UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Carbon isotope discrimination in Tsuga heterophylla and its relationship to mineral nutrition and growth Walia, Ankit 2004

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

Item Metadata

Download

Media
831-ubc_2005-0128.pdf [ 4.03MB ]
Metadata
JSON: 831-1.0075073.json
JSON-LD: 831-1.0075073-ld.json
RDF/XML (Pretty): 831-1.0075073-rdf.xml
RDF/JSON: 831-1.0075073-rdf.json
Turtle: 831-1.0075073-turtle.txt
N-Triples: 831-1.0075073-rdf-ntriples.txt
Original Record: 831-1.0075073-source.json
Full Text
831-1.0075073-fulltext.txt
Citation
831-1.0075073.ris

Full Text

Carbon Isotope discrimination in Tsuga heterophylla and its relationship to mineral nutrition and growth by  AN KIT WALIA  B . S c , Y S Parmar University of Horticulture and Forestry, INDIA  A T H E S I S S U B M I T T E D IN PARTIAL F U L F I L M E N T O F THE REQUIREMENTS FOR THE D E G R E E O F  MASTER OF SCIENCE in THE FACULTY O F GRADUATE STUDIES (Forestry)  T H E UNIVERSITY O F BRITISH C O L U M B I A December 2004 ©AnkitWalia  ABSTRACT  Western hemlock (Tsuga heterophylla) is one of the most important tree species in coastal British Columbia. Forest fertilization is a method used by foresters to enhance the growth of forest trees, but results are inconsistent for western hemlock. The objective of my research was to explore the use of carbon isotope analysis as a physiological tool to diagnose the nutritional status and potential response of western hemlock to fertilization. Normally cellulose is isolated from wood samples to be analyzed for carbon isotope analysis (expressed as 8 C values), but this is a tedious and likely unnecessary process. Reaction wood (high lignin content) and adjacent normal wood in two western hemlock saplings was analysed to evaluate the possible effects of wood composition on 8 C. The 8 C values of the lignin and cellulose fractions differed by 3.43 %c ± 0.26 (mean ± SD; n = 40). 8 C values of lignin and cellulose from different disk positions were more variable in one sapling than the other. The isotopic mass balance of whole wood was conserved and therefore did not vary with lignin content indicating that use of whole wood, rather than cellulose, is suitable for isotopic analysis. 13  13  13  Eight pure western hemlock stands selected and experimentally fertilized in an earlier study were used as source for foliage and stemwood samples. Fertilization treatments applied to these stands in mid to late May of 1995 were as follow: (1) control, (2) N (225 kg/ha), (3) N (225 kg/ha) + P (100 kg/ha), (4) N (225 kg/ha) + P (500 kg/ha), (5) N (225 kg/ha) + P (100 kg/ha) + blend (230 kg/ha), and (6) N (225 kg/ha) + P (500 kg/ha) + blend (230 kg/ha). The blend included additions of S, K, Mg, Zn and Cu. 8 C values within foliage and stemwood were analysed after fertilization. At the end of first growing season after fertilization, the effect of treatments on foliar 8 C was almost significant (P = 0.0539) and there was an interaction between sites and treatments. At the end of second growing season, no interaction between site and treatment was evident and the effect of treatments on foliar 8 C was not significant. 13  13  Foliar SO4-S (Sulfate) levels at the end of the first and second growing seasons following fertilization were reduced by either N or N + P fertilization treatments. The change in 8 C values of tree rings from before to after nutrient additions was significantly affected by treatment and site and there was an interaction between sites and treatments. N applied alone had no significant effect on change in wood 8 C. The greatest change (0.33 %o) was in the NP100 (treatment 3 above) level of treatment. 13  13  Relative change in six-year basal area increment (BAI) was significantly affected by treatments and sites and there was an interaction between sites and treatments. N applied alone had no significant effect on relative change in basal area increment whereas NP100B (treatment 5 above) had the greatest effect. The relative change in basal area increment was also significantly affected by sites.  ii  There was indeed a physiological effect of nutrient additions on trees which was evident as an increase in growth response and changes in foliar and wood 5 C values. However, the nutrient effect was small relative to variation associated with intrinsic site characteristics, weather parameters and inherent genetic variation of individual trees, in terms of effects on 5 C. Thus, use of carbon isotope analysis alone as a physiological tool to diagnose the nutritional status and potential response of trees to nutrients cannot be recommended. 13  13  iii  T A B L E OF CONTENTS  Abstract  •  n  List of Tables  v i  List of Figures  v n  Acknowledgements  viii  CHAPTER 1 INTRODUCTION  Western Hemlock Nutrition Review of Carbon Isotope Discrimination Whole Wood for Analysis Thesis Questions  2 4 7 8  CHAPTER 2 FRACTIONATION OF CARBON ISOTOPES IN WOOD OF WESTERN H E M L O C K AS A FUNCTION OF LIGNIN TO C E L L U L O S E RATIO  INTRODUCTION  10  MATERIALS AND METHODS  12  Western .Hemlock saplings Analysis of Wood Fractions  12 12  Statistical Analysis  14  RESULTS  15  Isotopic Composition of Wood Fractions Variation in 8 C between Fractions and Disks 13  15 16  DISCUSSION  23  CONCLUSION  25 iv  CHAPTER 3 EFFECT OF FERTILIZATION ON CARBON ISOTOPE DISCRIMINATION AND GROWTH OF EIGHT IMMATURE WESTERN H E M L O C K STANDS  INTRODUCTION  27  Objectives  30  METHODOLOGY  32  Stand Descriptions Treatments Plot Establishment Foliage Collection Determination of Isotopic Composition of Current-year Needles Foliar Sulfur Analysis Increment Core Collection Basal Area Increment Determination of Isotopic Composition within Individual Rings Calculation of Summer Dryness Index Statistical Analysis RESULTS  32 32 33 33 34 34 35 35 36 36 37 38  Foliar 8 C Foliage Sulfur Data Growth Response Stem Wood 5 C Pearson Correlation Analysis ,3  13  38 42 42 45 52  DISCUSSION  56  CONCLUSION  67  CHAPTER 4 THESIS CONCLUSION  Future Research  73  REFERENCES APPENDIX 1 APPENDIX 2 APPENDIX 3  75 85 86 86  v  LIST OF TABLES  Table 2.1  Comparison of 8 C values (%c) of saplings A and B. Data are means of four quarters representing disk positions with standard errors in parenthesis. 13  Table 2.2 Analysis of variance for effects of height (disk position) and  sections within quarters on 8 C of saplings A and B.  21  22  13  Table 3.1 Latitude, longitude and elevation (m) of eight study sites.  39  Table 3.2 Two-way analysis of variance (ANOVA) for the effect  40  of treatments and site on 5 C of foliage collected at the end of first growing season (1995). 13  Table 3.3 Two-way analysis of variance (ANOVA) for the effect of the  40  treatments and site on 8 C of foliage collected at the end of the second growing season (1996). 13  Table 3.4 Two-way analysis of variance (ANOVA) for the effect of  43  Table 3.5 Two-way analysis of variance (ANOVA) for the effect of  43  Table 3.6 Two-way analysis of variance (ANOVA) for the effect of  46  Table 3.7 Six-year basal area response relative to control by treatment  48  treatments and site on SO4-S (ppm) levels in foliage collected at the end of first growing season (1995).  treatments and site on SO4-S (ppm) levels in foliage collected at the end of second growing season (1996). treatments and sites on six year basal area increment.  and installation.  Table 3.8 Two-way analysis of variance (ANOVA) for the effect of  50  treatments and site on change in 8 C of tree rings. 13  Table 3.9 Pearson correlation coefficient and associated P values (below)  for different foliage nutrients for year 1995, foliage 8 C (1995), change in wood 8 C and basal area increment.  53  13  13  Table 3.10 Pearson correlation coefficient and associated P values (below)  for different foliage nutrients for year 1996, foliage 8 C (1996), change in wood 8 C and basal area increment. 13  13  vi  54  LIST OF FIGURES  Fig 2.1 Boles of two western hemlock saplings. Reaction wood forms along the lower side of the curved stem. Fig 2.2 Geometric mean regressions of 8 C as a function of lignin content for sapling A. Fig 2.3 Geometric mean regressions of 8 C as a function of lignin content for sapling B. 13  Fig 2.4 Geometric mean regression of calculated and observed 8 C values of whole wood. 13  Fig. 3.1 Study site located within the Nimpkish Valley on northern Vancouver Island. Fig 3.2 Carbon Isotope composition (8 C) of current-year foliage collected at the end of the first (1995) and second growing season (1996) following fertilization. 13  Fig 3.3  Sulfate SO4-S (ppm) levels in the foliage collected at the end of first (1995) and second growing season (1996) following fertilization.  Fig 3.4 Six year growth response (BAI) due to fertilizer treatments. Fig 3.5 General trend in 8 C of tree-rings. l3  Fig 3.6 Change in 5 C of tree rings upon fertilization. 13  Fig 3.7 Sum of three years basal area increment after fertilization as a function of the mean of three years wood 8 C prior to fertilization. 13  vii  ACKNOWLEDGEMENTS  I am grateful to my supervisor, Dr. Robert D. Guy, for his patience, insightful guidance, valuable comments and above all for providing me the opportunity to pursue graduate studies. I also wish to acknowledge the contributions of my committee members, Drs. Barry White and Steve Mitchell, who provided valuable comments regarding the project and improving the manuscript. I am thankful to Dr. Barry White for providing foliage samples, assistance with wood core samples and growth data analysis. I sincerely thank Dr. Mitchell for helpful discussions and providing western hemlock saplings. I am thankful to Limin Liao for his assistance in preparation of samples for 8 C and Ms. Jodie Krakowski for her help with statistical analysis. Mr. Tongli Wang from Centre for Forest Gene Conservation, UBC, assisted in modeling the climate parameters. Finally, I wish to acknowledge the contributions of my parents who provided invaluable guidance, moral and emotional support. 13  viii  CHAPTER 1  INTRODUCTION  Western hemlock {Tsuga heterophylla) is the climatic climax species in the Coastal Western Hemlock biogeoclimatic zone (CWH) that covers most of the lower-mid elevations west of the Coast Mountains in British Columbia (BC). Western hemlock is one of the most important tree species in this biogeoclimatic zone and is valued for its timber quality. Thus it is an important species for the forest industry.  Within BC, western hemlock is one of the primary species facing serious reductions in wood supply in the near future - due to a combination of unbalanced age-class distribution and a reduction in forested area that can be managed for timber production. There are only a very few options available to cope with the problem of reductions in wood supply. Fertilization is one of the pragmatic approaches that have been used by foresters to increase the growth of forest tree species. The determination of present nutritional status and, following on this, the prediction of potential response of a given stand to a fertilization treatment, are pivotal in an implementation of a successful fertilization program.  Unfortunately, operational fertilization of western hemlock is not currently an option as foresters lack necessary tools to accurately predict the nutritional status of a given stand and determine whether that stand will respond to fertilization  1  treatment (White, 2000; Carter et al., 2001; Brown, 2003). Although foliar analysis has some utility with respect to determining the nutritional status of a given stand, it is not a reliable technique with respect to western hemlock (Carter, 2001). The negative economic consequences of an incorrect or an ineffective prescription are substantial.  WESTERN HEMLOCK NUTRITION  The Regional Forest Nutrition Group located at the University of Washington initiated studies related to western hemlock fertilization in 1969. Similar research, exploring the response of Douglas-fir and western hemlock stands to thinning and fertilization was also carried out in B C (Darling and Omule, 1989). Several studies reported that the growth of western hemlock could be greatly enhanced by nitrogen (N) fertilization, although the response was erratic and many stands did not respond to the fertilization treatments (Radwan and DeBell, 1980; Heilman and Ekuan, 1980; Anderson et al., 1982; Shumway and Olson, 1992). Furthermore, the foliar nutrient concentrations were not related to the response.  In some studies the application of N was enhanced by the concomitant addition of phosphorus (P), but again the response was erratic (Gill and Lavender, 1983; Radwan et al., 1991). It has been speculated that this erratic response is a result of deficiencies in other mineral nutrients or a potential effect of urea in reducing the number of mycorrhizal roots - which may affect P uptake. The form of  2  nitrogen supplied did not appear to affect western hemlock response to fertilization (Radwan and DeBell, 1989). In summary, the response of western hemlock to fertilization treatment appears to be unrelated to foliar nutrient concentration or site quality indicators (Radwan and Shumway, 1983; Carter et al., 2001).  