UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

In situ wood quality assessment in interior spruce Fundová, Irena 2012

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

Item Metadata

Download

Media
24-ubc_2012_fall_fundova_irena.pdf [ 1.91MB ]
Metadata
JSON: 24-1.0073230.json
JSON-LD: 24-1.0073230-ld.json
RDF/XML (Pretty): 24-1.0073230-rdf.xml
RDF/JSON: 24-1.0073230-rdf.json
Turtle: 24-1.0073230-turtle.txt
N-Triples: 24-1.0073230-rdf-ntriples.txt
Original Record: 24-1.0073230-source.json
Full Text
24-1.0073230-fulltext.txt
Citation
24-1.0073230.ris

Full Text

IN SITU WOOD QUALITY ASSESSMENT IN INTERIOR SPRUCE by Irena Fundová Ing., Czech University of Life Sciences Prague, 2005 and 2006 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Forestry)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  September 2012 © Irena Fundová, 2012  Abstract Wood quality is of great importance for end-users but the concurrent conventional selection approaches targeted for growth rate are often associated with its decrease. The inclusion of wood quality into breeding programs requires finding a fast and inexpensive method that is capable of providing reasonably accurate estimates of wood quality attributes on standing trees without their significant injury during data collection. In the present study, wood density as the best single predictor of wood quality was estimated through drilling resistance using Resistograph IML F300; dynamic modulus of elasticity (MoEd) representing an important wood mechanical parameter was calculated from sound velocity measured by Director ST300TM. Twenty-five open-pollinated families of 37- and 38-year-old interior spruce (the complex of white spruce (Picea glauca (Moench) Voss), Engelmann spruce (Picea engelmannii Parry), and their hybrids) growing on three sites (1,146 trees) were included in this study. Narrow-sense heritabilities and phenotypic and genetic correlations were estimated for growth (height, diameter at breast height, and volume) and wood quality attributes (overall x-ray density, x-ray density of the first 15 rings, resistograph-based density, earlywood density, latewood density, latewood proportion, acoustic velocity, and MoEd). Phenotypic and genetic correlations were strongly related (correlation of 0.85 based on the Mantel test).  As anticipated for interior spruce, growth traits were negatively  correlated with wood density, but surprisingly not with MoEd. It suggests that in interior spruce selection for rapid growth would result in wood density reduction while MoE would remain unaffected, pointing at a low usefulness of MoE’s inclusion among the selection criteria. The Resistograph provided a reliable estimate of wood density of the whole profile (0.59 and 0.84 for phenotypic and genetic correlations, respectively) as well as of the first 15 rings (0.60 and 0.95, respectively) and thus demonstrated its suitability for testing young trees. Although the heritabilities for the wood quality attributes estimated by x-ray were mainly moderate (0.17–0.26), the heritability of resistograph-based density was low (0.15). The heritabilities for other traits were low to moderate. The Resistograph appears to be a reliable non-destructive tool for in situ wood density assessment in interior spruce.  ii  Table of Contents Abstract .................................................................................................................................... ii Table of Contents ................................................................................................................... iii List of Tables ........................................................................................................................... v List of Figures ......................................................................................................................... vi List of Tree Species ............................................................................................................... vii List of Abbreviations ........................................................................................................... viii Acknowledgements ................................................................................................................ ix Dedication ................................................................................................................................ x 1 Introduction ........................................................................................................................ 1 1.1 Interior spruce ............................................................................................................................ 1 1.2 Forest tree improvement ............................................................................................................ 3 1.2.1 Improvement of interior spruce in British Columbia ...................................................... 4 1.2.2 Interior spruce Seed Planning Zones ............................................................................... 5 1.3 Wood quality ............................................................................................................................. 6 1.3.1 Factors influencing wood quality .................................................................................... 8 1.3.1.1 Juvenile wood ..................................................................................................... 8 1.3.1.2 Compression wood, knots, decay, and extractives ............................................. 9 1.3.2 Relationship between wood density and growth ........................................................... 10 1.3.2.1 Correlations between wood density and growth traits ...................................... 12 1.3.2.2 Heritability of wood density ............................................................................. 13 1.3.3 Early selection ............................................................................................................... 14 1.3.4 Breeding and wood quality............................................................................................ 14 1.3.5 Variability of wood density ........................................................................................... 16 1.3.6 Modulus of elasticity and modulus of rupture ............................................................... 17 1.4 Non-destructive wood quality assessment ............................................................................... 20 1.4.1 Resistograph IML F300................................................................................................. 22 1.4.1.1 Description ....................................................................................................... 23 1.4.1.2 Function ............................................................................................................ 23 1.4.1.3 Comparison of in situ and laboratory estimates ............................................... 24 1.4.1.4 Factors causing bias of estimates...................................................................... 24 1.4.1.5 Impact of drilling on trees’ health .................................................................... 25 iii  1.4.2 Director ST300 .............................................................................................................. 25 1.4.2.1 Description and function .................................................................................. 26 1.4.2.2 Comparison of in situ and laboratory estimates ............................................... 27 1.4.2.3 Acoustic resonance ........................................................................................... 28 1.4.2.4 Factors causing bias of estimates...................................................................... 29 1.5 Objectives ................................................................................................................................ 30  2 Materials and Methods .................................................................................................... 31 2.1 Trial description ....................................................................................................................... 31 2.2 Data collection ......................................................................................................................... 32 2.3 Statistical analysis.................................................................................................................... 34  3 Results and Discussion ..................................................................................................... 37 3.1 Descriptive statistics ................................................................................................................ 37 3.2 Components of variance .......................................................................................................... 38 3.3 Heritability estimates ............................................................................................................... 39 3.4 Phenotypic and genetic correlations ........................................................................................ 45  4 Conclusion ......................................................................................................................... 58 References .............................................................................................................................. 60  iv  List of Tables Table 2.1 Geographical coordinates of the progeny trials .................................................... 31 Table 2.2 Response variables and their transformations ....................................................... 35 Table 3.1 Descriptive statistics for all measured variables ................................................... 38 Table 3.2 Variance components, percentage of variance explained by a factor or interaction, and heritabilities for growth traits ........................................................................ 41 Table 3.3 Variance components, percentage of variance explained by a factor or interaction, and heritabilities for main wood quality attributes .............................................. 41 Table 3.4 Variance components, percentage of variance explained by a factor or interaction, and heritabilities for other wood characteristics .................................................. 42 Table 3.5 Comparison of heritabilities for growth traits with other studies ......................... 43 Table 3.6 Comparison of heritabilities for wood density and its components with other studies ............................................................................................................................ 44 Table 3.7 Phenotypic and genetic correlations between studied traits .................................. 50 Table 3.8  Comparison of phenotypic and genetic correlations between overall  density and its components with other studies ........................................................................ 51 Table 3.9 Phenotypic and genetic correlations between growth traits and wood density for spruce species ....................................................................................................... 52 Table 3.10 Phenotypic and genetic correlations between overall x-ray density and growth traits for the three studied sites ................................................................................... 53 Table 3.11 Phenotypic and genetic correlations between density and standing-tree acoustic variables (acoustic velocity, squared acoustic velocity, and ToF) ........................... 53  v  List of Figures Figure 1.1 The natural distributions of white spruce (Picea glauca (Moench) Voss) and Engelmann spruce (P. engelmannii Parry)......................................................................... 2 Figure 1.2 Map of Seed Planning Zones for interior spruce in British Columbia .................. 6 Figure 2.1 Map of trials’ location ......................................................................................... 32 Figure 2.2 Depiction of tree sampling .................................................................................. 33 Figure 3.1  Relationship between overall x-ray density and earlywood density,  latewood density, and latewood proportion ............................................................................ 54 Figure 3.2 Relationship between resistograph-based density and overall x-ray and 15-year x-ray density .............................................................................................................. 55 Figure 3.3 Relationship between resistograph-based density and MoEd .............................. 55 Figure 3.4 Relationship between acoustic velocity and overall x-ray and 15-year x-ray density............................................................................................................................ 56 Figure 3.5 Relationship between acoustic velocity and resistograph-based density ............ 56 Figure 3.6 Visualization of the relationships among traits based on their phenotypic correlations .............................................................................................................................. 57 Figure 3.7 Visualization of the relationships among traits based on their genetic correlations .............................................................................................................................. 57  vi  List of Tree Species black spruce  Picea mariana (Mill.) BSP  Douglas-fir  Pseudotsuga menziesii (Mirb.) Franco  Engelmann spruce  Picea engelmannii Parry ex Engelm.  interior spruce  common name for white spruce, Engelmann spruce, and their hybrids  jack pine  Pinus banksiana Lamb.  loblolly pine  Pinus taeda L.  maritime pine  Pinus pinaster Ait.  Norway spruce  Picea abies (L.) H. Karst.  ponderosa pine  Pinus ponderosa Douglas ex C. Lawson  radiata pine  Pinus radiata D. Don  red spruce  Picea rubens Sarg.  Scots pine  Pinus sylvestris L.  Sitka spruce  Picea sitchensis (Bong.) Carr.  slash pine  Pinus taeda L.  subalpine fir  Abies lasiocarpa (Hook.) Nutt.  western hemlock  Tsuga heterophylla (Raf.) Sarg.  western redcedar  Thuja plicata Donn ex D. Don  white spruce  Picea glauca (Moench) Voss  yellow cypress  Callitropsis nootkatensis (D. Don) Orsted ex D.P. Little  vii  List of Abbreviations a.s.l.  Above sea level  dbh  Diameter at breast height (1.3 m)  FSP  Fiber saturation point  MoE  Modulus of elasticity  MoEd  Dynamic modulus of elasticity  MoEs  Static modulus of elasticity  MoR  Modulus of rupture  NIR  Near-infrared  PDA  Personal digital assistant  SPU  Seed planning unit  SPZ  Seed planning zone  ToF  Time of flight  viii  Acknowledgements I would like to thank everyone who helped me succeed with my masters program at UBC. First of all, I would like to thank my supervisor Yousry El-Kassaby for his support, encouragement, scientific guidance, and sharing his life wisdoms with me. I appreciate that Yousry was always available for me regardless of whether he was in his office or thousands of kilometers away. I am very grateful for the opportunity to be his graduate student and receive education at one of the best forestry faculties in the world. I thank my supervisory committee members Simon Ellis and Michael Stoehr for reviewing my thesis and helping me improve it as well as Christopher Chanway for his comments and suggestions. I would like to thank Valerie LeMay, Michael Whitlock, Simon Ellis, and Yousry El-Kassaby for amazing courses which helped me get a stronger background particularly in statistics and genetics. I am grateful to the Ministry of Forests, Lands and Natural Resources Operations for allowing me to access their progeny trials, namely to Barry Jaquish for facilitating the experimental sampling and to Bonnie Hooge for her assistance during field data collection. I thank Jaroslav Klapste for introducing the statistical package ASReml. My special thanks go to my husband Tomas Funda who was not only a loving and supporting husband, but also a knowledgeable colleague and mentor.  I much appreciate that he  accompanied me during my fieldwork and bravely faced raids of hungry mosquitoes while assessing trees. I would also like to acknowledge my brother Marek Permedla who assisted me with extracting tree cores during his summer holidays. I thank my parents for their remote moral support and all friends in Canada and elsewhere for remaining my friends despite having little time to respond to their emails. Last but not least, I would like to thank my lovely daughter Lucie who was always here to cheer me up! This research project was funded by the Canadian ForValueNet, the Natural Sciences and Engineering Research Council of Canada Strategic Network, whose contribution is greatly appreciated. I am also grateful for financial support from UBC, George S. Allen Memorial Scholarship, Donald S. McPhee Fellowship and Mary and David Macaree Fellowship.  ix  Dedication  To my little Lucie  x  1 Introduction 1.1  Interior spruce  Five spruce species (black, white, red, Engelmann, and Sitka spruce) are native to Canada (Farrar 1995), four of which grow in British Columbia (Roche 1969b). White spruce (Picea glauca (Moench) Voss), Engelmann spruce (P. engelmannii Parry ex Engelm.), and their hybrids are collectively called interior spruce (Coates et al. 1994), because they are difficult to distinguish just based on their morphological differentiation (Khasa and Dancik 1996). Interior spruce was the most commercially important species for the forest industry in British Columbia until the mid 80’s when it was surpassed by lodgepole pine (Coates et al. 1994). Out of the 55 million ha of British Columbia’s forested land, 22% is covered by spruce, a large proportion of which represents white spruce (Forintek 2007). Around 2,300 million m3 of spruce timber was harvested in 2007 (B.C. Ministry of Forests, Mines and Lands 2010). White spruce is the northernmost distributed spruce species and one of the most widely ranging conifers in North America (Figure 1.1a) (Musil 2003). It has a transcontinental distribution, from the Atlantic Ocean (Newfoundland and Labrador) west across Canada to the Pacific Ocean (south-western Alaska) and from the northern tree limit (69N in western North America) to the north-eastern part of the United States (44N). In British Columbia, it grows naturally in most of the province, except for areas less than ca 100 km from the Pacific coast and 200 km from the border with Washington State (Sutton 1969). As to the vertical range, white spruce occurs between sea level and ca 1500 m (Roche 1969b). The natural distribution of Engelmann spruce is much more limited and scattered (Figure 1.1b) as it is concentrated in higher elevations, primarily in the Rocky, Coast, and Cascade Mountains. The range extends from the northern part of central British Columbia and the eastern part of south Alberta south to New Mexico and Arizona and west to California.  The vertical  distribution varies from ca 600 m a.s.l. in the north to ca 3000 m a.s.l. in the central and south parts of Rocky Mountains (Coates et al. 1994). The two spruce species commonly undergo introgressive hybridization at mid-elevations (600–1500 m) where their distributions overlap (Dobbs et al. 1976; Fowler and Roche 1976).  1  Both white and Engelmann spruce are evergreen conifers with narrow spire-like crowns and shallow root systems. They are medium-sized trees, but Engelmann spruce generally reaches greater dimensions (Coates et al. 1994; Farrar 1995). In practice, the two species are usually distinguished based on their cone scale and twig characteristics (Garman 1957); however, it can with certainty be verified using species-specific genetic markers (Bousquet et al. 2007; Khasa and Dancik 1996).  (a)  Figure 1.1  (b)  The natural distributions of (a) white spruce (Picea glauca (Moench) Voss), and  (b) Engelmann spruce (P. engelmannii Parry) (Little 1971)  The wood structure of white and Engelmann spruce cannot be visually distinguished from each other (Perem et al. 1981); however, wood of Engelmann spruce is a little harder and stronger (Forintek 2007). The light, yellowish-brown wood of interior spruce is moderately soft with an average air dry density of 0.40 g·cm-3. There is little or no difference in colour between sapwood and heartwood. The texture is moderately fine with fairly even grain and distinct growth rings (Hoadley 2000). The transition between the much wider earlywood and narrower latewood is gradual (Isenberg 1943). The commercial utilization of white spruce started with the shipbuilding industry in New England (Ostrander 1974). Nowadays, interior spruce is mainly used as lumber and for pulp and paper production (Forintek 2007). Naturally long fibers, light colour, low microfibril angle, and low contents of thick-walled latewood fibers and resin make it an excellent base  2  material for the pulp industry (Ostrander 1974; Taylor et al. 1982). Its pulp is used for manufacturing paper with smooth surface and good printabilities (Duchesne and Zhang 2004). Since interior spruce wood has a moderate density, it is moderately stiff, tough, and hard (Alden 1997). Unusually strong wood for its weight (Forintek 2007) is employed mainly for light building constructions (Taylor et al. 1982) such as roofing, sheathing, or framing (Forintek 2007). Among softwood species, it is valued for its resilience, which makes it a favoured material for scaffolding planks and similar uses (Winters 1951); however, its resistance to bending and longitudinal compression is limited (Ostrander 1974). It dries easily, with moderate shrinkage, and is stable after drying (Alden 1997).  