve the microclimate in cold environments. Plants decrease wind velocities, creating an insulating, laminar boundary layer of air above the plant (Oke 1987). Plants also increase the amounts of short and long wave energy absorbed and emitted (Oke 1987), increasing temperatures and reducing snowpack duration in the immediate vicinity of the plant. Therefore, benefactor species tolerant of extreme cold and desiccation can modify the microenvironment sufficiently to allow the establishment and growth of beneficiary species. A deeper or more extensively rooted plant can serve to trap and stabilize the substrate for a less stable associate. Substrate stabilization can reduce mortality where wind velocities are sufficient to cause windfhrow (Kerr et al. 1992). Root grafts among species of different rooting habit have been observed to be more common in wetland than upland areas, suggesting stabilization benefits in wetland soils (Keeley 1988). Gusting winds can cause abrasion damage of tree crowns. Mixtures with a stratified canopy, like those of shade-tolerant and shade-intolerant species, serve to separate tree crowns and may be able to reduce crown abrasion (Man and Lieffers 1999). Incidence of Mortality Agent Attenuation in Hemlock-Redcedar Stands It is unlikely that hemlock and redcedar in mixture displace more vigorous competitors for each other, facilitating growth. Wi th similar shade tolerances (Minore 1990; Packee 1990; K l inka et al. 2000), and initial growth rates less than those of most shrub-hazard species (Curran and Dunsworth 1988), neither hemlock nor redcedar would be better suited than the other at competing with understory vegetation. Thus neither species would preferentially adopt the role of benefactor, displacing competing species and providing a favourable environment for regeneration. 39 Insect pests commonly attack hemlock and most of these pests are defoliators, such as the western blackheaded budworm (Acleris glover and), western hemlock looper (Lambdina fiscellaria lugubrosa), and the hemlock sawfly (Neodiprion tsugae) (Packee 1990). Hemlock is browsed to a lesser extent by elk, deer and hares. Alternately, redcedar suffers little damage from insects (Minore 1990), but seedlings and saplings are so often severely browsed by elk, deer and rodents that browse damage may be the most important stand establishment problem (Curran and Dunsworth 1988). Since the pests that attack hemlock and redcedar are different, herbivory reduction may be possible when the species are intermixed. The benefits of mutual camouflage and decreased herbivore/insect search efficiencies are those most likely to occur in hemlock-redcedar mixtures, as neither species produces armor or secondary compounds to combat herbivory. Unfortunately, no evidence of this exists in the study stands, as stands were selected to be free from disease or obvious herbivory damage. High temperatures and drought can damage both hemlock and redcedar seedlings, and redcedar seedlings are more susceptible to high and low temperatures than most of its common associates (Minore 1990). In areas of high insolation, seedlings of both species would benefit from shade, and are protection-requiring in this lifestage. However, neither hemlock nor redcedar is better adapted than the other to grow in environments with high temperature extremes (Minore 1990; Packee 1990). Thus neither species is suited to act as a regular benefactor to the other in these environments. Absolute minimum temperatures tolerated by hemlock are -39°C for the coast and -48°C for the interior (Packee 1990), while those tolerated by redcedar are -30°C and -47°C respectively (Krajina 1969). Both species tolerate similar cold extremes, leaving no clear role separation between benefactor and beneficiary, making this type of facilitation in hemlock -redcedar mixtures unlikely. 40 Hemlock and redcedar are not known to form root grafts with each other, and evidence is weak that redcedar has a deeper rooting habit. Hemlock is subject to severe windthrow on many exposed sites, while redcedar is relatively windfirm except on wet sites (Minore 1990; Packee 1990; K l i n k a et al. 2000). This suggests that on sites with a high windthrow hazard, redcedar may protect hemlock from windthrow. Therefore a hemlock-redcedar mixture may facilitate hemlock growth on sites subject to wind disturbance. N o support for this mechanism of facilitation can be established in the study stands, since the stands were selected to be free from damage such as that caused by windthrow. Redcedar is commonly overtopped by hemlock and then grows slowly in the understory (Minore 1990). This tends to stratify hemlock and redcedar into separate canopy positions, and increases the distance between tree crowns in each canopy layer. Such an arrangement could decrease canopy abrasion in windy conditions, facilitating both species in hemlock-redcedar stands. However, no evidence of this was found in the hemlock-redcedar study stands. 4.3.4 Additional Effects Pollination and Propagule Concentration Certain plants are relatively more attractive to pollinators than others, due to adaptations in flowering, coloration, and pheromone production. Greater pollination rates have been observed for less attractive species, when intermixed with attractive, co-flowering species (Dafni 1984). These benefactor plants are known as 'magnet species' (Laverty 1992). When plants concentrate the propagules of others, the result may simply be altered spatial distribution patterns. Propagule concentration can also have a facilitative effect, where the propagules are prevented from being lost to the system or destroyed (Callaway 1995). This phenomenon may act in concert with other facilitating mechanisms. Seeds concentrated at the 41 base of a benefactor species may receive the additional benefits of resource and substrate modification, and decreased impacts of mortality agents. Incidence of Additional Facilitating Effects in Hemlock-Redcedar Stands Hemlock and redcedar rely upon wind for pollination, and pollination is not subject to preferential visitation by insect vectors. Both species are prolific seed producers, with hemlock producing 19.8 mil l ion seeds per hectare from a 100 year-old coastal stand in a good year (Packee 1990), and redcedar producing 2.47 mil l ion seeds per hectare in coastal stands (Minore 1990). It is unlikely that either species is facilitated by having their seeds concentrated by the other, given the massive quantities of seeds produced. 4.4 The Dynamics and Implications of Positive Interactions among Plants Identifying competitive reduction as a driver for enhanced productivity in mixtures implies that competition occurs simultaneously with positive plant interactions. Therefore, the overall effect of one species upon another is likely the cumulative effect of multiple, complex interactions (Callaway 1995), involving the balance of positive plant interactions and competitive interactions. Ell ison and Houston (1958) demonstrated this balance in a study of understory herbs. The growth of open-grown herbs was compared that in a stand of aspen, and that in a stand of aspen in which aspen root competition was reduced by trenching. The herbs in the untrenched aspen stands grew the least, the open-grown herbs had intermediate growth, and the herbs in the trenched aspen stands grew the most. The two most prominent factors affecting the balance between facilitation and competition appear to be the life stage of the interacting species, and the severity of environmental extremes the species are subjected to (Callaway 1995). Facilitation commonly occurs in mixtures where the species are in different life stages, such as seedling establishment in the shade of a mature tree. Those benefactor species that enable the colonization and growth of a 42 beneficiary are often later overtopped and outlived by the species being nursed (Archer and Rykiel 1994). The balance shifts from facilitation to competition, with the shift in each species' life stage dynamics, often resulting in succession of the beneficiary. A s abiotic stress and consumer pressure increase, facilitation may become more important in plant communities, while competition gains importance as physical stress and consumer pressure decrease (Bertness and Callaway 1994). Callaway and Bertness (1994) examined relationships between whitebark pine (Pinus albicaulis Engelm.) and subalpine fir (Abies lasiocarpa (Hook.) Nutt.) in sheltered, low-stress habitats, and in exposed, high-stress habitats. In low-stress habitats, fir was randomly distributed with respect to pine, and fir growth improved as pine died. In high-stress habitats, fir was clumped around the pine, and fir growth decreased as the pine died. Facilitation appeared to be more important in the high-stress areas, while competition seemed to dominate in the low-stress areas. Similar relationships were found among marsh species, where increasing salinity was the stressor promoting facilitation (Bertness and Y e h 1994). Positive interactions may partly determine community spatial patterns, permit coexistence, enhance species and structural diversity, enhance productivity, and drive community dynamics (Callaway 1995). Benefactor plants may also provide regeneration niches for beneficiary species (Grubb 1977), allowing the beneficiary to incrementally creep into previously uncolonized areas. Therefore, the boundaries between plant communities may shift across the landscape, suggesting that facilitative processes drive dynamic change in the absence of large-scale disturbance (Callaway 1995). The early stages of primary and post-glacial succession are environmentally extreme, causing stressful conditions for plants. Facilitative mechanisms may then be more important during these successional stages than during secondary succession (Connell and Slatyer 1977; Chapin et al. 1994). 43 4.5 Methods of comparing productivity of single- and mixed-species stands Productivity studies of mixed-species stands have been based upon either additive or substitutive experimental designs (Harper 1977). Additive experiments compare multiple plots that contain a single species at a fixed density, and incrementally increasing densities of a second species. Substitutive, or replacement series, experiments keep total density the same for all plots, but alter the proportion of constituent species. Substitutive designs are more common than the additive experiments, as they allow yield comparisons among each species in mixture with its yield in monoculture (Harper 1977; Vandermeer 1989). These comparisons detect i f two species are relatively more or less productive in mixture than in single-species stands, supporting arguments for positive or negative interactions among those species. If both species use resources in mixture identically to that in monoculture, the R Y of each should be 0.5 and the R Y T should be 1.0. Where competition is reduced or species are facilitated, R Y T values greater than 1.0 are expected. Where the species compete, R Y T values of less than 1.0 are expected. However, it is necessary to compare absolute, and not relative, yield to identify the highest yielding stand type in these experiments (Kelty 1992). If one of the component species is much more productive than the other in monoculture, the R Y T of the mixture can be greater than 1.0, but not exceed the absolute yield of the more productive species in monoculture. The less productive species effectively dilutes the productivity potential of the more productive species (Figure 4.1). This characteristic of R Y comparisons means that both relative yields and absolute yields are necessary for meaningful productivity comparisons (Kelty 1992). 