A series of fertilization experiments in mixed species stands of western hemlock and Douglas-fir and in pure western hemlock stands was established on Vancouver Island in 1995 (White, 1999; 2000). The fertilization treatments were shown to have profound effects on stable carbon isotope ratios (expressed as 5 C values) in an 18-year-old stand of western hemlock and Douglas-fir during 13  each of three years following fertilization (White, 1999). Carbon isotope ratios were also affected in a 5-year-old western hemlock plantation growing on a nutrient stressed site on northern Vancouver Island, in response to fertilization treatments, leading to speculation that a potentially useful relationship exists between nutritional status and isotope discrimination (White, 2000). These studies explored the changes in carbon isotope discrimination values after fertilization with nitrogen, phosphorus, and blend fertilizer. The blend included additions of S, K, Mg, Zn and Cu. Nitrogen fertilization resulted in a strong reduction in discrimination against the heavier C isotope in Douglas-fir and western hemlock foliage. Furthermore, addition of P and the blend fertilizer each resulted in a further reduction in discrimination beyond that achieved with only N. Based on the studies of White (1999, 2000), it appears that analysis of carbon  3  isotopes in plant tissue has potential as a physiological tool to diagnose the nutritional status and potential response of western hemlock to fertilizer treatment.  CARBON ISOTOPE DISCRIMINATION  Plants discriminate against C 0 2 over the lighter 13  1 2  C0  during photosynthesis.  2  The magnitude of this discrimination varies with photosynthetic type, environment and genotype. Discrimination against  1 3  C 0 occurs during both the diffusion and 2  carboxylation components of gas exchange, resulting in modified  1 3  C  concentrations (expressed as 8 C values) in tissue. The overall effects are 13  integrated over the period in which carbon is assimilated resulting in a relationship between 8 C and mean intercellular C 0 concentration (Farquhar et 13  2  al., 1982, 1989). Stable carbon isotope discrimination, determined from tissue 8 C relative to the 8 C of the atmosphere, is related to the ratio of the 13  13  intercellular C 0 concentration (Cj) over the ambient (c ) such that a higher c / c 2  a  a  leads to greater discrimination. Stomatal conductance, along with photosynthetic activity, regulates c by controlling the rate at which C 0 diffuses along a s  2  concentration gradient into leaf tissues where the C 0  2  is ultimately fixed in the  photosynthetic carbon reduction cycle (Hubick et al., 1986; Condon et al., 1987; Ehleringer, 1990).  Water-use efficiency (WUE) is the amount of carbon gained per unit water loss. At the leaf level, W U E is determined by the diffusion gradient for C 0 going to the 2  4  leaf (inversely proportional to c/c ) divided by the diffusion gradient for water a  vapour exiting the leaf. Relative water-use efficiency is therefore determined by the balance between stomatal conductance and photosynthetic capacity and their combined effects on c/c . Since both isotope discrimination and W U E are a  related to c-,, W U E is highly correlated with 5 C . The 8 C in plant organic tissue 13  13  can therefore provide useful long- or short-term indicator of water-use efficiency (Sun et al.,1996; Livingston et al.,1999) and has been widely used as a tool for the selection of genotypes with improved water use efficiency and productivity in the field of plant biology (Farquhar and Richards, 1984; Xu et al., 2000).  Numerous studies exploring isotopic variations within stemwood have also been reported. For example, carbon isotope discrimination in the foliage of radiata pine (Pinus radiata D. Don) is reflected in stemwood (Walcroft et al., 1997; Marshall et al., 2001; Porte and Loustau, 2001). Similarly, relationships between 8 C in 13  annual rings, moisture supply, and temperature were reported in a study of spatial variation and species composition within the boreal forest (Brooks et al., 1998). Several studies have demonstrated the potential of tree-ring stable carbon isotope ratios for high-resolution climatic reconstruction, using both hardwood and softwood species (Loader and Switsur, 1996; Robertson et al., 1997; Hemming et al.,1998; McCarrol and Pawellek, 2001). The stable carbon isotope ratios (8 C) of whole wood, cellulose and lignin from annual latewood increment 13  from modern and sub-fossil wood of oak tree-rings were used as palaeoenvironmental indicators (Loader et al., 2003).  5  Use of carbon isotopes as a diagnostic tool for determining the nutritional status and potential response to a fertilization treatment has not been pursued before within the field of tree nutrition. This may be due to the fact that carbon isotope signatures are significantly influenced by other variables that influence c / c  a  and/or 8 C, such as light, water stress, degree of canopy closure (resulting in 13  C 0 recycling within/beneath canopies) and weather (Livingston et al., 1999; 2  Porte and Loustau, 2001; Helle and Schleser, 2004). Furthermore, some studies have suggested that age - or size - related changes in hydraulic conductivity affect discrimination in some species but perhaps not others (Cernusak and Marshall 2001; McDowell et al., 2002). The effects of some of these variables can be circumvented while collecting foliage samples. For example, needles can be sampled from the upper canopy to circumvent the potential effects of light and recycled CO2. Since, C W H biogeoclimatic zone in B C usually receives up to 4400 mm of precipitation and has a mesothermal climate with summers that are typically cool and moist (Pojar et al.,1991), water stress may not a factor limiting productivity. Thus, in this region at least, changes in 8 C values as a result of 13  water stress or vapour pressure deficit (i.e. stomatal constraints) should be relatively small.  Another approach to explore the potential relationships between nutritional status and isotope discrimination is the use of wood 8 C rather than foliage 8 C. 13  13  A  potential advantage of using wood 8 C over foliage is that wood is expected to 13  6  integrate the isotopic signal from the entire productive canopy. In a recent study exploring carbon isotope discrimination in forest and pasture ecosystems of the Amazon basin (Brazil), it was suggested that the major portion of recently respired C 0  2  from below-canopy structures was metabolized carbohydrate fixed  by the sun leaves at the top of the forest canopy (Jean et al., 2002). If this is generally true for stand-grown trees, then stemwood should primarily reflect photosynthesis taking place in the sun leaves.  WHOLE WOOD FOR ANALYSIS  Dry wood is comprised of lignin, cellulose and extractives each having different 8 C values. Although holocellulose is typically extracted from wood for stable 13  isotope analysis, the results obtained are qualitatively indistinguishable from whole wood (Guy and Holowachuk, 2001). Several studies have shown a very good correlation between the isotopic composition of whole wood and extracted cellulose and have recommended the use of whole wood rather than isolated cellulose for isotopic studies (Walcroft et al., 1997; Korol et al., 1999; Guy and Holowachuk, 2001). In fact, mass balance considerations suggest that error might be introduced by sampling only particular biochemical fractions. For example, because lignin is depleted in C relative to the carbon source, any 1 3  increase in its synthesis must be balanced by an isotopic enrichment of the remaining fractions, particularly cellulose. The other assumptions are that both lignin and cellulose are synthesized from carbon compounds in phloem  7  originating from the same source leaves, and there is no further fractionation caused by unaccounted sources of carbon or during respiration (Lin and Ehleringer, 1997).  In response to a gravity-based stimulus, there is formation of reaction wood on the underside of stems or branches in conifers (compression wood) mediating load-bearing functions (Westing, 1965, 1968). Compression wood has high lignin and low cellulose content. This fact can be exploited to test the hypothesis that wood 5 C values are insensitive to wood composition, that is, 8 C should be the 13  13  same for reaction wood and normal wood, irrespective of their different lignin and cellulose contents.  THESIS QUESTIONS  1. Is there potential to use mean 5 C values in western hemlock needles 13  and stemwood, prior to fertilization, to infer the initial nutritional status of trees? For example, can mean 5 C values in western hemlock be used in 13  the same fashion as conventional foliar analysis to determine the nutritional status of a stand?  2. Can changes in discrimination of foliage and stemwood in the years following fertilization relative to that prior to fertilization be related to nutritional status and long-term growth response?  8  3. Is use of whole wood 8 C more appropriate than using cellulose 5 C in 13  13  carbon isotope analysis?  During the past three decades, researchers have been studying western hemlock nutrition sporadically in the Pacific Northwest region. This research has indicated that fertilization has potential to enhance the productivity of western hemlock stands. However, fertilization of western hemlock cannot be recommended unless the certainty of gaining a specific fertilizer response is improved upon. From the physiological point of view it is important to understand the effects of nutrients on carbon isotope discrimination. This will further our understanding of response of western hemlock to nutrient applications. The overall objective of my thesis is to explore the use of carbon isotope analysis as a physiological tool to diagnose the nutritional status and potential response of trees to nutrients.  i  9  CHAPTER 2 FRACTIONATION OF CARBON ISOTOPES IN WOOD OF WESTERN HEMLOCK AS A FUNCTION OF LIGNIN TO CELLULOSE RATIO  INTRODUCTION  Wood consists mostly of cellulose and lignin, with a smaller proportion made up of various extractives each having different 8 C values (Park and Epstein, 1961). 13  Wood is also enriched in C relative to leaves (Benner et al., 1987). Being a 1 3  nonmobile wood component, cellulose is preferred over whole wood for isotopic studies of wood. It has been suggested that a bias may be introduced if whole wood is used for isotopic analysis due to varying proportions of organic constituents in whole wood, each with distinct isotopic composition (Park and Epstein, 1961; Leavitt and Long, 1986).  Although cellulose is typically extracted from wood for stable isotope analysis, the results are qualitatively indistinguishable from whole wood (Guy and Holowachuk, 2001). Several studies have corroborated the use of whole wood rather than cellulose for isotopic studies and shown good correlation between isotopic composition of whole wood and extracted cellulose (Livingston and Spittlehouse, 1996; Walcroft et al., 1997; Korol et al., 1999; Warren and Adams, 2000; Guy and Holowachuk, 2001). A recent study compared stable carbon isotope ratios in the whole wood, cellulose and lignin of oak tree-rings and found that whole wood retains the strongest climate signal (Loader et al., 2003).  10  Because the relative proportions of wood components (i.e. lignin, cellulose and extractives) vary with genotype, position on the tree, wood density, etc., there has been concern that wood composition may affect 8 C values. The natural 13  abundance of C relative to total carbon in lignin, cellulose and extractive 1 3  fractions differs. Because wood composition varies, a general assumption in isotopic analysis of tree rings for physiological, ecological or dendrochronological purposes has been that the isotopic signature of the cellulose fraction most closely reflects the carbon fixed in photosynthesis. This assumption neglects mass balance considerations. Because all of these components must ultimately be synthesized from the same carbon source arriving in the phloem, it is more likely that error is actually introduced by sampling only particular biochemical fractions. For example, because lignin is depleted in C relative to the carbon 1 3  source, any increase in its synthesis must be balanced by an isotopic enrichment of the remaining fractions, such as cellulose.  In conifers, reaction wood (i.e., compression wood) has high lignin and low cellulose content. It has been suggested that this mediates load-bearing functions (Westing 1965, 1968). Reaction wood is therefore ideal for testing mass-balance considerations. I tested the hypothesis that 8 C values will be 13  same for reaction wood and normal wood irrespective of their different lignin and cellulose contents by comparing isotopic signatures of major wood components in reaction wood and adjacent normal wood, from the same tree rings in western  11  hemlock saplings. I measured the 8 C of the most recent 10 years of whole 13  wood, lignin and cellulose from reaction wood (high lignin content) and adjacent normal wood (low lignin content) in the boles of two western hemlock saplings.  MATERIALS AND METHODS  Western hemlock saplings  Two western hemlock saplings (A and B; - 1 0 cm dbh) with sharply curved stem butts were felled (UBC Malcolm Knapp Research Forest) and sectioned into quarters at five positions (Disks 1 to 5; Fig. 2.1) within and above the zone of reaction wood formation (total = 40 wood samples). Disk position 1 was at the bottom of the stem, whereas position 5 was at the top of the stem (i.e., above the zone of curvature). Disks were quartered into 90° sections at different positions of the bole representing four cardinal directions, centered on the pith. Sapwood rings covering the period 1993-2002 were collectively isolated from the disk quarters.  Analysis of wood fractions Extractives were quantitatively prepared from large milled samples (Green, 1963). In brief, the reservoir flask of a Soxhlet extractor was filled with a 2:1 toluene/ethanol mixture and ran overnight (ca. 16-18h). After cooling, samples were poured from the Soxhlet to dry for 1-2 h before treating with deionized hot  12  water for 6h in a 1000-mL flask, to remove low molecular weight polysaccharides including gums and starch. After 2:1 toluene/ethanol mixture and boiling water treatments, extractives were dried down and weighed.  