Its  dimensional stability along with excellent gluing ability make interior spruce suitable for prefabrication (Forintek 2007). Interior spruce wood is easily worked. It is straight, even grained, soft, and finishes with a smooth satin-like surface (Alden 1997). However, it is not considered very durable in situations favourable to decay and is difficult to penetrate with preservatives (Perem et al. 1981). Owing to its superior resonance properties it is used for sounding boards for a variety of musical instruments. The wood is almost tasteless and odorless when dried and, therefore, it is suitable for food containers (Forintek 2007).  1.2  Forest tree improvement  Forest tree improvement comprises of three phases: tree breeding, production, and silviculture. It is based on exploiting natural genetic variation existing in forest tree species using genetic principles in order to improve traits of economic interest. The goal is not only to attain well-adapted, high-quality trees, but also to retain a sufficient level of genetic diversity so that trees’ resilience to changing environments is maintained (Zobel and Talbert 1984). The first step of each tree improvement program consists of phenotypic selection of candidate trees, called plus trees. The most commonly applied selection criteria are based on growth performance. They can be volume and stem straightness as in a maritime pine breeding program in France (Bouffier et al. 2009; Bouffier et al. 2008b; Pot et al. 2002), growth rate as in a Sitka spruce breeding program in Great Britain, where stem form and branching are controlled by silvicultural practices (Hubert and Lee 2005), or simply just  3  height as a proxy to volume. Other traits like wood properties or biotic and abiotic stress resistance are also of interest. Selection for multiple traits is sometimes implemented (Neale and Kremer 2011). Phenotypic selection for quantitative traits can be supplemented by genetic screening of the candidate parents, from which a smaller subset will be finally selected for further breeding based on their pooled genetic profiles (Hosius et al. 2000). The next step involves controlled crosses between the selected individuals, which is followed by testing their progenies so that the candidate parents’ genetic quality can be inferred. Since the breeding phase is a time-consuming and labor-intensive activity, a novel concept known as breeding without breeding was proposed (El-Kassaby et al. 2011a; El-Kassaby and Lstiburek 2009). This approach is capable of shortening the breeding cycle via posterior assemblage of naturally-occurred crosses and subsequent estimation of parental breeding values. The entire process results in the production phase which comprises production plantations, either seed orchards for the production of high genetic quality seed or stool beds for vegetative propagation such as in yellow-cedar. The material obtained during the production phase then serves as a means for the establishment of improved forest stands (Eriksson et al. 2006). Silviculture is a complex of activities that includes forest establishment, fertilization, thinning, pruning, etc. It helps the improved trees obtained from previous breeding and production stages to achieve even better performance and economic gain in forest stands (Jozsa and Middleton 1994; Nyland 1996). 1.2.1  Improvement of interior spruce in British Columbia  The tree improvement program with interior spruce in British Columbia was initiated in two phases. The first phase, comprising regions of Prince George, Smithers, and East Kootenay, started in the mid 1960s as a response to an urgent need for seed supply (Jaquish et al. 2009). It was a consequence of an extensive timber harvest in interior British Columbia along with inconsistent natural regeneration of interior spruce, especially in higher-elevation areas (Jaquish 2009).  However, a better understanding of the interior spruce genetics was  necessary. The first comprehensive study on this topic was provided by Roche (1969a) who  4  revealed an exceptional growth potential of two provenances – Birch Island and Horsefly Lake. Gyula Kiss was the first pioneer tree breeder who built a solid base for interior spruce improvement in British Columbia (Kiss 1967 in Jaquish 2009).  Trees were primarily  selected based on visual assessment of their size (height and diameter) and secondarily by stem form and branching (Kiss 1976 in Xie and Yanchuk 2002). Later, resistance to white pine terminal weevil (Pissodes strobi Peck) also became a more important trait (Kiss and Yeh 1988; Kiss and Yanchuk 1991). Clone banks and seed orchards using grafts from selected trees as well as their open-pollinated progeny tests were established (Jaquish 2009). The second phase covering other important regions began in mid 1970s. It initially focused just on quick production of large quantities of unimproved seed because progeny testing did not commence until 1984 (Jaquish 2009). Currently, 50–90 million seedlings produced from improved, first-generation seed orchards are planted every year. The second-generation of full-sib progeny tests is in progress (Jaquish et al. 2011). 1.2.2  Interior spruce Seed Planning Zones  Seed Planning Zones (SPZ) are geographic areas that follow both administrative and biogeoclimatic units (Xie 2003). In order to prevent maladaptation, special emphasis is given to homogeneous environmental conditions within delineated zones (Xie and Yanchuk 2002). Each SPZ is further divided into Seed Planning Units (SPU) according to altitude. SPZs and their SPUs are important for tree breeding planning as well as for seed production (Tree Improvement Branch 2012a). The first SPZs from the 1940s covered just a small part of British Columbia: Vancouver Island and the south coast of the mainland. They were extended into 67 SPZs within eight geographic regions in the 1970s following the European model, which prioritized local seed. Nevertheless, as the knowledge of the interior spruce’s genetics and biogeoclimatic characteristics increased, the number of SPZs kept reducing. The last reduction was done in 1996 and resulted in six SPZs (Bukley Valley, East Kootenay, Nelson, Prince George, Peace River, and Thompson Okanagan) and a number of transition zones (see Figure 1.2) (Jaquish 2009).  5  Figure 1.2 Map of Seed Planning Zones for interior spruce in British Columbia (Tree Improvement Branch 2012b)  1.3  Wood quality  The most widely used definition of wood quality was proposed by Mitchell (1961): “Wood quality is the resultant of physical and chemical characteristics possessed by a tree or a part of a tree that enable it to meet the property requirements for different end products.” Wood density is referred to as the best single predictor of wood quality (Armstrong et al. 1984; Zobel and Jett 1995; Zobel and Van Buijtenen 1989) and is therefore often assumed to be equivalent to wood quality (Zhang 2003). It is a reliable and easy-to-measure trait, which significantly affects wood suitability for different end-uses (Pot et al. 2002). In addition to that, wood density can be manipulated by genetic improvement and silvicultural practices (Zobel and Van Buijtenen 1989). It is well correlated with other wood properties such as strength or stiffness. High density is, however, more appropriate for structural lumber, while  6  lower density is better for pulp and paper manufacturing (Saranpää 2003). Although higher wood density is associated with greater pulp yield (Armstrong et al. 1984; Hatton and Cook 1992; Ivkovich and Koshy 2002; Zobel and Jett 1995), higher tear strength, or beating resistance, some pulp quality traits such as folding endurance and tensile and burst strength decrease with increased wood density (Zhang and Gingras 1998). Lower-density wood consists of more collapsible fibers, which bond together better (Jozsa and Middleton 1994), and results in denser, less bulky, smoother, softer, and less porous sheet (Wimmer et al. 2002).  Density positively influences wood machinability, particularly boring, turning,  shaping, and mortising (Hernández et al. 2001). However, wood with higher density has higher weight for pieces with equivalent dimensions and thus its manipulation is more labor demanding (Rozenberg and Cahalan 1997). Wood density is influenced by a complex of factors such as size of wood cells and thickness of their walls, proportion of earlywood and latewood, and lignin content (King et al. 1988). Average wood density in temperate softwood species essentially relies on density of earlywood and latewood and their proportions (Louzada and Fonseca 2002). It is also affected by ring width, tree age, and geographical origin (Corriveau et al. 1987). Wood density points well at fiber morphology as well as chemical composition of wood, but it poorly indicates fiber strength and heterogeneity of wood (Pot et al. 2002). It should be noted that wood is a hygroscopic material and thus changes its weight and volume according to surrounding air humidity. It is therefore important to know the moisture content prior to wood density estimation (Jozsa and Middleton 1994). Wood density is not the same trait as wood specific gravity; however, they are sometimes used interchangeably (Williamson and Wiemann 2010). Density (kgm-3) is mass per unit volume, both of which are measured at the same moisture content. Specific gravity (SG) (Equation 1.1), also known as relative density or basic density, is a unitless variable calculated as the ratio of a material’s density and density of water (1,000 kgm-3). There are different wood specific gravity measures according to the wood state in which the volume is determined – basic (green), air-dry, and oven-dry (true). Green volume as well as air-dry volume is influenced by moisture content (Williamson and Wiemann 2010).  7  [1.1]  1.3.1  Factors influencing wood quality  The current forest tree improvement strategies commonly focus solely on volume production. Improved trees grow faster, which means that they reach a merchantable size sooner; however, this approach is associated with a significant decrease in wood quality (Kennedy 1995; Kliger et al. 1995; Perstorper et al. 1995a, b; Petty et al. 1990; Zhou and Smith 1991), affecting not only mechanical properties of lumber but also pulp production efficiency (Pot et al. 2002). While the volume-oriented improvement can substantially shorten the rotation age, it might at the same time result in a higher proportion of juvenile wood (Bouffier et al. 2009; Kennedy 1995; Matziris and Zobel 1973). This is supported by a strong negative correlation (-0.84) between juvenile wood proportion and stand age reported by Zobel et al. (1965) in Matziris and Zobel (1973). Nevertheless, it is noteworthy that one would find very little disparity in juvenile wood proportion between improved plantation trees and unimproved slower natural-grown trees at the same age (Zobel and Sprague 1998). 1.3.1.1  Juvenile wood  Juvenile wood is formed from the pith outwards and usually encompasses the first 10 to 20 growth rings (Choi 1986; Zobel and Sprague 1998). Compared to mature wood, it has a different anatomical structure and properties. An extremely important and thus the most frequently cited difference is a lower density (Bouffier et al. 2009; Zobel and Sprague 1998; Zobel and Van Buijtenen 1989), which is primarily a consequence of a lower latewood proportion. Moreover in comparison to mature wood, juvenile wood has shorter longitudinal tracheids, thinner cell walls, lower cellulose contents, higher hemicelluloses and lignin contents, greater fibril angles and more spiral grain (Zobel and Sprague 1998), more compression wood (Kretschmann 1998), and more knots (Choi 1986). These characteristics are linked with lower mechanical strength, higher longitudinal shrinkage (Kretschmann 1998), higher propensity to warp during drying (Choi 1986), and lower pulp yield when using kraft pulping (Zobel and Sprague 1998) as well as other chemical processes (Jozsa and Middleton 1994).  8  Nevertheless, lower density of juvenile wood is not always a rule; it is species specific which also applies to the wood density pith-bark gradients (Jozsa and Middleton 1994). Matziris and Zobel (1973) mentioned that juvenile wood in loblolly pine sometimes has a higher density than mature wood. Also the juvenile wood of yellow cypress, western red cedar, and subalpine fir was found to have higher density compared to density of the mature wood (Jozsa and Middleton 1994).  Higher density of juvenile wood was also reported for  white/interior spruce; the highest density adjacent to the pith sharply decreased to its minimum (between 10–25 years) and then slightly increased or remained more or less constant (Corriveau et al. 1991; Jozsa and Middleton 1994; Taylor et al. 1982). A similar pattern of rapid decrease was observed in interior spruce for latewood percentage (Ivkovich et al. 2002a). Moreover, specific gravity in white spruce was found to increase in the stem upwards, which is in contrast with species from genera Larix, Pseudotsuga, Pinus, and Abies (Panshin and de Zeeuw 1980). 1.3.1.2  Compression wood, knots, decay, and extractives  Studying white spruce, Wang and Micko (1984) mentioned wood specific gravity to be higher close to the pith and to increase upwards along the stem. However, they found this to be true on some sites only; furthermore, the higher specific gravity close to the pith was never as high as that of mature wood. The contradicting results could be explained by environmental differences in which the studied trees grew.  Trees exposed to harsh  conditions often have to bend, which results in the formation of denser reaction wood close to the pith (Wang and Micko 1984). For example, Ivkovich et al. (2002a), examining increment cores of 22-year-old interior spruces taken above the sixth spindle of branches, found only 9–11 outermost rings of the total of 15–17 rings to be free of compression wood. Wang and Micko (1984) also pointed out that upper parts of a tree exhibit greater flexibility and at the same time are more affected by weather conditions compared to lower parts; therefore, they usually contain a higher portion of reaction wood, which contributes to longitudinal differences in the specific gravity. The compression wood (the reaction wood in softwoods) is an abnormal tissue formed underneath a bending stem (Jozsa and Middleton 1994; Keith and Kellogg 1981); nevertheless, its occurrence in juvenile wood of fast-growing trees in terms of height has also  9  been reported (Savidge 2003). The main factors activating the formation of compression wood are gravity and an imbalance of auxins.  In comparison with normal wood, the  compression wood is characterized by wider rings, a higher proportion of latewood, a higher lignin content, and a larger microfibril angle (Jozsa and Middleton 1994). The density of compression wood is also higher; however, it does not result in higher strength as one would expect (Keith and Kellogg 1981; Kennedy et al. 1968; Lindström et al. 2004). Knots have similar effects on strength as the compression wood. Their density is higher than that of the surrounding wood but their presence causes strength reduction because of grain deviation (Larsen 2001). In softwoods, the proportion of knots is about 0.5–2.0%; however, the amount of affected wood is much greater (Chauhan et al. 2006a). Decay is serious wood quality deterioration caused by fungi digesting the wood. There are two main types of decay: brown rot and white rot. Fungi causing the brown rot hydrolyze hemicelluloses and cellulose, while fungi causing white rot can also break down lignin (Dolenko et al. 1981). Narrow-sense heritabilities for brown and white rot resistance of white spruce were estimated to be 0.21 and 0.27, respectively (Yu et al. 2003). Another factor influencing wood density is the presence of extractives (Lindstrom 1996). Extractives are extraneous chemicals formed during sapwood-heartwood transition. Depending on species, their content in wood was reported to be up to 20% (Walker 2006); however, in white spruce it was found to be much lower, only 0.2–3.2% (Wang and Micko 1984). Extractives may contribute to wood’s natural resistance to decay and insects (Jozsa and Middleton 1994) and they also favorably influence wood density and mechanical properties including modulus of elasticity and modulus of rupture (Grabner et al. 2005). Nevertheless, a low negative correlation between extractive content and specific gravity (-0.18) was reported for white spruce (Wang and Micko 1984), suggesting that extractives may have little impact on wood quality in this species. 1.3.2  Relationship between wood density and growth  Generally, increased growth rate is associated with a decrease in wood density (Chang and Kennedy 1967; Corriveau et al. 1991; Corriveau et al. 1987; Park et al. 2012; Taylor et al. 1982; Wang et al. 1985; Zhang 1995), affecting both the juvenile and mature wood  10  (Corriveau et al. 1991). Radial growth was reported to be a stronger factor than height growth in affecting wood density (Vargas-Hernandez and Adams 1991; Zhang and Morgenstern 1995). The overall wood density reduction in fast-growing trees is then caused by the decrease in both earlywood and latewood densities (Bouffier et al. 2009). According to Zhu et al. (2008) only latewood density decreases whilst earlywood density increases which results in a more uniform overall density. However, opposite results to Zhu et al. (2008) were obtained previously by Zhang and Morgenstern (1995). Furthermore, earlywood width increases proportionally with increasing ring width, but the denser latewood width increases only slightly (Zhang 1995; Zhang et al. 1996). Therefore, latewood proportion and in turn the overall density is reduced (Bouffier et al. 2009). This seems to be true for conifer genera Picea and Abies but not for others such as Larix or Pseudotsuga (Zobel and Van Buijtenen 1989). In the latter two genera, latewood width increases at the same rate as the ring width; therefore, the overall wood density is not affected by growth rate because the latewood proportion remains more or less the same (Zhang 1995). A decrease in specific gravity/wood density as a consequence of fast growth was found in white spruce (Chang and Kennedy 1967; Park et al. 2012) and Sitka spruce (Brazier 1970). Studying 15-year-old maritime pines in France, Bouffier et al. (2009) discovered a significant decrease in wood quality in populations improved for volume production and stem form compared with an unimproved population. In contrast, a similar wood density was observed in volume-improved and unimproved maritime pine trees in western Australia (Hill 2000). Li and Wu (2005) reported a slight decrease in wood density in radiata pine improved for growth rate and stem form. Zobel and Jett (1995) reviewed several studies on different species reporting no impact of growth-based selection on wood density. Bouffier et al. (2009) concluded that wood density is either not at all or just slightly affected by breeding for growth traits, regardless of species. They supported this statement by a low correlation between wood density and growth traits as well as by low wood density heritability. This explanation is, however, not in accordance with many other studies (see Section 1.3.2.1).  11  1.3.2.1  Correlations between wood density and growth traits  The relationship between wood density and growth traits has been studied in many species; however, the results seem to be inconsistent. For example, in a review by Zobel and Van Buijtenen (1989) more than one half of 58 research papers dealing with hard pines reported no relationship between wood density and growth rate and four of them even found a positive relationship. On the other hand, a negative relationship seems to be the norm for other conifers (Zobel and Van Buijtenen 1989), including the genus Picea (Rozenberg and Cahalan 1997; Zobel and Jett 1995). Significant positive correlations between specific gravity and growth traits were reported for loblolly pine by Matziris and Zobel (1973); phenotypic/genetic correlation between specific gravity on one side and height, volume, and diameter on the other side were estimated to be 0.27/0.23, 0.24/0.46, and 0.23/0.02, respectively. Also for maritime pine, Louzada (2003) published positive correlations between ring density and ring width, which were, moreover, consistent over the first thirteen years (phenotypic 0.03–0.18, genetic 0.10–0.32). However, non-significant or weak genetic correlations between wood density and growth traits for the same species were reported by Bouffier et al. (2009). Non-significant or weak phenotypic and/or genetic correlations were also reported for Scots pine (Hannrup et al. 2000), radiata pine (Nicholls et al. 1980), Douglas-fir (Abdel-Gadir et al. 1993), or black spruce (Zhang 1998; Zhang and Morgenstern 1995; Zhang et al. 1996). On the other hand, negative relationship between wood density and growth rate was found in radiata pine (-0.80) (Li and Wu 2005), western hemlock (R2 = 0.39) (DeBell et al. 1994), black spruce (ranging between -0.26 and -0.41) (Zhang and Morgenstern 1995), or Douglas-fir (R2 = 0.83) (McKimmy and Campbell 1982), (-0.53) (King et al. 1988), (ranging between -0.19 and -0.63) (VargasHernandez and Adams 1991), (ranging between -0.16 and -0.19) (Ukrainetz et al. 2008). Focusing on interior spruce (or just white spruce), Yanchuk and Kiss (1993) reported negative phenotypic (-0.40, -0.46) and non-significant genetic (0.00 ± 0.28, 0.08 ± 0.41) correlations for wood density versus height and diameter, respectively. Lenz et al. (2010) found very strong genetic correlations between wood density and height (earlywood -0.72 ± 0.17, latewood -1.07 ± 0.08) whereas phenotypic correlations for the two traits were low or non-significant.  