44 'TO 2 CD 0.0 Species A (%) o 25 50 75 100 Species B (%) 100 75 50 25 0 Species A Species B Combined 0 100 25 75 50 50 75 25 100 0 Figure 4.1 (From Kelty 1992) Results from hypothetical replacement series experiment, expressed as relative and absolute yield. Dashed lines represent expected yield i f intra- and inter-specific interactions were equivalent. Solid lines represent experimental yields. Comparisons based on relative land output (RLO) also compare yields based on replacement series experiments, but involve equal populations and equal areas allocated to mixture and monocultures (Jolliffe 1997) (Figure 4.1). Mixture a R L O b • • • • • • • • • • • • Monocultures c R Y T d • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • Figure 4.2 (From Jolliffe 1997) Example plots in which the same area is planted in a (a) 2:1 mixture of two species, (b) 2:1 ratio of pure sub-plots, (c) pure stand of species 1, and (d) pure stand of species 2. 45 Higher yields may not necessarily be realized in mixed-species stands where competitive reduction and facilitation are expected to occur. Positive interactions can be muted by inappropriate species proportions, densities, spatial arrangement of individual trees, or site conditions where resource limitations are not reduced by the species chosen (Kelty 1992; M a n and Lieffers 1999). A review of yield studies suggests that it is uncommon to find mixtures significantly more productive than the highest yielding constituent species in monoculture (Trenbath 1974). However, further analysis of these studies revealed that species in the mixtures were not necessarily chosen for their niche separation (Harper 1977). A recent review of yield studies using R L O comparisons, found the average productivity of mixtures to be 12% higher than that of monocultures based on 202 direct observations, and 13% higher than that of monocultures based on 604 yield - density relationships (Jolliffe 1997). Comparisons in replacement series experiments require a fixed planting density, site quality, and time frame (Kelty 1992), requiring controlled plantations. Such plantation experiments are limited by the number of species combinations possible, and the time span required (the lifespan of the species involved). Therefore, it is useful to augment plantation experiments with empirical studies involving existing single-and mixed-species stands. However, natural variation in sites (climate and soils) and stands makes the empirical studies challenging, but not impossible. There are opportunities among some species to locate an adequate number of acceptable study stands on ecologically-equivalent sites, given the biogeoclimatic ecosystem classification in B C , and the variety of existing, naturally regenerated, unmanaged, second growth stands on a variety of sites. This study is an example of such an opportunity. 46 4.6 Summary and Conclusions Positive plant interactions are those where competition between species is reduced, and/or the presence of one species facilitates the growth of another. Competition between species can be reduced both above- and below-ground, where the species exploit relative differences in physiology, physical attributes, phenology, or successional roles. A plant species can facilitate the growth of another by creating favourable substrate and nutrient conditions, decreasing exposure to mortality agents, and increasing visitations by pollinators. Theory on the mechanisms of competitive reduction and facilitation suggest that hemlock and redcedar in mixture may experience positive plant interactions through: (1) vertical canopy separation and a random spatial pattern of trees, (2) longevity differences, which spread and separate resource demands in time, (3) preferential uptake of different forms of nitrogen, (4) reducing herbivore search efficiencies, and (5) increased windfirmness. Based on stand selection criteria from the hemlock-redcedar study stands, evidence of longevity differences, herbivory reduction, and increased windfirmness is unexpected in the study stands. However, both species in the study stands may experience positive interactions through: (1) vertical canopy separation and a random spatial pattern of trees, and (2) preferential uptake of different nitrogen forms. 47 5. THE PRODUCTIVITY OF SINGLE- AND MIXED-SPECIES STANDS OF H E M L O C K AND REDCEDAR Abstract I compared the productivity of western hemlock and western redcedar growing in even-aged, unmanaged, second-growth, single- and mixed-species stands on intermediate (fresh and nutrient-medium) sites, within the perhumid mesothermal climate of southern coastal British Columbia. The relative yield of hemlock and redcedar was lower in mixture, as compared to single-species stands, due to the effects of competition with each other and non-study species. Mean annual increment and wood volume production increased with increasing presence of hemlock. Basal area increased with increasing presence of redcedar, and the redcedar stand type was also the densest. Redcedar trees in the redcedar stand type were similar in height, diameter, and mean annual increment, relative to those in the hemlock-redcedar stand type. Hemlock trees in the hemlock stand type were taller, had larger diameters, and had higher mean annual increments, than those in the hemlock-redcedar stand type. These differences in hemlock and redcedar growth in each stand type are thought to be responsible for the differences in mean productivity detected among stand types. 5.1 Introduction Theory on positive plant interactions suggests that a mixture of two species can be more productive than either species in monoculture, i f those two species have high ecological combining ability (Harper 1977; Kelty 1992). Hemlock and redcedar may have high ecological combining ability, due, in part, to vertical canopy separation, spatial randomness in their distribution patterns, and uptake of different nitrogen forms. The first step in assessing whether 48 or not hemlock and redcedar interact positively in mixture is to compare productivity between mixed- and single-species stands of these species. The objective of this study was to determine whether there are differences in productivity between single- and mixed-species stands of hemlock and redcedar, with the mixed-species stands containing approximately an equal proportion of each species, on a basal area basis. Productivity was measured by mean annual increment. Similar comparisons were also made with stand volume, and basal area. Specifically, this study addressed two questions: (1) Is there a difference in the relative yield o f hemlock and redcedar in mixture, or in mean annual increment between mixed- and single-species stands? (2) What are the possible causal factors for any differences observed? 5.2 Results Uncorrected mean annual increment and quadratic mean diameter increased with increasing presence of hemlock, while the opposite trend was found for mean stems per hectare (Table 5.1), suggesting greater annual volume per tree in the hemlock stands. The highest mean basal area was also found in hemlock stands, although this may be misleading considering the influence of stand H7, which appears substantially higher in basal area than the other hemlock stands. Without the contribution of H7, the mean basal area of the hemlock stands was 9 1 9 1 intermediary (x = 80.0 m ha", 5 £ = 5.5 m ha") between the hemlock-redcedar and redcedar stands. Stand H7 had a higher mean annual increment than the other hemlock stands, likely due to its relatively higher age, site index, and density values (Table 5.1), but these values were not found to be outliers. Mean dominant tree height, site index, volume, and mean annual increment decreased with increasing presence of redcedar (Table 5.1). However, increasing presence of redcedar was also associated with higher mean basal area (excluding the influence of H7) and stems per 49 hectare. These findings suggest that redcedar stands have relatively higher numbers of shorter, more tapered stems, relative to the other stand types. 50 o o J3 <-l s <•-» * s pq w O to * 1 "3 ^ OH tn CO a o (D • 9 ? • H oo « r i m c i a m \ q i n « O N N O C N N O I H O N O N N O 00 N© N O r - r-- r>-o o r-~ <N v d m tfi oo O N oo oo N O r - H r-- i n <s o \ n H d H <—i M (N t N f S r>- NO - H IT) C \ > Os r n <S M M c o m C i O 00 o r -Hw) Hw) N O co co CN N O CN o oo >n i n r - NO i r i >n i n i n V ) co co oo >n i n r - r » 2S N O N O g O O N ® CN NO O O co © CN CN * h « h H N O i > r> o oo • M N ( S C l N O • — ( N r o o u u u ( S i n H \ o ^ H m t co i n i n \ f m » c M t i n NO 00 i n o i n ON o ON NO oo NO 00 o r~-O N O N i — i i n N O 00 O N O N CN O CN 00 y—1 © CN O N i—1 00 NO © i n co r - 00 NO CN i n N O O N CN © co co CN co CN co co fO O N CN 00 ^ ° CN O N ID O NO NO ^ (N c o Q x , _ O N O N H J O O O N S ^ S ^ 1 ^ O N O O N q o o N v q ^ N O N O - < f r © N q c N C O l - ; C N C N C N C O C N C N C N f N O ^ H i n o ^ H O O o o N O N O i n i n i n m m m » f ) o o r ^ i n r - ^ c o O N r n o • ^ f r C N c o ^ c o ^ H C O f } c o c o c o c o c o c o c o f ' i O N co O N r—i i — i co o N O N O o oo co C N C N co C N co C N © t> NO NO O 00 00 ^ — i C N co ->^ - <n N O r~ ^ r \ q ^ r — < r - - — < — < N O i n ' K K ^ i n ' o o r < i h * - H r - H * - H CN 1-H © i — i oo cn N O 00 oo © i n © © © i n r n o O N i n co co co 00 co o O N oo O N oo N O O N 00 O N N O o o CN ^ O N O Tj-o o o q N ^ r~ i n O N F H • cn o - H C N I I N O o 00 oo CO (N o i n CN CO CO oo m o oo oo oo r~ oo o\ CO O i n O N O o CN CN O N t> 00 o CN o CO C N p c N ^ c o ^ o i n c N ^ r c N c o r ^ i n o N o ' c o o o ^ C O C O C O C O C O C O N O c O »—i o o o O N r~ Tt 00 N O O N r> t> CO C N o o C N O C O O N oo m oo — i i n r f ; t > o o N i n o o o a O O r H l n ' r - i ^ O O M C N C N c O C O C O C O C O f O i n r ~ ~ ^ i n N O ^ H O T r N O i n N O N O N O N O I > v o i n r-~- C N oo - H i n H co co co ^ - co F H 00 r> 00 CO CO CN i n H ( N j m i - i n * > H 5.2.1 Productivity Comparisons - Relative Yield The total relative yield of the hemlock-redcedar stands, when the effects of non-study species have been removed, was greater than 1.0 (Equations [3.1-3.2]) (Table 5.2). This suggests that the mixtures are more productive than single-species stands of either species. Hemlock and redcedar both had R Y values greater than 0.5, implying that they benefit from each other's presence (Table 5.2). Table 5.2 Relative yield ( R Y ) and total relative yield ( R Y T ) values comparing volume of hemlock and redcedar stands with stands of hemlock-redcedar mixtures, with and without the volume contributions of non-study species. Contribution of non-study species RYRedcedar RYHemlock R Y T N o 0.63 0.61 1.24 Yes 0.40 0.50 0.90 However, yields of hemlock and redcedar in mixture were relatively lower than those in monoculture when the volume contributions of non-study species were accounted for. Although hemlock still achieves the same yield in mixture with redcedar relative to what it would in a single-species stand, redcedar achieves lower yield in mixture with hemlock relative to what it would in a single-species stand. Both species appear to suffer from competitive interactions with the non-study species, but the yield of redcedar is more severely impacted. This impact on redcedar yield decreases total relative yield in hemlock-redcedar mixtures below 1.0, indicating that the mixtures are experiencing competitive inter-species interactions. Although plantation studies have never had to account for the impacts of non-study species, these impacts in existing, unmanaged mixtures cannot be ignored. 