Lignin extraction followed the method of Goering and Van Soest (1970). Milled hemlock wood samples (each sample -250 mg) were soaked in 7 2 % sulphuric acid (4.35 mL) for 16 hours, then the acid concentration was adjusted to 3% and the mixture boiled for 4 hours in a water bath. The lignin obtained was collected on a glass fibre filter (Type A/E, PALL Life Sciences, Michigan), thoroughly washed with deionized water (3 x 50 mL), dried at 105 °C and weighed.  Cellulose extraction followed the method of Green (1963). Wood samples (each sample -250 mg) were suspended in 5 mL of deionized water in a round-bottom flask (25 mL) with a glass stopper. The flask was then submerged in a water bath maintained at 90 °C. The reaction in the flask was initiated by adding 1.25 mL of sodium chlorite/acetic acid solution. At 30-min intervals, 1.25 mL of sodium chlorite/acetic acid solution was added to the reaction, for a total of 5 mL. At the end of 2 hr (total of four additions), the reaction was cooled and filtered through a glass fibre filter. The resulting cellulose was washed with deionized water, dried at 105 °C, and weighed.  Isotopic composition of carbon (5 C values) of whole wood, lignin, cellulose and 13  extractives were determined on a model 1106 Elemental Analyser (Carlo Erba,  13  Valencia, CA, USA) interfaced to Prism triple-collecting ratio mass spectrometer (VG Isotech, Middlewich, UK) in the Department of Earth & Ocean Sciences, University of British Columbia. The 5 C value of the sample is expressed as: 1 3  5 C 1 3  (%o) = [(Pisample  —  Rstandard)/Rstandard] X  1000  where R mpie and Rstandard are the ratios of C / C respectively in the original 1 3  1 2  sa  wood samples and the arbitrary standard, Vienna Pee Dee belemnite ( V - P D B ) .  Statistical analysis The results were analyzed statistically by two-way analysis of variance (ANOVA) by using SigmaStat statistical package (Version 2.0). Differences in mean values were regarded as significantly different at p<0.05. Lignin content was calculated as mg lignin/mg dry mass. Pearson correlation coefficients were determined between calculated and observed 5 C values of whole wood in saplings A and B 1 3  The regressions were calculated by geometric mean regression because error in measurement occurs in both the dependent and independent variables.  14  RESULTS  Isotopic Composition of wood fractions The total extractive content of wood samples removed from the two hemlock saplings was approximately 3% (alcohol and water-soluble fractions combined). The 5 C values of extractive-free wood samples were very close (-28.36) to 13  whole wood. Therefore extractives could be safely ignored in mass balance calculations.  With an increase in the lignin content, the 8 C values of cellulose and lignin 13  fractions both became less negative (Fig. 2.2 and 2.3). For sapling A, the correlation was significant for the separate lignin (r = 0.857, P < 0.001) and cellulose (r = 0.569, P = 0.009) fractions, whereas for whole wood there was no change with wood composition (r = 0.263, P = 0.262).  The same trends occurred in sapling B for lignin (r = 0.762, P < 0.001), cellulose (r = 0.582, P = 0.007), and whole wood (r = 0.318, P = 0.171). Despite of an increase in the lignin content, the overall 5 C of wood remains constant 13  supporting the notion that any change in one fraction must be isotopically balanced by the remaining fractions.  15  Variation in S C between fractions and disks 13  As expected, lignin was depleted in C relative to whole wood, whereas 1 3  cellulose was enriched. The 5 C values of the two fractions differed by 3.43 %o ± 13  0.26 (mean ± SD; n = 40). The correlation between calculated (i.e. reconstructed from the 8 C values of component fractions) and observed 8 C of whole wood 13  13  was significant in both saplings (r = 0.9173, P < 0.001) (Fig. 2.4).  For sapling A, 5 C values of lignin and cellulose from different disk positions 13  were more variable. Whole wood 5 C values from different disk positions were 13  quite stable within each sapling, as were 5 C values for the lignin and cellulose 13  fractions from sapling B (Table 2.1). The differences in the mean values of whole wood among the different levels of height (disk position) and sections (within quarters) were not significant (Table 2.2).  16  Fig 2.1 Boles of two western hemlock saplings. Reaction wood forms along the lower side of the curved stem. Ribbons show position where disks were removed for analysis of wood rings.  17  Sapling A -26  -32 -I 0.26  ,  0.28  r  0.30  ,  1  0.32  0.34  , 0.36  , 0.38  , 0.40  , 0.42  1 0.44  Lignin content  Fig 2.2 Geometric mean regressions of 8 C as a function of lignin content for sapling A. Lignin content was calculated as mg lignin/mg dry mass. Following are the regression equation for the lignin fraction (•) (y = 7.39x - 33.51), the cellulose fraction (•) (y = 4.33x - 28.93), and whole wood (A) (y = 1.65x - 29.19). The correlation was significant for the lignin fraction (•) (r = 0.857, P < 0.001) and the cellulose fraction (•) (r = 0.569, P = 0.009), whereas for whole wood (A) it was not significant (r = 0.263, P = 0.262). 13  18  SaplingB  -26  0.26  0.28  0.30  0.32  0.34  0.36  0.38  0.40  0.42  0.44  Lignin Content  Fig 2.3 Geometric mean regressions of 8 C as a function of lignin content for sapling B. Lignin content was calculated as mg lignin/mg dry mass. Following are the regression equations for the lignin fraction (•) (y = 7.27x - 33.73), the cellulose fraction (•) (y = 4.06x - 29.29)) and for whole wood (A) (y = 1.64x - 29.617).The correlation was significant for the lignin fraction (•) (r = 0.762, P < 0.001) and the cellulose fraction (•) (r = 0.582, P = 0.007), whereas for whole wood (A) it was not significant (r = 0.318, P = 0.171). 13  19  Fig 2.4 Geometric mean regression of calculated and observed 8 C values of whole wood. The regression equation is (y = 0.95x - 1.25). The correlation is significant (r = 0.9173, P < 0.001). Data are calculated and observed 8 C values of whole wood representing sapling A (•) and sapling B (•). The solid line indicates the expected 1:1 relationship between calculated and observed 8 C of whole wood. 13  13  20  Table 2.1 Comparison of 5 C values (%o) of saplings A and B. Disk position 1 was at the 13  bottom of the stem, whereas position 5 was from above the zone of curvature. Maximum curvature was at positions 2 and 3. Reaction wood was essentially absent from disks collected at position 5. Data are means of four quarters representing disk positions with standard errors in parenthesis. Lignin '  Cellulose  Whole Wood  Disk Position  Sapling A  Sapling B  Sapling A  Sapling B  Sapling A  1  -30.663(.133)  -31.159(.235)  -27.039(.179)  -27.822(.161)  -28.406(.188)  -29.015(.102)  2  -30.756(.295)  -31.403(.296)  -27.423(.195)  -27.981(.223)  -28.666(.145)  -29.157(.122)  3  -31.176(.063)  -31.335(.332)  -27.516(.077)  -28.000(.271)  -28.718(.115)  -29.065(.201)  4  -31.251(.096)  -31.022(.107)  -27.664(.093)  -27.869(.028)  -28.718(.130)  -29.105(.110)  5  -31.227(.089)  -31.093(.204)  -27.680(.074)  -27.789(.133)  -28.677(.117)  -28.883(.086)  21  Sapling B  Analysis of variance for effects of height (disk position) and sections within quarters on 8 C of saplings A and B. Table 2.2  13  Sapling A  Height Section Residual Total  Sapling B  Height Section Residual Total  DF  SS  MS  F  P  4 3 12 19  0.272 0.311 0.903 1.486  0.0681 0.104 0.0752 0.0782  0.904 1.376  0.492 0.297  4 3 12 19  0.175 0.238 0.790 1.203  0.0438 0.0794 0.0658 0.0633  0.665 1.207  0.628 0.349  22  DISCUSSION  The influence of wood composition on 5 C was examined by measuring the 13  isotopic signatures of major wood components in reaction wood and adjacent normal wood, from the same tree rings. In conifers, reaction wood (i.e., compression wood) has high lignin and low cellulose content (Westing, 1965; 1968). Consequently, lignin/cellulose ratio should decrease with height in our saplings. In our analysis, lignin to cellulose ratios did appear to vary with height (e.g. through the zone of reaction wood formation), but did not show any clear radial or longitudinal trend (Appendix 1). The cell wall of the compression wood is more heavily lignified adding to the compressive strength of the wood. On the lower side of an inclined stem, the stress may be very high and compression wood is adapted to resist such stresses in the basal part of the stem. It has been suggested that extensive lignification of compression wood mediates loadbearing functions (see Timell 1986).  In a variety of vascular plants and their different tissues, lignin was depleted in 1 3  C by 2-6 % o relative to whole-plant material and by 4-7 % o relative to cellulose  (Benner et al., 1987). Aromatic amino acids phenylalanine and tyrosine are biosynthetic precursors of lignin (Sarkanen and Ludwig, 1971) and it has been suggested that most of the discrimination against C in lignin occurs during 1 3  biosynthesis of phenylalanine and tyrosine (Benner et al., 1987). Among amino acids, tyrosine in particular is relatively depleted in C (Macko, 1987) and is an 1 3  23  important precursor of lignin in grasses. Though tyrosine is not an efficient precursor of lignin in conifers and woody angiosperms (Sarkanen and Ludwig, 1971), the relative contribution of phenylalanine and tyrosine to lignin biosynthesis appears to influence the observed depletion of C in lignin in 1 3  conifers and grasses (Benner et al., 1987).  In addition, methylation of lignin precursors also has potential to deplete lignin in 1 3  C . The relative mass difference between  1 2  C H and  1 3  3  C H (15 vs. 16) is greater 3  than the relative mass difference between C 0 2 and C 0 2 (44 vs. 45). Thus, the 12  1 3  potential for isotope discrimination is quite high. For example, isolated glycine betaine from C 3 and C 4 halophytes was approximately 20% lighter relative to the o  rest of tissue (personal communication Dr. Rob Guy). Glycine betaine has three methyl groups and it may be these methyl groups were contributing to the depletion of  1 3  C.  The 5 C of whole wood did not vary with lignin/cellulose ratio. In contrast, the 13  5 C values of both lignin and cellulose became less negative as the lignin 13  content increased. The 8 C values of the two fractions differed by a constant 13  3.43%o ± 0.26 (mean ± SD, n=40). The excellent correlation between the calculated and observed 8 C of whole wood indicates that virtually all the carbon 13  has been accounted for (i.e., cellulose and lignin) and remaining fractions (i.e., extractives) have little influence on isotopic composition (Fig 2.4). Data presented in figure 2.4 indicates that there are repeatable differences around and through  24  the stem with regards to whole wood 8 C values. These differences probably 13  reflect a lack of circumferential mixing of carbohydrates from the crown of saplings through the phloem. There was, however, no clear pattern with circumferential or longitudinal position (Tables 2.1 and 2.2). Spiral grain would partially scramble these positional signals to varying degree down the stem.  CONCLUSION  Because lignin and cellulose differ in isotopic composition, it has been commonly assumed that differences in the lignin/cellulose ratio might affect whole wood 8 C values. Our analysis using reaction wood and adjacent normal wood reveals 13  that the overall 8 C of whole wood remains constant irrespective of differences 13  in lignin and cellulose content (Fig 2.2 and 2.3, Table 2.1). Based on mass balance considerations this is to be expected if both lignin and cellulose are synthesized from the same ultimate carbon source arriving by way of the phloem. These results reveal that extraction of cellulose prior to isotope analysis actually may introduce, rather than remove, bias resulting from changes in lignin content.  Although wall formation and lignification are temporally separated, most carbon not used to synthesize cellulose must ultimately be incorporated into lignin. If there is no discrimination in cellulose synthesis, and if cellulose synthesis precedes lignin synthesis (Fritts, 1976), then there should be no difference between lignin and cellulose if all the carbon entering a developing xylem cell  25  stays in that cell. The degree of mixing between cells might therefore determine the difference in 5 C between lignin and cellulose (personal communication, Dr. 13  Rob Guy).  26  CHAPTER 3 EFFECT OF FERTILIZATION ON CARBON ISOTOPE DISCRIMINATION AND GROWTH OF EIGHT IMMATURE WESTERN HEMLOCK STANDS INTRODUCTION  The degree to which nutrients are limiting growth, and the capacity of individual trees to respond to nutrient inputs along with other growth limiting factors will determine the response to fertilization. Recognising and examining the factors that play a role in nutrient response are important in that regard. For example, White (2000) reported evidence of growth response to N-only additions in western hemlock stands based on three-year basal area increment. Similarly, physiological studies using P indicated that following N-only additions, a 3 2  secondary deficiency in P was induced. Furthermore, in 5-year-old western hemlock plantation trials, N, P and blend (the blend included additions of S, K, Mg, Zn and Cu) fertilizers increased photosynthetic rates and reduced carbon isotope discrimination (White, 2000).  The relationship between photosynthetic capacity and carbon isotope discrimination is well-established. Several studies have correlated photosynthetic capacity with differences in 5 C values (Sun et al., 1996; White, 2000; Xu et al., 13  2000; Marshall et al., 2001). A n increase in the photosynthetic capacity acts as a sink to draw down intracellular carbon dioxide concentrations which in turn affect  27  tissue 8 C values. In other words, enhanced photosynthetic capacity leads to 13  less discrimination by Rubisco (ribulose-1, 5-bisphosphate carboxylase/oxygenase), resulting in less negative tissue 5 C values. It should 13  be noted that intracellular C 0 concentrations are also determined by stomatal 2  conductance and, therefore, 8 C values reflect the net influence of both stomatal 13  constraints and photosynthetic capacity.  