Wang and Micko (1984) published a significant phenotypic  12  correlation for specific gravity and radial growth rate (-0.42).  Significant phenotypic  (genetic) correlations between relative density on one side and height, diameter, and volume on the other side were estimated to be -0.17 (0.35 ± 0.25), -0.36 (-0.32 ± 0.27), and -0.33 (-0.26 ± 0.26), respectively (Corriveau et al. 1991).  Moreover, significant negative  correlations between ring width and juvenile and mature wood density (-0.24 and -0.49, respectively) were published (Corriveau et al. 1987).  However, Taylor et al. (1982)  investigated profiles of ten interior spruce trees at different heights and found negative but in most cases non-significant correlations between wood density and growth rate. The relationship between wood density and growth rate seems to fluctuate, to some extent, with environment and genotype (Zhang 1995; Zhang et al. 1996). Significant differences in the relationship between the two traits among sites and/or genotypes were found in various species including interior spruce (Ivkovich et al. 2002a), black spruce (Zhang et al. 1996), Norway spruce (Rozenberg and VandeSype 1996), and Douglas-fir (Abdel-Gadir et al. 1993). For species where the negative correlation between wood density and growth rate is the norm, weaker relationship in trees growing in more favorable environments was reported (Abdel-Gadir et al. 1993; Zhang et al. 1996). Thus, the relationship found on one site cannot be extrapolated for the whole species (Louzada 2003; Zhang et al. 1996). Moreover, the relationship between wood density and growth rate can be manipulated by silvicultural practices. Fertilization and thinning support growth rate, but cause a decrease in wood density when applied simultaneously. However, cases when thinning itself had no negative impact on wood density had also been reported (Zobel and Van Buijtenen 1989). 1.3.2.2  Heritability of wood density  Heritability is a measure of phenotypic variance explainable by genetic differences (Falconer and Mackay 1996). All heritabilities discussed in this study are narrow-sense heritabilities (  ), i.e., the ratios of additive genetic variance over phenotypic variance. The term  stands for individual-tree heritability while  stands for family-mean heritability.  Cornelius (1994) reviewed 67 research papers dealing with different forest tree species. In 87% of them, individual-tree heritabilities of specific gravity were estimated between 0.3 and 1.0 (median 0.48, mean 0.50). High wood density heritabilities were reported for example for radiata pine,  = 0.47 (Nicholls et al. 1980), 13  = 0.30 and  = 0.60 (Li and Wu 2005),  loblolly pine, 2002),  = 0.45 (Talbert et al. 1983), maritime pine,  = 0.63 ± 0.19 (Gaspar et al. 2008), and Douglas-fir,  = 0.59 ± 0.12  and  = 0.55 ± 0.03  (Vargas-Hernandez  > 0.52 (Louzada and Fonseca = 0.90 (King et al. 1988), and  Adams  1991),  = 0.47 ± 0.15 (Ukrainetz et al. 2008). In their review Rozenberg and Cahalan (1997) also reported a high wood density heritability for genus Picea. Wood density heritabilities were estimated to be for black spruce (Zhang and Morgenstern 1995), (Hannrup et al. 2004), and  = 0.34 and  = 0.60 and  = 0.56  = 0.53 ± 0.21 for Norway spruce  = 0.72 for white spruce (specific gravity) (Khalil  1985b). Yanchuk and Kiss (1993) estimated specific gravity heritability for interior spruce to be  = 0.47 ± 0.03.  Similar results were obtained for 15-year-old interior spruce by  Corriveau et al. (1987) ( relative density of  = 0.47 and  = 0.67). Individual-tree heritabilities of mean  = 0.52 ± 0.14 for East Kootenay and  = 0.36 ± 0.13 for Prince  George interior spruce progenies were estimated by Ivkovich et al. (2002a). Heritabilities obtained by Lenz et al. (2010) for wood density of white spruce were greatly different for earlywood and latewood, 1.3.3  = 0.69 ± 0.26 and 0.13 ± 0.11, respectively.  Early selection  A significant genetic correlation between density of juvenile wood and mature wood (0.57) was published by Corriveau et al. (1987) for white spruce, indicating the feasibility of early selection for wood density. Matziris and Zobel (1973) reported a significant correlation of 0.69 between loblolly pine parents and their 5-year-old progenies’ juvenile wood specific gravity. The possibility of selection based on 5-year performance was also confirmed for white spruce (Khalil 1985a).  Nevertheless, selection after only one year would be  inappropriate because of significant interaction between family and growing season. Furthermore, selection at a young age should be done with caution, considering the possible presence of compression wood (Ivkovich et al. 2002a). 1.3.4  Breeding and wood quality  Many authors emphasize a need to include wood quality along with wood quantity into tree improvement programs (Vargas-Hernandez and Adams 1991; Zhang and Morgenstern 1995;  14  Zobel and Jett 1995). However, using wood density as a primary selection criterion and at the same time maintaining current levels of gain for growth traits is an unrealistic goal. Moreover, breeding only for wood density would indeed result in a higher pulp yield and lower production cost but it would cause lower fiber dimension and higher fiber curl index (Pot et al. 2002). Therefore, a two-step tree improvement, comprising selection for growth traits with a subsequent selection for wood quality traits, appears to be a more appropriate approach (Akachuku 1984; Corriveau et al. 1987; Pot et al. 2002). Because a high variability in wood density occurs among fast growing families (Matziris and Zobel 1973; Zhang 1995), it is possible to select families with both fast growth and the required wood properties. However, adaptive traits should not be omitted from the first round of selection (Akachuku 1984; Zobel and Jett 1995). Some studies examining wood density in more detail found that information about earlywood density is very advantageous and proposed its inclusion into existing selection criteria (Louzada 2003; Louzada and Fonseca 2002; Ukrainetz et al. 2008). In a maritime pine study, earlywood density reached the highest and the most age-age stable heritabilities compared to other wood density components (Louzada and Fonseca 2002). Moreover, high and favorable correlations of earlywood density with other wood density characteristics suggest that selection for earlywood density would lead to simultaneous responses in other density components (Louzada 2003). In particular, increased earlywood density results in higher overall density and lower wood heterogeneity (Louzada and Fonseca 2002). Ivkovich et al. (2002b) mentioned the potential of indirect selection for ring width and earlywood percentage instead of wood density itself in interior spruce. On the other hand, latewood density in the above-mentioned maritime pine study showed the lowest and most unstable heritabilities over time (Louzada and Fonseca 2002). Therefore, latewood density seems to be more or less dependent on the surrounding environment and its inclusion in the selection criteria would not bring any significant advantage (Louzada and Fonseca 2002; Zhang and Morgenstern 1995). Latewood percentage was also reported to be under poor genetic control by Zhang et al. (1996). Some other studies did not find any of the wood density components to be helpful for selection and remain using the average ring density (Aubry et al. 1998; Nicholls et al. 1980; Vargas-Hernandez and Adams 1991).  15  Important wood characteristics for the pulp and paper industry are those regarding longitudinal tracheids especially their length (Kennedy 1995). Tracheids in softwoods are naturally long enough for good quality of final products and their length does not need to be the subject of genetic improvement (Beaulieu 2003; Kennedy 1995). However, it may not be true in the future, as fast growing trees tend to have higher juvenile wood proportion which, compared to mature wood, consists of shorter tracheids. It is to be considered whether or not to include tracheid length among the selection criteria as it is not related to wood density (Beaulieu 2003). Unfortunately, tracheid length is under moderate (Ivkovich et al. 2002b) to low genetic control (Beaulieu 2003; Matziris and Zobel 1973). 1.3.5  Variability of wood density  Wood is a very complex and highly variable material in terms of anatomical structure and chemical composition. It varies not only between species but also within species and even within a single tree (Pot et al. 2002). The variability provides an opportunity to utilize wood in many different ways but at the same time it limits its performance in each of them (Pot et al. 2002). Therefore, the lack of uniformity is often viewed as a great wood quality problem (Zobel and Jett 1995). Large variations in wood density were reported among stands (Ivkovich et al. 2002a; Taylor et al. 1982; Wang and Micko 1984), families (Ivkovich et al. 2002a; Yanchuk and Kiss 1993), and single trees (Taylor et al. 1982; Zobel and Jett 1995). Holst (1960) in Zobel and Van Buijtenen (1989) observed variability among trees within one population to be almost as large as that among trees from the biggest possible geographical distances. Nevertheless, genetic wood density variation in spruce was estimated by numerous authors as moderate (reviewed by Rozenberg and Cahalan 1997). Corriveau et al. (1987) reported that about one fifth of the total variation of wood density in white spruce in Quebec could be explained by differences among populations, while the rest could be attributed to differences among trees and experimental error. Large differences in wood density among white spruce trees were also observed in central and southern Alberta, whereas no significant differences were found in northern parts of the province (Taylor et al. 1982).  16  Because of a complex nature of wood density, it is not easy to recognize sources of its variation (Louzada and Fonseca 2002). However, it was ascertained that genetic differences in wood density among trees mostly reflect differences in earlywood density but not in overall density (Zhang and Jiang 1998; Zhang and Morgenstern 1995). Moreover, latewood proportion was found to have a substantially higher phenotypic variation than any other intraring wood density characteristics (Abdel-Gadir et al. 1993; Louzada 2003; Zhang 1998). Nevertheless, variability in wood density was found to be lower than in other traits (Cornelius 1994; Zobel and Van Buijtenen 1989). Coefficient of variation for relative wood density estimated by Corriveau et al. (1991) was close to 8%; juvenile wood was more variable than mature wood. 1.3.6  Modulus of elasticity and modulus of rupture  Modulus of elasticity (MoE) and modulus of rupture (MoR) are two commonly used parameters for estimation of wood mechanical properties (Bouffier et al. 2009); MoE is generally more important than MoR in terms of mechanical performance (Kliger 2000 in Beaulieu et al. 2006).  The former, also called stiffness, is a numerical expression of  non-permanent sample deformation when a load is applied. The latter, also known as strength, is equal to the stress needed to cause a failure (Desch and Dinwoodie 1996). MoE depends on the firmest part of the wood which is latewood, while MoR depends on the weakest part of the wood which is earlywood (McKimmy 1985 in Rozenberg et al. 1999). Mechanical properties are often assumed to be quite well related to wood density and therefore well predictable through it (Jozsa and Middleton 1994; Porter 1981; Zhang et al. 2004). On the other hand, there are also studies reporting that wood density is a poor indicator of mechanical properties, especially in softwoods (Cave and Walker 1994). The relationship however differs from species to species, including species from the same genus. Besides, MoE does not seem to be related to specific gravity as much as MoR (Zhang 1997). Zhang (1995) estimated moderate negative correlations between MoE and specific gravity (-0.4 < r < -0.7) for softwoods with gradual earlywood-latewood transition, where spruce belongs.  In contrast, a strong positive relationship between MoE and specific gravity  (0.5 < R2 < 0.8) has been reported by numerous authors for various species including spruce  17  (Rozenberg and Cahalan 1997; Rozenberg et al. 1999). Armstrong et al. (1984) used worldwide data focusing on economically important tree species and calculated correlations for specific gravity versus MoE in static bending and MoR to be 0.78 (green MoE), 0.79 (air-dry MoE), 0.91 (green MoR), and 0.89 (air-dry MoR). In maritime pine correlations for wood density versus MoE and MoR were estimated to be 0.61 and 0.56, respectively (Reuling 2005 in Bouffier et al. 2009). A weaker but positive relationship between relative density on one side and MoE and MoR on the other side was reported for white spruce (R2 = 0.30 and 0.23, respectively) (Zhou and Smith 1991).  Unlike these studies, Beaulieu et al. (2006),  investigating young fast-grown white spruces, did not find any relationship between wood density and lumber MoE/MoR; however, the two mechanical properties were strongly correlated (0.71). The relationship between MoR and specific gravity was found to be almost linear, whereas curvilinear dependence appears to more suitably predict MoE through specific gravity. Moreover, softwoods showed larger variation in mechanical properties and wood density compared to hardwoods. According to a study made on sixteen tree species with different wood characteristics, specific gravity was capable of explaining about 50 and 30% of variation in MoR and MoE, respectively (Zhang 1997). Several studies also scrutinized the relationship between ring-density parameters and the two wood mechanical properties.  Choi (1986) found the MoE and MoR to be positively  correlated with latewood proportion (R2 = 0.54 and 0.45) as well as with earlywood density (R2 = 0.20 and 0.27), and negatively with ring width (R2 = 0.54 and 0.35); correlations with latewood density were close to zero.  Mamdy et al. (1999) discovered the strongest  relationship between trunk MoE and mean ring density (R2 = 0.42) and between board MoE and latewood width (R2 = 0.37). Unlike Choi (1986), Mamdy et al. (1999) reported the same results for board MoE versus latewood proportion and latewood density (R2 = 0.26); the relationship of board MoE with earlywood density was non-significant (R2 = 0.06). Both of these studies were done on Douglas-fir. A considerable decrease in wood mechanical properties with increased growth rate has been published (Beaulieu et al. 2006; Zhang 1995; Zhou and Smith 1991). Lower MoE and MoR values associated with higher juvenile wood proportion (Kennedy 1995; Kretschmann and  18  Bendtsen 1992) and higher microfibril angle (Cave and Walker 1994) have been observed in fast-growing softwoods. A much greater variability was reported for MoE and MoR in comparison with wood density (26%, 46%, and 9%, respectively). It indicates that only a minor decrease in wood density will result in an appreciable deterioration of the mechanical properties (Bouffier et al. 2009) where MoR will be more affected than MoE. It also implies that changes in wood mechanical properties due to faster growth cannot be entirely explained by the relationship between specific gravity and growth rate (Zhang 1995). MoE and MoR were estimated to be under strong genetic control in radiata pine (  = 0.42  and 0.72, respectively) (Matheson et al. 1997a,b in Beaulieu et al. 2006), whereas in white spruce it was distinctly lower (  = 0.28 ± 0.32 and 0.02 ± 0.17, respectively) (Beaulieu et al.  2006). Different heritabilities of MoE in white spruce were reported for earlywood and latewood (  = 0.27 ± 0.15 and 0.41 ± 0.19, respectively) (Lenz et al. 2010).  Environmental conditions along with silvicultural practices may strongly influence mechanical wood properties. For example, Raymond et al. (2008) reported a decrease in wood stiffness by 3% on thinned sites. Co-dominant and intermediate trees demonstrate a slower growth rate along with higher wood density and strength compared with dominant and suppressed trees (Zhou and Smith 1991). MoE and MoR are most affected by wood density, juvenile wood proportion, and number of knots (Beaulieu et al. 2006). MoR was found to be more affected by the presence of knots than MoE (correlations 0.60 and 0.47, respectively) (Zhou and Smith 1991). Higher microfibril angle, the common feature of juvenile wood, was found to be associated with lower dimensional stability and mechanical performance (Park et al. 2012) when MoE was more affected than MoR (Cave and Walker 1994). However, wood density remained the same with higher microfibril angle (Jozsa and Middleton 1994). In white spruce, genetic correlations between MoE and microfibril angle were reported to be strong and negative (earlywood -0.71 ± 0.19, latewood -0.81 ± 0.12) whilst between microfibril angle and wood density weak and non-significant (Lenz et al. 2010). Nevertheless, the combination of the two traits appears to be suitable for MoE prediction (Cave and Walker 1994).  19  1.4  Non-destructive wood quality assessment  Non-destructive assessment is the “examination of an object with technology that does not affect the object’s future usefulness” (American Society of Nondestructive Testing 2000 in Shull 2002); in other words, it is a procedure where an object under examination is not damaged (Bucur 2003) and its physical properties remain unchanged (Shull 2002). Finding a non-destructive, reliable, and inexpensive method for standing-tree wood quality evaluation is of great interest for breeders and foresters. A variety of tools for indirect wood quality assessment have been developed during the last few decades; however, not all of them are capable of giving an estimate right on site. Wood density, the best single predictor of wood quality, is traditionally determined by a destructive volumetric method. It is calculated based on mass and volume of a sample with definite moisture content. A popular modification of this approach is maximum moisture content method (Smith 1954), which is suitable for very small samples such as increment cores or their parts and can be therefore considered non-destructive. Mass of oven-dry and water-saturated samples along with density of wood substance (1.53 g·cm-3) are then used for wood density calculations (Smith 1954). Wood quality can be examined using various types of radiation (x, β, or γ rays) (Bucur 2003).  Aside from wood density, other growth-ring characteristics can be read from  increment cores based on rays’ attenuation after passing through wood (Evans and Ilic 2001). Although some tools for in situ applications are available, radiation techniques prevail under laboratory conditions (Wessels et al. 2011). Recently, x-ray densitometry has been considered as one of the most reliable non-destructive method for assessment of wood density (Li and Wu 2005), providing a strong correlation with the volumetric approach (0.85) (Isik and Li 2003). Density and microfibril angle are estimated using direct x-ray scanning (Evans and Ilic 2001) where x-ray absorption and diffraction relate to density and microfibril angle, respectively (Zink-Sharp 2004). Pith-tobark increment cores represent a sufficient amount of wood for the analysis (Echols 1973). Their preparation prior to scanning includes resin extraction using methanol, moisture content adjustment, and thickness reduction to 2 mm. A sample attached to a movable carriage is then exposed to an x-ray beam and the degree of the beam’s attenuation after  20  passing through the wood is recorded (Cown and Clement 1983). Either a calibrated x-ray film or camera can be used for recording (Zink-Sharp 2004). X-ray densitometry is based on Beer-Lambert law [1.2]  where I is the intensity of a beam transmitted through a wood layer with thickness x [cm], I0 is the intensity of incident beam, and μ is the linear attenuation coefficient [cm-1] which depends on a material’s composition and density (Macedo et al. 2002). Considering that wood is a heterogeneous material consisting of solid wood substance, air, and water, μ can be expressed as [1.3]  where μms and μmw are mass attenuation coefficients [cm2·g-1] for wood substance and water, respectively (μm=μ/ρ), ρg is dry bulk density [g·cm-3], and w is volumetric water content, i.e., the ratio between water volume present in wood and total sample volume (Macedo et al. 2002). Knowing moisture content and μms estimated from samples of known density, bulk density can be calculated using Eq. 1.4 such that [1.4]  X-ray densitometry is a reliable but relatively expensive and time consuming method (Bouffier et al. 2008a). Besides the demanding sample preparation, it requires calibration prior to x-raying. Another recently introduced method of neutron imaging utilizes the same principles like the x-ray method, except that no calibration is needed (Mannes et al. 2007). Indirect estimation of wood density in situ can be done by penetrometers such as Pilodyn or Resistograph (see Chapter 1.4.1). Pilodyn measures penetration depth of a steel pin shot into the wood with an exact force (Taylor 1981). In order to get unbiased estimates, a bark patch has to be removed ahead of the measurement (Hansen 2000); however, it can significantly damage young trees (Greaves et al. 1996). Pins with three different diameters of 2.0, 2.5, and 3.0 mm can be used, the smallest of which penetrates deeper than the larger ones (Sprague et al. 1983). As the maximum possible depth is 40 mm, measurements are only limited to the  21  outer growth rings (Hansen 2000). Despite this disadvantage, Greaves et al. (1996) obtained sufficiently accurate estimates with just two measurements per tree. A strong correlation was reported between the Pilodyn estimates and the actual density (-0.95) (Feio et al. 2007). However, the Pilodyn may underestimate wood density of large-diameter trees (Wang et al. 1999) and, in some studies, it did not provide satisfactory results (Isik and Li 2003). MoE and MoR, giving an estimate of wood mechanical properties, are traditionally measured using large boards or small clear specimens from felled trees.  