52 Non-study species were present in all stands, although their presence did vary slightly among stand types. Non-study species contributed most to the basal area of hemlock stands, and least to the redcedar stands, while the mixtures had intermediate contributions (Table 5.3). Table 5.3 Mean basal areas, standard errors (in parentheses), and percent contribution of hemlock, redcedar, and non-study species to total stand basal area, stratified according to stand type. Stand type Hemlock Redcedar Non- study Total species Basal area ( n ^ h a 1 ) (%) Basal area ( n^ha 1 ) (%) Basal area (n^ha" 1) (%) Basal area ( n^ha 1 ) (%) Hemlock 78.5 85.1 5.8 7.2 6.9 7.7 91.2 100.0 n = 7 (32.4) (4.8) (4.6) (30.0) Hemlock- 36.9 47.8 36.6 46.9 4.2 5.3 77.8 100.0 redcedar (4.2) (7.3) (3.2) (7.7) n = 7 Redcedar 21.8 25.2 60.4 71.0 3.1 3.8 85.1 100.0 n = 4 (7.2) (4.7) (3.4) (7.8) 5.2.2 Productivity Comparisons - Absolute Yield Analysis of variance revealed that mean annual increment did not significantly differ among the locations (Table 5.4). This model indicates that non-significant differences among replicates exist. These differences were corrected for in analyses of covariance among stand types. 53 3 1 1 Table 5.4 Analysis of variance summary for mean annual increment (m ha" yr" ), where: hemlock, hemlock-redcedar, and redcedar are stand types; Capilano, Malcolm Knapp, and Miss ion are locations. Source Sum of squares Degrees of freedom Mean square F value P > F Stand type 54.7 2 27.3 9.037 0.007 Location 5.7 2 2.8 0.939 0.426 Stand type by 22.2 4 5.5 1.833 0.207 location interaction Error 27.2 9 3.0 Total 109.8 17 Increasing hemlock presence was associated with increasing mean annual increment (Table 5.5) (equation 3.7). Increasing amounts of variation in mean annual increment were explained as the complexity of the density covariate increased, allowing more sensitive testing for significant differences among the stand types. When density was expressed in the A N C O V A model as stems per hectare, a non-significant, increasing trend of productivity with hemlock presence was observed. Quadratic mean diameter explained more variation than stems per hectare, allowing the hemlock stand type to be found significantly more productive than the redcedar stand type. The model using relative density explained the most variation in stand type mean annual increment, and significant differences in increasing productivity were found for each stand type, with increasing presence of hemlock. 54 Table 5.5 Least squares means and standard errors (in parentheses) of mean annual increment (MAI) , volume, and basal area ( B A ) of 18 study stands, stratified according to density covariate and stand type. Values in the same row with the same superscript are not significantly different (a = 0.05). Productivity measure Density Stand productivity according to stand type covariate Hemlock n = 7 Hemlock-redcedar n = 7 Redcedar n = 4 Stand M A I (m 3 ha"1 yr"1) Stems per 16.3 a 14.9 a 14.3 a hectare (0.7) (0.6) (0.8) Quadratic 16.5 a 15.1 a ' b 14.1 b mean diameter (0.6) (0.5) (0-7) Relative 17.0 a 15.1 b 13.5C density (0.4) (0.3) (0.4) Stand volume (m ha" ) Stems per 954.8 a 911.6 3 874.2 a hectare (72.1) (41.0) (46.1) Quadratic 949.1 a 915.2 3 876.3 a mean diameter (62.6) (37.5) (37.9) Relative 1053.3 3 881.2 b 766.4C density (38.0) (21.0) (28.1) Stand B A (m 2ha"') Stems per 70.4 a 87.5 a 97.8 a hectare (8.2) (6.7) (8.4) Quadratic 78.7 a 85.9 a 90.6 a mean diameter (4.5) (3.5) (4.5) Relative 85.0 a 8 5 . l a 83.3 a density (3.2) (2.3) (3.2) Wood volume production also increased with increasing presence of hemlock (Table 5.5). The analysis of covariance model for volume performed similarly with the addition of different density covariates as the model used for mean annual increment. Increasing amounts of variation in volume were explained as the complexity of the density covariate increased, although significant differences among stand types were only found using relative density. In 55 this model each stand type was found to be significantly different, with an increasing trend in volume with progression from the redcedar to the hemlock-redcedar to the hemlock stand types. Basal area was not found to significantly differ among stand types (Table 5.5). Like the models for mean annual increment and volume, the analysis of covariance model for basal area explained increasing variation with increasing complexity of the density covariate. However, this was also associated with increasingly similar predictions of basal area among stand types. The redcedar trees in the single-species stands were similar in most respects to those in the mixed-species stands (Table 5.6). N o significant differences in mean diameter at breast height, tree height, height to live crown, and mean annual increment were found in the redcedar trees of either stand type. The mean height to live crown of redcedar in the hemlock-redcedar stands was significantly greater than that in the redcedar stands. N o differences were found among the cumulative distribution functions of redcedar in each of the tree characteristics compared. In the single-species stands the hemlock trees had greater mean annual increment, and significantly greater mean diameter, height, and height to live crown, than the hemlock in mixed-species stands (Table 5.6). N o differences were found among the cumulative distribution functions of hemlock in each of the tree characteristics compared. 56 Table 5.6 Individual mean tree characteristics, and standard errors of the means (in parentheses), o f each species in single- and mixed-species stands. Redcedar tree values in the same row with the same lowercase superscript are not significantly different, and hemlock tree values in the same row with the same uppercase superscript are not significantly different (a = 0.05). Tree characteristic Redcedar Trees Hemlock trees Redcedar stand type Hemlock-redcedar stand type Hemlock stand type Hemlock-redcedar stand type Diameter at breast height (cm) 28.4 a (14.1) 28.1 a (11.6) 33.3 A (11.5) 29.4 u (10.9) Total height (m) 22.2 a (5.7) 22.8 a (5.3) 3 1 . 0 A (6.6) 26.6 B (6.5) Height to live crown (m) 13.0 a (3.4) 14.4 b (3.5) 18.1 A (4.1) 15.3 B (4.5) Mean annual increment (mVyr' 1) 0.012 a (0.013) 0 .011 a (0.011) 0.022 A (.0.17) 0 .017 A (0.013) 5.3 Discussion 5.3.1 Productivity Comparisons - Relative Yield Relative yield comparisons in existing mixed-species stands are more difficult to interpret than those in plantation studies. Non-study species can influence the results and interpretation of relative yield (Table 5.2). Since most existing single or mixed-species stands commonly contain inclusions of other species, this influence on the study species cannot be ignored. In the single-and mixed-species stands of hemlock and redcedar, total relative yield changed from 1.24, indicating positive interactions, to 0.90, indicating competitive interactions, when the yield contributions of non-study species were factored into calculations. Comparing relative yield calculations with and without non-study species may indicate the degree to which non-study species compete, or positively interact, with the study species. 57 If non-study species compete with study species, the relative yield of the mixture being studied w i l l decrease. More competitive study species may decrease the negative impact the non-study species have on total relative yield. If non-study species interact positively with the study species, then total relative yield of the study mixture should be higher than would be observed in a plantation study. Although this increased yield may be desirable in economic terms, the effects of non-study species make it difficult to compare findings among other studies in existing, unmanaged mixtures. Hemlock and redcedar both appear to compete with non-study species. Redcedar's relative yield decreases below 0.5 with the influence of non-study species (Table 5.2), suggesting that redcedar competes with these species, and that this competition mutes the benefits of any positive interactions redcedar experiences with hemlock. It should be noted that redcedar stands are uncommon, and locating such stands was difficult. The result was redcedar stands averaging 25.2% basal area contribution by hemlock, and 3.8% by non-study species (Table 5.3). Conversely, relatively pure hemlock stands were much easier to locate, and these stands averaged only 7.2% and 7.7% basal area contributions by redcedar and non-study species respectively. Thus hemlock stands contained 15% inclusions and redcedar stands contained 29% inclusions, providing evidence of relatively greater competition in the redcedar stands. Overall, the competition between hemlock and redcedar with non-study species resulted in a net decrease in the yield of mixed- versus single-species stands. Although hemlock's relative yield decreases under the influence of non-study species, it still remains at 0.5 (Table 5.2). Hemlock appears to achieve the same productivity it would in a single-species stand while contending with non-study species. However, the relative yields of hemlock and redcedar in mixture must be interpreted with caution. Although the basal area contributions of hemlock and redcedar in the mixed-species stands were roughly equal, each of these contributions was not exactly half of the basal area of 58 each species in single-species stands. In fact, the basal area contribution of hemlock in hemlock-redcedar stands was 56% of the contribution of hemlock in hemlock stands. The basal area contribution of redcedar in hemlock-redcedar stands was 66% of the contribution of redcedar in redcedar stands. Based on these percentages, the relative yields of hemlock and redcedar are biased toward higher values, and their true relative yields are lower than those calculated with the equations for relative yield for plantation experiments. This suggests that the relative yields greater than 0.5 calculated without non-study species are inaccurate, and that the true relative yields for each species are lower. Since including non-study species in calculations further decreases relative yield, it must be concluded that the mixed-species stands are dominated by competition among hemlock, redcedar, and the non-study species. 5.3.2 Productivity Comparisons - Absolute Yield Mean annual increment was found to be similar among study locations (Table 5.4). Both mean annual increment and volume increased from redcedar, to hemlock-redcedar, to hemlock stands (Table 5.5). The mean annual increment model was more sensitive than the model detecting differences in volume. Mean annual increment eliminates the need for age correction, and may have resulted in a more sensitive analysis of variance model. The models for mean annual increment and volume became more sensitive with increasing complexity of the density covariates used in each model. Since relative density incorporates more stand information, including quadratic mean diameter and basal area, this density measure was expected to explain more variance in productivity among stand types than the density measures containing relatively less stand information. Analyses of covariance for each of mean annual increment and stand volume reveal that hemlock stands produce relatively higher wood volumes, and mean annual increments, than hemlock-redcedar stands or redcedar stands of equivalent age, site, and density. 