Increased photosynthetic rates following fertilization of western hemlock led to a reduction in carbon isotope discrimination in a severely nutrient stressed 5-yearold plantation (White, 2000). These results prompted the question as to whether nutrient applications in immature stands would lead to similar changes in foliage and stem-wood 8 C. 13  In that regard, White (2000) wondered whether there  would be significant differences in foliage and stem-wood 8 C as a result of 13  different combinations of N, P and blend nutrients.  It has been reported that addition of N may induce deficiencies of other nutrients, notably sulfate levels, which may explain the lack of response to N treatment (White, 2000). For example, in N-fertilized lodgepole pine, a major factor limiting the growth response has been S deficiency induced by N fertilization (Brockley 1990, 1995; Kishchuk et al., 2002). Similarly, studies with Douglas-fir have reported an antagonistic relationship between foliar N and sulfate levels (Turner et al., 1977; Turner et al., 1979; Blake, 1988; Carter et al.,1998). In that regard, determining the available sulfate-S in foliage collected over the first two years  28  after nutrient additions (i.e. 1995 and 1996) would shed more light on the potential relation between N nutrition and available sulfate levels in western hemlock.  I extended my investigation regarding the relationship between nutrient status, growth and carbon isotope discrimination from foliage to stem wood. The isotopic variation within stem wood has also been widely used in ecophysiological and dendrochronological studies (Leavitt, 1993; Walcroft et al., 1997; Porte and Loustau, 2001; Monserud and Marshall, 2001), however its potential in the field of tree nutrition has not been explored. I speculated that stem wood would provide better integration of the expected variation in isotope discrimination through and across the crown as a function of light availability. Comparison of the changes in foliage 5 C with stem-wood 8 C following nutrient applications will 13  13  further our understanding of 8 C dynamics within trees and shed light on the 13  usefulness of stem-wood as a potential diagnostic tool to assess nutrient status.  The relationship between nutrient additions and carbon isotope discrimination (White, 2000) raised the question if 8 C values in western hemlock prior to 13  fertilization could be used to ascertain nutrient status. Can mean 8 C values (of 13  wood or foliage) be used in a similar fashion as foliar analysis? For example, if the mean 8 C value of a particular stand is around -30.00, is this stand more 13  likely to respond to nutrient additions (owing to a greater increase in photosynthetic capacity) than a stand having a mean 8 C value of -27.00? I 13  29  hypothesized that initial 5 C values of foliage or stemwood would reflect initial 13  nutrient status of trees and changes in isotope discrimination after fertilization would reflect growth response and would differ amongst stands depending on their initial nutrient status.  Three-year basal area increment was significantly affected by N only additions and White (2000) recommended six-year basal area increment be used as a superior indicator of the long-term growth response of these trees to nutrient additions.  OBJECTIVES  1) To determine the effect of fertilization treatments on carbon isotope ratios of current-year foliage collected at the end of the first and second growing season following fertilization.  2) To determine the sulfate levels in the foliage collected at the end of the second growing season following fertilization and to investigate the effect of N and P treatments on sulfate levels.  3) To determine the effect of nutrient additions on the isotopic composition of stem-wood on a year-by-year basis.  30  4) To investigate the magnitude and response of nutrient additions on six-year basal area increment.  5) To test whether changes in isotopic composition of stem wood correlates with growth response to fertilization.  31  METHODOLOGY  Stand description  Eight pure western hemlock stands previously selected in an experiment by White (2000) were used to collect samples. Each of these stands had been fertilized in 1990 (Carter et al., undated) and then again in 1995 (White, 2000). They ranged in age between 30 and 50 years when fertilized in 1995, and the site indexes were between 27 and 30. Stands were located on northern Vancouver Island: one near Eve River, two near Port Alice (PA1 & PA2), two near Port McNeill (PN1 & PN2), one near Zeballos, and one in the Nimpkish Valley. The eighth stand is located on the sunshine coast near Sechelt. Each of the stands had undergone pre-commercial thinning around 1983-84. (Fig 3.1).  Treatments  No new fertilization treatments were applied for the present study. Fertilization treatments applied by Dr. White in mid to late May of 1995 were as follows: (1) control, (2) N (225 kg/ha), (3) N (225 kg/ha) + p (100 kg/ha), (4) N (225 kg/ha) + P (500 kg/ha), (5) N (225 kg/ha) + P (100 kg/ha) + blend (230 kg/ha), and (6) N (225 kg/ha) + P (500 kg/ha) + blend (230 kg/ha). Nitrogen and phosphorous were applied as urea and triple-super-phosphate, respectively. The blend fertilizer included additions of S, K, Mg, Zn and Cu. The blend application included 60  32  kg/ha K applied as potassium sulfate, 40 kg/ha Mg applied as magnesium sulfate, 19 kg/ha C u applied as copper sulfate and 20 kg/ha Zn applied as zinc sulfate. This resulted in the addition of approximately 100 kg/ha S in the form of sulfate (White, 2000).  Plot Establishment  The treatments were applied to thirty-six single-tree plots (6 treatments x 6 replicates) at each of the eight installations (i.e. a total of 288 individual trees). The plot radius around each tree was 3.3 m and a 10 m unfertilized buffer was established between all trees (White, 2000).  Foliage Collection  Current-year foliage collected during October 1995 and 1996 (i.e. at the end of the first and second growing seasons following fertilization) were provided by Dr. White and subsequently used for carbon isotope analysis. These foliage samples (i.e. from current-year shoots) were collected from the base of the upper onethird of the live crown (White, 2000).  33  Determination of Isotopic Composition of Current-year Needles  Foliage 5 C values were determined on current year needles collected at the 13  end of the first and second growing seasons after fertilization (White, 2000). The isotopic compositions of approximately 1 mg sub-samples of the ground and pulverized needles packed in tin capsules were determined on an isotope ratio mass spectrometer (Model No. Europa Hydra 20/20) at the Stable Isotope Facility of the University of California, Davis. The 8 C value of the sample is 13  expressed as:  8 C 13  (%o) = [(Rsample  _  Rstandard)/ Rstandard] X  1000  where R mpie and Rstandard are the ratios of C / C respectively in the original 1 3  1 2  sa  foliage samples and the standard. Vienna Pee Dee belemnite (V-PDB) is the arbitrary standard (in practice, a working standard of known composition such as acetanilide is used).  Foliar Sulfur Analysis  Total S and sulfate levels representing the first growing season (1995) were provided (White, 2000). Current-year foliage representing the second growing season (1996) was analyzed for total sulfur and sulfate concentrations. The analysis was carried out by staff at the Pacific Soil Analysis laboratory, Richmond,  34  British Columbia. Total S was determined by combustion with a Leco SC-132 sulphur analyzer (Gutherie and Lowe, 1984). Available sulfate was extracted with 0.1 mol/L HCI (1g foliage per 20 mL of HCI boiled for 20 min) followed by the hydriotic acid bismuth reduction of the extract and bismuth colorimetry using the procedure of Johnson and Nishita (1952).  Increment Core Collection  Increment cores were collected in the winter of 2003 (i.e., eight growing seasons after fertilization) from the single-tree plots at all eight sites. A minimum of two cores representing different directions were taken at breast height from each tree.  Basal Area Increment  Cores were shipped to the Alberta Research Council (Vegreville, AB) for determination of six-year basal area increment following fertilization in 1995. Windendro (version 6.04) was used to measure annual ring increment during the 1992, 93, 94 (three years prior to fertilization) and 1995, 96, 97, 98, 99 and 2000 (six years following fertilization) growing seasons. The radius measured from the pith to a growth ring boundary was used to calculate basal area in any given year. Basal area increment was the difference in area between successive years. Relative change in basal area increment was expressed as a ratio of mean of annual increment during the six years following fertilization (1995-2000) to the  35  mean of annual increment during the three years preceding fertilization (19921994). Determination of relative change in basal area increment would also allow to determine the magnitude and duration of the response to nutrient additions.  Determination of Isotopic Composition within Individual Rings  Measured cores were returned to U B C for preparation of whole wood samples to determine isotope compositions of annual rings. One wood core per tree was destructively sampled to determine 5 C of rings representing three years 13  preceding fertilization (1992, '93, and '94) and three years immediately following fertilization (1995, '96, and '97). Each tree-ring representing above-mentioned years was hand filed from the core and then mixed homogeneously. Care was taken to avoid false rings. Approximately 1 mg of sample was packed in tincapsules and sent for analysis. Samples were analyzed on an isotope ratio mass spectrometer (Model No. Europa Hydra 20/20) at the Stable Isotope Facility, University of California, Davis.  Calculation of summer dryness Index (SDI)  Mean temperature of the warmest month (MTWM) and mean summer precipitation (MSP) were estimated from latitude (LAT), longitude (LONG), and elevation (ELEV) of the eight sites (Table 3.1) using the P R I S M Model (Hamann  36  and Wong, 2004). These data were used to calculate a summer dryness index (Appendix 2) for each of the eight sites (Guy and Holowachuk, 2001):  SDI = (e s[MTWM] A 100)/MSP  where precipitation is in millimeters, and e is the saturation vapor pressure in s  kPa at the MTWM, calculated according to Buck (1981):  e  s n  = 0.61121 x (1.007 + (0.0000346 x P)) x exp((17.502 x 7)/(240.97 + 7))  where P is atmospheric pressure in kPa calculated from elevation (m) after Yin (1998):  P = exp(-ELEV/8000) x 100  Statistical Analysis  The foliage 8 C, change in wood 5 C, relative change in basal area increment 13  13  and sulfate data were analysed statistically by A N O V A using the P R O C General Linear Model (GLM) procedure in S A S (SAS Institute, Cary, N C , USA). Installation and treatment were each considered fixed variables. Differences in mean values were regarded as significantly different at p<0.05. Duncan's multiple range test was used to compare the means. Foliar nutrients levels at the end of  37  the first and second growing seasons recorded by White (2000) were correlated with foliar 8 C, change in wood 8 C and relative change in basal area increment 13  13  using Pearson correlation analysis to generate correlation coefficients and associated P values. Simple linear regressions were performed using the SigmaStat statistical package (version 2.0).  RESULTS Foliar 8 C 13  Differences between sites were significant, whereas treatments were almost significant (P = 0.0539). The interaction between treatments and sites for 8 C of 13  foliage was significant (Table 3.2) at the end of first growing season (1995). The 8 C values became less negative after the addition of P combined with N 13  (NP100) and the blend fertilizer (NP100B) as compared to untreated controls (C) (Fig 3.2), although all the treatments were not statistically different from C. Addition of P with (NP500B) or without (NP500) the blend fertilizer had no detectable effect on 8 C as compared to NP100B and NP100, respectively. 13  Differences in foliage 8 C values at the end of second growing season (1996) 13  were not statistically significant among the different treatments but were significant among the different sites (Table 3.3). No interaction between site and treatment was evident. The 8 C values in 1996 were more negative than in 1995 13  but showed the same general patterns across sites and treatments (Fig 3.2).  38  Fig. 3.1 Example of study site located within the Nimpkish Valley on northern Vancouver Island.  Table 3.1 Latitude, longitude and elevation (m) of the eight study sites.  Location  Latitude  Longitude  Port Alice #1 Port Alice #2 Nimpkish Port McNeill # 1 Port McNeill # 2 Eve River Zeballos Sechelt  50°43' 50°43' 50°29' 50°55' 50°55' 50°43' 50°01' 49°56'  127°39' 127°38' 127°77' 127°23' 127°24' 127°27' 127°09' 124 72' 0  39  Elevation 90 75 240 150 150 150 420 1000  Table 3.2 Two-way analysis of variance (ANOVA) for the effect of treatments and site  on 8 C of foliage collected at the end of first growing season following fertilization (1995). 13  Source of Variance  DF  Treatment Site SiteTreatment  5 7 35  F  P  2.22 19.67 1.82  0.0539 <.0001 0.0057  Table 3.3 Two-way analysis of variance (ANOVA) for the effect of treatments and site  on 8 C of foliage collected at the end of the second growing season following fertilization (1996). 