Near-infrared (NIR)  spectroscopy, a method utilizing electromagnetic radiation with wavelengths between 800 and 2,500 nm, is one of the methods suitable for indirect estimation of wood properties (Wessels et al. 2011). NIR spectroscopy is cheap, fast, and easy; however, proper calibration is necessary. A strong relationship has been reported between NIR reflections on one side and density, MoE, and MoR on the other (R2 = 0.93, 0.90, and 0.77, respectively) (Schimleck et al. 2001). Another non-destructive method developed for indirect assessment of MoE is based on measuring acoustic velocity (Director ST300, see Chapter 1.4.2). Although methods using increment cores are non-destructive, they are generally assumed to be time, cost, and labor intensive; especially core extraction and preparation (Wessels et al. 2011). Penetrometer Resistograph IML F300 (Chapter 1.4.1) and Director ST300 measuring acoustic velocity (Chapter 1.4.2) may satisfy the requirements for a non-destructive, reliable, and inexpensive tool suitable for in situ wood quality assessment. 1.4.1  Resistograph IML F300  Resistograph IML F300 is an instrument manufactured by a German company Instrumenta Mechanik Labor, GmbH (Hein 2008).  It belongs to a family of penetrometers, which  measure drilling resistance, i.e., the power consumption during penetration of a thin probe through wood (Nicolotti and Miglietta 1998).  The Resistograph provides detailed  information about internal structure of woody materials as well as live trees and estimates the relative wood density. It easily reveals small cavities, cracks, or insects’ bore holes (Kahl et al. 2009) and is widely used for decay detection (Johnstone et al. 2007). The Resistograph was originally developed for tree care in urban environments (Brashaw and Ross 2002), but was later recognized as a reliable tool for wood quality assessment on standing trees (Isik and  22  Li 2003) as well as for dendrochronological surveys (Nicolotti and Miglietta 1998). Kubus (2009) qualified the Resistograph as the most harmless invasive method for live trees inspection. 1.4.1.1  Description  The Resistograph consists of an electronic-drilling unit mounted onto a Bosch hand drill (IML 2008d). The rotation speed can be set in two levels – low speed of 400 rpm for tree assessment and high speed of 1200 rpm for utility pole examination (IML 2008b) and drilling speed is about 30 cm per minute (Ukrainetz and O'Neill 2010). The attached electronicdrilling unit has a tubular housing, which supports a rotating and axially moving drill needle. The Resistograph enables adjusting for softwood and hardwood (IML 2008b). Measurements are stored in the Resistograph’s internal memory and can be downloaded to a computer (IML 2008a). They are also simultaneously visualized at a 1:1 scale on the top part of the housing (IML 2008a).  The Resistograph is considered to be a very sensitive  instrument providing an accurate picture of wood conditions as density is recorded every 0.1 mm (IML 2008c). The elastic drill needle, 1.5 mm in diameter and 30–50 cm in length, is made of steel (IML 2008d) with a special alloy surface (IML 2008a). The hardened needle’s tip has a fish-tale shape with a diameter of 3 mm, which bores a wider hole than the needle’s diameter itself. It ensures that the penetration is not influenced by the needle’s friction with walls of the drill hole (Nicolotti and Miglietta 1998) and the drilling resistance is centralized at the wider tip (Rinn 1996). After drilling, the tiny hole stays full of shavings and wood dust, which makes this tool to be classified as non-destructive (Nicolotti and Miglietta 1998). 1.4.1.2  Function  The slender drill needle goes through the wood with a uniform speed and the energy consumption needed for bark to bark penetration is calculated by the Resistograph. A variety of factors that could cause a bias are taken into account. Measurements of individual annual rings are weighted, assuming that outside rings are longer than those in the center. Also, bark thickness on both sides of the trunk is omitted with the aim to get wood assessment only. The total reading is adjusted to eliminate possible additional friction on the needle’s  23  tip.  The resulting number is called the density number.  This single unitless number  represents a relative wood density and, in many cases, it is considered to be a sufficient output (IML 2008c). 1.4.1.3  Comparison of in situ and laboratory estimates  Strong relationships have been reported between the volumetric and Resistograph-based densities by a number of authors: R2 > 0.8 (Rinn et al. 1996), r = 0.82 (Ceraldi et al. 2001), r = 0.76 (Lin et al. 2003), r = 0.9 (Feio et al. 2007), R2 = 0.74 (El-Kassaby et al. 2011b). In loblolly pine, Isik and Li (2003) obtained low to moderate phenotypic (0.29–0.65) and strong genetic (0.95 ± 0.04) individual-tree correlations between the two variables. Correlations between Resistograph- and x-ray-based densities were moderate to high; Bouffier et al. (2008a) estimated strong genetic and phenotypic correlations for maritime pine (> 0.80), similarly Eckard et al. (2010) obtained a strong genetic correlation for loblolly pine (0.92 ± 0.06), while Chantre and Rozenberg (1997) reported moderate correlation for Douglas-fir (0.62). In addition to that, graphical profiles produced by the Resistograph had a very similar pattern as the x-ray charts (Rinn et al. 1996). Published correlations between the Resistograph’s readings and wood mechanical properties were moderate. Feio et al. (2007) studying chestnut reported correlations of 0.62–0.78 for MoE and 0.64–0.77 for MoR (both compression parallel to grain). Similar results were obtained by Eckard et al. (2010) for loblolly pine: genetic correlations 0.45 ± 0.30 and 0.68 ± 0.31 for MoE and MoR, respectively.  On the other hand, Ceraldi et al. (2001)  estimated a low correlation between Resistograph-based density and compressive strength (0.22) using sandwiches of various species’ wood. Heritabilities for Resistograph-based density were moderate to high: 0.32 ± 0.18 and 0.43 ± 0.16 on two different sites (Bouffier et al. 2008a), 0.25 ± 0.10 (El-Kassaby et al. 2011b), and 0.42 ± 0.46 (Ratcliffe 2012). 1.4.1.4  Factors causing bias of estimates  Moisture content is one of the most detrimental factors influencing drilling results (Rinn 1996). Moderate significant correlations between the drilling resistance and moisture content (0.56) were reported by Lin et al. (2003). The presence of water in the wood results in  24  increased drilling resistance and, consequently, in a higher density number (Kahl et al. 2009). Air temperature also strongly influences the Resistograph’s readings. When the results are to be compared, they should be measured under similar temperatures. Also, measurements should be avoided when air temperature drops below 0°C (Ukrainetz and O'Neill 2010). The drill needles’ flexibility can cause some bias since it does not always go straight through the wood.  It can result in additional friction and in turn in an imprecise density number  (Nicolotti and Miglietta 1998). Drill bit flexion caused by an operator’s movements can also significantly influence the results; therefore, the operator should be as stable as possible during the drilling process (Ukrainetz and O'Neill 2010). Knots represent another factor as their density is approximately two fold bigger than that of normal wood (Sahlberg 1995). Ukrainetz and O'Neill (2010) reported that the influence of knots is insignificant in distances longer than 3 cm. 1.4.1.5  Impact of drilling on trees’ health  The drill holes filled with shavings and wood dust provide a favorable environment for further decay expansion in trees with pre-existing decay (Kersten and Schwarze 2005). However, van Mantgem and Stephenson (2004) did not register any mortality due to coring during their 12-year study. The internal decay has a potential to invade the drill hole slightly but only with a local effect. Also, no pathogenic microorganisms coming from outside into the drill hole and surviving there were observed (Weber and Mattheck 2006). 1.4.2  Director ST300  Director ST300TM, a non-destructive tool for measuring acoustic velocity in standing trees, is a product of Fibre-gen company, New Zealand (Fibre-gen 2010). It is based on a stress wave method invented by Sobue in 1986 (Bucur 2006a). Time-of-flight (ToF) of a stress wave induced by a mechanical shock is measured between two probes imbedded in the wood (Bucur 2006b; Wang et al. 2000a).  This approach, introduced by Hungarian Fakopp  Enterprise (Chauhan et al. 2006b), was initially used for product quality control but later was extended for log grading and standing tree assessment (Pellerin and Ross 2002; Wang et al. 2007). The stress wave method is considered to be an efficient method for estimation of intrinsic wood properties (Kumar et al. 2002; Lindström et al. 2004); however, it can be also  25  used for measuring of moisture content in logs (Wang et al. 2007), internal decay detection (Andrews 2002; Chauhan et al. 2005), determination of juvenile wood proportion (Laverty 2001 in Bucur 2006b), or microfibril angle estimation (Andrews 2002). 1.4.2.1  Description and function  Director ST300 consists of two sensor probes (receiver and transmitter) that are hammered 2–3 cm into the stem under 45° angle and ca 0.5–1.5 m apart (Fibre-gen 2004). A visual laser beam helps to align the probes vertically while the exact distance between them is measured by ultrasound. As the transmitter probe is tapped with a steel hammer, stress waves start to propagate through wood. ToF is then recorded by the receiver probe and wirelessly sent to a personal digital assistant (PDA) unit, which calculates acoustic velocity (speed of the stress waves) from the ToF and distance (Carter et al. 2005). Each estimate represents an average of eight taps; however, in order for the estimate to be reliable, it is recommended that the operator obtained three similar averages (Fibre-gen 2004); i.e., the transmitter probe must be tapped at least 24 times. Additional information such as stand and tree number, age, or dbh can be entered and stored in the PDA device along with the measurements. All data can be easily downloaded into an MS Excel spreadsheet. Using ToF approach, dynamic modulus of elasticity (MoEd) [GPa] can be calculated for the column of measured wood (Auty and Achim 2008): [1.5]  where ρ is wood density [g·cm-3] and v is acoustic velocity [km·s-1]. The formula assumes that one-dimensional plane stress waves travel through rod-like homogeneous and isotropic material (Jones and Emms 2010; Wang et al. 2001a). Stress waves induced by mechanical shock have high wavelength and low frequency, which ensures that any heterogeneity (including wood defects) present in a stem profile is not distracting for the method’s utilization. In other words, owing to the high wavelength wood can meet the assumption of homogeneity despite being heterogeneous (Chauhan et al. 2006b). Furthermore, the formula assumes that acoustic waves are plane (i.e., have constant frequency and amplitude with wave-fronts parallel to a plane); however, elements closer to the spot of impact travel faster than those located further. Acoustic waves eventually become plane if they can reverberate,  26  which is possible in felled trees but not in standing trees (Andrews 2002).  Also the  assumption of stress waves traveling through a rod-like object is not quite true for a stem and therefore some deviations can be expected (Jones and Emms 2010). Stress wave propagation in standing trees is influenced by both physical and mechanical wood properties; however, the mechanism has not been fully understood (Wang et al. 2000b). Zhang et al. (2009) described that stress waves tend to continuously change their course of propagation from initial direction of mechanical impulse (45°) towards the longitudinal direction. Because of wood’s anisotropic nature, the velocity of sound traveling through wood varies in different directions (longitudinal : radial : tangential = 15:5:3) (Horáček 1998). 1.4.2.2  Comparison of in situ and laboratory estimates  The accuracies of standing-tree MoEd estimates have been tested under laboratory conditions; however, the results were not consistent. Wang et al. (2000b) obtained moderate to strong correlations between laboratory-measured MoEd, MoEs, and MoR on one side and MoEd of standing trees (38–70 years old) on the other side to be 0.73, 0.63, and 0.65, respectively, for western hemlock and 0.77, 0.78, and 0.63, respectively, for Sitka spruce. Strong correlation was reported between laboratory-measured dynamic and static MoE (>0.90) (Achim and Carter 2005; Wang et al. 2000b) as well as between laboratory-measured and standing-tree acoustic velocities (0.83) (Wang et al. 2000a; Wang et al. 2001b). In Scots pine, Auty and Achim (2008) found acoustic velocity to be more strongly related to laboratory-measured MoEs and MoR than standing-tree MoEd (0.73, 0.77, and 0.58, 0.68, respectively).  Strong or moderate correlations between laboratory-measured MoEs and  standing-tree MoEd were also reported for 3-year-old radiata pine, 0.81 (Lindström et al. 2004), 28–43-year old radiata pine, 0.79 (Raymond et al. 2008), 32-year-old coastal Douglasfir, 0.83 (El-Kassaby et al. 2011b), 25-year-old Douglas fir, 0.45 (Cherry et al. 2008), or 14– 19-year old loblolly pine, 0.81 (Mora et al. 2009). Matheson et al. (2002) obtained a significant moderate correlation between standing-tree acoustic velocity and MoEs for a control stand (0.33) but a non-significant correlation for stand established using seed orchard material (0.01).  Kumar et al. (2002) published moderate negative correlations (both  phenotypic and genetic) between ToF and laboratory-measured MoEs for 12-year-old radiata  27  pine (-0.47 and -0.69 ± 0.17, respectively). Note that although wood defects do not influence acoustic velocity in standing or felled trees, they can affect laboratory measurements, resulting in smaller values of bending MoE (Andrews 2002). Heritabilities for all of the discussed acoustic traits were generally moderate to high: 0.30– 0.76 for acoustic velocity (Isik et al. 2011), 0.46 ± 0.17 (Kumar et al. 2002) and 0.16 ± 0.14 (Matheson et al. 2002) for ToF, and 0.30–0.86 for MoEd (Isik et al. 2011). 1.4.2.3  Acoustic resonance  Unlike for standing trees, the mechanical properties of felled trees can be assessed using a resonance technique because the cut ends serve as good sound wave reflectors (Andrews 2002).  Acoustic velocity is then calculated from the frequency of an acoustic plane wave  reverberating within a log as [1.6]  where l is a log’s length and f is the fundamental resonance frequency (Chauhan et al. 2005). This approach gives an assessment for the whole log’s profile and therefore is considered to be more accurate than the ToF approach, which is restricted just to the column of outer wood (Auty and Achim 2008; Lindström et al. 2004). Considerably higher acoustic velocities were measured by ToF compared to acoustic resonance (Auty and Achim 2008; Grabianowski et al. 2006; Ross and Wang 2005; Wang et al. 2007) but the cause has not been clarified yet (Ross and Wang 2005). Explanation proposed by Grabianowski et al. (2006) says that areaweighted average velocity measured by acoustic resonance accounts also for bark, which increases cross-section area but not wood stiffness. Another hypothesis suggests that higher ToF in standing trees involves waves that are not plane (Andrews 2002; Wang et al. 2007). Nevertheless, deviations between log and standing-tree acoustic velocities can be adjusted using an appropriate species-specific model (Ross and Wang 2005; Wang et al. 2007). Favourable relationship between ToF and acoustic resonance has been reported by a number of studies. Wang et al. (2007) obtained strong correlations between the two variables in Sitka spruce, jack pine, radiata pine, ponderosa pine, and western hemlock (0.84–0.96). Strong correlations were also reported for young radiata pine, 0.96 (Grabianowski et al.  28  2006), loblolly pine, 0.90 (Mora et al. 2009), or five tropical tree species, 0.81 (Baar et al. 2011). 1.4.2.4  Factors causing bias of estimates  Acoustic velocity values determined can be affected by a number of factors such as environmental conditions, natural wood variability, or acoustic velocity measurement itself. One of the most important factors is temperature.  Acoustic velocity shows a gradual  decreasing trend with increasing temperature above 0°C; nevertheless, the decrease is more abrupt below 0°C (Gao et al. 2012). Strong discontinuity just above and below 0°C is caused by freezing of capillary water (Bachle and Walker 2006). Moreover, acoustic velocity decrease is more pronounced in green wood compared to dry wood (Gao et al. 2012). Another factor affecting acoustic velocity and MoEd is moisture content. Increasing moisture content is associated with a rapid decrease in acoustic velocity below the fiber saturation point (FSP); however, the decrease is more gradual above the FSP. MoEd decreases with increasing moisture content below FSP but slightly increases above FSP; MoEd is calculated based on density, which increases with moisture content (Wang and Chuang 2000). Differences in temperature and moisture content could be sufficiently compensated for by appropriate models. Acoustic velocity may vary highly around the stem; therefore, an average of two values from opposite directions is recommended in order to get a reliable estimate (Grabianowski et al. 2006). It also varies along the stem (Carter 2007) as well as from pith to bark (Xu and Walker 2004). Raymond et al. (2008) reported a decreasing MoE in the stem upwards. In contrast, Xu and Walker (2004) found a large amount of low-stiffness wood at the bottom (up to 2.7 m above the ground) while the stiffness above was constant. Acoustic velocity is also influenced by the presence of knots, compression wood (Carter 2007), and last but not least by age (Grabianowski et al. 2006). A strong correlation between squared acoustic velocity and age (0.9) (Toulmin and Raymond 2007) indicates that sound travels faster in older trees. It is crucial to obtain precise measurements of acoustic velocity because MoE is directly proportional to its squared value. Bias in acoustic velocity can be related to variability in tap intensity (Carter 2007), angle of hammer blow, or angle of probes’ imbedding (Andrews  29  2000 in Lindström et al. 2004). Besides, Andrews (2002) pointed out that tools used for measuring acoustic velocity have a limited accuracy.  In standing trees, measuring of  acoustic velocity is restricted to a column of outer wood in the bottom part of the tree and therefore it does not seem to represent corewood quality (Raymond et al. 2008); nevertheless, Grabianowski et al. (2006) reported a strong correlation between outerwood and corewood ToF (0.86).  1.5  Objectives  The objectives of this thesis are to 1) determine the underlying genetic control of wood density attributes in interior spruce and 2) evaluate the suitability of Resistograph IML F300 and Director ST300 for in situ wood quality assessment in interior spruce.  