59 A trend of decreasing basal area with increasing hemlock presence was detected when density was represented by stems per hectare in analysis of covariance (Table 5.5). The trend was more weakly expressed when quadratic mean diameter was used in the model, and disappeared when relative density was used. Like the models for mean annual increment and volume, the basal area model explained more variation and had lower error when the complexity of the density measure increased. However, the best predictor of basal area among stand types is probably the basal area model employing stems per hectare. Stems per hectare does not incorporate measures of stand basal area, as do quadratic mean diameter (equation [3.8]) and relative density (equation [3.9]). Using quadratic mean diameter and relative density, forced these models, that were designed to detect differences in basal area, to reflect stand types of equivalent basal area. Other observations support the output of the model employing stems per hectare, where increasing basal area is associated with increasing redcedar presence. The redcedar stands had similar quadratic mean diameters as the other stand types, but had higher stems per hectare (Table 5.1). Given roughly equal tree diameters in each stand type, redcedar stands simply seem to contain more trees than hemlock-redcedar or hemlock stands. Based on the observations of stand characteristics (Table 5.1), relative yield calculations (Table 5.2), and analyses of covariance (Tables 5.3-5.5), it can be concluded that the hemlock stands produce higher wood volume and mean annual increment, on less basal area, than the hemlock-redcedar or redcedar stands. Although hemlock stands have fewer stems, they appear to achieve superior productivity through much taller trees than the other stand types, and higher quadratic mean hemlock diameters than those in mixture. This suggests that, within the single-species stands, hemlock trees are taller and of better form than redcedar trees. Hemlock trees are less productive in mixed-species stands, while redcedar trees are equally productive, compared to their growth in single-species stands. The reductions in diameter at breast height, tree height, height to live crown, volume, and mean annual increment 60 that hemlock trees exhibit in mixture with redcedar support the hypotheses, based on relative yield, that hemlock experiences a decrease in productivity in mixture with redcedar. However, redcedar in hemlock-redcedar stands suffers no appreciable decrease in diameter, height, height to live crown, or mean annual increment, compared to that in redcedar stands. In fact, the increased height to live crown redcedar exhibits in mixture may improve stem form factor. These characteristics o f redcedar in mixture contradict the hypotheses from relative yield calculations that the relative yield redcedar suffers due to competition. The intermediate productivity of the hemlock-redcedar stands appears to be due to relative decreases in the height and diameter growth of hemlock trees, as a result of competition with redcedar and non-study species. This competition appears to more than offset any potential yield increases due to positive interactions among hemlock and redcedar in mixture. When grown in single-species stands, hemlock seems to have stronger capacity to self-thin than redcedar. The result is hemlock of higher diameter growing in stands of lower density. Hemlock also exhibits superior height growth, at the age of trees sampled, resulting in higher productivity stands. Redcedar in these stands tends to grow at higher density, and, when mixed with hemlock, mutes hemlock's tendency to self-thin. In mixture, hemlock experiences the effects of higher density, enhancing the negative effects of competition on tree growth. This combination of factors results in increases in productivity with increasing hemlock presence. Where maximum wood volume production is desired, single-species stands of hemlock should be established on sites similar to those studied. 5.4 Conclusions In mixtures of hemlock and redcedar, the relative yields of each species in single- versus mixed-species stands appear to suffer from competitive interactions. The natural occurrence of other species in existing stands further decreases the relative yields of hemlock and redcedar in 61 mixture. Mean annual increment and wood volume production increased with increasing presence of hemlock, while basal area increased with increasing presence of redcedar. The hemlock stand type was most productive because hemlock's height growth was greater than that of redcedar, and its height and diameter growth were greater in single-species stands. This greater diameter growth may be due to hemlock's superior ability to self-thin on these sites. The redcedar stand type was the least productive due to redcedar growing in high numbers of short, small diameter stems. Although hemlock was shorter and of smaller diameter in mixed-species stands, redcedar grew equally well in mixed- and single-species stands. Maximum wood volume production on similar sites would be achieved by establishing hemlock stands with a minimal component of other species. 62 6. ARE INCREASING CANOPY SEPARATION AND SPATIAL RANDOMNESS ASSOCIATED WITH INCREASED PRODUCTIVITY IN MIXED-SPECIES STANDS OF HEMLOCK AND REDCEDAR? Abstract I examined mixed-species stands of western hemlock and western redcedar to determine i f increasing mean annual increment was associated with increasing canopy separation, or increasing spatial randomness in the arrangement, of the two species. The second-growth stands were located in southern coastal British Columbia on intermediate sites, and were unmanaged, and even-aged, with approximately equal proportions of hemlock and redcedar on a basal area basis. Increasing stand productivity was associated with increasing vertical separation of hemlock and redcedar canopies, where the area of hemlock canopy overtopping that of redcedar, and crown depths of both species, were maximized. Increased productivity was not associated with increasing levels of randomness in stand spatial tree patterns, but was associated with decreasing stand density, and a tendency towards regularity in the stand spatial tree patterns. Maximum productivity in fully stocked hemlock-redcedar mixtures can be attained with low density stands in which hemlock does not experience an establishment lag. 6.1 Introduction Aboveground competitive reduction can be logically subdivided into physiological, spatial, and temporal components. Whatever the mechanism of aboveground competitive reduction, the manifestation is a stand canopy vertically separated into clearly defined layers of each species, spatially arranged to maximize the benefits of this separation. The objective of this chapter was to determine whether aboveground competitive reduction is associated with a gradient of increasing productivity in hemlock-redcedar stands. Specifically, this chapter sought to address the following question: Is increasing mean annual 63 increment (MAI) in hemlock-redcedar stands associated with (1) increasing vertical separation of hemlock and redcedar canopies, and/or (2) increasing randomness in tree spatial patterns? Previous studies suggest that increased canopy stratification is associated with increased stand productivity (Kelty 1992), and that clustering of two species indicates facilitation, while regularity in the pattern of two species indicates competition (Moeur 1993). Where this competition is reduced among the species, a random pattern intermediate between regularity and clustering is expected, and so spatial randomness w i l l be presumed to indicate a reduction in competition between the species. 6.2 Results 6.2.1 Canopy stratification Significant differences in mean annual increment per tree were detected among the hemlock-redcedar stands (Table 6.1). When corrected for age, site, and density, the average mean annual increment was greatest for trees in stand M 2 , and lowest in stand M 3 . Table 6.1 Least squares means and standard errors (in parentheses) of mean annual increment per tree, stratified according to hemlock-redcedar stand. Values in the same row with the same superscript are not significantly different (a =0.05). Productivity Stand Identification measure M 2 M 6 M 4 M 7 M l M 5 M 3 M e a n , 0.01646 3 0.01623 a b 0.01618 a b 0.01360 a b c 0.01303 a b c 0.01274 b c 0.01208 c mcrement (0-00133) (0.00127) (0.00156) (0.00133) (0.00134) (0.00109) (0.00127) per tree ( m y 1 ) Hemlock had larger quadratic mean diameters than redcedar in all but one of the hemlock-redcedar stands (Table 6.2). Hemlock was significantly taller than redcedar in all hemlock-redcedar stands. The height to the live crown of hemlock was significantly greater than that of 64 redcedar in all but 2 of the mixed-species stands. Ratios of hemlock and redcedar crown overlap (Equation [3.15]) ranged from 31 to 92% (Table 6.2). N o significant differences in cumulative distribution functions of tree height, or height to live crown, were found among hemlock and redcedar. Table 6.2. Percentage of basal area contribution ( % B A ) , quadratic mean diameter, mean tree height, mean height to live crown, and crown overlap ratio (COR), of hemlock and redcedar in each of the hemlock-redcedar stands. Stands are arranged in order of decreasing productivity. Standard errors of means appear in parentheses. Tree height values in the same row with the same lowercase superscript are not significantly different, and height to live crown values in the same row with the same uppercase superscript are not significantly different (a = 0.05). Stand % B A Quadratic mean Tree height (m) Height to live crown C O R diameter (cm) (m) C w H w C w H w C w H w C w H w M 2 52 40 32.3 35.8 21.6 a 26.5" 13.4A 14.7B 0.319 (5.8) (5.2) (2.9) (2.3) M l 55 39 31.1 32.5 22.9 a 26.5 b 12.5 A 12.8A 0.472 (5.0) (8.0) (2.4) (3.6) M 6 40 51 27.8 29.6 23.6 a 28.3 b 15.7A 17.9 B 0.314 (4.6) (4.4) (2.3) (2.8) M 7 39 52 27.9 29.5 22.0 a 26.0 b 15.2A 16.7 B 0.428 (5.0) (6.6) (2.8) (3.7) M 4 48 52 40.3 31.9 22.1 a 21.7 a 8.6A 7.8 A 0.916 (7.4) (9.1) (3.2) (3.0) M 5 42 58 31.1 33.9 22.6 a 28.1 b 15.2 A 18.8 B 0.369 (5.1) (5.3) (2.7) (2.4) M 3 52 43 26.0 27.0 23.9 a 21.1° 16.7A 17.6B 0.477 (4.6) (4.8) (2.3) (2.2) The hemlock in the most productive mixed-species stand (M2) had a larger quadratic mean diameter, and significantly lower height to live crown, than those in the least productive mixture (M3) (Table 6.3). Hemlock was shorter in stand M 2 than in stand M 3 , but not significantly so. The redcedar in stand M 2 had larger quadratic mean diameters, and had 65 significantly lower mean tree height and mean height to live crown, than the redcedar in stand M 3 (Table 6.3). Table 6.3. Percentage of basal area contribution ( % B A ) , quadratic mean diameter, mean tree height, and mean height to live crown, of hemlock and redcedar in the most (M2) and least (M3) productive hemlock-redcedar stands. Standard errors of means appear in parentheses. Tree height values in the same row with the same lowercase superscript are not significantly different, and height to live crown values in the same row with the same uppercase superscript are not significantly different ( a = 0.05). Species % B A Quadratic mean Tree height (m) Height to live crown diameter (cm) (m) M 2 M 3 M 2 M 3 M 2 M 3 M 2 M 3 Hemlock 40 43 35.8 27.0 26.5 a 27.7 a 14.7A 17.6B (5.2) (4.8) (2.3) (2.2) Redcedar 52 52 32.3 26.0 21.6 a 23.9 b 13.4A 16.7B (5.8) (4.6) (2.9) (2.3) 6.2.2 Spatial randomness Refined nearest neighbour analysis Figure 6.1 shows how empirical nearest neighbour distance changes as a function of distance, for the most (M2) and least (M3) productive hemlock-redcedar stands. The nearest neighbour function of stand M 2 showed the highest degree of regularity in the distribution function of nearest neighbour distances of any hemlock-redcedar stand. This regularity reaches a maximum approximately 0.7 m from trees in stand M 2 (Figure 6.1). A t distances greater than 4.4 m from trees in stand M 2 , nearest neighbouring trees were found at the same distance as would be found in a stand of randomly arranged trees. 