13  Source of Variation  DF  Treatment Site SiteTreatment  5 7 35  F  P  1.94 11.27 1.40  0.0901 <.0001 0.0807  40  -26.0  -29.0  -| N  r  1 NP100  NP100B  NP500  NP500B  Treatment  Fig 3.2 Carbon isotope composition (5 C) of current-year foliage collected at the end of the first (1995) and second growing season (1996) following fertilization (all sites pooled). Each value represents mean ± SE. 13  41  Foliar Sulfate data  The interaction between treatments and sites for available foliage SO4-S (sulfate) levels at the end of the first and second growing seasons following fertilization was significant. In both growing seasons (1995 and 1996) treatments and sites were also significantly different (Table 3.4 and 3.5). Available SO4-S levels were significantly different in N, NP100, and NP100B levels of treatments as compared to C (Fig 3.3). SO4-S levels were reduced by either N or N + P fertilization treatments; whereas blend treatments had elevated S 0 - S levels relative to 4  controls.  Growth Response  The relative change in basal area increment was expressed as a ratio of the mean of annual increment during the six years following fertilization (1995-2000) to the mean of annual increment during the three years preceding fertilization (1992-94). The interaction between treatments and sites was significant for relative change in basal area increment following fertilization. The treatments and sites were also significant (Table 3.6). N applied alone had no significant effect on relative change in basal area increment whereas the NP100 level of treatment had a significant effect as compared to C. Overall, the NP100B level of treatment had the greatest effect on the relative change in basal area increment (Fig 3.4).  42  Two-way analysis of variance (ANOVA) for the effect of treatments and site on SO4-S (ppm) levels in foliage collected at the end of first growing season following fertilization (1995). Table 3.4  Source of Variance  DF  Treatment Site SiteTreatment  5 7 35  F 4.83 14.96 2.43  P 0.0003 <.0001 <.0001  Two-way analysis of variance (ANOVA) for the effect of treatments and site on SO4-S (ppm) levels in foliage collected at the end of second growing season following fertilization (1996). Table 3.5  Source of Variance  DF  F  Treatment Site SiteTreatment  5 7 35  43.37 16.09 3.95  43  P <.0001 <.0001 <.0001  450 400 350 -  • 1995 m  1996  300 -  °Uo  ••Bp  O  C/> 200  ft*  P  150  be  100  d d  cd  cd  cd  ab ••32  50  3sK  0  c  NP100  NP100B  NP500  NP500B  Treatment  Fig 3.3 Sulfate S0 (ppm) levels in the foliage collected at the end of first (1995) and second growing season (1996) following fertilization. Each value represents mean ± SE (all sites pooled). Means within a given year with the same letter are not significantly different at a = 0.05. 4  44  Increasing the P levels from 100 kg/ha (NP100) to 500 kg/ha (NP500) had no further significant effect on growth response (Fig 3.4).  The relative change in basal area increment was also significantly affected by site (Table 3.7). The Port McNeill #1 and Nimpkish sites showed strong response to the NP100B level of treatment followed by Eve River and Sechelt, whereas Port McNeill #2, Port Alice #1 and #2 and the Zeballos sites were generally unresponsive.  Stem Wood 5 C 13  The trend in the 5 C of tree-rings representing three years prior to fertilization 13  and three years following fertilization is presented in Fig 3.5. The change in 8 C 13  of tree rings upon fertilization was significantly different among different levels of treatments and sites. Furthermore, there was an interaction between sites and treatments (Table 3.8). N applied alone had no significant effect on change in wood 5 C, whereas NP100 level of treatment had a significant effect as 13  compared to the control. The greatest change occurred under the NP100 level of treatment and further additions of P at the 500 kg/ha alone or combined with the blend treatments were not significantly different from the NP100 level of treatment (Fig 3.6).  45  Table 3.6 Two-way analysis of variance (ANOVA) for the effect of treatments and sites on the relative change in basal area increment. Relative change in basal area increment was expressed as a ratio of mean of annual increment during the six years following fertilization (1995-2000) to the mean of annual increment during the three years preceding fertilization (1992-1994). Source of Variation  DF  F  P  Site Treatment Site*Treatment  7 5 35  7.57 7.27 1.56  <.0001 <.0001 0.0326  46  NP100  NP100B  NP500  NP500B  Treatment  Fig 3.4 Relative change in basal area increment due to fertilizer treatments. Relative change in basal area increment was expressed as a ratio of mean of annual increment during the six years following fertilization (1995-2000) to the mean of annual increment during the three years preceding fertilization (1992-1994). Each value represents mean ± SE. Means with the same letter are not significantly different at a = 0.05.  47  Table 3.7 Relative change in basal area increment relative to control by treatment and site.  Treatment EVE  NIMP  PA1  PA2  PN1  PN2  ZEB  SEC  Control  1.00  1.00  1.00  1.00  1.00  1.00  1.00  1.00  N  1.37  0.89  0.93  1.04  1.05  1.14  1.14  1.23  NP100  1.29  1.13  0.95  0.93  1.52  1.17  1.22  1.23  NP100B  1.40  1.47  1.00  1.12  1.70  1.19  1.14  1.29  NP500  1.44  1.43  1.05  0.97  1.51  1.50  1.11  1.38  NP500B  1.40  1.52  1.23  1.04  1.75  0.94  1.14  1.11  48  Fig 3.5 General trend in 5 C of tree-rings representing three years before fertilization (1992, 93, 94) and three years following fertilization (1995, 96, 97). The stands were fertilized in May of 1995. Each value represents mean ± SE. 13  49  Table 3.8 Two-way analysis of variance (ANOVA) for the effect of treatments and site on change in 8 C of treeringsfollowing fertilization. The change in 8 C was calculated by subtracting the mean of three years prior to fertilization (1992, 93, 94) from the mean of three years after fertilization (1995, 96, 97). 13  13  Source of variation  DF  F  P  Site Treatment SiteTreatment  7 5 35  4.67 3.03 1.56  <.0001 0.0117 0.0321  50  -0.5 C  N  NP100  NP10QB  NP500  NP500B  Treatment  Fig 3.6 Change in 5 C of treeringsupon fertilization. Each value represents 8 C (mean ± SE) calculated by subtracting the mean of three years prior to fertilization (1992, 93, 94) from the mean of three years after fertilization (1995, 96, 97). Means with the same letters are not significantly different at a = 0.05. 13  13  51  Pearson Correlation Analysis  N, P, K, Mg and Mn showed significant relationships (Table 3.9) with foliar 8 C 13  (1995), although at the end of second growing season (1996), the relationships with K and Mg were not significant (Table 3.10). At the end of first growing season (1995), foliar C a and Mn levels were significantly correlated with changes in wood 8 C (Table 3.9) but a similar relationship was not evident at the end of 13  the second growing season (Table 3.10). Total foliar P and Pi (inorganic phosphate) concentrations were significantly correlated with relative change in basal area increment along with Mn at the end of the first growing season, whereas P, Mg and SO4 correlated significantly with relative change in basal area increment at the end of the second growing season (Table 3.9 and 3.10). Change in wood 8 C significantly correlated with relative change in basal area 13  increment (r = 0.23; P = 0.0002), although mean of the three year wood 8 C 13  prior to fertilization showed poor correlation with the relative change in basal area increment (r = -0.0039; P = 0.95). All other foliar nutrients did not show significant correlations with foliar 8 C , change in wood 5 C and relative change 13  13  in basal area increment (data not shown).  52  Table 3.9 Pearson correlation coefficients and associated P values (below) across •I  < a  treatments and sites for different foliar nutrients for year 1995, foliage 8 C (1995), change in wood 8 C (Mean of year 1992, 93, 94 subtracted from mean of years 1995, 96, 97), and relative change in basal area increment (expressed as ratio of mean of annual increment during years 1995-2000 to mean of annual increment during the years 19921994). Significant relationships at P<0.001 (Bonferroni adjustment) are shown in bold. IJ  13  8 C (Foliage) 0.3900 <.0001 0.2600 <.0001 13  N P  A8 C (wood) 13  0.0500 0.4207 0.1400 0.0283 0.1600 0.0114 -0.1400 0.0248  Relative change in BAI  0.1400 0.0263 0.2600 <.0001 0.2700 <.0001  Pi  0.1400 0.0257  K  0.3000 <.0001  Ca  -0.1500 0.0131  Mg  0.3300 <.0001 -0.2600 <.0001  -0.0600 0.3131  0.1100 0.0676 0.2000 0.0011 -0.1300 0.0397  0.2100 0.0009  0.3100 <.0001  -0.1700 0.0051  0.0200 0.7503 -0.1400 0.0257  0.1036 0.1070 0.0600 0.3251  Mn S0  4  8"C (Foliage) A8 C (wood)  0.2100 0.0008  0.2300 0.0002  13  53  Table 3.10 Pearson correlation coefficients and associated P values (below) across  treatments and sites for different foliar nutrients for year 1996, foliage 5 C (1996), change in wood 8 C (Mean of year 1992, 93, 94 subtracted from mean of years 1995, 96, 97), and relative change in basal area increment (expressed as ratio of mean of annual increment during years 1995-2000 to mean of annual increment during years 1992-1994). Significant relationships at P<0.001 (Bonferroni adjustment) are shown in bold. 13  13  5 C Foliage 0.3600 <.0001 0.2800 <.0001 13  - N P K Ca Mg  0.1800 0.0037 -0.1200 0.0440 0.2000 0.0013  Mn  -0.2900 <.0001  S0  0.0640 0.3100  4  5 C (Foliage) 13  A8 C (wood) 13  -0.1000 0.1000 0.1300 0.0300 -0.1500 0.0156 0.1700 0.0052 -0.1100 0.0745 0.1900 0.0024 0.0700 0.2206 -0.1300 0.0367  Relative change in BAI  -0.0300 0.5888 0.3100 <.0001  0.1500 0.0148 0.1700 0.0071 -0.2700 <.0001  0.1500 0.0167 0.2100 0.0010  0.1300 0.0321 0.2300 0.0002  A5 C(wood) 13  54  600 500  <  400 H  CQ  g  • c s N A NP100 X NP100B A NP500 o NP500B  X  4  t  X  4  .o  300 200  A OA  100 -29.0  •  X  o  -28.5  -28.0  -27.5  -27.0  -26:5  -260  -25.5  -25.0  -24.5  -24.0  §13C  Fig 3.7 Sum of six years basal area increment after fertilization (1995-2000) as a function of the mean of three years wood 8 C prior to fertilization (1992, 93, 94). Each data point represents one individual tree. The correlation is significant (r = 0.43;P<0.001). ' 13  55  DISCUSSION It is now well-established in the field of tree nutrition that nutrient deficiencies can reduce the photosynthetic capacity. The main reason for this reduced photosynthetic capacity is due to the activity of the enzyme Rubisco that is affected by the availability of N, P and other nutrients (Clearwater and Meinzer, 2001; Warren and Adams, 2002). These results were corroborated by the studies on western hemlock nutrition where nutrient deficiencies significantly reduced the photosynthetic capacity of western hemlock (White, 2000). Furthermore, this reduction in photosynthetic capacity was most strongly affected by severe N deficiency. Similarly, the photosynthetic capacity of western hemlock was limited by a secondary deficiency in P (White, 2000). In a recent study on the changes in water and carbon relations of western hemlock to dwarf mistletoe infection, it was shown that leaf nitrogen content was 35% lower and photosynthetic rates were approximately 50% lower in infected western hemlock trees as compared to uninfected trees (Meinzer et al., 2004). Furthermore, the foliar 8 C values were 13  2.8 % more negative in infected than in uninfected trees indicating a decline in 0  photosynthesis as a result of N-stress.  Treatment effects on foliar 8 C were almost significant (P = 0.0539) at the end of 13  first growing season after fertilization. This result suggests that there may have been a physiological effect of treatments on 5 C , even though the different 13  treatment levels could not be separated from the control. Furthermore, in 1995  56  there was an interaction between sites and treatments indicating that different sites respond to treatments differently. Although at the end of second growing season treatment effects were not significant.  The effect of N on photosynthetic capacity and carbon isotope discrimination is well-established (Guy et al., 1993; Ellsworth and Liu, 1994; Dietz and Harris, 1997; Clearwater and Meinzer, 2001; McDowell et al., 2002). It has been suggested that addition of N alone can decrease the foliar concentration of both total P and inorganic P (Pi) thus inducing a secondary P deficiency (White, 2000). There was no significant difference between N alone and N + P combinations on foliar 8 C suggesting that P deficiency did not influence 8 C values. 13  13  It has been suggested that addition of P will increase the photosynthetic rate in fertilization scenarios (Rao and Terry, 1986, 1990; Reich and Schoettle, 1988; White, 2000; Rausch and Bucher, 2002; Brown and Courtin, 2003). P deficiency has been shown to affect chloroplast ultrastructure and photosynthesis rate (White, 2000). In Arabidopsis,  a chloroplast phosphate transporter was identified  that influences allocation of phosphate within the plant and phosphate-starvation responses (Versaw and Harrison, 2002). P would be needed for growth too, so the effect on photosynthesis could actually be through enhanced sink strength. Future studies should explore the physiological dynamics of N and P homeostasis and its effects on photosynthesis and carbon isotope discrimination.  57  The addition of P increased height growth, photosynthetic rates and reduced carbon discrimination in a 5-year-old western hemlock plantation (White, 2000). In that study the mean foliar 5 C values of control tissue were -31.30 and -30.99 13  at the end of first and second growing season, respectively. The mean foliar 8 C 13  values of control tissue in our studies were -27.90 and -28.34 at the end of first and second growing season, respectively. The mean %N of control foliage in our studies across all sites were 1.10 and 1.02 at the end of first and second growing season, respectively. In 5-year-old plantation study, the mean %N of control foliage were 0.73 and 0.76 at the end of first and second growing season, respectively (White, 2000).  