30  2 Materials and Methods 2.1  Trial description  In total, 25 open-pollinated interior spruce families were selected for this study based on their parental breeding values (individual tree volume at rotation age of 80 years). All parents originated from low- to mid-elevations (ca 650–1500 m) east and south-east of Prince George. The area covers the McGregor (MGR), Quesnel Lake (QL) and Cariboo Transition (CT) zones (currently Prince George and Prince George Nelson Transition class A zone). The parents are mostly represented by white spruce and to a lesser extent by hybrids of white and Engelmann spruce (Kiss 1976 in Xie and Yanchuk 2002). The study includes three progeny trials: 1) Aleza Lake, 2) Prince George Tree Improvement Station (PGTIS), and 3) Quesnel (see Table 2.1 for a detailed description and Figure 2.1 for a map of the trials’ locations). Aleza Lake was established in 1972 and PGTIS and Quesnel in 1973. All three trials were planted with 2+1 bare root seedlings and 2.5 x 2.5 m spacing (Xie and Yanchuk 2002) and were established using a complete randomized block design with five or ten blocks and ten- or fifteen-tree row plots. From each site, four trees per family from four blocks were randomly sampled, resulting in a total of 1,146 trees. Table 2.1 Geographical coordinates of the progeny trials Elevation [m a.s.l.]  Soil description  Site preparation  N 54 03’ 15.7” W 122 06’ 35.4”  700  Silty clay loama  Deep discing  PGTIS  N 53 46’ 17.9” W 122 43’ 07.6”  610  Alluvial gravelly sandy loama  Deep discing  Quesnel  N 52 59’ 27.2” W 122 12’ 30.6”  915  Gravely clay loamb  Prescribed burning  Progeny trial  Geographical coordinates  Aleza Lake  PGTIS – Prince George Tree Improvement Station, a Dawson (1989), b Lord and Mackintosh (1982)  31  Prince George    Aleza Lake  PGTIS   Wells Quesnel   Quesnel 20 km  Figure 2.1 Map of trials’ location  2.2  Data collection  In June 2009, the selected trees were measured for height [m] using an ultrasonic clinometer VertexTM III (Haglöf, Sweden) and stem circumference [cm] at breast height with a measuring tape. Diameter at breast height (dbh; [cm]) of each tree was then calculated as [2.1]  and stem volume [m3] was estimated using the following formula by Millman (1976) [2.2].  The same trees were also assessed for wood quality by two different non-destructive methods (drilling and acoustic) and cored for further x-ray densitometry. The drilling method estimates wood density by means of wood drilling resistance. It was applied using Resistograph IML F300 (Instrumenta Mechanic Labor System GmbH, Germany).  Every tree was drilled at height of 1 m (see Fig. 2.2) in two mutually 32  perpendicular directions with an effort to avoid drilling through knots, as they are denser than the surrounding wood and therefore could cause a bias in the density estimate (Larsen 2001). Readings from the two directions were averaged, providing a relative density of a given tree. Director ST300 (Fiber-gen, New Zealand) represents the acoustic method. It is based on measuring sound velocity, which is considered to reflect wood stiffness and provides an indirect estimate of MoE. The sound velocity was measured between two probes hammered in a stem ca 1.2 m apart (see Fig. 2.2) and three readings in two directions corresponding to those of the Resistograph were recorded and subsequently averaged.  MoEd was then  calculated using acoustic velocity and x-ray density instead of green density (Equation 1.5).  Director ST300 Receiver probe  1.3 m  1.2 m  core for x-ray densitometry  1.0 m  Resistograph IML F300  Director ST300 Transmitter probe  Figure 2.2 Depiction of tree sampling  X-ray wood density data, obtained by FPInnovations – Forintek from tree cores, served as the benchmark for evaluation of the reliability of the two non-destructive methods. Bark-to-bark cores were extracted from each tree at breast height (see Fig. 2.2) in the north-south direction by 5-mm increment borers (Haglöf, Sweden) and stored in a cool place. Their preparation  33  prior to x-ray scanning included soaking in a cyclohexane-ethanol solution (2:1) for 24 hours and in hot water for another 24 hours, conditioning to 8% moisture content, sawing to 1.60 mm thickness, and manual dating. The x-ray scanning was performed by QTRS-01X Tree Ring Scanner (Quintek Measurement Systems Inc., USA) using 0.080 mm vertical collimator and 0.080 mm step size. Scan speed was around 20 mm/min. After scanning, ring boundaries were checked with manual dating marks and adjusted when necessary. Five variables obtained by the x-ray densitometry were collected for this study: overall core density [g·cm-3], density of the first fifteen rings [g·cm-3], earlywood density [g·cm-3], latewood density [g·cm-3], and percentage of latewood [%].  2.3  Statistical analysis  Prior to the statistical analysis, eleven response variables, summarized in Table 2.2, were checked using the statistical package SAS® 9.1.3 (SAS Institute Inc., Cary, NC) whether they meet the underlying assumptions of the regression analysis and, when necessary, proper data transformation was applied (Table 2.2). The assumption of normality was tested by the Shapiro-Wilk normality test due to the relative small data set (sample size < 2,000) as well as using the normal probability plot (Park 2008). Data are considered to be normally distributed if the null hypothesis assuming that residuals follow normal distribution is accepted by the normality test and if data points show a straight line on the normal probability plot (Sabin and Stafford 1990). The significance level for the normality test was set to 0.05. The assumption of equal variances of residuals was visually assessed on a residual plot. Variances are considered to be equal if individual residuals in relation to the predicted values are distributed evenly around zero value. All data points were assumed to be independent of time and space (Sabin and Stafford 1990). The response variables were fitted into the following linear regression model [2.3]  where  is the overall mean of a given variable regardless of family, site and block; j is the  number of sites S; k is the number of families F; l is the number of blocks B (nested within site);  is an estimate of the site effect for site j;  34  is an estimate of the family effect for  family k;  is an estimate of the block effect for block l nested within site j;  estimate of the site  family interaction effect, nested within site interaction effect,  is an  is an estimate of the family  block  is an estimate of the experimental error term, and  is an estimate of the sampling error term. All of the three factors (family, site, and block nested within site) as well as their interactions (family  site and family  block nested within site) were considered to be random effects. Table 2.2 Response variables and their transformations Response variable (y)  Transformation —  Height Dbh Volume  —  Acoustic velocity Resistograph-based density X-ray overall density Density of the first 15 rings Earlywood density  —  Latewood density Latewood proportion MoEd  Utilizing univariate linear models in statistical package ASRemlTM (Gilmour et al. 2009), components of variance and narrow-sense heritabilities for each variable were estimated. Considering open-pollinated families to be truly half-sib families, the heritabilities were calculated as [2.4]  where  , and  are the variance components of family, site, block  nested within site, site by family, family by block nested within site, and error, respectively. Genetic correlations (Equation 2.5) were estimated using ASReml for each pair of variables (Gilmour et al. 2009, page 217) and compared with phenotypic correlations (Equation 2.6) calculated in MS Excel. The phenotypic correlation is a measure of association between two  35  traits of interest that can be directly scored. It is influenced jointly by inheritance and environment and thus can be partitioned into genetic and environmental components. The genetic correlation is then the relationship of an individual’s breeding values for the two traits (Falconer and Mackay 1996).  Both genetic and phenotypic correlations were  calculated as Pearson’s product-moment correlation coefficients. [2.5]  where  and  are genetic variances of traits x and y, respectively, and  genetic covariance and and  [2.6]  and  is their  are phenotypic variances of traits x and y, respectively,  is their phenotypic covariance (Falconer and Mackay 1996).  Significance of the phenotypic correlation coefficients was examined using the t-test (H0: rxy = 0.0; Equation 2.7) where t-values follow t-distribution with n-2 degrees of freedom. [2.7]  The magnitude of differences between phenotypic and genetic correlation matrices was estimated using the Mantel test (Mantel 1967) in the statistical package R (R Development Core Team 2012). Relationships between all traits based on their significant phenotypic and genetic correlations were visualized in the software program Pajek (Batagelj and Mrvar 2012). The likelihood ratio test (Equation 2.8) was used to examine the significance of factors included in the linear model (H0: σ2 = 0). The difference in log-likelihood (log L) between the full (F) and restricted (R) models served as the test statistic (λLR) where the latter was always reduced by the factor in question along with its interactions. The test statistic of follows chi-square (χ2) distribution with the number of degrees of freedom equal to the number of removed model members. [2.8]  36  3 Results and Discussion 3.1  Descriptive statistics  A total of 1146 trees were scored for dbh, acoustic velocity, and drilling resistance, while eight trees were excluded from height measurement due to their broken tops. Consequently, volume, which was derived from height and diameter, resulted in the same sample size as that of height. Variables estimated by x-ray densitometry (overall core density, density of the first 15 rings, earlywood and latewood density, and percentage of latewood) were represented by 1142 trees; as it was also the sample size for the MoE because it was estimated from x-ray density and acoustic velocity data. Tree size varied from 6 m in height and 11 cm in diameter to 20 m in height and 37 cm in diameter. Nevertheless, average trees were rather small with heights around 13 m, diameter of 21 cm, and volume of 0.2 m3. The relative variability of data expressed by the coefficient of variation (CV) for all variables is listed in Table 3.1. Overall wood density estimated by x-ray densitometry closely corresponded with the density of the first 15 rings and was much closer to the density of earlywood than latewood, which was a result of a fairly low latewood proportion (16.5%) (Table 3.1). Overall wood density (0.33 g·cm-3) was similar to that of other studies dealing with interior or white spruce: 0.32 (Beaulieu et al. 2006), 0.33 (Zhang et al. 2004), 0.34 (Taylor et al. 1982), 0.35 (Kennedy et al. 1968), and was not far from wood density reported by Jozsa and Middleton (1994) for old-growth (0.36). However, there were also studies presenting higher values: 0.38–0.35 (Ivkovich et al. 2002a) and 0.38–0.44 (Yanchuk and Kiss 1993).  37  Table 3.1 Descriptive statistics for all measured variables Variable  Units  N  Mean  SD  CV [%]  Min  Max  Height  M  1138  13.10  2.111  16.1  6.00  20.10  Dbh  cm  1146  21.19  4.240  20.0  11.30  37.24  3  m  1138  0.21  0.111  52.1  0.04  0.80  km·s  1146  3.59  0.420  11.7  2.02  4.90  –  Volume Acoustic velocity Resistograph-based density X-ray density Density 15 Earlywood density Latewood density  1146  18.14  5.692  31.4  3.70  37.55  -3  1142  0.33  0.032  9.5  0.26  0.49  -3  1142  0.34  0.034  9.9  0.26  0.47  -3  1142  0.29  0.023  7.9  0.23  0.38  -3  g·cm  1142  0.56  0.020  3.5  0.50  0.64  %  1142  16.49  5.362  32.5  4.92  54.24  GPa  1142  4.38  1.184  27.0  1.40  9.24  g·cm g·cm g·cm  Percentage of latewood MoEd  N – sample size, SD – standard deviation, CV – coefficient of variation, Dbh – diameter in breast height, Density 15 – x-ray density of the first 15 rings, MoEd – dynamic modulus of elasticity  3.2  Components of variance  Family, site, and family  block nested within site interaction were significant for all studied traits. With the exception of dbh for block nested within site, all traits were significant and this trend was reversed for site  family where all traits were non-significant while dbh was significant (Tables 3.2-3.4). Family explained only 0.8–6.5% of the total variance. Nevertheless, it should be noted that the measured trees were progenies of parents selected based on their growth performance and that only 25 families with the highest breeding value for height were included in this study. This fact might result in the reduction of the total phenotypic variance and in turn affect its proportionate components.  Therefore, considerable underestimation of additive genetic  variance for the growth traits can be expected (Falconer and Mackay 1996). In general, more variance was explained by site; however, the amount varied from as much as 45.5% for resistograph-based wood density to as little as 1.2% for latewood density. Site  family interaction did not account for more than 5% of the total variance for any of the traits and was only significant for dbh (2.3%). Block nested within site showed similar results to the family effect (0.5–6.7%).  Family  block nested within site interaction  38  accounted for 3.8% (density 15) to 14.6% (volume) of the total variance. As expected, most of the total variance remained unexplained in the residual term (range: 33.1% (resistographbased wood density) to 81.9% (latewood density)) (Table 3.2). Random error has been reported to account for most of the total variance in similar studies (El-Kassaby et al. 2011b; Park et al. 2012; Ukrainetz et al. 2008).  3.3  Heritability estimates  Individual-tree narrow-sense heritabilities (Tables 3.2–3.4) estimated for dbh and volume were very low (0.03 and 0.07, respectively) (note, dbh is a component of volume estimation). Heritabilities for other variables were low to moderate, ranging from 0.15 (resistographbased wood density) to 0.26 (height, acoustic velocity, latewood density, and percentage of latewood). Unfortunately, standard errors of the heritability estimates were relatively large, thus reducing their reliability. Table 3.5 summarizes the comparison of growth trait heritabilities with other studies. Very low heritabilities for dbh and volume were similar to those published by El-Kassaby et al. (2011b) in Douglas-fir. Low heritabilities were also reported for dbh in interior/white spruce (Xie and Yanchuk 2002; Yanchuk and Kiss 1993; Ying and Morgenstern 1979) and in Norway spruce (Hannrup et al. 2004; Steffenrem et al. 2009).  Other studies obtained  moderate to high heritabilities for both dbh and volume. The moderate heritability estimated for height in this study was in a good accordance with other studies on interior spruce (Kiss and Yeh 1988; Yanchuk and Kiss 1993) or white spruce only (Lenz et al. 2010). Moderate heritabilities for height were also found in Douglas-fir (Ukrainetz et al. 2008), western larch (Ratcliffe 2012), and many other species reviewed by Cornelius (1994). Nevertheless, a number of studies reported a high heritability for this trait (see Table 3.5). The comparison of heritabilities for wood density and its components is listed in Table 3.6. Generally, the heritability for wood density is high; nevertheless, in the present study it was found to be substantially lower (0.21 ± 0.17) than in most previous studies. Similarly low to moderate heritabilities were observed for black spruce (Zhang 1998) and western larch (Ratcliffe 2012). Wood density heritabilities for interior/white spruce obtained by other studies ranged from 0.36 to 0.52 (Ivkovich et al. 2002a). Bouffier et al. (2008b) and Louzada  39  and Fonseca (2002) reported earlywood density to be distinctly more heritable than latewood density, with latewood proportion in between. This finding is consistent with results of Ukrainetz et al. (2008) or Gaspar et al. (2008); however, heritabilities for these three wood density characteristics estimated in this study were similarly moderate, with a slightly lower heritability for earlywood density.  Of note is a high standard error associated with  heritability for latewood density (0.26 ± 0.82); thus, this estimate should be interpreted with caution. Heritability for wood density of the first 15 rings (0.17 ± 0.17) was a little lower than that of overall wood density and considerably lower compared to heritabilities of wood density for the same species in studies with similar age. Heritability for resistograph-based wood density (0.15 ± 0.12) was rather low while heritabilities for acoustic velocity and MoE (0.26 ± 0.50 and 0.23 ± 0.38, respectively) were moderate but unreliable due to high standard errors. Moderate heritability was reported for resistograph-based wood density in Douglas-fir (0.25 ± 0.10) (El-Kassaby et al. 2011b) and for Pilodyn-based density in interior spruce (0.22 ± 0.11) (Yanchuk and Kiss 1993). Low heritability for acoustic velocity (0.14 ± 0.06) and moderate for MoE (0.30 ± 0.11) were obtained by El-Kassaby et al. (2011b). High but non-significant heritabilities for the three variables were estimated by Ratcliffe (2012) for western larch.  40  Table 3.2 Variance components, percentage of variance explained by a factor or interaction, and heritabilities ( Source  Height  Dbh  Volume  Estimate ± SE  %  Estimate ± SE  %  Estimate ± SE  %  Fam  0.327 ± 0.136 **  6.5  0.00035 ± 0.00054 **  0.8  0.0048 ± 0.0041 **  1.7  Site  2.187 ± 2.209 **  43.3  0.01880 ± 0.01918 **  41.8  0.1304 ± 0.1317 **  45.4  Site*Fam  0.095 ± 0.088 ns  1.9  0.00105 ± 0.00084 **  2.3  0.0069 ± 0.0053 ns  2.4  Block/Site  0.027 ± 0.036 **  0.5  0.00055 ± 0.00047 ns  1.2  0.0031 ± 0.0028 **  1.1  Fam*Block/Site  0.732 ± 0.117 **  14.5  0.00623 ± 0.00109 **  13.9  0.0419 ± 0.0068 **  14.6  Residual  1.686 ± 0.083 **  33.4  0.01795 ± 0.00087 **  39.9  0.1003 ± 0.0049 **  34.9  0.26 ± 0.20  ) for growth traits  0.03 ± 0.09  0.07 ± 0.11  ** significantly different at p ≤ 0.0001, * significantly different at 0.05 ≤ p > 0.0001, ns – non-significant Note that the estimated variances are for transformed variables.  Table 3.3 Variance components, percentage of variance explained by a factor or interaction, and heritabilities ( Source  Resistograph-based density  X-ray density  Acoustic velocity  ) for main wood quality attributes MoEd  Estimate ± SE  %  Estimate ± SE  %  Estimate ± SE  %  Estimate ± SE  %  Fam  0.0194 ± 0.0088 **  3.6  0.0050 ± 0.0020 **  5.3  0.0117 ± 0.0052 **  6.5  0.00061 ± 0.00028 **  5.8  Site  0.2424 ± 0.2525 **  45.5  0.0353 ± 0.0364 **  37.4  0.0107 ± 0.0143 **  6.0  0.00134 ± 0.00151 **  12.7  Site*Fam  0.0054 ± 0.0068 ns  1.0  0.0039 ± 0.0021 ns  4.1  0.0033 ± 0.0038 ns  1.8  0.00021 ± 0.00021 ns  2.0  Block/Site  0.0356 ± 0.0187 **  6.7  0.0012 ± 0.0011 **  1.3  0.0112 ± 0.0064 **  6.2  0.00053 ± 0.00031 **  5.0  Fam*Block/Site  0.0531 ± 0.0099 **  10.0  0.0042 ± 0.0016 *  4.4  0.0222 ± 0.0055 **  12.3  0.00116 ± 0.00030 **  11.0  Residual  0.1764 ± 0.0086 **  33.1  0.0449 ± 0.0022 ** 0.21 ± 0.17  47.5  0.1208 ± 0.0059 **  67.1  0.00671 ± 0.00033 ** 0.23 ± 0.38  63.5  0.15 ± 0.12  0.26 ± 0.50  ** significantly different at p ≤ 0.0001, * significantly different at 0.05 ≤ p > 0.0001, ns – non-significant Note that the estimated variances are for transformed variables.  41  Table 3.4 Variance components, percentage of variance explained by a factor or interaction, and heritabilities ( Source  Density 15  Earlywood density  Latewood density  ) for other wood characteristics Latewood %  Estimate ± SE  %  Estimate ± SE  %  Estimate ± SE  %  Estimate ± SE  %  Fam  0.0039 ± 0.0018 **  4.2  0.0044 ± 0.0018 **  5.0  0.000025 ± 0.000011 **  6.5  0.0071 ± 0.0027 **  6.5  Site  0.0290 ± 0.0305 **  30.9  0.0393 ± 0.0405 **  44.1  0.000005 ± 0.000007 **  1.2  0.0299 ± 0.0312 **  27.2  Site*Fam  0.0018 ± 0.0013 ns  1.9  0.0013 ± 0.0010 ns  1.5  0.000011 ± 0.000009 ns  3.0  0.0007 ± 0.0014 ns  0.6  Block/Site  0.0057 ± 0.0030 **  6.0  0.0037 ± 0.0020 **  4.1  0.000001 ± 0.000003 *  0.3  0.0036 ± 0.0021 **  3.2  Fam*Block/Site  0.0036 ± 0.0017 *  3.8  0.0043 ± 0.0014 *  4.8  0.000027 ± 0.000011 *  7.1  0.0051 ± 0.0022 *  4.6  Residual  0.0498 ± 0.0024 ** 0.17 ± 0.17  53.1  0.0363 ± 0.0018 ** 0.20 ± 0.15  40.6  0.000316 ± 0.000015 ** 0.26 ± 0.82  81.9  0.0637 ± 0.0031 ** 0.26 ± 0.25  57.9  ** significantly different at p ≤ 0.0001, * significantly different at 0.05 ≤ p > 0.0001, ns – non-significant Note that the estimated variances are for transformed variables.  