66 M2 0.05 0.00 clustering Figure 6.1. Output of refined nearest neighbour analysis for the most (M2) and least (M3) productive hemlock-redcedar stands, in terms of mean annual increment (m3ha"1yr"1). The horizontal line with Y intercept = 0 represents the expected And 2 nearest neighbour function for a Poisson forest [ G(d) = l-e ' ]. 67 The nearest neighbour function of stand M 3 spiked toward a clustered pattern at a distance of approximately 0.3 m. This clustering was followed by a rapid trend toward regularity, but the regularity was less strongly expressed than in stand M 2 (Figure 6.1). This regularity was most strongly expressed at distances of 0.7 m , 1.0 m, and 1.7 m from trees in stand M 3 . A t distances greater than 3.6 m from trees in stand M 3 , nearest neighbouring trees found at the same distance as would be found in a stand of randomly arranged trees. Combined count and distance analysis In the most productive stand (M2) inter-tree distances deviated significantly from those expected from completely random tree spatial patterns (Figure 6.2A-C). Significant clustering was detected in the stand at distances less than 1 m from any tree (Figure 6.2A). A t inter-tree distances in the stand between 1.5 m and 7 m, the trees closely follow a random spatial pattern. Significant regularity in inter-tree distances was found between 9 m and 11m. Beyond 1 1 m inter-tree distances were considered random, but displayed a persistent tendency toward regularity. Redcedar in stand M 2 were highly clustered at distances less than 1 m (Figure 6.2B). The redcedar then closely follow a random pattern to a distance of 9 m, and remain random with a regular tendency to 15 m, beyond which they remain significantly regular. The hemlocks in stand M 2 were also clustered within a i m inter-tree distance (Figure 6.2C). Between 1 m and 4 m the arrangement of hemlocks closely follows a random pattern. Beyond 4 m, hemlock tends toward regularity, and beyond 7 m hemlock maintains a random inter-tree distance with a tendency towards regularity. 68 M2 Figure 6.2 £ § 0 redcedar clustering (B) regularity 10 15 d(m) stand clustering (A) ! regularity d(m) E § 0 hemlock clustering (C) regularity 3 5 10 15 20 d(m) Ripley's Ln (d) results for composite (stand - A ) , and individual (redcedar - B ; hemlock - C) , spatial point patterns for the most productive hemlock-redcedar stand (M2). The horizontal line with Y intercept = 0 represents the expected value [L 12(d) = 0] o f a random spatial point pattern, and the dotted lines represent the upper and lower pointwise 95% confidence envelope boundaries. In the least productive stand (M3) inter-tree distances closely approximated those expected of a completely random spatial pattern of trees (Figure 6.3A-C). The combined spatial pattern o f hemlock and redcedar trees is somewhat regular between distances of 2 m to 4 m, beyond which the pattern maintains a random pattern with a slight tendency toward regularity (Figure 6.3A). The pattern of redcedar in stand M 3 spiked toward regularity twice at distances less than 1.5 m (Figure 6.2B). The redcedar then approach and closely follow a random pattern. 69 The hemlock in stand M 3 are highly regular between 1 m and 5 m, then are arranged approximately randomly at all distances beyond 5 m (Figure 6.3C). M3 clustering regularity stand 10 d(m) (A) Figure 6.3 redcedar clustering (B) regularity o 5 10 15 20 d(m) E, S o clustering regularity hemlock (C) d(m) Ripley's Ln (d) results for composite (stand - A ) , and individual (redcedar - B ; hemlock - C) , spatial point patterns for the least productive hemlock-redcedar stand (M3). The horizontal line with Y intercept = 0 represents the expected value [L 12(d) = 0] o f a random spatial point pattern, and the dotted lines represent the upper and lower pointwise 95% confidence envelope boundaries. 70 6.3 Discussion 6.3.1 Canopy stratification The trend of increasing productivity in the hemlock-redcedar stands was not associated with an increase in the percentage of basal area contribution by hemlock (Table 6.2). Although hemlock stands were more productive than hemlock-redcedar and redcedar stands, and this productivity was due to hemlock's superior height and diameter growth, increasing productivity in hemlock-redcedar stands was not simply due to increasing hemlock content. This suggests that hemlock and redcedar do exert influences on each other in mixture, and that these influences have a direct bearing on the productivity of these mixtures. Where the influences that hemlock and redcedar have on each other are competitive, separating each species spatially can reduce this competition. Within the hemlock redcedar stands, increasing separation of hemlock and redcedar canopies appeared to be associated with increased productivity in hemlock redcedar mixtures. The characteristics of hemlock and redcedar in the most (M2) and least (M3) productive stands indicate that differences between these stands exist in canopy architecture. Both hemlock and redcedar in stand M 2 have greater mean crown depths than those in stand M 3 . In addition to the deeper crowns in stand M 2 , there is a greater difference in height between hemlock and redcedar, relative to that in stand M 3 . The hemlock is 4.9 m taller than redcedar in stand M 2 , while hemlock is only 3.8 m taller than redcedar in stand M 3 . Therefore, the hemlock in stand M 2 has a greater crown area free from competition with redcedar than the hemlock in stand M 3 . The crown overlap ratios of each stand support these findings, confirming that the crown of redcedar overlaps that of hemlock approximately 15% more in stand M 3 than in stand M 2 (Table 6.2). However, no clear trend between increasing productivity and crown overlap ratio was evident in the hemlock-redcedar stands (Table 6.2). This implies that either crown overlap 71 among redcedar and hemlock is not a strong indicator of competition, or that above ground competition is not sufficiently explained by crown overlap ratio alone, or that the crown overlap equation does not accurately explain crown overlap. It seems likely that some other factors act in concert with crown overlap to affect the productivity of hemlock-redcedar mixtures. In addition to lower crown overlap, the hemlock and redcedar in stand M 2 have a greater leaf area than the hemlock and redcedar in stand M 3 . Although the hemlock and redcedar are taller in stand M 3 , the heights to their live crowns were also greater than in stand M 2 . The greater leaf areas of hemlock and redcedar in stand M 2 explain the greater quadratic mean diameters in this stand, and these greater diameters may partially explain the greater productivity of this stand. Wood volume production was calculated as a function of both diameter and height (equations [3.3-3.6]). Although the hemlock and redcedar in stand M 2 were shorter than those in stand M 3 , stand M 2 still had higher wood volume and mean annual increment. It appears that the greater quadratic mean diameters of hemlock and redcedar in stand M 2 more than offset their lower mean heights, resulting in trees of greater volume than in stand M 3 . The deeper crowns of hemlock and redcedar in stand M 2 indicates that the trees in this stand are more widely spaced than in stand M 3 . In fact, at 944 stems per hectare, stand M 2 is more widely spaced than stand M 3 , which has 1356 stems per hectare (Table 5.1). However, stand M 2 is of higher relative density than stand M 3 (Table 5.1). These findings suggest that the wider spacing of stand M 2 produces wood volume more efficiently than the narrower spacing of stand M 3 . Therefore, the productivity of hemlock-redcedar stands appears to be a function of stand density and crown overlap between the two species. Given a fully stocked stand, decreasing stems per hectare results in a higher quadratic mean diameter, without a major impact on mean height growth (Davis and Johnson 1987). Where hemlock does not experience an establishment 72 lag, it w i l l overtop redcedar and reduce crown overlap. The result is a stand that produces wood volume more efficiently per unit basal area, than a stand of higher stems per hectare and crown overlap. 6.3.2 Spatial randomness It should be noted that pattern classifications (e.g.: clustered, random, or regular) reflect the level of measurement, or scale used for examination (Chen and Bradshaw 1999). A s a result, what appears clustered at one scale may seem regular at another (Upton and Fingleton 1985). However, refined nearest neighbour and Ripley's analyses are powerful tools, as they cover the entire range of scales bounded by the limits restricted by sampling boundaries (Moeur 1993). Interpretations of similar stands can be made to the scale of the study stands. Refined nearest neighbour analysis Stand data for nearest neighbour analysis included the point locations of dead stems in each stand. Stand M 2 had a total of 345 dead stems per hectare, while stand M 3 had 800 dead stems per hectare. It may be that the initial spike of clustering observed in stand M 3 at a distance of approximately 0.3 m (Figure 6.1) is caused, in part, by the presence of the dead stems. During stand mapping dead stems appeared clustered around surviving stems, especially in the case of redcedar, and particularly in stand M 3 , which had a high number of stems per hectare. If the clustering observed in stand M 3 represents dead stems, then the minimum nearest neighbour distance (where minimum nearest neighbour distance is defined as the first distance at which G(d) - G(d) is positive (Moeur 1993)) observed in this stand would represent the minimum physical distance between clustered trees prior to the understory reinitiation stage of stand development. The minimum physical distance between the stems represents the zone of inhibition between the trees in the cluster (Moeur 1993). However, there is no minimum nearest neighbour distance in the plot for stand M 2 (Figure 6.1), and this may demonstrate the weakness 73 of spatial analyses based only on nearest neighbours. However, the zone of inhibition in stand M 2 can likely be inferred from the spike toward regularity between 0 .1m and 1.8 m. The visual evidence of dead stems clustered around live ones, and the nearest neighbour plots, suggests that the zone of inhibition around trees in the cluster increases with stand age, and only the most successful trees in each cluster w i l l survive at the expense of their neighbours. The zone of inhibition also increases, and becomes more strongly expressed, as stand density decreases from stand M 3 to stand M 2 . It seems likely that this phenomenon is the result of a uniquely regular seedling establishment pattern in stand M 2 , due to the lack of evidence of clustered dead stems in this stand. Combined count and distance analysis The significant clustering in stand M 2 at distances less than 1 m (Figure 6.2A-C) directly contradicts the output of nearest neighbour analyses for this stand (Figure 6.1). The results of Ripley's analyses are likely to be more accurate due to the increased complexity of spatial pattern information incorporated into these analyses. Based on Ripley's analyses, the hemlock and redcedar in stand M 2 both tend to cluster at distances less than 1 m, and this clustering occurs within and among the species (Figures 6.2A-C and 6.4). The combined spatial pattern of hemlock and redcedar, and each of their individual spatial patterns, reflects a random arrangement of stems between approximately 1 m and 7 m, followed by a significant departure toward regularity at 10 m. These findings suggest that hemlock and redcedar in stand M 2 are arranged in inter-mixed groups. These groups tend to have smaller stems clustered within a meter of larger stems, but the groups are randomly arranged and are roughly 10 m in radius. Hemlock in this stand is likely to occur randomly outside this 10 m radius, while redcedar is much more likely to be found within each group (Figures 6 .2A-C and 6.4). Hemlock tends to arrange itself in a regular pattern as close as 5 m, while redcedar does not do so until 9 m. This suggests that either hemlock competes more vigorously with conspecifics than redcedar, or that 74 hemlock seedlings originally established themselves according to a more regular pattern than redcedar. 