It appears then that these stands were not as severely deficient in nutrients as the 5-year-old plantation examined by White (2000). Furthermore, the site indexes of our sites were between 27 and 30 suggesting that these are productive sites. This may explain why foliage 5 C values were not much 13  affected by nutrient additions at the end of first and second growing season in the older stands in the present study, although a repeatable trend was evident (i.e. 8 C values becoming less negative). These results indicate that N, P and blend 13  nutrients have a physiological effect on foliage 8 C , but the magnitude of this 13  response may vary according to site and weather parameters. These results also indicate that the effect of nutrients on foliar 8 C is not a sustained response, 13  in particular, when the stands are not severely deficient.  58  Several stands showed increased growth following the addition of other nutrients combined with N and P (White, 2000). Application of only N may induce deficiencies of other nutrients, which may explain some lack of response to N treatment (White, 2000; Brockley, 2004). If the sulfate concentrations for control foliage in lodgepole pine is between 40 - 60 ppm, it is considered to be moderately to severely deficient and growth response following N fertilization is unlikely unless S is added in combination (Brockley, 2001). Similar critical threshold levels in western hemlock for sulfate deficiency has not been yet established (Brown, 2001). Carter et al., (2001) reported decrease in foliar sulfate levels by N-only fertilization at the end of first-year growing season in western hemlock, compared with controls (43 vs. 198 mg/kg). Foliar sulfate ( S 0 4  S) levels determined at the end of the first (49.44 vs. 81.99) and second growing season (54.81 vs. 133.63) were significantly reduced in N only treatment, compared with controls (Fig 3.3) which in turn may explain lack of response to N additions.  Further studies exploring the physiological interactions between N and S would shed more light on this antagonistic relation. Nevertheless, correlation analysis did not show any significant relationship between foliage SO4-S levels and foliage 8 C or change in wood 8 C . Sulfate levels appear to play an important role in 13  13  growth of trees since foliage SO4-S levels at the end of second growing season after fertilization had a significant relationship with growth response (Table 3.10). The physiological role of S in maintaining photosynthetic apparatus is well  59  understood. For example, the transcriptional profile of several genes regulating photosynthesis was perturbed in Arabidopsis  under S deficiency (Hirai et al.,  2003) suggesting that S, like P, is also needed for growth.  White (2000) reported that the three-year basal area response was significantly affected by N-only treatment and none of the eight stands responded to P additions at either rate (100 kg/ha or 500 kg/ha). The need for six-year basal area response was emphasised, as it could detect not only the magnitude but also the duration of the response. The addition of N alone did not result in a growth response (Fig 3.4), supporting the notion that the response of western hemlock following N fertilization is often limited by secondary deficiencies. Addition of blend (NP100B) may have alleviated S deficiency and may explain why the best growth response to fertilization was achieved with NP100B treatment.  These results are not consistent with the three-year basal area response data reported by White (2000) in which stands responded to N-only treatments. This apparent contradiction may be due to the fact that three-year basal area response is less sensitive than six-year basal area response since an increase in basal area must be preceded by one or more years of crown expansion. Gough et al., (2004) reported a temporary increase, followed by a reduction, in foliar N and photosynthetic capacity following fertilization of loblolly pine, indicating that the timing of measurements after fertilization is critical to detect the effect of N on  60  photosynthetic capacity. It appears that the growth response to the N-only treatment over three-years in western hemlock was short-term, whereas over sixyears N levels may have been 'diluted' and hence the growth response is no longer perceptible.  The growth of western hemlock following N fertilization was limited by P deficiency and application of P at 100 kg/ha was not sufficient to relieve this deficiency (White, 2000). Furthermore, P concentrations were low in current-year foliage of either the control or N treated plots at all installations indicating a P deficiency (White, 2000). The relative change in basal area increment data indicated that application of P at 500 kg/ha (NP500) had no further significant effect on growth response (Fig 3.4) suggesting that from silvicultural point of view, application of P at 100 kg/ha may be sufficient. In that regard, the best response through NP100B can be explained as a result of balanced nutrition. Future studies should include a blend treatment without P and a P only treatment (as control) to conclusively address the role of P nutrition.  The tree-ring 6 C values were quite variable even before the treatments (i.e. 13  year 1992, 93, 94) (Fig 3.5). It appears that inherent genetic variation and weather play a significant role in influencing stem wood 5 C . For example, the 13  1997 growing season (May-September) was relatively wet as compared to the growing seasons of 1992, 93, 94, 95 and 96 (Appendix 3) and hence control 5 C 13  values were more negative. It appears that changes in 8 C values as a result of 13  61  tree water relations (i.e. water availability, water-stress or vapour pressure deficit) are still highly significant despite coastal B C receiving relatively abundant precipitation (contrary to our assumption). Another factor contributing to a low "signal-to-noise ratio" in the data could be that sample size (i.e. number of trees per treatment) may not have been sufficient to account for the inherent genetic variation in 5 C . 13  Data presented in chapter 2 clearly showed that the 5 C values of stemwood 13  can vary from one position to another on the trunk. For this chapter, wood rings were analysed from only one position on each tree stem. Therefore, heterogeneity in stem-wood 5 C as a function of position may be yet another 13  factor reducing the "signal-to-noise ratio". This heterogeneity presumably results from a lack of physiological integration of phloem around the circumference of the stem relative to source tissues in the canopy. Different sides of the crown may feed different sectors around the stem. In addition, spiral grain would partially scramble these positional signals (i.e. carbohydrates) to varying degree down the stem. Because the isotopic composition of western hemlock wood is quite variable, several cores may need to be sampled to get a representative value for each tree. Similarly, to encompass the differences in the circumferential variation, several cores from different positions on the bole may be needed.  Although foliar 8 C values were correlated with tree-ring 5 C values for year 13  13  1995 (r = 0.26, P <0.001) and year 1996 (r = 0.24, P <0.001), respectively,  62  correlation coefficients were low. These results may again highlight the lack of circumferential mixing of carbohydrates from the crown of trees through the phloem. There may also be a temporal lag with respect to the distribution of carbohydrates. In a recent study with broad-leaf deciduous tree species, a triphase carbon isotope pattern in tree rings was evident, which cannot be explained by the common model of carbon isotope fractionation during photosynthesis. This tri-phase pattern has been suggested to result from changes in downstream processes of carbohydrate metabolism such as accumulation and remobilization of storage material, carbohydrate partitioning etc. (Helle and Schleser, 2004). Complex downstream processes of carbohydrate metabolism may explain, in our studies, the poor correlation between foliar 8 C and stem-wood 8 C values. However, further studies with 13  13  regard to 8 C patterns within a tree may be quite useful to explore carbon 13  transfer in trees.  Data presented in figure 3.7 show a correlation between water-use efficiency and growth, corroborating other previous studies (Livingston et al., 1999; Guy and Holowachuk, 2001). A similar pattern was evident when the mean of three years wood 8 C prior to fertilization was correlated with the three years basal area 13  increment prior to fertilization (data not shown). These data indicate that the fertilization effect is not that pronounced when compared to natural within-stand variation in growth potential or nutritional status as a function of genetics or microsite at the individual tree level. If we exclude shade and soil C 0  63  2  effects on  8 C , the variation in values is not likely to be due to differences in transpiration 13  but, rather, assimilation. Thus, genetic differences in sink strength or the nutritional status of the individual trees may account for differences in 5 C . If 13  nutrition, then tree-to-tree variation in nutritional status must far exceed the impact of the fertilizer treatments (in combination with the circumferential noise in 5 C values of the stem wood). Tree-to-tree genetic variation in sink strength 13  would obscure the treatment patterns in a similar fashion. Experiments to manipulate sink strength independently of photosynthesis would shed more light on this relationship. For example Guy et al., (2001) were able to increase 8 C , 13  height and stem dry weight growth in greenhouse grown lodgepole pine with applications of gibberellins.  The change in the means of tree-ring 8 C representing three years after and 13  prior to fertilization was significant (Table 3.8). Most of the change can be attributed to the combined effect of N and P. These results indicate that treatments indeed had a physiological impact on trees with regard to 8 C (Fig 13  3.6). Since our tree-ring 8 C values do reflect a physiological change, isotope 13  information can be used for dendrochronological studies to supplement growth data (i.e. ring width), which may be less sensitive.  Site means (e.g. mean of 8 C of controls foliage (1995 -1996), mean of wood 13  8 C (1992 - 94), mean of N content of controls foliage (1995 -1996), mean of P 13  content of controls foliage (1995 -1996) and mean foliar SG^of controls foliage)  64  did not correlate significantly with summer dryness index (calculated from climate variables modelled by PRISM)(Data not shown). These results indicate that there is no obvious genetic adaptation to site in terms of W U E and the likelihood of drought. However, provenance differences in isotope discrimination may not be apparent unless populations are grown in a common garden with uniform moisture and nutrient conditions.  To compliment nutritional data (i.e. foliar nutrients) into the photosynthetic response following fertilization, as determined through foliage and wood 5 C, 13  correlation analyses were performed (Table 3.9 and 3.10). Correlation analysis indicated that several other mineral nutrients, besides N, P, and S, showed significant relationships with foliage 5 C, change in wood 8 C, and growth 13  13  response. Forward stepwise multiple regressions using all foliar nutrients, foliage and wood 8 C to predict the relative change in basal area increment on a tree13  by-tree basis were also performed. Nothing additional came out of these multiple regressions, however, beyond what was already apparent from Pearson's correlation (Data not shown).  Several physiological studies have demonstrated that K, Ca, Mg, Mn are nutrients that play a role in mediating chloroplast functions and photosynthesis. For example, leaf photosynthesis, leaf expansion and leaf transpiration were reduced after only 1 day of C a starvation in tomato plants (Amor et al., 2003). Ishijima et al., (2003) in their studies with spinach demonstrated that free Mg in  65  chloroplasts may contribute to the regulation of photosynthetic enzymes. Similarly, perturbations in C a and Mg homeostasis led to defects in chloroplast functions in tobacco and spinach (Sai and Johnson, 2002; Ishijima et al., 2003). In beech seedlings, low C a and Mg nutrition lead to a strong dysfunction and reduced photosynthesis (Michele and Jean-Pierre, 2000). These studies highlight the importance of C a and Mg homeostasis in photosynthesis.  Using chlorophyll fluorescence, high levels of Mn were shown to result in a decline in photosynthetic rate in deciduous broad-leaved trees (Kitao et al., 1997; Kitao et al., 1998). Similarly, K plays a role in photosynthesis (Sun and Payn, 1999; Goh et al., 2002) and other metabolic processes of plants (Emanuel and Bloom, 2004). Since these mineral nutrients are involved in chloroplast functions, it is possible that they are influencing tissue 5 C partly due to their effects on the 13  photosynthetic apparatus.  The soils on Vancouver Island are derived from rocks that are rich in C a and Mg (Krajina, 1969). Therefore, soil type and properties should be considered while interpreting the effects of these nutrients, in particular C a and Mg, on tissue 8 C 13  and growth response. Precise climate data may further our understanding regarding the role of nutrients (i.e. nutrient uptake and mobilization rates as a function of tree water relations) in influencing tissue 5 C and tree growth. Taken 13  together, N, P, K, C a , Mn and Mg appear to have an influence on carbon isotope discrimination potentially through changes in the photosynthetic capacity. Further  66  physiological studies are needed to understand the dynamics of these nutrients involved in mediating photosynthetic rate and growth.  