42  Table 3.5 Comparison of heritabilities (  ) for growth traits with other studies Number of families  Age  Height  Dbh  Volume  interior spruce  25  37, 38  0.26 ± 0.20  0.03 ± 0.09  0.07 ± 0.11  interior spruce  80  22  0.56 ± 0.15  Ivkovich et al. (2002a) (EK)  interior spruce  80  20  0.74 ± 0.16  0.34 ± 0.12  Xie and Yanchuk (2002)  interior spruce  406  15, 20  0.15 to 0.34  0.10 to 0.23  Yanchuk and Kiss (1993)  interior spruce  40  15  0.30 ± 0.14  0.11 ± 0.11  Kiss and Yeh (1988)  interior spruce  174  3  0.52 ± 0.26  6  0.36 ± 0.19  10  0.29 ± 0.16 0.20 ± 0.13  Study  Species  Present study Ivkovich et al. (2002a) (PG)  Lenz et al. (2010)  white spruce  25  26  Merrill and Mohn (1985)  white spruce  50  20  Ying and Morgenstern (1979)  white spruce  –  22  0.01 to 0.32  0.04 to 0.10  Steffenrem et al. (2009)  Norway spruce  13  33  0.32 ± 0.19  0.07 ± 0.09  Hannrup et al. (2004)  Norway spruce  40  19  0.17 ± 0.16  0.13 ± 0.14  0.13 ± 0.14  Zhang and Morgenstern (1995)  black spruce  40  15  0.38  0.41  0.52  El-Kassaby et al. (2011b)  Douglas-fir  20  32  0.08 ± 0.07  0.04 ± 0.06  0.06 ± 0.06  Ukrainetz et al. (2008)  Douglas-fir  15  26  0.23 ± 0.09  0.25 ± 0.10  0.30 ± 0.11  Ratcliffe (2012)  western larch  25  20  0.31 ± 0.29  0.26 ± 0.42  0.27 ± 0.36  Pot et al. (2002)  maritime pine  73  14  0.46 ± 0.14  Matziris and Zobel (1973)  loblolly pine  –  5  0.44  0.29  0.28  Cornelius (1994)  various species  –  –  0.28  0.23  0.21  0.14  PG – Prince George progenies, EK – East Kootenay progenies  43  Table 3.6 Comparison of heritabilities (  ) for wood density and its components with other studies Number of families  Age  Earlywood density  Latewood density  Percentage of latewood  Overall density  interior spruce  25  37, 38  0.20 ± 0.15  0.26 ± 0.82  0.26 ± 0.25  0.21 ± 0.17  interior spruce  80  22  0.56 ± 0.18  0.36 ± 0.13  Ivkovich et al. (2002a, b) (EK)  interior spruce  80  20  0.55 ± 0.17  0.52 ± 0.14  Yanchuk and Kiss (1993)  interior spruce  40  15  Lenz et al. (2010)  white spruce  25  26  Beaulieu et al. (2006)  white spruce  39  36  0.45 ± 0.26  Zhang et al. (2004)  white spruce  35  36  0.37 ± 0.20  Corriveau et al. (1991)  white spruce  39  19  0.46 ± 0.13  Steffenrem et al. (2009)  Norway spruce  13  33  Hannrup et al. (2004)  Norway spruce  40  19  0.31 ± 0.17  0.74 ± 0.24  Zhang (1998)  black spruce  40  15  0.25  0.07  Zhang and Morgenstern (1995)  black spruce  40  15  0.86  0.65  0.18  El-Kassaby et al. (2011b)  Douglas-fir  20  32  Ukrainetz et al. (2008)  Douglas-fir  15  26  0.54 ± 0.16  0.21 ± 0.09  0.29 ± 0.11  0.47 ± 0.15  Vargas-Hernandez and Adams (1991)  Douglas-fir  60  15  0.47 ± 0.11  0.36 ± 0.10  0.24 ± 0.08  0.59 ± 0.12  Ratcliffe (2012)  western larch  25  20  Gaspar et al. (2008)  maritime pine  46  17  Pot et al. (2002)  maritime pine  73  14  Hodge and Purnell (1993)  slash pine  56  15, 25  Nicholls et al. (1980)  radiata pine  17  –  0.43  Matziris and Zobel (1973)  loblolly pine  –  5  0.47  Cornelius (1994)  various species  –  –  0.50  Study  Species  Present study Ivkovich et al. (2002a, b) (PG)  0.47 ± 0.16 0.69 ± 0.26  0.13 ± 0.11  0.50 ± 0.26 0.64 ± 0.23  0.53 ± 0.21 0.23 0.60 0.68 ± 0.16  0.22 ± 0.26 0.60 ± 0.19  0.26 ± 0.14  0.46 ± 0.17  0.63 ± 0.19 0.30 ± 0.13  PG – Prince George progenies, EK – East Kootenay progenies  44  0.13 ± 0.14  0.16 ± 0.16  3.4  Phenotypic and genetic correlations  Using the Mantel test, a strong and significant correlation between the phenotypic and genetic correlations was estimated (r = 0.85, p = 0.002), which shows that they are closely related and thus, on average the genetic control is high for the studied traits (i.e., the environmental effect is small). Phenotypic and genetic correlations (Table 3.7) within groups of growth traits and wood attributes as well as between derived variables and their sources corresponded with anticipation. Height, dbh, and volume showed strong positive correlations with each other in this study. Strong significant phenotypic and genetic correlations between diameter and height in interior spruce were also published by Ivkovich et al. (2002a) (> 0.77 and > 0.92, respectively) and Yanchuk and Kiss (1993) (0.85 and 0.94, respectively). Merrill and Mohn (1985) obtained a moderate phenotypic correlation for the two traits (0.55) for white spruce, while Beaulieu et al. (2006) reported a very weak and non-significant correlation (0.06). Bouffier et al. (2009) studying maritime pine also obtained a very weak correlation between height and diameter (< 0.1). A strong phenotypic correlation between dbh and volume (0.92) and moderate correlation between height and volume (0.43) were reported for white spruce by Beaulieu et al. (2006). MoEd calculated from acoustic velocity and overall x-ray density (Equation 1.5) was strongly positively correlated with the former and only moderately positively correlated with the latter. Correlation between overall x-ray density and density of the first 15 rings was strong, positive and significant (phenotypic 0.88 and genetic ~1), indicating that early testing for wood density of interior spruce using this method would be reliable and that the x-ray densitometry could, in older stands, be applied just to a part of a core. Overall x-ray density was also strongly positively correlated with earlywood density and latewood proportion (both phenotypic and genetic correlations were equal to or bigger than 0.90); however, its correlation with latewood density was only moderate (phenotypic 0.42 and genetic 0.69) (Figure 3.1). This result indicates that an overall x-ray density decrease would be primarily a result of both an earlywood density decrease and a latewood proportion 45  decrease. The density of latewood would decrease as well, but to a lesser extent. It indicates that wood density was primarily determined by earlywood density, due to the fairly low latewood proportion. Because latewood has a higher density than earlywood, a decrease in its proportion naturally leads to a decrease in overall x-ray density. Therefore, latewood proportion is one of the key characteristics of wood density (Louzada and Fonseca 2002). Low phenotypic and moderate genetic correlations between earlywood density and latewood density were obtained in this study (0.19 and 0.30 ± 0.27, respectively) (Table 3.7). A similar pattern with a strong genetic correlation of overall x-ray density with earlywood density and latewood proportion and slightly lower genetic correlation with latewood density was also revealed for Douglas-fir (Ukrainetz et al. 2008; Vargas-Hernandez and Adams 1991) and maritime pine (Gaspar et al. 2008) (Table 3.8) and a strong positive genetic correlation between x-ray density and latewood proportion (0.97) was obtained for white spruce (Park et al. 2012). In contrast, strong negative (Zhang and Morgenstern 1995), weak negative (Zhang 1998), and non-significant (Louzada 2003) genetic correlations were reported for overall density with latewood density (Table 3.8). Such results suggest that latewood density increases or remains the same when overall density decreases. Overall x-ray density was negatively correlated with growth traits. Phenotypic correlations were moderate, while genetic correlation was moderate with height and strong with dbh and volume. These results were consistent with other studies of spruce species (Table 3.9) (Rozenberg and Cahalan 1997; Zobel and Jett 1995). However, Beaulieu et al. (2006) found weak phenotypic correlations for all three variables, one of which (wood density versus dbh) was non-significant. Yanchuk and Kiss (1993) reported non-significant genetic correlations involving specific gravity, height, and dbh and Park et al. (2012) published a non-significant genetic correlation between x-ray density and height. Moreover, one of the reviewed studies of spruce revealed a moderate positive genetic correlation for relative density versus height (Corriveau et al. 1991). Although positive correlations between wood density and growth traits are an exception in spruce, they are quite common in hard pines (Zobel and Van Buijtenen 1989) (see Chapter 1.3.2.1 for more information). The statement that height growth affects wood density less than radial growth or volume (Vargas-Hernandez and Adams 1991; Zhang and Morgenstern 1995) concurred with all studies listed in Table 3.9 except for Beaulieu et al. (2006). 46  Differences in the relationship between overall x-ray density and growth traits were detected at all three sites, which is in agreement with other studies conducted on spruce (Ivkovich et al. 2002a; Rozenberg and VandeSype 1996; Zhang et al. 1996). The most and least favorable correlations were obtained for Quesnel and PGTIS, respectively (Table 3.10), which suggests that Quesnel provided a more favorable environment for growth of interior spruce than PGTIS. Similarly as in the case of the overall x-ray density, the resistograph-based wood density provided negative correlations with growth traits; the phenotypic correlation was weak with height and moderate with dbh and volume while the genetic correlation was moderate with height and strong with dbh and volume. The resistograph-based wood density was positively correlated with all wood attributes except for MoEd with which the genetic correlation was negative and non-significant. Phenotypic correlations with overall x-ray density, density of the first 15 rings (Figure 3.2), earlywood density, and percentage of latewood were moderate (0.53–0.60), whereas genetic correlations were strong (0.74–0.95). Phenotypic correlation with latewood density was weak and genetic correlation was weak and non-significant. The relationship between resistograph-based wood density and MoEd was also weak (Figure 3.3). Although contradictory results have been reported for wood density and MoE (see Chapter 1.3.6), the relationship between the two variables is generally poor in softwoods (Cave and Walker 1994).  In white spruce, Zhou and Smith (1991) detected a moderately strong  relationship between wood density and MoE (R2 = 0.30) but Beaulieu et al. (2006) did not find any relationship. The strong genetic correlation (0.84) between resistograph-based wood density and overall x-ray wood density suggests that the Resistograph is a suitable tool for in situ wood density assessment. Moreover, the very strong genetic correlation between the resistograph-based wood density and x-ray density of the first 15 rings (0.95) indicates that the Resistograph could be a suitable tool for estimating wood density in young interior spruce trees (as young as 15 years). The strong genetic correlation between resistograph-based wood density and x-ray wood density was also reported by Eckard et al. (2010) for loblolly pine (0.92), El-Kassaby et al. (2011b) for Douglas-fir (0.85), and by Ratcliffe (2012) for western larch (1.00).  47  Acoustic velocity (estimated by Director ST300) and growth traits provided weak negative phenotypic correlations and moderate positive genetic correlations. Correlations between acoustic velocity and wood attributes were also weak to moderate. Phenotypic correlations with overall x-ray density and density of the first 15 rings were 0.30 and 0.15, respectively (Figure 3.4) (genetic correlations were not significant). Similar results were also reported by El-Kassaby et al. (2011b) for 32-year-old coastal Douglas-fir and Chauhan and Walker (2006) for 16-year-old radiata pine (Table 3.11). Ratcliffe (2012) studying 20-year-old western larch obtained non-significant phenotypic (-0.04) and weak genetic (0.25) correlations between x-ray wood density and acoustic velocity. No relationship between acoustic velocity and density was reported by Ilic (2003) in 55 different tree species, Eckard et al. (2010) in loblolly pine, and Baar et al. (2011) in five tropical tree species, whilst Kumar et al. (2002) obtained weak negative correlations in radiata pine (Table 3.11). Although low or non-significant correlations suggest that acoustic velocity may not be a good predictor of wood density, a strong relationship between acoustic velocity and both MoEs and MoR has been detected (Auty and Achim 2008). Moreover, MoEd calculated using acoustic velocity was reported to be strongly correlated with laboratory-estimated MoEs (El-Kassaby et al. 2011b; Lindström et al. 2004; Raymond et al. 2008; Wang et al. 2000b) (Chapter 1.4.2.2). This finding implies that Director ST300 is not suitable for estimating wood density but that it is an excellent tool for assessing wood mechanical properties. Both phenotypic and genetic correlations between resistograph-based wood density and acoustic velocity were significant, providing values of -0.08 and -0.40, respectively (Figure 3.5). The moderate negative genetic correlation was however surprising: since the acoustic velocity was not well related to x-ray density, one would expect similar results for the resistograph-based density.  Nevertheless, some authors obtained non-significant genetic  correlations between the resistograph-based density and acoustic velocity (Eckard et al. 2010; El-Kassaby et al. 2011b; Ratcliffe 2012), which was in accordance with expectations. Ratcliffe (2012) also obtained a non-significant phenotypic correlation while El-Kassaby et al. (2011b) reported a moderate positive phenotypic correlation (0.35). Correlations of MoEd with growth traits were weak to moderate. Phenotypic correlations were negative and significant, whereas genetic correlations were positive and non-significant. Unlike wood density, it seems that MoEd may not be greatly affected by growth rate. MoEd 48  was moderately positively correlated with all wood density traits except for latewood density and latewood percentage, for which genetic correlations were non-significant (Table 3.7). Contrasting results were obtained in a number of studies investigating the relationship between MoE and wood density, where moderate negative correlations as well as strong positive correlations were both reported (see Chapter 1.3.6). A positive correlation between wood density and laboratory-estimated MoEs was obtained e.g. for maritime pine, 0.61 (Reuling 2005 in Bouffier et al. 2009) or Douglas-fir, 0.72 (El-Kassaby et al. 2011b). Moderate phenotypic and high genetic correlations between wood density and MoEd derived from acoustic velocity were reported for Douglas-fir (0.66 and 0.81, respectively) (ElKassaby et al. 2011b). Moderate phenotypic but weak genetic correlations between the two variables were estimated for western larch (0.34, and 0.25, respectively) (Ratcliffe 2012). Lindström et al. (2004) also obtained moderate phenotypic correlation (0.32) for radiata pine. Relationships among all traits were visualized based on their phenotypic and genetic correlations (Figures 3.6 and 3.7). The similarity between the two types of correlations suggests that there is a high genetic control for these traits, which was confirmed by the Mantel test. Two apparent clusters appeared, comprising growth traits on one side and wood traits on the other. Traits within the clusters were positively correlated whilst those from opposing clusters, i.e. wood quality traits vs. growth traits, were negatively correlated. As discussed above, the overall x-ray density was closely related to density of the first 15 rings, earlywood density and latewood proportion, while latewood density was slightly more distant. In addition acoustic velocity and MoEd were situated further from the cluster’s centre. Moreover, in case of the genetic correlations’ visualization, acoustic velocity stood in between the clusters, being positively correlated with growth traits. Resistograph-based density was not far from the centre, but it was surprisingly closer to density of the first 15 rings than to the overall x-ray density.  49  Table 3.7 Phenotypic (above diagonal) and genetic correlations (below diagonal) between studied traits Height Height  Dbh  Volume  v  D res  D 15  ED  LD  L%  D x-ray  MoEd  0.71 **  0.85 **  -0.03 ns  -0.29 **  -0.31 **  -0.42 **  0.06 ns  -0.20 **  -0.34 **  -0.13 **  0.97 **  -0.26 **  -0.41 **  -0.51 **  -0.59 **  -0.13 *  -0.46 **  -0.57 **  -0.41 **  -0.21 **  -0.40 **  -0.48 **  -0.58 **  -0.08 *  -0.41 **  -0.54 **  -0.36 **  -0.08 *  0.15 **  0.28 **  0.32 **  0.21 **  0.30 **  0.94 **  0.60 **  0.58 **  0.18 **  0.53 **  0.59 **  0.14 **  0.80 **  0.39 **  0.84 **  0.88 **  0.41 **  0.19 **  0.74 **  0.94 **  0.55 **  0.49 **  0.42 **  0.41 **  0.91 **  0.48 **  Dbh  0.80 ± 0.24  Volume  0.94 ± 0.07  0.92 ± 0.04  V  0.42 ± 0.26  0.49 ± 0.56  D res  0.47 ± 0.40  -0.55 ± 0.24 -0.99 ± 0.64 -0.81 ± 0.35 -0.40 ± 0.27 a  D 15  -0.50 ± 0.25  -0.96 ± 0.32  0.03 ± 0.32  0.95 ± 0.13  ED  -0.37 ± 0.25 -0.62 ± 0.37 -0.57 ± 0.31  0.43 ± 0.24  0.74 ± 0.19  0.95 ± 0.09  LD  -0.34 ± 0.28 -0.16 ± 0.55 -0.31 ± 0.41 -0.08 ± 0.30  0.27 ± 0.29  0.48 ± 0.23  0.30 ± 0.27  L%  -0.53 ± 0.22 -0.85 ± 0.37 -0.80 ± 0.26 -0.06 ± 0.28  0.83 ± 0.16  0.90 ± 0.07  0.62 ± 0.16  0.82 ± 0.12  0.84 ± 0.16  a  0.90 ± 0.05  0.69 ± 0.18  0.90 ± 0.05  0.37 ± 0.28  0.68 ± 0.17  0.15 ± 0.29  0.25 ± 0.26  D x-ray MoEd  -0.999  -0.50 ± 0.23 -0.71 ± 0.34 -0.71 ± 0.27 0.24 ± 0.29  0.12 ± 0.52  0.17 ± 0.42  0.20 ± 0.27  0.96 ± 0.02 -0.11 ± 0.31  0.999  0.58 ** 0.51 ± 0.21  Dbh – diameter in breast height, v – acoustic velocity, D res – resistograph-based wood density, D 15 – density of the first fifteen rings, ED – early- wood density, LD – latewood density, L% – percentage of latewood, D x-ray – wood density measured by x-rays, MoEd – dynamic modulus of elasticity ** significantly different at p ≤ 0.0001, * significantly different at 0.05 ≤ p > 0.0001, ns – non-significant; a – correlation fixed at a boundary, standard error cannot be estimated  50  Table 3.8 Comparison of phenotypic and genetic correlations between overall density and its components with other studies Number Study  Species  of  Wood  families Present study  interior spruce  25  37, 38  Hannrup et al. (2004)  Norway spruce  40  19  Zhang (1998)  black spruce  40  15  Zhang and Morgenstern (1995)  black spruce  40  15  Ukrainetz et al. (2008)  Douglas-fir  15  26  Vargas-Hernandez and Adams (1991)  Douglas-fir  60  15  Gaspar et al. (2008)  maritime pine  46  17  Louzada (2003)  maritime pine  15  13  Correlation  density  Age  trait  phenotypic  genetic  ED LD L% ED LD L% ED LD L% ED LD L% ED LD L% ED LD L% ED LD L% ED LD L%  0.94 * 0.42 * 0.91 *  0.90 ± 0.05 0.69 ± 0.18 0.90 ± 0.05 0.97 * 0.49 * 0.69 * 0.59 / -0.19 / 0.32 / 0.72 / -0.73 / 0.21 / 0.91 ± 0.04 0.71 ± 0.13 0.93 ± 0.03 0.94 ± 0.02 0.74 ± 0.10 0.95 ± 0.06 0.96 ± 0.06 0.79 ± 0.13 0.96 ± 0.03 0.99 ± 0.00 0.05 ± 0.27 1.06  0.89 / 0.14 / 0.72 / 0.82 / -0.09 / 0.49 / 0.74 * 0.47 * 0.85 *  0.67 * 0.72 * 0.90 * 0.96 / 0.52 / 0.97 /  * significantly different at p ≤ 0.05, ns – non-significant, ED – earlywood density, LD – latewood density, L% – percentage of latewood  51  Table 3.9 Phenotypic and genetic correlations between growth traits and wood density for spruce species Number Study  Species  of families  Age  Present study  interior spruce  25  37, 38  Ivkovich et al. (2002a) (PG)  interior spruce  80  22  Ivkovich et al. (2002a) (EK)  interior spruce  80  20  Yanchuk and Kiss (1993)  interior spruce  40  15  Park et al. (2012)  white spruce  5  14  Beaulieu et al. (2006)  white spruce  39  36  Corriveau et al. (1991)  white spruce  39  19  Zhang and Morgenstern (1995)  black spruce  40  15  Hannrup et al. (2004)  Norway spruce  40  19  Hylen (1997)  Norway spruce  47  28  Growth trait Height Dbh Volume Height Dbh Height Dbh Height Dbh Height Volume Height Dbh Volume Height Dbh Volume Height Dbh Volume Height Dbh Volume Height Dbh  Correlation phenotypic  genetic  -0.34 * -0.57 * -0.54 * -0.45 * -0.58 * -0.39 * -0.52 * -0.40 * -0.46 *  -0.50 ± 0.23 -0.71 ± 0.34 -0.71 ± 0.27 -0.36 ± 0.20 -0.49 ± 0.23 -0.49 ± 0.13 -0.67 ± 0.11 0.00 ± 0.28 0.08 ± 0.41 -0.34 ns -0.75 *  -0.19 * -0.10 ns/ -0.17 * -0.17 * -0.36 * -0.33 * -0.26 / -0.39 / -0.37 /  0.35 ± 0.25 -0.32 ± 0.27 -0.26 ± 0.26  -0.52*, -0.82* -0.84*, -0.94* -0.77*, -0.96* -0.19 * -0.45 *  * significantly different at p ≤ 0.05, ns – non-significant, PG – Prince George progenies, EK – East Kootenay progenies  52  Table 3.10 Phenotypic (lower number) and genetic (upper number) correlations between overall x-ray  density and growth traits for the three studied sites PGTIS  Aleza lake  Quesnel  Across sites  Height  -0.46 ± 0.47 -0.24**  -0.67 ± 0.3 -0.26**  -0.42 ± 0.25 -0.19*  -0.50 ± 0.23 -0.34 **  -0.95 ± 0.76 -0.43**  -0.67 ± 0.29 -0.43**  -0.69 ± 0.21 -0.48**  -0.71 ± 0.34  Dbh  -0.93 ± 0.81  -0.85 ± 0.28  -0.61 ± 0.21  -0.71 ± 0.27  -0.39**  -0.43**  -0.43**  -0.54 **  Volume  -0.57 **  ** significantly different at p ≤ 0.0001, * significantly different at 0.05 ≤ p > 0.0001, ns – non-significant, PGTIS – Prince George Tree Improvement Station  Table 3.11 Phenotypic and genetic correlations between density and standing-tree acoustic variables (acoustic velocity, squared acoustic velocity, and ToF) Study  Species  Present study  interior spruce  El-Kassaby et al. (2011b)  Douglas-fir  Correlation  Age  Acoustic variable  phenotypic  genetic  37, 38  v  0.30 *  0.20 ± 0.27  32  v  0.35 *  0.28 ± 0.25  ns  0.25 ± 0.24  Ratcliffe (2012)  western larch  20  v  Chauhan and Walker (2006)  radiata pine  8  v  -0.04  0.14 /  16  v  0.26 /  25  v  0.42 /  2  Eckard et al. (2010)  loblolly pine  8  v  Kumar et al. (2002)  radiata pine  12  ToF  -0.03 ± 0.15 -0.27 /  * significantly different at p ≤ 0.05, ns – non-significant, v – acoustic velocity, ToF – time of flight  53  -0.26 ± 0.23  0.70  45  R² = 0.88 R² = 0.18 0.65  R² = 0.83  40  0.60 35  0.55  ED and LD [g·cm-3]  0.50 25 0.45 20 0.40  Latewood proportion [%]  30  15 0.35  10 0.30  5  0.25  0.20 0.25  0.30  0.35  0.40  0.45  0 0.50  Overall x-ray density [g·cm-3]  Figure 3.1 Relationship between overall x-ray density and earlywood density (ED, blue), latewood density (LD, black), and latewood proportion (L%, red)  54  0.50  R² = 0.35 R² = 0.36  X-ray denisty [g·cm-3]  0.45  0.40  0.35  0.30  0.25 0  5  10  15  20  25  30  35  40  Resistograph-based density  Figure 3.2  Relationship between resistograph-based density and  overall x-ray (black) and 15-year x-ray density (blue)  10  R² = 0.02  9 8 MoEd [GPa]  7 6 5 4 3 2 1 0 0  5  10  15  20  25  30  35  40  Resistograph-based density  Figure 3.3  Relationship between resistograph-based density and  MoEd  55  0.50  R² = 0.09 R² = 0.02  