25 10 5 • o • Ci U 1 * o 9 O o e • 0 • • • o 0 0 • oO ' I < • 0 D c O ) » • 1 • • • o 0 • 0 O 0 • ° 0 • • • O • • o • o O • o O 0 • • 4 a 0 > • • o • 0 5 10 15 20 25 3 X Axis (m) O Cw © Cw dead • Hw # Hw dead Figure 6.4 Top down view of the most productive hemlock-redcedar stand M 2 . Each circle represents an individual stem, stratified by colour according to species, and the diameter of each circle is based on the diameter at breast height of each tree. In stand M 3 , the clustering detected in nearest neighbour analyses at distances less than 1 m was also detected with Ripley's analyses, but only in the case of the hemlock trees (Figure 6.3C). Since the stand data used for Ripley's Kn (d) and Ln (d) analyses did not include the spatial point patterns of dead stems, comparison of Ln (d) and nearest neighbour plots over distances less than 1 m should reveal the degree to which dead stems contributed to the 75 clustering at these distances. Therefore, it can be assumed that the trees clustered at distances less than 1 m in stand M 3 are either dead stems or live hemlock trees (Figures 6.1, 6 .3A-C, and 6.5). The spatial patterns of hemlock, redcedar, and their combined pattern, all approach significant regularity between distances of 2 m and 3 m, beyond which a random, i f not slightly regular, pattern is maintained. The trees in stand M 3 appear to be arranged in a random pattern of individuals. Where larger hemlock trees are clustered with smaller trees, these trees are likely to be smaller hemlocks or dead trees. Where larger redcedar trees are clustered with smaller trees, these tend to be dead trees only. Live trees tend to be 2 m to 3 m from each other, and the area within this radius delineates each tree's zone of inhibition. 76 30 25 20 E, .52 15 10 • • % O • • • • r o • > • o o • o o • o • o ft • • o 1 o o e o • • 0 o • • o e • € O o 9' ° • o 0 o . °« 0 0 c • 0 • , • • • • b to < o 0 o ) • ° ° CP . O o O Q ' " 0 e / o. o • o • • • 0 c • 0 o # > o o O • o • • • o • 0 • • e o • • o e O e o o • • • • o 10 15 20 25 30 X Axis (m) OCw OCw dead • Hw • H w dead Figure 6.5 Top down view of the least productive hemlock-redcedar stand M 3 . Each circle represents an individual stem, stratified by colour according to species, and the diameter o f each circle is based on the diameter at breast height o f each tree. Moeur (1993) found that competitively inferior trees in a stand tended towards clustering while successful competitors tend toward regularity. Since the most productive stand (M2) is characterized by regularity, and the least productive stand (M3) by randomness, this implies that the trees in the most productive stand are competing more successfully with each other than those in the least productive stand. If the total of dead and live stems in each stand represent the density of each stand before stem exclusion, the least productive stand is still more dense than the most productive stand was at the beginning of stem exclusion. Given the assumption that the least productive stand w i l l self-thin towards the density of the most productive stand, the least 77 productive stand represents a relatively earlier stage of stand development than the most productive stand, in which inter-tree competition is more intense. It seems likely that stands progress from clustering to regularity as they age due to competition-induced mortality, and that trees primarily compete with their immediate neighbours. These conclusions agree with findings from other studies involving spatial analyses of forest stands (Ford and Diggle 1981; Kenkel 1988; Moeur 1993). Based on the findings of nearest neighbour and Ripley's analyses, productivity was not observed to increase with increasing spatial randomness in the hemlock-redcedar stand type. A trend of increasing regularity was found from the least to the most productive hemlock-redcedar stand. Thus decreasing density and increasing regularity in spatial pattern are associated with increasing productivity in hemlock-redcedar mixtures. 6.4 Conclusions Increasing productivity in the hemlock-redcedar stand type was associated with increasing vertical separation of hemlock and redcedar canopies. In particular, productivity appeared to increase with increasing area of hemlock canopy overtopping that of redcedar, and with increasing canopy depths of both species. Increasing spatial randomness in the spatial patterns of hemlock, redcedar, or their combination, was not associated with increasing productivity. However, productivity did increase with increases in the regularity of the spatial patterns of trees in hemlock-redcedar stands. These findings suggest that density plays a key role in the productivity of these stands. Given full stocking and no establishment lag for hemlock, reductions in density w i l l promote slightly shorter trees of higher quadratic mean diameter, and 78 deeper crowns with reduced overlap, in the stand. These tree and stand characteristics allow more efficient wood volume production per unit basal area. It can be concluded that hemlock and redcedar in mixture may experience some reduction in competition due to vertical canopy separation, but that competition probably dominates the dynamics of how hemlock and redcedar arrange themselves on the ground. 79 7. GENERAL CONCLUSIONS Given the silvical characteristics of hemlock and redcedar, theory on positive plant interactions suggests that mixed-species stands might have a higher productivity than single-species stands. Hemlock-redcedar stands may have higher productivity due to canopy separation, longevity differences, preferential uptake of different forms of nitrogen, beneficial modification of soil nutrients, reduced herbivore search efficiencies, and increased windfirmness. Evidence has been found that suggests hemlock and redcedar may benefit from positive plant interactions. Both species have been shown to uptake different forms of nitrogen, and analyses indicated that increasing differentiation of hemlock and redcedar canopies was associated with increasing productivity. However, the benefits associated with growing in mixture appear to be offset by competition for growing space, particularly in dense stands. This competition results in hemlock and redcedar producing lower relative yields in mixture compared to single-species stands. In terms of mean annual increment, the hemlock stand type was more productive than the hemlock-redcedar, and redcedar stand types. The increasing trend in productivity with increasing presence of hemlock seemed to be due to hemlock's superior height growth compared to redcedar, and hemlock's superior height and diameter growth in single-species stands compared to mixtures. Given the choice between single- and mixed-species stands of hemlock and redcedar, hemlock stands should be established i f maximum wood production is the management objective. If maintaining a mixture, or improving site nutrient status, is the management objective, then hemlock-redcedar stands should be established. Max imum wood production in a hemlock-redcedar stand w i l l be achieved where stands are fully stocked, but established at low density. These recommendations are for intermediate (zonal, H w - B a Blueberry) sites in the 80 Submontane Very Wet Maritime Coastal Western Hemlock ( C W H v m 1) variant, for stands less than 80 years old. 81 8. LITERATURE CITED Aarssen, L . W . and G . A . Epp. 1990. Neighbour manipulations in natural vegetation: A review. J. Veg. Sci . 1: 13-30. Alban, D . H . 1967. The influence of western hemlock and western redcedar on soil properties. Ph.D. Thesis, Washington State University, A n n Arbor, M I . Alban, D . H . 1969. The influence of western hemlock and western redcedar on soil properties. Soi l Science Society of America Proceedings. 33: 453-457. Anonymous. 1976. Ministry of forests volume tables. B . C . Ministry of Forests, Victoria, B . C . Pp. 431-433. Anonymous. 1995. Establishment to free growing guidebook, Vancouver Forest Region. Forest Practices Code. B . C . Ministry of Forests and Ministry of Environment, Victoria, B . C . Archer, S. and E . J . Rykie l . 1994. Similarities in vegetation cluster formation processes in contrasting savanna ecosystems. I S E M Symposium: Facilitation, nurse plants, and nucleation: Vegetation cluster processes driving plant succession. Assman, E . 1970. The principles of forest yield study. Pergamon Press Ltd. , Toronto, O N T . 506p. Bakuzis, E . V . 1969. Forestry viewed in an ecosystem perspective. Pp. 189-258. In: V a n Dyne, G . M . (Ed.). 1969. The ecosystem concept in natural resource management. Academic Press, N e w York, N Y . Bartlett, M . S . 1937. Properties of sufficiency and statistical tests. Proceedings of the Royal Society of London, Series A 160: 268-282. Berendse, F . and R. Aerts. 1984. Competition between Erica tetralix L . and Molina caerulea (L.) Moench as affected by the availability of nutrients. Acta Oecol. 5:3-14. Bertness, M . D . and R . M . Callaway. 1994. Positive interactions in communities. Trends Ecol . Evo l . 9: 191-193. Bertness, M . D . and S . M . Yeh . 1994. Cooperative and competitive interactions in the recruitment of marsh elders. Ecology. 75:2416-2429. Besag, J.E., and P.J. Diggle. 1977. Simple Monte Carlo tests for spatial pattern. App l . Stat. 26: 327-333. Biging, G.S. , and M . Dobbertin. 1995. Evaluation of competition indices in individual tree growth models. For. Sci . 41: 360-377. Bil l ings, W . D . 1969. Vegetational pattern near alpine timberline as affected by fire -82 snowdepth interactions. Vegetatio 19: 192-207. Binkley, D . 1995. The influence of tree species on forest soils: Processes and patterns. Pp 1-34. In: Mead, D . J. and Cornforth, I. S. 1995. Proceedings of the Trees and Soi l Workshop. Lincoln University Press, Canterbury, N e w Zealand, Agronomy Society of New Zealand Special Publication N o . 10. Binkley, D . and S. Hart. 1989. The components of nitrogen availability assessments in forest soils. Advances in Soil Science 10: 57-112. Binkley, D . , K . A . Dunkin, D . DeBel l and M . G . Ryan. 1992. Production and nutrient cycling in mixed plantations o f Eucalyptus and Albizia in Hawaii . For. Sci . 