CONCLUSION  1) Although there was a physiological effect of treatments on foliar 8 C, the 13  magnitude of this effect was small.  2) Foliar S 0 - S (Sulfate) levels at the end of the first and second growing 4  seasons following fertilization were reduced by either N or N + P fertilization treatments.  3) N applied alone had no significant effect on change in wood 5 C. The 13  greatest change was in the NP100 level of treatment.  4) N applied alone had no significant effect on relative change in basal area increment whereas NP100B had the greatest effect. The relative change in basal area increment was also significantly affected by sites. Port McNeill #1 and Nimpkish were the most responsive sites.  5) Change in wood 8 C significantly correlated with relative change in basal 13  area response but of not large enough magnitude to be used as a promising diagnostic tool.  67  CHAPTER 4  THESIS CONCLUSION  The objective of my thesis was to explore the use of carbon isotope analysis as a physiological tool to diagnose the nutritional status and response of trees to fertilization. There was indeed a physiological effect on trees as a result of nutrient additions which was evident through an increase in growth and in changes in foliage and wood 5 C values. These eight stands had been subjected 13  to previous fertilization, with similar treatments, by Carter and Klinka (undated) in 1990. Four of the stands responded to prior treatments while the other four did not show evidence of response. These sites were selected by White (2000) on the basis of this previous response (i.e. they were not randomly selected). Hence, the site by treatment interaction was expected (personnel communication Dr. Barry White). The variability among stands in their response to new treatments (White, 2000) indicates that stands vary in nutrient status, capacity to nutrient response, external factors like year-by-year weather fluctuations or in nutrient uptake.  The nutrient effect on 8 C was small relative to annual weather patterns and, 13  most probably, inherent genetic variation within trees. Furthermore, intrinsic site characteristics (i.e. interaction between sites and treatments) also influenced nutrient effects on 8 C in a similar fashion. 13  68  The question I asked was "can mean 8 C values in western hemlock be 13  used in a similar fashion to that which conventional foliar analysis is used to determine the nutritional status of a stand?" Furthermore, "can changes in discrimination of foliage and stemwood in the years following fertilization relative to that prior to fertilization be related to the nutritional status and long-term growth response?"  Despite the physiological effect on foliage 5 C due to nutrient additions, it 13  appears that mean 8 C values in western hemlock foliage cannot be used in a 13  similar fashion to that of conventional foliar analysis to determine the nutritional status of trees. These stands appear to be moderately deficient in nutrients as compared to severely deficient 5-year-old western hemlock plantation and that may partly explain why there were no significant treatment effects on foliage 8 C 13  values. The effect of nutrients on foliage 6 C is not a sustained response either 1 3  and is further masked by intrinsic site characteristics and weather parameters. However, carbon isotope analysis may yet have potential to be used as a physiological tool to diagnose nutritional status of severely deficient trees.  Since the tree-ring 5 C values were quite variable even before the fertilizer 1 3  treatments, it appears that annual fluctuation in climate parameters, site characteristics and inherent genetic variation among trees are playing a significant role in influencing tree-ring 6 C . This complex interaction may have 13  masked the effect of nutrients on tissue 5 C . Furthermore, poor correlations 13  between foliage and wood 5 C values indicate that either 1). the potential for 1 3  69  discrimination exists after the process of carboxylation, 2). there is a temporal lag with regard to carbohydrate mobilization or 3). stem wood may not integrate canopy processes all that well. The mean of three years wood 5 C prior to 1 3  fertilization did not show a significant relationship with the relative change in basal area response following fertilization. Thus, despite a physiological effect on wood 5 C , analysis of wood 5 C from a single core cannot be used to 13  1 3  determine nutritional status prior to fertilization.  The reduced foliar sulfate levels in the N and N + P combination of treatments indicate an antagonistic relationship between N and SO4 levels consistent with studies on Douglas-fir and lodgepole pine. Sulfate is an important nutrient required by trees for their growth. Application of N reduced S 0 levels which may 4  explain lack of growth response to further N additions.  The best growth response to nutrient additions was obtained with the NP100B. This appears to be an effect of a more balanced nutrition since the blend included sulfate and other nutrients. Application of P at 500 kg/h did not appear to have further significant effect on growth as compared with application of P at 100 kg/h. Based on growth response data, Port McNeill #1 and Nimpkish were the most responsive sites followed by Eve River and Sechelt.  Another question I asked was "if use of whole wood 8 C is more pertinent 13  than using cellulose 8 C in carbon isotope analysis of wood?" 13  70  To address this question my experiments using reaction wood and adjacent normal wood in two western hemlock sapling indicated that lignin was depleted in 1 3  C relative to whole wood, whereas cellulose was enriched. The isotopic mass  balance of whole wood was conserved (i.e., isotopically light C not used to synthesize cellulose must end up in lignin) and therefore did not vary with lignin content. Whole wood 5 C values varied from spot to spot on the bole but there 13  was no obvious pattern to this variation. Contrary to current opinion, extraction of cellulose prior to isotope analysis may actually introduce, rather than remove, bias resulting from changes in lignin content.  Taken together, .the use of 5 C values in western hemlock foliage and stem13  wood prior to fertilization, is not a promising technology to operationally infer the nutritional status of trees. Because of complex patterns of inherent genetic variation, weather parameters and intrinsic site characteristics, changes in foliage and stem-wood 5 C values due to nutrient additions cannot be easily related to 13  the long-term response of the stands. However my studies with carbon isotopes provided another piece of information which could be useful in context of other studies related to western hemlock and have several implications. Carbon isotope information in combination with growth data can be used for dendrochronological studies (i.e. climate reconstruction). For example, there may not be much variation in ring-width from year-to-year but good physiological variation in 5 C dynamics. Exploring 5 C patterns within a tree may be useful in 13  13  studies of carbon transfer in trees which may provide valuable information to  71  explore carbon allocation patterns with respect to potential climate change. Similarly, isotope information could be used to understand W U E , N U E and their relationship with growth which may be quite useful with respect to breeding superior genotypes of western hemlock.  72  FUTURE RESEARCH  Future research should explore the physiological dynamics of N, P and sulfate homeostasis and its effects on photosynthesis and carbon isotope discrimination. Most of this work can be complimented with studies at the molecular and cellular level. The availability of Arabidopsis  mutants in N, P and Sulfate transporters  could be exploited to address the physiological dynamics and then pertinent findings can be extrapolated to perform relevant experiments in trees. A similar approach could be taken to determine the physiological roles of K, Mg and C a . From a physiological point of view, it is important to understand the dynamics and interactions between different nutrients along with their synergistic or antagonistic relations on photosynthetic capacity or other biological process. The recent (Sept. 2004) sequencing of a genome of a tree species  trichocarpa)  (Populus  and availability of spruce gene chips are promising technologies for  the pursuit of tree nutrition studies at the molecular level.  From a physiological point of view, in future experiments, treatments that include blend with and without P (as controls) are needed to conclusively address the role of P nutrition in western hemlock. Data on climate parameters from different study-sites may aid in addressing the precise role of nutrients in 5 C dynamics 13  and potential relationships with growth response in western hemlock. Weather data may also be useful to further understand the potential relationship between water-use efficiency, nitrogen-use efficiency and growth response in western  73  hemlock. Similarly, several study-sites should be selected to represent different biogeoclimatic subzones of the C W H biogeoclimatic zone of B C to conclusively address the relationship between carbon isotopes, nutrient status and potential growth response due to nutrient additions in western hemlock.  The heterogeneity in stem-wood with regard to 5 C values indicates that wood is 13  quite variable. To capture this heterogeneity several cores sampled from different positions around the bole of the tree may be necessary for isotope analysis. Furthermore investigations into the physiological integration of phloem transport and the spiral grain pattern should be pursued by examining the longitudinal and circumferential trends in 5 C using the same tree rings from several different 13  vertical positions on the bole.  These are exciting times in the field of plant biology. A multidisciplinary approach utilizing ecological, silvicultural and molecular techniques may prove to be pivotal in understanding the physiological dynamics of tree nutrition. Knowledge gained from physiological studies can be incorporated to design silvicultural practices to improve the growth of forest trees.  74  REFERENCES  Amor del FM, Marcelis L F M (2003) Regulation of nutrient uptake, water uptake and growth under calcium starvation and recovery. J Hort Sci & Biotech 78: 343-349 Anderson S, Zososki RJ, Gessel S P (1982) Phosphorus and lime response of Sitka spruce, western hemlock seedlings and romaine lettuce on two coastal Washington soils. Can J For Res 12: 985-991 Benner R, Fogel ML, Sprague EK, Hodson R E (1987) Depletion of C in lignin and its implications for stable carbon isotope studies. Nature 329: 708710 1 3  Blake Jl, Webster SR, Gessel S P (1988) Soil sulfate-sulfur and growth responses of nitrogen fertilized Douglas-fir to sulfur. Soil Sci Soc Am J 52: 1141-1147  Brockley R P (1990) Response of thinned, immature lodgepole pine to nitrogen and boron fertilization. Can J For Res 20: 579-585  Brockley R P (1995) Effects of nitrogen source and season of the application on the nutrition and growth of lodgepole pine. Can J For Res 25: 516-526  Brockley R P (2001) Foliar sampling guidelines and nutrient interpretative criteria for lodgepole pine. British Columbia Ministry of Forests, Victoria, B C . Ext. Note 52  Brockley R P (2004) Effects of different sources and rates of sulphur on the growth and foliar nutrition of nitrogen-fertilized lodgepole pine. Can J For Res 34: 728-743  Brooks JR, Flanagan LB, Ehleringer J R (1998) Responses of boreal conifers to climate fluctuations: indications from tree ring widths and carbon isotope analyses. C a n J For Res 28: 524-533  75  Brown K (2003) Growth and nutritional responses of western hemlock to fertilization: A review (http://www.forrex.org/iem/2003/vol3/no2/art3.pdf)  Brown K, Courtin P J (2003) Effects of phosphorus fertilization and liming on growth, mineral nutrition, and gas exchange of Alnus rubra seedlings grown in soils from mature alluvial Alnus stands. Can J For.Res 33: 2089-2096  Buck A L (1981) New calculations for computing vapour pressure and enhancement factor. J Appl Meteorol 20: 1527-1532  Carter RE, McWilliams E R G , Klinka K (Undated) Effects of climate, soil water nutrients on growth of coastal western hemlock: response to fertilization. Report to the Science Council of British Columbia.  Carter RE, McWilliams E R G , Klinka K (1998) Predicting response of coastal Douglas-fir to fertilizer treatments. For Eco Manage 107: 275-289  Carter RE, McWilliams E R G , Klinka K (2001) Does Coastal Western Hemlock respond to fertilization? Scientia Silvica Extension Series No.44  Cernusak LA, Marshall J D (2001) Responses of foliar delta13C, gas exchange and leaf morphology to reduced hydraulic conductivity in Pinus monticola branches. Tree Physiol 21: 1215-1222.  Clearwater MJ, Meinzer F C (2001) Relationships between hydraulic architecture and leaf photosynthetic capacity in nitrogen-fertilized Eucalyptus grandis trees. Tree Physiol 21: 683-690  Condon A G , Richards RA, Farquhar G D (1987) Carbon isotope discrimination is positively correlated with grain yield and dry matter production in field-grown wheat. Crop Sci 27: 996-1001  Darling LM, Omule S A Y (1989) Extensive studies of fertilizing and thinning coastal Douglas-fir and western hemlock: an establishment report. B C Min For Res Br F R D A Rep No 054, Victoria, B C 17p  76  Dietz K, Harris G C (1997) Photosynthesis under nutrient deficiency. In: Pessaraki M, Dekker M (eds) Handbook of Photosynthesis. New York, pp 951972  Ehleringer J R (1990) Correlations between carbon isotope discrimination and leaf conductance to water vapour in common beans. Plant Physiol 93: 14221425 Ellsworth DS, Liu X (1994) Photosynthesis and canopy nutrition of four sugar maple forests on acid soils in northern Vermont. Can J For Res 24: 2118-2127  Emanuel E, Bloom A J (2004) Mineral Nutrition of Plants: Principles and Perspectives. Sinauer Associates Books, 380p.  