X-ray denisty [g·cm-3]  0.45  0.40  0.35  0.30  0.25 2.0  2.5  3.0  3.5  4.0  4.5  5.0  Acoustic velocity [km·s-1]  Figure 3.4 Relationship between acoustic velocity and overall x-ray (black) and 15-year x-ray density (blue)  45  R² = 0.006  Resistograph-based density  40 35 30 25 20 15 10 5 0 2.0  2.5  3.0  3.5  Acoustic velocity  4.0  4.5  5.0  [km.s-1]  Figure 3.5 Relationship between acoustic velocity and resistographbased density  56  Figure 3.6 Visualization of the relationships among traits based on their phenotypic correlations*  Figure 3.7 Visualization of the relationships among traits based on their genetic correlations* * Solid lines represent positive correlations while dashed lines represent negative correlations, Vol – volume, Dens – x-ray overall density, Dens15 – density of the first 15 rings, Res – resistographbased density, v – acoustic velocity, ED – earlywood density, LD – latewood density, L% – latewood proportion  57  4 Conclusion The present study is based on two fundamental assumptions: 1- wood mechanical properties and wood density are heritable traits and 2- these traits can be quantified (assessed) using indirect, non-destructive methods applied in situ. Wood density as the best single predictor of wood quality greatly affects wood suitability for different end-uses. Therefore, it is of vital importance to incorporate this trait among existing selection criteria; however, this task brings about a need for finding a fast, reliable, non-destructive, and inexpensive tool for the assessment of live standing trees. Two different tools were evaluated in this study: 1- Resistograph IML F300 estimating wood density through drilling resistance of wood and 2- Director ST300 providing an indirect estimate of modulus of elasticity calculated from sound velocity. This study confirms that growth traits in interior spruce are negatively correlated with wood density, but not with MoE. It implies that selection based on tree volume would result in wood density reduction, while MoE would remain unchanged. Height growth affects wood density less than radial growth; therefore, using just height as a proxy to volume appears to be a better selection approach than volume itself. Moreover, heritability of height was moderate whereas heritabilities for volume and radial growth were very low. The study further shows that the Resistograph is a reliable tool for estimating wood density in interior spruce and, moreover, that it is suitable for testing young trees (15 years). However, although heritability of wood density estimated by x-ray was moderate, heritability of resistograph-based density was low. The Director seems to be a useful tool for estimating MoEd; however, its reliability cannot be proven by this study because MoEd is not well correlated with any of the measured variables and no laboratory testing was performed. Moreover, since the negative relationship between fast growth and MoE was not confirmed, the inclusion of MoE into selection criteria may not bring significant improvement. The strong and significant correlation between phenotypic and genetic correlations estimated using the Mantel test suggests that the studied traits are only influenced by the environment to a limited extent. When visualized using the Pajek networking software, the phenotypic correlations mirrored genetic correlations, producing two apparent clusters. The cluster of 58  positively correlated growth traits was negatively correlated with that of positively correlated wood traits, with the exception of the positive genetic correlations of acoustic velocity with height and volume. It can be concluded that the Resistograph is an appropriate tool for indirect non-destructive assessment of wood density in interior spruce.  59  References Abdel-Gadir, A.Y., Krahmer, R.L., and McKimmy, M.D. 1993. Relationship between intraring variables in mature Douglas-fir trees from provenance plantations. Wood and Fiber Science 25(2): 182-191. Achim, A., and Carter, P. 2005. Assessing the potential of acoustic tools to predict sawn batten stiffness in Sitka spruce. Available from http://www.fibre-gen.com/hm200.html [accessed 20.6. 2012]. Akachuku, A.E. 1984. The possibility of tree selection and breeding for genetic improvement of wood properties of Gmelina arborea. Forest Science 30(2): 275-283. Alden, H.A. 1997. Softwoods of North America. U.S. Department of Agriculture, Forest Service, Forest Products Laboratory, Madison, Wisconsin. Andrews, M. 2002. Wood quality measurement – son et lumière. New Zealand Journal of Forestry 47(3): 19-21. Armstrong, J.P., Skaar, C., and Dezeeuw, C. 1984. The effect of specific gravity on several mechanical properties of some world woods. Wood Science and Technology 18(2): 137-146. Aubry, C.A., Adams, W.T., and Fahey, T.D. 1998. Determination of relative economic weights for multitrait selection in coastal Douglas-fir. Canadian Journal of Forest Research 28(8): 1164-1170. Auty, D., and Achim, A. 2008. The relationship between standing tree acoustic assessment and timber quality in Scots pine and the practical implications for assessing timber quality from naturally regenerated stands. Forestry 81(4): 475-487. B.C. Ministry of Forests, M.a.L. 2010. The State of British Columbia's Forests. 3rd ed. Forest Practices and Investment Branch, Victoria, B.C. Baar, J., Tippner, J., and Gryc, V. 2011. The influence of wood density on longitudinal wave velocity determined by the ultrasound method in comparison to the resonance longitudinal method. European Journal of Wood and Wood Products. Available from http://www.springerlink.com/content/k6p63566n3u2l5wt/.  60  Bachle, H., and Walker, J. 2006. The influence of temperature on the velocity of sound in green pine wood. Holz als Roh- und Werkstoff 64(5): 429-430. Batagelj, V., and Mrvar, A. 2012. Pajek and Pajek-XXL - programs for analysis and visualization of very large networks. Reference manual. Ljubljana, Slovenia. Available from http://pajek.imfm.si/doku.php?id=pajek. Beaulieu, J. 2003. Genetic variation in tracheid length and relationships with growth and wood traits in eastern white spruce (Picea glauca). Wood and Fiber Science 35(4): 609-616. Beaulieu, J., Zhang, S.Y., Yu, Q.B., and Rainville, A. 2006. Comparison between genetic and environmental influences on lumber bending properties in young white spruce. Wood and Fiber Science 38(3): 553-564. Bouffier, L., Charlot, C., Raffin, A., Rozenberg, P., and Kremer, A. 2008a. Can wood density be efficiently selected at early stage in maritime pine (Pinus pinaster Ait.)? Annals of Forest Science 65(1): 106-106. Bouffier, L., Raffin, A., Rozenberg, P., Meredieu, C., and Kremer, A. 2009. What are the consequences of growth selection on wood density in the French maritime pine breeding programme? Tree Genetics & Genomes 5(1): 11-25. Bouffier, L., Rozenberg, P., Raffin, A., and Kremer, A. 2008b. Wood density variability in successive breeding populations of maritime pine. Canadian Journal of Forest Research 38(8): 2148-2158. Bousquet, J., Isabel, N., Pelgas, B., Cottrell, J., Rungis, D., and Ritland, K. 2007. Spruce. In Genome Mapping and Molecular Breeding in Plants. Edited by Chittaranjan Kole. Springer-Verlag, Berlin, Heidelberg. pp. 93-114. Brashaw, B.K., and Ross, R.J. 2002. Probing, drilling an coring - nondestructive evaluation methods. In Nondestructive evaluation of wood. Edited by Roy F. Pellerin, and Robert J. Ross. Forest product society, USA. p. 55. Brazier, J.D. 1970. Timber improvement II. The effect of vigour on young-growth Sitka spruce. Forestry 43(2): 135-150.  61  Bucur, V. 2003. Nondestructive characterization and imaging of wood. Springer-Verlag Berlin Heidelberg, Germany. Bucur, V. 2006a. Acoustics of wood. 2nd ed. Springer-Verlag Berlin Heidelberg, Germany. Bucur, V. 2006b. Urban forest acoustics. Springer-Verlag Berlin Heidelberg, Germany. Carter, P. 2007. Breeding for wood quality; acoustic tools and technology. In AFG & IUFRO SPWG Joint Conference, Hobart, Tasmania. Carter, P., Briggs, D., Ross, R.J., and Wang, X. 2005. Acoustic testing to enhance Western forest values and meet customer wood quality needs. In Productivity of Western forests: a forest products focus. General technical report PNW-GTR-642. Edited by Constance A. Harrington, and Stephen H. Schoenholtz. U.S. Department of Agriculture, Forest Service, Pacific Northwest Research, Portland. pp. 121-129. Cave, I.D., and Walker, J.C.F. 1994. Stiffness of wood in fast-grown plantation softwoods: The influence of microfibrial angle. Forest Products Journal 44(5): 43-48. Ceraldi, C., Mormone, V., and Ermolli, E.R. 2001. Resistographic inspection of ancient timber structures for the evaluation of mechanical characteristics. Materials and Structures 34(235): 59-64. Chang, C.I., and Kennedy, R.W. 1967. Influence of specific gravity and growth rate on dry wood production in plantation-grown white spruce. Forestry Chronicle 43(2): 165-173. Chantre, G., and Rozenberg, P. 1997. Can drill resistance profiles (Resistograph) lead to within-profile and within-ring density parameters in Douglas-fir wood? In Timber management toward wood quality and end-product values: workshop highlights and related infromation. Edited by S. Y. Zhang, R. Gosselin, and G. Chauret. Forintek Canada Corp., Quebec. pp. 41-47. Chauhan, S., Donnelly, R., Huang, C.L., Nakada, R., Yafang, Y., and Walker, J.C.F. 2006a. Wood quality: in context. In Primary wood processing: principles and practice. Edited by J.C.F. Walker. Springer, Dordrecht, The Netherlands. pp. 121-158.  62  Chauhan, S., Donnelly, R., Huang, C.L., Nakada, R., Yafang, Y., and Walker, J.C.F. 2006b. Wood quality: multifaceted opportunities. In Primary wood processing: principles and practice. Edited by J.C.F. Walker. Springer, Dordrecht, The Netherlands. pp. 159-202. Chauhan, S.S., Entwistle, K.M., and Walker, J.C.F. 2005. Differences in acoustic velocity by resonance and transit-time methods in an anisotropic laminated wood medium. Holzforschung 59(4): 428-434. Chauhan, S.S., and Walker, J.C.F. 2006. Variations in acoustic velocity and density with age, and their interrelationships in radiata pine. Forest Ecology and Management 229(1-3): 388-394. Cherry, M.L., Vikram, V., Briggs, D., Cress, D.W., and Howe, G.T. 2008. Genetic variation in direct and indirect measures of wood stiffness in coastal Douglas-fir. Canadian Journal of Forest Research 38(9): 2476-2486. Choi, A.S.C. 1986. Correlation between mechanical strength of wood and annual ring characteristics in Douglas-fir juvenile and mature wood (Master's thesis). Oregon State University, Corvallis, OR. p. 84. Available from http://ir.library.oregonstate.edu /xmlui/handle/1957/18026. Coates, K.D., Haeussler, S., Lindeburgh, S., Pojar, R., and Stock, A.J. 1994. Ecology and silviculture of interior spruce in British Columbia. FRDA Report 220. B.C. Ministry of Forests, Victoria, Canada. Cornelius, J. 1994. Heritabilities and additive genetic coefficients of variation in forest trees. Canadian Journal of Forest Research 24(2): 372-379. Corriveau, A., Beaulieu, J., and Daoust, G. 1991. Heritability and genetic correlations of wood characteristics of Upper Ottawa Valley in white spruce population grown in Quebec. Forestry Chronicle 67(6): 698-705. Corriveau, A., Beaulieu, J., and Mothe, F. 1987. Wood density of natural white spruce population in Quebec. Canadian Journal of Forest Research 17(7): 675-682. Cown, D.J., and Clement, B.C. 1983. A wood densitometer using direct scanning with xrays. Wood Science and Technology 17(2): 91-99.  63  Dawson, A.B. 1989. Soils of the Prince George - McLeod Lake area. BC soil survey. Technical report no. 23. Ministry of Environment and Ministry of Agriculture and Fisheries, Victoria, B.C. DeBell, J.D., Tappeiner, J.C., and Krahmer, R.L. 1994. Wood density of western hemlock: Effect of ring width. Canadian Journal of Forest Research 24(3): 638-641. Desch, H.E., and Dinwoodie, J.M. 1996. Timber: structure, properties, conversion, and use. 7th ed. Food Products Press, New York. Dobbs, R.C., Edwards, D.G.W., Konishi, J., Wallinger, D., and Wiens, J.C. 1976. Guideline to collecting cones of B.C. conifers. British Columbia Forest Service / Canada, Forestry Service, join report no. 3, Victoria, B.C. Dolenko, A.J., Shields, J.K., King, F.W., Roff, J.W., and Ostaff, D. 1981. Wood protection. In Canadian woods: their properties and uses. Edited by E.J. Mullins, and T.S. McKnight. University of Toronto Press, Toronto, Canada. pp. 177-223. Duchesne, I., and Zhang, S.Y. 2004. Variation in tree growth, wood density, and pulp fiber properties of 35 white spruce (Picea glauca (Moench) Voss) families grown in Quebec. Wood and Fiber Science 36(4): 467-475. Echols, R.M. 1973. Uniformity of wood density assessed from x-rays of increment cores. Wood Science and Technology 7(1): 34-44. Eckard, J.T., Isik, F., Bullock, B., Li, B.L., and Gumpertz, M. 2010. Selection efficiency for solid wood traits in Pinus taeda using time-of-flight acoustic and micro-drill resistance methods. Forest Science 56(3): 233-241. El-Kassaby, Y.A., Cappa, E.P., Liewlaksaneeyanawin, C., Klapste, J., and Lstiburek, M. 2011a. Breeding without Breeding: Is a complete pedigree necessary for efficient breeding? Plos One 6(10): 11. El-Kassaby, Y.A., and Lstiburek, M. 2009. Breeding without breeding. Genetics Research 91(2): 111-120. El-Kassaby, Y.A., Mansfield, S., Isik, F., and Stoehr, M. 2011b. In situ wood quality assessment in Douglas-fir. Tree Genetics & Genomes 7(3): 553-561.  64  Eriksson, G., Ekberg, I., and Clapham, D. 2006. An introduction to forest genetics. Genetic Center, Department of Plant Biology and Forest Genetics, SLU, Uppsala, Sweden. Evans, R., and Ilic, J. 2001. Rapid prediction of wood stiffness from microfibril, angle and density. Forest Products Journal 51(3): 53-57. Falconer, D.S., and Mackay, T.F.C. 1996. Introduction to quantitative genetics. Longman, Harlow, England. Farrar, J.L. 1995. Trees in Canada. Fitzhenry & Whiteside Ltd. and the Canadian Forest Service, Ottawa. Feio, A.O., Lourenco, P.B., and Machado, J.S. 2007. Non-destructive evaluation of the mechanical behavior of chestnut wood in tension and compression parallel to grain. International Journal of Architectural Heritage 1(3): 272-292. Fibre-gen. 2004. Director ST300 - operating manual. Available from http://www.fibregen.com/documents/ST300OperatingManual.pdf [accessed 5.3. 2012]. Fibre-gen. 2010. Hitman. Available from http://www.fibre-gen.com/index.html [accessed 5.3. 2012]. Forintek.  2007.  White  spruce  &  Engelmann  spruce.  Available  from  http://www.myendnoteweb.com/EndNoteWeb.html?func=new& [accessed 20.9. 2011]. Fowler, D.P., and Roche, L. 1976. Genetics of Engelmann spruce. U.S. Forest Service Research Paper WO(30): 1-13. Gao, S., Wang, X., Wang, L., and Allison, R.B. 2012. Effect of temperature on acoustic evaluation of standing trees and logs. Part 1. Laboratory investigation. Wood and Fiber Science 44(3): 1-12. Garman, E.H. 1957. Occurrence of spruce in the interior of British Columbia. Department of Lands and Forests, British Columbia Forest Service, Victoria, B.C. Gaspar, M.J., Louzada, J.L., Aguiar, A., and Almeida, M.H. 2008. Genetic correlations between wood quality traits of Pinus pinaster Ait. Annals of Forest Science 65(7).  65  Gilmour, A.R., Gogel, B.J., Cullis, B.R., and Thompson, R. 2009. ASReml user guide. Release 3.0. VSN International Ltd, Hemel Hempstead, UK. Available from www.vsni.co.uk. Grabianowski, M., Manley, B., and Walker, J.C.F. 2006. Acoustic measurements on standing trees, logs and green lumber. Wood Science and Technology 40(3): 205-216. Grabner, M., Muller, U., Gierlinger, N., and Wimmer, R. 2005. Effects of heartwood extractives on mechanical properties of larch. Iawa Journal 26(2): 211-220. Greaves, B.L., Borralho, N.M.G., Raymond, C.A., and Farrington, A. 1996. Use of a Pilodyn for the indirect selection of basic density in Eucalyptus nitens. Canadian Journal of Forest Research 26(9): 1643-1650. Hannrup, B., Cahalan, C., Chantre, G., Grabner, M., Karlsson, B., Le Bayon, I., Jones, G.L., Muller, U., Pereira, H., Rodrigues, J.C., Rosner, S., Rozenberg, P., Wilhelmsson, L., and Wimmer, R. 2004. Genetic parameters of growth and wood quality traits in Picea abies. Scandinavian Journal of Forest Research 19(1): 14-29. Hannrup, B., Ekberg, I., and Persson, A. 2000. Genetic correlations among wood, growth capacity and stem traits in Pinus sylvestris. Scandinavian Journal of Forest Research 15(2): 161-170. Hansen, C.P. 2000. Application of the Pilodyn in forest tree improvement. Technical note no. 55. Danida Forest Seed Centre. Hatton, J.V., and Cook, J. 1992. Kraft pulps from second-growth Douglas-fir: relationships between wood, fiber, pulp, and handsheet properties. Tappi Journal 75(1): 137-144. Hein, O. 2008. Internal tree structure assessment with the IML Resistograph. Instrumenta Mechanik Labor System GmbH. Hernández, R.E., Bustos, C., Fortin, Y., and Beaulieu, J. 2001. Wood machining properties of white spruce from plantation forests. Forest Products Journal 51(6): 82-88. Hill, P. 2000. Wood density of improved compared with unimproved maritime pine (Pinus pinaster). CALMScience 3(3): 309-315.  66  Hoadley, R.B. 2000. Understanding wood: a craftsman's guide to wood technology. Taunton Press, Newtown, CT. Hodge, G.R., and Purnell, R.C. 1993. Genetic parameter estimates for wood density, transition age, and radial growth in slash pine. Canadian Journal of Forest Research 23(9): 1881-1891. Horáček, P. 1998. Fyzikální a mechanické vlastnosti dřeva. Department of Wood Science, Mendel  University  in  Brno,  Czech  Republic.  Available  from  http://wood.mendelu.cz/cz/sections/Props/?q=node/51. Hosius, B., Bergmann, F., Konnert, M., and Henkel, W. 2000. A concept for seed orchards based on isoenzyme gene markers. Forest Ecology and Management 131(1-3): 143152. Hubert, J., and Lee, S. 2005. A review of the relative roles of silviculture and tree breeding in tree improvement: the example of Sitka spruce in Britain and possible lessons for hardwood breeding. Forestry 78(2): 109-120. Hylen, G. 1997. Genetic variation of wood density and its relationship with growth traits in young Norway spruce. Silvae Genetica 46(1): 55-&. Ilic, J. 2003. Dynamic MOE of 55 species using small wood beams. Holz als Roh- und Werkstoff 61(3): 167-172. IML. 2008a. Applications for pole inspections (drilling resistance measurement). IML Resi F300, F400, F500. Edited by Erich Hunger. Instrumenta Mechanic Labor System GmbH, Wiesloch, Germany. IML. 2008b. Test- and measuring instruments for trees and timber structures (drilling restistance measurement). IML Resi F300, F400, F500. Instrumenta Mechanik Labor System GmbH, Wiesloch, Germany. IML. 2008c. Testing wood quality? The Resistograph F series - Q: The right tool for the job. Instrumenta Mechanik Labor System GmbH, Wiesloch, Germany. IML. 2008d. Tools for an internal view. Instrumenta Mechanik Labor System GmbH, Wiesloch, Germany.  67  Isenberg, I.H. 1943. Properties of pulpwoods. The institure of paper chemistry. Isik, F., and Li, B.L. 2003. Rapid assessment of wood density of live trees using the Resistograph for selection in tree improvement programs. Canadian Journal of Forest Research 33(12): 2426-2435. Isik, F., Mora, C.R., and Schimleck, L.R. 2011. Genetic variation in Pinus taeda wood properties predicted using non-destructive techniques. Annals of Forest Science 68(2): 283-293. Ivkovich, M., and Koshy, M. 2002. Optimization of multiple trait selection in western hemlock (Tsuga heterophylla (Raf.) Sarg.) including pulp and paper properties. Annals of Forest Science 59(5-6): 577-582. Ivkovich, M., Namkoong, G., and Koshy, M. 2002a. Genetic variation in wood properties of interior spruce. I. Growth, latewood percentage, and wood density. Canadian Journal of Forest Research 32(12): 2116-2127. Ivkovich, M., Namkoong, G., and Koshy, M. 2002b. Genetic variation in wood properties of interior spruce. II. Tracheid characteristics. Canadian Journal of Forest Research 32(12): 2128-2139. Jaquish, B. 2009. British Columbia Interior spruce tree improvement program summary (unpublished report). Jaquish, B., Ashley, V., Phillips, G., and Hooge, B. 2009. Interior spruce tree breeding program. Forest Genetics Council of BC. Jaquish, B., Ashley, V., Phillips, G., and Hooge, B. 2011. Interior spruce tree breeding program. Forest Genetics Council of BC. Johnstone, D.M., Ades, P.K., Moore, G.M., and Smith, I.W. 2007. Predicting wood decay in eucalypts using an expert system and the IML-Resistograph drill. Arboriculture & Urban Forestry 33(2): 76-82. Jones, T.G., and Emms, G.W. 2010. Influence of acoustic velocity, density, and knots on the stiffness grade outturn of radiata pine logs. Wood and Fiber Science 42(1): 1-9.  