38: 393-408. Boardman, N . K . 1977. Comparative photosynthesis of sun and shade plants. Ann . Rev. Plant Physiol. 28: 355-377. Borman, F . H . 1966. The structure, function and ecological significance of root grafts in Pinus strobus L . Eco l . Monogr. 36: 1-26. Borman, F . H . and B . F . Graham, Jr. 1959. The occurrence of natural root grafting in eastern white pine Pinus strobus L . , and its ecological implications. Ecology. 40: 677-691. Brady, N . C . and R.R. Wei l . 1996. The nature and properties of soils, 12 t h Ed . Prentice-Hall Inc., Upper Saddle River, N J . 88lp . Brooke, R . C . , E . B . Peterson and V . J . Krajina. 1970. The subalpine mountain hemlock zone. Ecol . Western N . Amer. 2: 148-239. Burschel, P., H . E l Kateb, and R. Mosandl. 1992. Experiments in mixed mountain forests in Bavaria. Pp. 183-215. In: M . J . Kelty (Ed.). 1992. The ecology and silviculture of mixed-species forests. Kluwer Academic Publishers, Netherlands. 287p. Cajander, A . K . 1926. The theory of forest types. Acta For. Fenn., 2: 11-108. Caldwell, M . M . and J .H . Richards. 1989. Hydraulic lift: Water efflux from upper roots improves effectiveness of water uptake by deep roots. Oecologia. 79: 1-5. Callaway, R . M . 1992. Effect of shrubs on recruitment of Quercus douglasii and Quercus lobata in California. Ecology. 73:2118-2128. Callaway, R . M . 1994. Facilitative and interfering effects of Arthrocnemum subterminale on winter annuals in a California salt marsh. Ecology. 75: 681-686. Callaway, R . M . 1995. Positive interactions among plants. The Botanical Rev. V o l . 61, 4: 306-349. Callaway, R . M . and M . D . Bertness. 1994. Gradients of physical stress and the relative importance of facilitation and interference. I S E M Symposium: Facilitation, nurse plants, and nucleation: Vegetation cluster processes driving plant succession. 83 Callaway, R . M . , N . M . Nadkarni and B . E . Mahal l . 1991. Facilitation and interference of Quercus douglasii on understory productivity in central California. Ecology. 72: 1484-1499. Chapin, F .S. HI, L . R . Walker, C . L . Fastie, and L . C . Sharman. 1994. Mechanisms of primary succession following deglaciation at Glacier Bay, Alaska. Eco l . Monogr. 64: 149-175. Chen, J., and G . A . Bradshaw. 1999. Forest structure in space: A case study of an old growth spruce-fir forest in Changbaishan Natural Reserve, P R China. For. Eco l . Manag. 120: 219-233. Clark, P. J., and F . C . Evans. 1954. Distance to nearest neighbour as a measure of spatial relationships in populations. Ecol . 35: 45-453. Collins, D . B . , M . C . Feller, K . Kl inka , and L . M . de Montigny. In review. Forest floor nutrient properties in single- and mixed-species, second growth stands of western hemlock and western redcedar. Northwest Science. Connell, J .H. 1983. On the prevalence and relative importance of interspecific competition: Evidence from field experiments. Amer. Naturalist. 122: 661-696. Connell, J .H . and R .O . Slatyer. 1977. Mechanisms of succession in natural communities and their role in community stability and organization. Amer. Naturalist. I l l : 1119-1124. Curran, M . P . and B . G . Dunsworth. 1988. Coastal western redcedar regeneration: Problems and potentials. Pp. 20-32 In: N . J . Smith (Ed.). 1988. Western redcedar—does it have a future? Proc. Conf , Vancouver, B C . University of British Columbia, Faculty of Forestry. Curtis, R .O . 1982. A simple index of stand density for Douglas fir. Forest Sci . 28:92-94. Dafhi, A . 1984. Mimicry and deception in pollination. Annual Rev. Eco l . Syst. 15:259-278. Daubenmire, R. 1953. Nutrient content of leaf litter of trees in the northern Rocky Mountains. Ecology 34: 786-793. Davis, L . S . and K . N . Johnson. 1987. Forest management. 3 r d Edition. M c G r a w - H i l l Book Company, N e w York, N Y . 790 pp. Dawson, T .E . 1993. Hydraulic lift and water use by plants: Implications for water balance, performance and plant-plant interactions. Oecologia. 95:565-574. DeAngelis, D . L . , W . M . Post and C . C . Travis. 1986. Positive feedback in natural systems. Springer-Verlag, N e w York, N Y . DeBel l , D.S. , T . G . Cole and C D . Whitesell. 1997. Growth, development, and yield, in pure and mixed stands of Eucalyptus and Albizia. For. Sci. 43:286-298. 84 Diggle, P.J. 1979. Statistical methods for spatial point patterns in ecology. Pp. 95-150 In: Cormack, R . M . , and J .K. Ord (Eds.). 1979. Spatial and Temporal Analysis in Ecology. International Cooperative Publishing House. Fairland, M D . Diggle, P.J. 1983. Statistical Analysis of Spatial Point Patterns. Academic Press. New York, N Y . Eis , S. 1973. Root system morphology of western hemlock, western red cedar, and Douglas fir. Can. J. For. Res. 4:28-38. Eis , S. 1987. Root systems of older immature hemlock, cedar, and Douglas fir. Can. J. For. Res. 17: 1348-1354. Ell ison, L . and W . R . Houston. 1958. Production of herbaceous vegetation in openings and under canopies of western aspen. Ecology. 39: 338-345. Evens, J.R., S. V o n Caemmerer, and W . W . Adams, HI (Eds.). 1988. Ecology of photosynthesis in sun and shade. C S I R O , Australia. 358p. Feller, M . C . , and K . Kl inka . 1998. Seedfall, seed germination, and initial survival and growth o f seedlings of Thuja plicata in southwestern British Columbia. Northwest Science, 72: 157-169. Ford, E . D . , and P.J. Diggle. 1981. Competition for light in a plant monoculture modeled as a spatial stochastic process. Ann . Bot. 48:481-500. Fowler, N . F . 1986. The role of competition in plant communities in arid and semiarid regions. Ann . Rev. Ecol . Syst. 17: 89-110. Gashwiler, J.S. 1970. Further study of conifer seed survival in a western Oregon clearcut. Ecology. 51: 849-854. Gordon, A . G . 1983. Nutrient cycling dynamics in differing spruce and mixedwood ecosystems in Ontario and the effects of nutrient removals through harvesting. Pp. 97-118 In: R . W . Wein, R .R. Riewe and I.R. Mathven (Eds.). 1983. Resources and dynamics of the boreal zone, Proc. Conf., Thunder Bay, Ont. Aug . 1982. Assoc. Can. Univ . North. Stud. Ottawa, O N . Grace and Tilman 1990. Perspectives on plant competition. Academic Press, Inc., New York, N Y . Green, B . and Kl inka , K 1994. A field guide for site identification and interpretation for the Vancouver Forest Region. Land Management Handbook Number 28, B . C . Ministry of Forests. Crown Publications Inc., Victoria, B . C . Green, R . N . , R . L . Trowbridge, and K . Kl inka . 1993. Towards a taxonomic classification of humus forms. Forest Science Monograph 29: 1-48. Green, R . N . , P.J. Courtin, K . Kl inka , and F . C . Nuszdorfer. 1984. Site diagnosis, tree species 85 selection, and slashburning guidelines for the Vancouver Forest Region. B . C . Ministry of Forests, Victoria, B . C . 143p. Grubb, P.J. 1977. The maintenance of species - richness in plant communities: The importance of the regeneration niche. B i o l . Rev. 52:107-145. Habek, J.R. 1968. Forest succession in the Glacier Park cedar - hemlock forests. Ecology. 49: 872-880. Habek, J.R. 1978. A study of climax western redcedar (Thuja plicata Donn) forest communities in the Selway - Bitterroot Wilderness, Idaho. Northwest Sci . Ecology. 52:67-76. Hamilton, G.J . , and J . M . Christie. 1971. Forest management tables (metric). Forestry Commission Booklet N o . 34, Her Majesty's Stationery Office. 201 p. Harmon, M . E . , and J.F. Franklin. 1989. Tree seedlings on logs in Picea - Tsuga forests of Oregon and Washington. Ecology. 70: 48-59. Harper, J .L . 1977. Population biology of plants. Academic Press, N e w York, N Y . Hendrickson, O.Q. 1985. Variation in the C : N ratio o f substrate mineralized during forest humus decomposition. Soil Biology and Biochemistry 17:435-440. Holmes, R . D . and K . Jepson-Innes. 1989. A neighborhood analysis of herbivory in Bouteloua gracilis. Ecology. 70: 971-976. Hunter, A . F . and L . W . Aarssen. 1988. Plants helping plants. Bioscience. 38:34-40. Joffre, R. and S. Rambal. 1993. H o w tree cover influences the water balance o f Mediterranean rangelands. Ecology. 74: 570-582. Jolliffe, P . A . 1997. Are mixed populations of plant species more productive than pure stands? Oikos. 80:595-602. Kaluzny, S.P., S.C. Vega, T.P. Cardoso, and A . A . Shelly. 1998. S+ Spatial Stats: User's Manual for Windows and U N I X . Springer-Verlag. New York, N Y . 327p. Kayahara, G.J . , K . Kl inka , and A . C . Schroff. 1997. The relationship of site index to synoptic estimates o f soil moisture and nutrients for western redcedar (Thuja plicata) in southern coastal British Columbia. Northwest Science. 71:167-173 Keeley, J .H . 1988. Population variation in root grafting and a hypothesis. Oikos. 52:364-366. Keenan, R.J . , C . E . Prescott, and J.P. Kimmins. 1995. Litter production and nutrient resorption in western red cedar and western hemlock forests on northern Vancouver Island, British Columbia. Canadian Journal of Forest Research 25:1850-1857. Kelty, M . J . 1989. Productivity of New England hemlock/hardwood stands as affected by species composition and canopy structure. For. Ecol . Manag. 28: 237-257. 86 Kelty, M . J . 1992. Comparative productivity of monocultures and mixed-species stands. Pp. 125-141. In: M . J . Kelty (Ed.). 1992. The ecology and silviculture of mixed-species forests. Kluwer Academic Publishers, Netherlands. 287p. Kelty, M . J . and I.R. Cameron. 1995. Plot designs for the analysis of species interactions in mixed stands. Commonwealth For. Rev. 74: 322-332. Kenkel, N . C . 1988. Pattern of self-thinning in jack pine: Testing the random mortality hypothesis. Ecology. 69: 1017-1024. Kerr, G . , C.J . N i x o n and R . W . Matthews. 1992. Silviculture and yield of mixed-species stands: The U K experience. Pp. 35-51 In: M . G . R . Cannel, D . C . Ma lco lm and P . A . Robertson (Eds.). 1992. The ecology of mixed-species stands of trees. Blackwell Scientific Publications. 312p. Keuls, M . , H.J . Over, and C.T. de Wit. 1963. The distance method for estimating densities. Stat. Neerlandica. 17:71-91. Kl inka , K . 1977. Guide for the tree species selection and prescribed burning in the Vancouver Forest District. B . C . Ministry of Forests, Victoria, B . C . 42p. Kl inka , K . and M . C . Feller. 1984. Principles used in selecting tree species for regeneration of forest sites in southwestern British Columbia. For. Chron 60: 77-85. Kl inka , K . , J. Worrall , L . Skoda, and P. Varga. 2000. The distribution and synopsis of ecological and silvical characteristics of tree species of British Columbia's forests. Canadian Cartographies Ltd. , Coquitlam, B C . Pp. 101-108. Krajina, V . J . 1969. Ecology of forest trees in British Columbia, pp 1-146. In Krajina, V . J . , and Brooke, R . C . 1969. Ecology of western North America. Volume 2, Number 1. Mitchell Press Ltd. , Vancouver, B . C . Krajina, V . J . , S. Madoc-Jones and G . Mellor . 1973. Ammonium and nitrate in the nitrogen economy of some conifers growing in Douglas-fir communities of the Pacific Northwest of America. Soi l B i o l . Biochem. 5: 143-147. Latham, P .A . , H .R . Zuuring, and D . W . Coble. 1998. A method for quantifying vertical forest structure. For. Ecol . Manag. 104:157-170. Laverty, T . M . 1992. Plant interactions for pollinator visits: A test of the magnet species effect. Oecologia. 89:502-508. Lawrence, D . B . , R . E . Schoenike, A . Quispel and G . Bonds. 1967. The role of Dryas drummondii in vegetation development following ice recession at Glacier Bay, Alaska, with special reference to its nitrogen fixation by root nodules. J. Ecol . 