Environment Canada Website (http://www.climate.weatheroffice.ec.gc.ca).  Farquhar GD, O'Leary MH, Berry J A (1982) On the relationship between carbon isotope discrimination and the intercellular carbon dioxide concentration in leaves. Aust J Plant Physiol 9: 121-137  Farquhar GD, Richards RA (1984) Isotopic composition of plant carbon correlates with water-use efficiency of wheat genotypes. Aust J Plant Physiol 11: 539-552  Farquhar GD, Ehleringer JR, Hubick KT (1989) Carbon isotope discrimination and photosynthesis. Annu Rev Plant Physiol Plant Mol Biol 40: 503-537  Fritts HC (1976) Tree Rings and Climate, Academic Press, New York  Gill R, Lavender DP (1983) Urea fertilization: effects on primary root mortality and mycorrhizal development of young-growth western hemlock. For Sci 29: 751-760  77  Goering HK, Van Soest (1970) Agriculture Handbook No 379 U S Department of Agriculture, Washington, pp 1 -20  Goh C H , Dietrich P, Steinmeyer R, Schreiber U, Nam HG, Hedrich R (2002) Parallel recordings of photosynthetic electron transport and K+ channel activity in single guard cells. Plant J 32: 623-630 Gough C M , Seiler JR, Maier C A (2004) Short-term effects of fertilization on loblolly pine (Pinus taeda L ) . physiology. Plant, Cell and Environ 27: 876886 Green JW (1963) Wood cellulose. In: Whistler RL (ed), Methods in Carbohydrate Chemistry. Academic Press, New York, pp 9-21 Guthrie TP, Lowe L E (1984) A comparison of methods for total sulphur analysis of tree foliage. C a n J For Res 14: 470-473 Guy RD, Fogel ML, Berry J A (1993) Photosynthetic fractionation of the stable isotopes of oxygen and carbon. Plant Physiol 101: 37-47  Guy RD, Holowachuk DL (2001) Population differences in stable carbon isotope ratio of Pinus contorta Dougl. Ex Loud: relationship to environment, climate of origin, and growth potential. Can J Bot 79: 274-283  Guy RD, Pharis RP, Aitken S N , Zhang R, Fung L (2001) Finding the best trees for the job: realizing the full potential of afforestation in Canada. Climate Change2: Canadian Technology Development Conference, Proceedings. Canadian Nuclear Society, Toronto, O N . 12p  Hamann A, Wang T (2004) Models of climatic normals for genecology and climate change studies in British Columbia. Agricultural and Forest Meteorology (In-press)  Heilman P E , Ekuan G (1980) Phosphorus response of western hemlock seedlings on Pacific coastal soils from Washington. Soil Sci Soc Am J 44: 392-395  78  Helle G, Schleser G H (2004) Beyond C0 -fixation by Rubisco - an interpretation of C / C variations in tree rings from novel intra-seasonal studies on broad-leaf trees. Plant, Cell & Environment 27: 367-380 2  1 3  1 2  Hemming DL, Switsur VR, Waterhouse J S , Heaton T H E , Carter A H C (1998) Climate variation and the stable carbon isotope composition of tree-ring cellulose: an intercomparison of Quercus  robur, Fagus  sylvatica  and  Pinus  sylvestris. Tellus 50B; 25-33 Hirai MY, Fujiwara T, Awazuhara M, Kimura T, Noji M, Saito K (2003) Global expression profiling of sulphur-starved Arabidopsis by DNA macroarray reveals the role of O-acetyl-l-serine as a general regulator of gene expression in response to sulphur nutrition. Plant J 33: 651-663  Hubick KT, Farquhar GD, Shorter R (1986) Correlation between water-use efficiency and carbon isotope discrimination in diverse peanut germplasm. Aust J Plant Physiol 13: 803-816  Ishijima S, Uchibori A, Takagi H, Maki R, Ohnishi M (2003) Light-induced increase in free Mg2+ concentration in spinach chloroplasts: measurement of free Mg2+ by using a fluorescent probe and necessity of stromal alkalinization. Arch Biochem Biophys 412: 126-132  Jean PHB, Flanagan LB, Martinelli LA, Moreira MZ, Higuchi N, Ehleringer J R (2002) Carbon isotope discrimination in forest and pasture ecosystems of the Amazon Basin, Brazil. Global Biogeochem Cycles/16: X-1 X-10.  Johnson HH, Nishita H (1952) Microestimation of sulphur in plant materials, soils and irrigation waters. Anal Chem 24: 736-742  Kitao M, Lei TT, Koike T (1997) Comparison of photosynthetic responses to manganese toxicity of deciduous broad-leaved trees in northern Japan. Environ Pollut 97: 113-118  Kitao M, Lei TT, Koike T (1998) Application of chlorophyll fluorescence to evaluate Mn tolerance of deciduous broad-leaved tree seedlings native to northern Japan. Tree Physiol 18: 135-140  79  Kishchuk BE, Brockley R P (2002) Sulphur availability on lodgepole pine sites in British Columbia. Soil Sci Soc Am J 66: 1325-1333  Korol RL, Kirschbaum MU, Farquhar GD, Jeffreys M (1999) Effects of water status and soil fertility on the C-isotope signature in Pinus radiata. Tree Physiol 19: 551-562 Krajina V J (1969) Ecology of forest trees in British Columbia. In: Krajina V J (ed), Ecology of Western North America. University of British Columbia Department of Botany, Vancouver B C . pp 1-146  Leavitt SW, Long A (1986) Stable-carbon isotope variability in tree foliage and wood. Ecology 67: 1001-1010  Lin G, Ehleringer J R (1997) No carbon isotopic discrimination during dark respiration in C and C plants. Plant Physiol 114: 391-394 3  4  Livingston NJ, Guy RD, Sun Z J , Ethier G J (1999) The effects of nitrogen stress on the stable carbon isotope composition, productivity and water use efficiency of white spruce seedlings. Plant Cell Environ 22: 281-289  Loader NJ, Switsur V R (1996) Reconstructing past environmental change using stable isotopes in tree-rings. Bot J Scotl 48: 65-78  Loader NJ, Robertson I, McCarroll D (2003) Comparison of stable carbon isotope ratios in the whole wood, cellulose and lignin of oak tree-rings. Palaeogeog Palaeoclimatol Palaeoecol 196: 395-407 McDowell NG, Phillips N, Lunch C, Bond BJ, Ryan M G (2002) An investigation of hydraulic limitation and compensation in large, old Douglas-fir trees. Tree Physiol 22: 763-774  McCarroll D, Pawellek F (2001) Stable carbon isotope ratios of Pinus sylvestris from northern Finland and the potential for extracting a climate signal from long Fennoscandian chronologies. Holocene 11: 517-526  80  Macko SA, Fogel ML, Hare PE, Hoering T C (1987) Isotopic fractionation of nitrogen and carbon in the synthesis of amino acids by microorganisms. Chem Geol 65: 79-92  Marshall JD, Rehfeldt G E , Monserud RA (2001) Family differences in height growth and photosynthetic traits in three conifers. Tree Physiol 21: 727-734 Meinzer FC, Woodruff DR, Shaw DC (2004) Integrated responses of hydraulic architecture, water and carbon relations of western hemlock to dwarf mistletoe infection. Plant, Cell and Environ 27: 937-946  Michele R, Jean-Pierre G (2000) Consequences of an excess Al and a deficiency in C a and Mg for stomatal functioning and net carbon assimilation of beech leaves. Annals of Forest Science 57: 209-218  Monserud RA, Marshall J D (2001) Time-series analysis of delta 13C from tree rings. I. Time trends and autocorrelation. Tree Physiol 21: 1087-1092  Park R, Epstein S (1961) Metabolic fractionation of C and C in plants. Plant Physiol 36: 133-138 1 3  1 2  Pojar J K, Klinka K, Demarchi DA (1991) In: Meidinger D, Pojar J (ed), Ecosystems of British Columbia. B C Ministry of Forests, Victoria, pp 95-111  Porte A, Loustau D (2001) Seasonal and interannual variations in carbon isotope discrimination in a maritime pine stand assesses from the isotopic composition of cellulose in annual rings. Tree Physiol 21: 861-868  Radwan MA, DeBell DS (1980) Site index, growth and foliar chemical composition of western hemlock. For Sci 26: 283-290  Radwan MA, Shumway J S (1983) Soil nitrogen, sulphur, and phosphorus in relation to growth response of western hemlock fertilization. For Sci 29: 469477  81  Radwan MA, DeBell D S (1989) Effects of different urea fertilizers on soil and trees in a young thinned stand of western hemlock. Soil Sci Soc Am J 53: 941-946 Radwan MA, Shumway J S , DeBell DS, Kraft J M (1991) Variance in relation to pole-size trees and seedlings of Douglas-fir and western hemlock to nitrogen and phosphorus fertilizers. Can J For Res 21: 1431-1438  Rao IM, Abadia J , Terry N (1986) Leaf phosphate status and photosynthesis in vivo: changes in light scattering and chlorophyll fluorescence during photosynthetic induction in sugar beet leaves. Plant Sci 44: 133-137  Rao IM, Fredeen AL, Terry N (1990) Leaf phosphate status, photosynthesis, and carbon partitioning in sugar beet. Plant Physiol 92: 29-36  Rausch C, Bucher M (2002) Molecular mechanisms of phosphate transport in plants. Planta 216: 23-37 Reich PB, Schoettle A W (1988) Role of phosphorus and nitrogen in photosynthetic and whole plant carbon gain and nutrient use efficiency in eastern white pine. Oecologia 77: 25-33  Robertson I, Switsur VR, Carter AHC, Barker A C , Waterhouse J S , Briffa KR, Jones PD (1997) Signal strength and climate relationships in the C / C ratios of tree-ring cellulose from oak in east England. J Geophys Res 102:19507-19516 1 3  1 2  Sai J , Johnson C H (2002) Dark-stimulated calcium ion fluxes in the chloroplast stroma and cytosol. Plant Cell 14: 1279-1291  Sarkanen KV, Ludwig C H (1971) Lignins: Occurrence, Formation, Structure and Reactions. Wiley, New York  Schleser G H (1992) 8 C pattern in a forest tree as an indicator of carbon transfer in trees. Ecology. 73: 1922-1925 13  82  Shumway J , Olson J (1992) Stand selection criteria for nitrogen fertilization: current practices and future needs. In: Chappel HN, Weetman GF, Miller R E (eds) Forest Fertilization: Sustaining and Improving Nutrition and Growth of Western Forests Res Contr No 73, Uni Wash Col For Res Seattle, pp 162-167  Sun ZJ, Livingston NJ, Guy RD, Ethier G J (1996) Stable isotopes as indicators of increased water use efficiency and productivity in white spruce seedlings. Plant Cell Environ 19: 887-894  Sun OJ, Payn TW (2002) Magnesium nutrition and photosynthesis in Pinus radiata: clonal variation and influence of potassium. Tree Physiol 19: 535-540 Timell T E (1986) Compression wood in gymnosperms. Springer, Berlin Tokyo Turner J , Lambert MJ, Gessel S P (1977) Use of foliage sulfate concentrations to predict response to urea application by Douglas-fir. C a n J For Res 7: 476-480 Turner J , Lambert M J , Gessel S P (1979) Sulfur requirement of nitrogen fertilized Douglas-fir. For Sci 25: 461-467  Versaw WK, Harrison M J (2002) A chloroplast phosphate transporter, PHT2.1, influences allocation of phosphate within the plant and phosphatestarvation responses. Plant Cell 14: 1751-1766  Walcroft AS, Silvester WB, Whitehead D, Kelliher FM (1997) Seasonal changes in stable carbon isotopes ratios within annual rings of Pinus radiata reflect environmental regulation of growth processes. Aust J Plant Physiol 24: 57-68  Warren CR, Adams MA (2000) Water availability and branch length determine 5 C in foliage of Pinus pinaster. Tree Physiol 20: 637-643 1 3  Warren CR, McGrath JF, Adams MA (2001) Water availability and carbon isotope discrimination in conifers. Oecologia 127: 476-486  83  Warren CR, Adams MA (2002) Phosphorus affects growth and partitioning of nitrogen to Rubisco in Pinus pinaster. Tree Physiol 22: 11-19  Westing A H (1965) Formation and function of compression wood in gymnosperms. Bot Rev 31: 381-480  Westing A H (1968) Formation and function of compression wood in gymnosperms II. Bot Rev 43: 51-78  White JB, Silim S, Weetman G F (1999) A comprehensive study of western hemlock nutrition. Report Prepared for the Sciences Council of British Columbia, pp 103  White J B (2000) Studies of Western Hemlock Nutrition. PhD thesis. Department of Forestry, U B C Vancouver, B C  Xu ZH, Saffigna P G , Farquhar GD, Simpson JA, Haines RJ, Walker S, Osborne DO, Guinto D (2000) Carbon isotope discrimination and oxygen isotope composition in clones of the F(1) hybrid between slash pine and Caribbean pine in relation to tree growth, water-use efficiency and foliar nutrient concentration. Tree Physiol 20: 1209-1217  Yin X (1998) The albedo of vegetated land surfaces: systems analysis and mathematical modelling. Theor Appl Climatol 60: 121-140  84  APPENDIX 1. Lignin/cellulose ratios of disks removed from two western hemlock saplings (A and B) through zones of extensive reaction wood deposition. Disk position 1 was at the bottom of the stem, whereas position 5 was from above the zone of curvature. Maximum curvature was at position 2 and 3 (Fig 2.1). Reaction wood was essentially absent from disks collected at position 5. Each bar represents the mean (±SE) of four separately extracted and analyzed quarters from each disk.  0.7  85  APPENDIX 2. Summer dryness index (SDI) calculated from climate parameters for respective sites. Mean annual temperature (MAT), mean warmest month temperature (MWMT), mean coldest month temperature (MCMT), mean annual precipitation in mm (MAP) and mean summer precipitation in mm (MSP) were modelled for each site using PRISM model.  SITE  MAT  MWMT  MCMT  MAP  MSP  SDI  PA1  8.3  14.92  2.23  3146.2  PA2  547.5  0.31  8.38  14.98  2.34  3146.2  NIMP  547.5  8.22  14.4  0.31  3.5  3995.9  922.8  PN1  0.18  8.02  14.53  2.58  PN2  8.02  2543.5  512.3  0.33  14.53  EVE  2.58  2543.5  512.3  0.33  7.96  15.19  1.24  1899.4  ZEB  450  0.39  8.55  15.83  2.56  4891.8  SEC  932.6  0.19  4.6  13.39  -4.38  1314.7  218.9  0.7  APPENDIX 3. Means of total precipitation (mm) received in the growing season (MaySeptember) in respective years. Data were taken from Port Alice climate station to calculate means and summer dryness index.  Year  Total ppt.  SDI  1992  432.1  0.45  1993  444.1  0.44  1994  555.1  0.35  1995  339.3  0.55  1996  480  0.38  1997  950.6  0.21  86  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0075073/manifest

Comment

Related Items