68  Jozsa, L.A., and Middleton, G.R. 1994. A Discussion of wood quality attributes and their practical implications. In Special publication no. SP-34. Forintek Canada Corp., Western Laboratory, Vancouver, B.C. p. 42. Kahl, T., Wirth, C., Mund, M., Boehnisch, G., and Schulze, E.D. 2009. Using drill resistance to quantify the density in coarse woody debris of Norway spruce. European Journal of Forest Research 128(5): 467-473. Keith, C.T., and Kellogg, R.M. 1981. The structure of wood. In Canadian woods: their properties and uses. Edited by E.J. Mullins, and T.S. McKnight. University of Toronto Press, Toronto, Canada. pp. 41-70. Kennedy, E.I., Jessome, A.P., and Petro, F.J. 1968. Specific gravity survey of eastern Canadian woods. Queen's printer and controller of stationery, Ottawa, Canada. Kennedy, R.W. 1995. Coniferous wood quality in the future - concerns and strategies. Wood Science and Technology 29(5): 321-338. Kersten, W., and Schwarze, F.W.M.R. 2005. Development of decay in the sapwood of trees wounded by the use of decay detecting devices. Arboricultural Journal 28(3): 165-181. Khalil, M.A.K. 1985a. Genetic variation in eastern white spruce (Picea glauca (Moench) Voss) populations. Canadian Journal of Forest Research 15(2): 444-452. Khalil, M.A.K. 1985b. Heritability of juvenile characters of white spruce (Picea glauca (Moench.) Voss.) in central Newfoundland, Canada. Theoretical and Applied Genetics 69(3): 247-251. Khasa, P.D., and Dancik, B.P. 1996. Rapid identification of white-Engelmann spruce species by RAPD markers. Theoretical and Applied Genetics 92(1): 46-52. King, J.N., Yeh, F.C., Heaman, J.C., and Dancik, B.P. 1988. Selection of wood density and diameter in controlled crosses of coastal Douglas-fir. Silvae Genetica 37(3-4): 152-157. Kiss, G., and Yeh, F.C. 1988. Heritability estimates for height for young interior spruce in British Columbia. Canadian Journal of Forest Research 18(2): 158-162.  69  Kiss, G.K., and Yanchuk, A.D. 1991. Preliminary evaluation of genetic variation of weevil resistance in interior spruce in British Columbia. Canadian Journal of Forest Research 21(2): 230-234. Kliger, I.R., Perstorper, M., Johansson, G., and Pellicane, P.J. 1995. Quality of timber products from Norway spruce, Part 3. Influence of spatial position and growth characteristics on bending stiffness and strength. Wood Science and Technology 29(6): 397-410. Kretschmann, D.E. 1998. Properties and use of wood, composites, and fiber products Properties of juvenile wood. Available from http://www.fpl.fs.fed.us/documnts/techline /properties-of-juvenile-wood.pdf [accessed 8.2. 2012]. Kretschmann, D.E., and Bendtsen, B.A. 1992. Ultimate tensile stress and modulus of elasticity of fast-grown plantation loblolly pine lumber. Wood and Fiber Science 24(2): 189-203. Kubus, M. 2009. The Evaluation of Using Resistograph when Specifying the Health Condition of a Monumental Tree. Notulae Botanicae Horti Agrobotanici Cluj-Napoca 37(1): 157-164. Kumar, S., Jayawickrama, K.J.S., Lee, J., and Lausberg, M. 2002. Direct and indirect measures of stiffness and strength show high heritability in a wind-pollinated radiata pine progeny test in New Zealand. Silvae Genetica 51(5-6): 256-261. Larsen, H.J. 2001. Properties affecting reliability design of timber structures. In Reliability of timber structures, Coimbra, Portugal. Lenz, P., Cloutier, A., MacKay, J., and Beaulieu, J. 2010. Genetic control of wood properties in Picea glauca - an analysis of trends with cambial age. Canadian Journal of Forest Research 40(4): 703-715. Li, L., and Wu, H.X. 2005. Efficiency of early selection for rotation-aged growth and wood density traits in Pinus radiata. Canadian Journal of Forest Research 35(8): 2019-2029. Lin, C.J., Wang, S.Y., Lin, F.C., and Chiu, C.M. 2003. Effect of moisture content on the drill resistance value in Taiwania plantation wood. Wood and Fiber Science 35(2): 234-238.  70  Lindstrom, H. 1996. Basic density in Norway spruce. III. Development from pith outwards. Wood and Fiber Science 28(4): 391-405. Lindström, H., Harris, P., Sorensson, C.T., and Evans, R. 2004. Stiffness and wood variation of 3-year old Pinus radiata clones. Wood Science and Technology 38(8): 579-597. Little, E.L. 1971. Atlas of United States trees - conifers and important hardwoods. U.S. Department of Agriculture, Washington, http://esp.cr.usgs.gov/data/atlas/little. Lord, T.M., and Mackintosh, E.E. 1982. Soils of the Quesnel area, British Columbia. BC soil survey. Technical report no. 31. Supply and Services Canada, Ottawa, ON. Louzada, J. 2003. Genetic correlations between wood density components in Pinus pinaster Ait. Annals of Forest Science 60(3): 285-294. Louzada, J., and Fonseca, F.M.A. 2002. The heritability of wood density components in Pinus pinaster Ait. and the implications for tree breeding. Annals of Forest Science 59(8): 867-873. Macedo, A., Vaz, C.M.P., Pereira, J.C.D., Naime, J.M., Cruvinel, P.E., and Crestana, S. 2002. Wood density determination by X- and gamma-ray tomography. Holzforschung 56(5): 535-540. Mamdy, C., Rozenberg, P., Franc, A., Launay, J., Schermann, N., and Bastien, J.C. 1999. Genetic control of stiffness of standing Douglas fir; from the standing stem to the standardised wood sample, relationships between modulus of elasticity and wood density parameters. Part I. Annals of Forest Science 56(2): 133-143. Mannes, D., Lehmann, E., Cherubini, P., and Niemz, P. 2007. Neutron imaging versus standard X-ray densitometry as method to measure tree-ring wood density. Trees Structure and Function 21(6): 605-612. Mantel, N. 1967. Detection of disease clustering and generalized regression approach. Cancer Research 27(2P1): 209-220. Matheson, A.C., Dickson, R.L., Spencer, D.J., Joe, B., and Ilic, J. 2002. Acoustic segregation of Pinus radiata logs according to stiffness. Annals of Forest Science 59(5-6): 471477.  71  Matziris, D.I., and Zobel, B.J. 1973. Inheritance and correlations of juvenile characteristics in loblolly pine (Pinus taeda L.). Silvae Genetica 22(1-2): 38–45. McKimmy, M.D., and Campbell, R.K. 1982. Genetic variation in the wood density and ring width trend in coastal Douglas-fir. Silvae Genetica 31(2-3): 43-51. Merrill, R.E., and Mohn, C.A. 1985. Heritability and genetic correlations for stem diameter and branch characteristics in white spruce. Canadian Journal of Forest Research 15(3): 494-497. Millman, M. 1976. Metric volume and V-bar tables derived from the BCFS whole stem cubic metre volume equations., Vancouver. Mitchell, H.L. 1961. A concept of intrinsic wood quality and nondestructive methods for determining quality in standing timber. Report No. 2233. Forest Products Laboratory, Madison, Wisconsin. Mora, C.R., Schimleck, L.R., Isik, F., Mahon, J.M., Jr., Clark, A., III, and Daniels, R.F. 2009. Relationships between acoustic variables and different measures of stiffness in standing Pinus taeda trees. Canadian Journal of Forest Research 39(8): 1421-1429. Musil, I. 2003. Lesnická dendrologie 1 - jehličnaté dřeviny, Prague, Czech Republic. Neale, D.B., and Kremer, A. 2011. Forest tree genomics: growing resources and applications. Nature Reviews Genetics 12(2): 111-122. Nicholls, J.W.P., Morris, J.D., and Pederick, L.A. 1980. Heritability estimates of density characteristics in juvenile Pinus radiata wood. Silvae Genetica 29(2): 54-61. Nicolotti, G., and Miglietta, P. 1998. Using high-technology instruments to assess defects in trees. Journal of Arboriculture 24(6): 297-302. Nyland, R.D. 1996. Silviculture: concepts and applications. McGraw-Hill Companies, New York. Ostrander, M.D. 1974. Eastern spruce - an American wood. U.S. Department of Agriculture, Washington, D.C.  72  Panshin, A.J., and de Zeeuw, C. 1980. Textbook of wood technology: structure, identification, properties, and uses of the commercial woods of the United States and Canada. 4th ed. McGraw-Hill, New York. Park, H.M. 2008. Univariate analysis and normality test using SAS, Stata and SPSS. The University Information Technology Services (UITS) Center for Statistical and Mathematical Computing, Indiana University. Park, Y.S., Weng, Y.H., and Mansfield, S.D. 2012. Genetic effects on wood quality traits of plantation-grown white spruce (Picea glauca) and their relationships with growth. Tree Genetics & Genomes 8(2): 303-311. Pellerin, R.F., and Ross, R.J. 2002. Nondestructive evaluation of wood. Forest product society, USA. Perem, E., McBride, C.F., and Keith, C.T. 1981. Commercial woods. In Canadian woods: their properties and uses. Edited by E.J. Mullins, and T.S. McKnight. University of Toronto Press, Toronto, Canada. Perstorper, M., Pellicane, P.J., Kliger, I.R., and Johansson, G. 1995a. Quality of timber products from Norway spruce, Part 1. Optimization, key variables and experimental study. Wood Science and Technology 29(3): 157-170. Perstorper, M., Pellicane, P.J., Kliger, I.R., and Johansson, G. 1995b. Quality of timber products from Norway spruce, Part 2. Influence of spatial position and growth characteristics on warp. Wood Science and Technology 29(5): 339-352. Petty, J.A., Macmillan, D.C., and Steward, C.M. 1990. Variation of density and growth ring width in stems of sitka and Norway spruce. Forestry 63(1): 39-49. Porter, A.W. 1981. Strength and physical properties of wood. In Canadian woods: their properties and uses. Edited by E.J. Mullins, and T.S. McKnight. University of Toronto Press, Toronto, Canada. pp. 71-96. Pot, D., Chantre, G., Rozenberg, P., Rodrigues, J.C., Jones, G.L., Pereira, H., Hannrup, B., Cahalan, C., and Plomion, C. 2002. Genetic control of pulp and timber properties in maritime pine (Pinus pinaster Ait.). Annals of Forest Science 59(5-6): 563-575.  73  Ratcliffe, B. 2012. Genetics of wood quality attributes in western larch (Unpublished undergraduate thesis). Department of Forest Sciences, University of British Columbia, Vancouver, B.C. Raymond, C.A., Joe, B., Anderson, D.W., and Watt, D.J. 2008. Effect of thinning on relationships between three measures of wood stiffness in Pinus radiata: standing trees vs. logs vs. short clear specimens. Canadian Journal of Forest Research 38(11): 28702879. Rinn, F. 1996. Resistographic visualization of tree-ring density variations. University of Arizona, Tuscon. Rinn, F., Schweingruber, F.H., and Schar, E. 1996. Resistograph and X-ray density charts of wood comparative evaluation of drill resistance profiles and X-ray density charts of different wood species. Holzforschung 50(4): 303-311. Roche, L. 1969a. A genecological study of genus  Picea in British Columbia. New  Phytologist 68(2): 505. Roche, L. 1969b. A Genecological Study of Genus Picea in British Columbia. New Phytologist 68(2): 505. Ross, R.J., and Wang, X. 2005. A review of the use of acoustic speed to assess standing timber quality. Available from http://www.fibre-gen.com/st300.html [accessed 20.6. 2012]. Rozenberg, P., and Cahalan, C. 1997. Spruce and wood quality: Genetic aspects (A review). Silvae Genetica 46(5): 270-279. Rozenberg, P., Franc, A., Mamdy, C., Launay, J., Schermann, N., and Bastien, J.C. 1999. Genetic control of stiffness of standing Douglas fir; from the standing stem to the standardised wood sample, relationships between modulus of elasticity and wood density parameters. Part II. Annals of Forest Science 56(2): 145-154. Rozenberg, P., and VandeSype, H. 1996. Genetic variation of the pilodyn-girth relationship in Norway spruce (Picea abies L. [Karst.]). Annales Des Sciences Forestieres 53(6): 1153-1166.  74  R Development Core Team. 2012. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available from http://www.rproject.org/. Sabin, T.E., and Stafford, S.G. 1990. Assessing the need for transformation of response variables. Forest Research Lab, Oregon State University. Sahlberg, U. 1995. Influence of knot fibers on Tmp properties. Tappi Journal 78(5): 162-168. Saranpää, P. 2003. Wood density and growth. In Wood quality and its biological basis. Edited by John R. Barnett, and George Jeronimidis. Blackwell Publishing Ltd. p. 87. Savidge, R.A. 2003. Tree growth and wood quality. In Wood quality and its biological basis. Edited by John R. Barnett, and George Jeronimidis. Blackwell Publishing, Oxford, UK. pp. 1-29. Schimleck, L.R., Evans, R., and Ilic, J. 2001. Estimation of Eucalyptus delegatensis wood properties by near infrared spectroscopy. Canadian Journal of Forest Research 31(10): 1671-1675. Shull, P.J. 2002. Nondestructive evaluation - theory, techniques and application. Marcel Dekker, Inc., USA. Smith, D.M. 1954. Maximum moisture content method for determining specific gravity. Report 2014. Forest Product Laboratory, Forest Service, U.S. Department of Agriculture. Sprague, J.R., Talbert, J.T., Jett, J.B., and Bryant, R.L. 1983. Utility of the Pilodyn in selection for mature wood specific gravity in loblolly pine. Forest Science 29(4): 696701. Steffenrem, A., Kvaalen, H., Hoibo, O.A., Edvardsen, O.M., and Skroppa, T. 2009. Genetic variation of wood quality traits and relationships with growth in Picea abies. Scandinavian Journal of Forest Research 24(1): 15-27. Sutton, R.F. 1969. Silvics of white spruce (Picea glauca (Moench) Voss). Canadian Forestry Service, Ottawa, ON.  75  Talbert, J.T., Jett, J.B., and Bryant, R.L. 1983. Inheritance of wood specific gravity in an unimproved loblolly pine population: 20 years of results. Silvae Genetica 32(1-2): 3337. Taylor, F.W. 1981. Rapid determination of southern pine specific gravity with a Pilodyn tester. Forest Science 27(1): 59-61. Taylor, F.W., Wang, E.I.C., Yanchuk, A., and Micko, M.M. 1982. Specific gravity and tracheid length variation of white spruce in Alberta. Canadian Journal of Forest Research 12(3): 561-566. Toulmin, M.J., and Raymond, C.A. 2007. Developing a sampling strategy for measuring acoustic velocity in standing Pinus radiata using the treetap time of flight tool. New Zealand Journal of Forestry Science 37(1): 96-111. Tree Improvement Branch. 2012a. Seed Planning Units (SPU) & Species Plans. Available from http://www.for.gov.bc.ca/hti/speciesplan/index.htm [accessed 3.6. 2012]. Tree Improvement Branch. 2012b. Seed Planning Zone Maps. Available from http://www.for.gov.bc.ca/hti/seedplanning/GRM-Map-Gallery.htm  [accessed  1.4.  2012]. Ukrainetz, N.K., Kang, K.-Y., Aitken, S.N., Stoehr, M., and Mansfield, S.D. 2008. Heritability and phenotypic and genetic correlations of coastal Douglas-fir (Pseudotsuga menziesii) wood quality traits. Canadian Journal of Forest Research 38(6): 1536-1546. Ukrainetz, N.K., and O'Neill, G.A. 2010. An analysis of sensitivities contributing measurement error to Resistograph values. Canadian Journal of Forest Research 40(4): 806-811. van Mantgem, P.J., and Stephenson, N.L. 2004. Does coring contribute to tree mortality? Canadian Journal of Forest Research-Revue Canadienne De Recherche Forestiere 34(11): 2394-2398. Vargas-Hernandez, J., and Adams, W.T. 1991. Genetic variation of wood density components in young coastal Douglas-fir: Implications for tree breeding. Canadian Journal of Forest Research 21(12): 1801-1807. 76  Walker, J.C.F. 2006. Basic wood chemistry and cell wall ultrastructure. In Primary wood processing: principles and practice. Edited by J.C.F. Walker. Springer, Dordrecht, The Netherlands. pp. 23-68. Wang, E.I.C., and Micko, M.M. 1984. Wood quality of white spruce from north central Alberta. Canadian Journal of Forest Research 14(2): 181-185. Wang, E.I.C., Mueller, T., and Micko, M.M. 1985. Drainage effect on growth and wood quality of some bog grown trees in Alberta. Forestry Chronicle 61(6): 489-493. Wang, S.Y., and Chuang, S.T. 2000. Experimental data correction of the dynamic elastic moduli, velocity and density of solid wood as a function of moisture content above the fiber saturation point. Holzforschung 54(3): 309-314. Wang, T., Aitken, S.N., Rozenberg, P., and Carlson, M.R. 1999. Selection for height growth and Pilodyn pin penetration in lodgepole pine: effects on growth traits, wood properties, and their relationships. Canadian Journal of Forest Research 29(4): 434-445. Wang, X., Ross, R.J., Erickson, J.R., and Ligon, J.B. 2000a. Nondestructive evaluation of trees. Experimental Techniques 24(6): 27-29. Wang, X., Ross, R.J., Mattson, J.A., Erickson, J.R., Forsman, J.W., Geske, E.A., and Wehr, M.A. 2001a. Several nondestructive evaluation techniques for assessing stiffness and MOE of small-diameter logs. Research paper FPL-RP-600. U.S. Department of Agriculture, Forest Service, Forest Products Laboratory, Madison, WI. p. 12. Wang, X., Ross, R.J., McClellan, M., Barbour, R.J., Erickson, J.R., Forsman, J.W., and McGinnis, G.D. 2000b. Strength and stiffness assessment of standing trees using a nondestructive stress wave technique. Research paper FPL-RP-585. U.S. Department of Agriculture, Forest Service, Forest Products Laboratory, Madison, WI. p. 9. Wang, X.P., Carter, P., Ross, R.J., and Brashaw, B.K. 2007. Acoustic assessment of wood quality of raw forest materials - A path to increased profitability. Forest Products Journal 57(5): 6-14. Wang, X.P., Ross, R.J., McClellan, M., Barbour, R.J., Erickson, J.R., Forsman, J.W., and McGinnis, G.D. 2001b. Nondestructive evaluation of standing trees with a stress wave method. Wood and Fiber Science 33(4): 522-533. 77  Weber, K., and Mattheck, C. 2006. The effects of excessive drilling diagnosis on decay propagation in trees. Trees-Structure and Function 20(2): 224-228. Wessels, C.B., Malan, F.S., and Rypstra, T. 2011. A review of measurement methods used on standing trees for the prediction of some mechanical properties of timber. European Journal of Forest Research 130(6): 881-893. Williamson, G.B., and Wiemann, M.C. 2010. Measuring wood specific gravity...correctly. American Journal of Botany 97(3): 519-524. Wimmer, R., Downes, G.M., Evans, R., Rasmussen, G., and French, J. 2002. Direct effects of wood characteristics on pulp and handsheet properties of Eucalyptus globulus. Holzforschung 56(3): 244-252. Winters, R.H. 1951. Canadian woods, their properties and uses. Forestry Branch, Forest Products Laboratories Division, Ottawa. Xie, C.Y. 2003. Genotype by environment interaction and its implications for genetic improvement of interior spruce in British Columbia. Canadian Journal of Forest Research 33(9): 1635-1643. Xie, C.Y., and Yanchuk, A.D. 2002. Genetic parameters of height and diameter of interior spruce in British Columbia. Forest Genetics 9(1): 1-10. Xu, P., and Walker, J.C.F. 2004. Stiffness gradients in radiata pine trees. Wood Science and Technology 38(1): 1-9. Yanchuk, A.D., and Kiss, G.K. 1993. Genetic variation in growth and wood specific gravity and its utility in the improvement of interior spruce in British Columbia. Silvae Genetica 42(2-3): 141-148. Ying, C.C., and Morgenstern, E.K. 1979. Correlations of height growth and heritabilities at different ages in white spruce. Silvae Genetica 28(5-6): 181-185. Yu, Q.B., Yang, D.Q., Zhang, S.Y., Beaulieu, J., and Duchesne, I. 2003. Genetic variation in decay resistance and its correlation to wood density and growth in white spruce. Canadian Journal of Forest Research 33(11): 2177-2183.  78  Zhang, H., Wang, X., and Ross, R. 2009. Stress wave propagation on standing trees. Part 1. Time-of-flight measurement and 2D stress wave contour maps. In Proceedings of the 16th nondestructive evaluation of wood symposium. Edited by Houjian Zhang, and Xiping Wang, Beijing Forestry University, Beijing, China. pp. 53-58. Zhang, S.Y. 1995. Effect of growth rate on wood specific gravity and selected mechanical properties in individual species from distinct wood categories. Wood Science and Technology 29(6): 451-465. Zhang, S.Y. 1997. Wood specific gravity mechanical property relationship at species level. Wood Science and Technology 31(3): 181-191. Zhang, S.Y. 1998. Effect of age on the variation, correlations and inheritance of selected wood characteristics in black spruce (Picea mariana). Wood Science and Technology 32(3): 197-204. Zhang, S.Y. 2003. Wood quality attributes and their impacts on wood utilization. In XII World Forestry Congress, Quebec City, Canada. Zhang, S.Y., and Gingras, J.F. 1998. Timber management toward wood quality and endproduct value: workshop highlights and related information. FERIC/Forintek joint report. Forest Engineering Research Institute of Canada, Forintek Canada Corp, PointeClaire, uébec. Zhang, S.Y., and Jiang, Z.H. 1998. Variability of selected wood characteristics in 40 half-sib families of black spruce (Picea mariana). Wood Science and Technology 32(1): 71-82. Zhang, S.Y., and Morgenstern, E.K. 1995. Genetic variation and inheritance of wood density in black spruce (Picea mariana) and its relationship with growth: Implications for tree breeding. Wood Science and Technology 30(1): 63-75. Zhang, S.Y., Simpson, D., and Morgenstern, E.K. 1996. Variation in the relationship of wood density with growth in 40 black spruce (Picea mariana) families grown in New Brunswick. Wood and Fiber Science 28(1): 91-99. Zhang, S.Y., Yu, Q.B., and Beaulieu, J. 2004. Genetic variation in veneer quality and its correlation to growth in white spruce. Canadian Journal of Forest Research 34(6): 1311-1318. 79  Zhou, H., and Smith, I. 1991. Factors influencing bending properties of white spruce lumber. Wood and Fiber Science 23(4): 483-500. Zhu, J.Y., Vahey, D.W., Scott, C.T., and Myers, G.C. 2008. Effect of tree-growth rate on papermaking fibre properties. Appita Journal 61(2): 141-147 and 155. Zink-Sharp, A. 2004. Assessment of wood quality - techniques & definitions. Wood Science and Forest Products, Virginia Tech. Zobel, B.J., and Jett, J.B. 1995. Genetics of wood production. Springer-Verlag, Berlin, Germany. Zobel, B.J., and Sprague, J.R. 1998. Juvenile wood in forest trees. Springer-Verlag, Berlin, Germany. Zobel, B.J., and Talbert, J.T. 1984. Applied forest tree improvement. John Wiley & Sons, New York. Zobel, B.J., and Van Buijtenen, J.P. 1989. Wood variation: its causes and control. SpringerVerlag, Berlin, Germany.  80  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0073230/manifest

Comment

Related Items