55: 793-813. Leaphart, C D . and M . A . Grismere. 1974. Extent of roots in the forest soil mantle. J. For. 72: 358-359. 87 Lotwick, H . W . and B . W . Silverman. 1982. Methods of analyzing spatial processes of several types of points. J. Roy. Stat. Soc. B , 44: 406-413. MacLean, D . A . 1996. Forest management strategies to reduce spruce budworm damage in the Fundy model forest. For. Chron. 72:399-405. Major, J. 1977. California climate in relation to vegetation. Pp. 11-74. In: Barbour, M . G . and J. Major (Eds.). 1977. Terrestrial vegetation of California. John Wiley and Sons, Inc. New York, N Y . Man, R. and V . J . Lieffers. 1997. Seasonal variations of white spruce (Picea glauca) and jack pine (Pinus banksiana) saplings. Can. J. Bot. 75:1766-1771. Man , R. and V . J . Lieffers. 1999. Are mixtures of aspen and white spruce more productive than single species stands? For. Chron. 75:505-513. Mathsoftlnc. 1999. S-Plus 2000 Guide to Statistics, Vols. 1 and 2. Data Analysis Products Division, Mathsoft. Seattle, W A . May, M . M . and S. Ahmad. 1983. Host location in the Colorado potato beetle: Searching mechanisms in relation to oligophagy. Pp. 173-199 In: S. Ahmad (Ed.). 1983. Herbivorous insects: Host - seeking behavior and mechanisms. Academic Press, New York, N Y . McNaughton, S.J. 1978. Serengeti ungulates: Feeding selectivity influences the effectiveness of plant defense guilds. Science. 199:806-807. Messier, C. 1993. Factors limiting early growth of western redcedar, western hemlock and Sitka spruce seedlings on ericaceous-dominated clearcut sites in coastal British Columbia. Forest Ecology and Management 60: 181-206. Minore, D . 1979. Comparative autoecological attributes northwestern tree species - a literature review. U S D A Forest Service, General Technical Report PNW-87 . Pacific Northwest Forest and Range Experiment Station, Portland, OR. 72p. Minore, D . 1990. Thuja plicata Donn ex. D Don, western redcedar. Pp. 590-600 In: R . M . Burns and B . H . Honkala (Tech. Coords.). 1990. Silvics of North America: Volume 1. Conifers. Agriculture Handbook 654. U .S . Department of Agriculture, Forest Service, Washington, D C . V o l . l ,675p. Minore, D . and C . E . Smith. 1971. Occurrence and growth o f four northwestern tree species over shallow water tables. U S D A Forest Service, research Note PNW-160. Pacific Northwest Forest and Range Experiment Station, Portland, OR. 9p. Minore, D . and J.C. Zasada. 1990. Acer macrophyllum Pursh, bigleaf maple. Pp. 33-40 In: R . M . Burns and B . H . Honkala (Tech. Coords.). 1990. Silvics of North America: Volume 2. Hardwoods. Agriculture Handbook 654. U . S . Department of Agriculture, Forest Service, Washington, D C . V o l . 2, 877p. 88 Moeur, M . 1993. Characterizing spatial patterns of trees using stem-mapped data. For. Sci. 39: 756-775. Morris, W . G . 1970. Effects of slash burning in overmature stands of the Douglas-fir region. For. Sci . 16:258-270. Myrold , D . D . 1987. Relationship between microbial biomass nitrogen and a nitrogen availability index. Soil Science Society of America Journal 51:1047-1049. Neter, J., Kutner, M . H . , Nachtsheim, C.J . , and Wasserman, W . 1996. Appl ied linear statistical models. Richard D . Irwin Inc., Toronto, O N . , 1408 pp. Oke, T.R. 1987. Boundary layer climates, 2 n d E d . The University Press, Cambridge. 435p. Oke, T. and Hay, J. 1994. The climate of Vancouver, 2 n d Ed . pp 1-71. In: Bovis , M , and Barnes, T. 1994. B . C . Geographical Series, Number 50. University of British Columbia, Department of Geography, Vancouver, B . C . Oliver, C D . and Larson, B . C . 1996. Forest stand dynamics, Update Edition. M c G r a w - H i l l Inc., N e w York, N Y . , 519 pp. Ovington, J.D. and Madgwick, H . A . I . 1957. Afforestation and soil reaction. Journal of Soil Science 8: 141-149. Packee, E . C 1990. Tsuga heterophylla (Raf.) Sarg., western hemlock. Pp. 613-622 In: R . M . Burns and B . H . Honkala (Tech. Coords.). 1990. Silvics of North America: Volume 1. Conifers. Agriculture Handbook 654. U . S . Department o f Agriculture, Forest Service, Washington, D C . V o l . l ,675p. Parker, T. 1979. Natural regeneration of western redcedar in northern Idaho. M . S c . Thesis, University of Idaho, Moscow, ID. 50p. Parker, T. 1986. Ecology of western redcedar groves. Ph.D. Thesis, University of Idaho, Moscow, ID. 187p. Parkinson, J .A. and Al len , S.E. 1975. A wet oxidation procedure for the determination of nitrogen and mineral nutrients in biological material. Communications in Soil Science and Plant Analysis. 6: 1-11. Pfister, C . A . and M . E . Hay. 1988. Associational plant refuges: Convergent patterns in marine and terrestrial communities result from differing mechanisms. Oecologia. 77: 118-129. Pharis, R.P. , R . L . Barnes and A . W . Naylor. 1964. Effects of nitrogen level, calcium level, and nitrogen source on growth and composition of Pinus taeda L . Physiol. Plant. 17: 560-572. Powers, R .F . 1980. Mineralizable soil nitrogen as an index of nitrogen availability to forest trees. Soil Science Society of America Journal 44:1314-1320. 89 Prescott, C . E . and C M . Preston. 1994. Nitrogen mineralization and decomposition in forest floors in adjacent plantations of western red cedar, western hemlock, and Douglas-fir. Canadian Journal of Forest Research 24: 2424-2431. Prescott, C . E . , K . D . Thomas, and G.F. Weetman. 1995. The influence of tree species on nitrogen mineralization in the forest floor: Lessons from three retrospective studies. Pp 59-68 In: Mead, D . J. and Cornforth, I. S. 1995. Proceedings of the Trees and Soil Workshop. Lincoln University Press, Canterbury, N e w Zealand, Agronomy Society of N e w Zealand Special Publication N o . 10. Prescott, C .E . , L . Vesterdal, J. Pratt, K . H . Venner, L . M . de Montigny and J .A. Trofymow. In review. Nutrient availability in forest floors of single species conifer plantations in coastal British Columbia. Canadian Journal of Forest Research. Pritchet, W . L . and R .F . Fisher. 1987. Properties and management of forest soils, 2 n d Ed . John Wiley and Sons, Inc. N e w York, N Y . 494p. Richards, J .H. and M . M . Caldwell . 1987. Hydraulic lift: Substantial nocturnal water transport between soil layers by Artemisia tridentata roots. Oecologia. 73: 486-489. Ripley, B . D . 1976. The second-order analysis of stationary point processes. J. App l . Prob. 13: 255-266. Ripley, B . D . 1977. Modeling spatial patterns (with discussion). J. Roy. Stat. Soc. B 39: 172-212. Roder, D . 1975. A procedure for assessing point patterns without reference to area or density. Prof. Geographer. 27: 432-440. Runge, M . and Rode, M . W . 1991. Effects of soil acidity on plant associations. Pp. 183-199. In Ulr ich , B . , and Sumner, M . E . 1991. Soil acidity. Springer-Verlag, Heidelberg, Germany Rygiewicz, P.T., C S . Bledsoe and R. Zazoski. 1984a. Effects of ectomycorrhizae and solution p H on 15N ammonium uptake by coniferous seedlings. Can. J. For. Res. 14: 885-892. Rygiewicz, P.T., C S . Bledsoe and R. Zazoski. 1984b. Effects of ectomycorrhizae and solution p H on 15N ammonium uptake by coniferous seedlings. Can. J. For. Res. 14: 893-899. S A S Institute Inc. 1989. S A S / S T A T User's Guide, Version 6, 4th Edition, Volumes 1 and 2. Cary, N C , S A S Institute Inc. Schaefer, D . G . and Nikleva, S. N . 1973. Mean precipitation and snowfall maps for a mountainous area of potential urban development, pp 80-89. In 1973. Proceedings of the Western Snow Conference. Grand Junction, Colorado. Simard, S.W., D . A . Perry, M . D . Jones, D . D . Myrold , D . M . Durall and R. Mol ina . 1997. Net transfer of carbon between ectomycorrhizal tree species in the field. Nature. 388: 579-582. 90 Site Productivity Working Group. 1997. Site Index Estimates by Site Series for Coniferous Tree Species in British Columbia. Forest Renewal British Columbia and the B . C . Ministry of Forests, Victoria, B C . Pp. 166 & 175. Soil Classification Working Group. 1998. The Canadian System of Soil Classification. Agriculture and Agri-Food Canada Publication 1646 (Revised). 187pp. Tarrant, R.F . , Isaac, L . A . , and Chandler, R .F . 1951. Observations on litter fall and foliage nutrient content of some Pacific Northwest tree species. Journal of Forestry pp 914-915. Tilman, D . 1988. Plant strategies and the dynamics and structure of plant communities. Princeton University Press, Princeton, N J . Trenbath, B . R . 1974. Biomass productivity of mixtures. A d v . Agron. 26:177-210. Turner, D.P . and Franz, E . H . 1985. The influence of Western hemlock and western redcedar on microbial numbers, nitrogen mineralization, and nitrification. Plant and Soil . 88: 259-267. Upton, G.J .G. , and B . Fingleton. 1985. Spatial Data Analysis by Example. Volume I: Point Pattern and Quantitative Data. John Wiley and Sons. Toronto, O N . 41 Op. Vandermeer, J. 1989. The ecology of intercropping. Cambridge University Press. Cambridge, M A . 237p. Vasek, F . C . and L . J . Lund. 1980. Soil characteristics associated with a primary plant succession on a Mojave Desert dry lake. Ecology. 61:1013-1028. Venables, W . N . , and B . D . Ripley. 1999. Modern Applied Statistics with S-Plus (3 r d Ed.). Springer-Verlag. N e w York, N Y . 501p. Walker, L . R . and F.S. Chapin HI. 1986. Physiological controls over seedling growth in primary succession on an Alaskan floodplain. Ecology. 67:1508-1523. Wang, G . G . , H . Qian, and K . Kl inka . 1994. Growth of Thuja plicata seedlings along a light gradient. Can. J. Bot. 72:1749-1757. Wang, J.R. 1997. Competition between paper birch and Douglas-fir in two different biogeoclimatic zones of British Columbia. Ph.D. Thesis, University of British Columbia, Vancouver, B . C . Waring, R . H . and W . H . Schlesinger. 1985. Forest ecosystems: Concepts and management. Academic Press, N e w York, N Y . 340p. Weetman, G.F. 1962. M o r humus: A problem in a black spruce stand at Iroquois Falls, Ontario. Pulp Pap. Res. Inst. Can., Woodl . Res. Index 130. 18p. Weetman, G.F. , M . A . Radwan, J. K u m i and E . Schnorbus. 1988. Nutrition and fertilization of western redcedar. Pp. 47-60 In: N . J . Smith (Ed.). 1988. Western redcedar—does it have 91 a future? Proc. Conf., Vancouver, B C . University of British Columbia, Faculty of Forestry. Weetman, G.F. , R. Fournier, E . Schnorbus-Panozzo, and J. Barker. 1990. Post-burn nitrogen and phosphorus availability of deep humus soils in coastal British Columbia cedar/hemlock forests and the use of fertilization and salal eradication to restore productivity. Pp. 451-459 In: Gessel, S.P., D.S . Lacate, G.F . Weetman, and R.F . Powers (Eds.). 1990. Sustained productivity of forest soils. Proceedings of the 7 t h North American Forest Soils Conference, University of British Columbia, Faculty of Forestry Publication, Vancouver, B C . Wiley, K . N . 1976. Site index and yield of western hemlock. Pp. 228-235 In: W . A . Atkinson and R.J . Zasoski (Eds.). 1976. Western hemlock management conference. Proc. Conf , Seattle W A . University of Washington, College of Forest Resources. Will iams, K . , M . M . Caldwell and J .H. Richards. 1993. The influence of shade and clouds on water potential: The buffered behavior of hydraulic lift. PI. Soi l . 157: 83-95. Wooldridge, D . D . 1961. Environmental factors related to growth and management of western hemlock Tsuga heterophylla (Raf.) Sarg. Ph.D. Thesis, University of Washington, Seattle, W A . 241p. 92 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0099569/manifest

Comment

Related Items