Longitudinal Variation in Wood Accumulation along the Stem of Populus Grandidentata; Implications for Forest Carbon Monitoring
AuthorChiriboga, April Therese
AdvisorSheppard, Paul R.
Leavitt, Steven W.
MetadataShow full item record
PublisherThe University of Arizona.
RightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
AbstractThe world's forests sequester roughly a quarter of anthropogenic emissions of carbon dioxide and store it in wood. Assessing this carbon sink includes quantifying annual wood production, establishing baselines, and characterizing both long-term trends and inter-annual variability. Direct measures of forest wood production are often based on measures of individual tree growth along the stem, often taken at a single height: basal height (1.3 meters). This assumes that a measurement of wood production at a single height is representative of wood production along the whole stem. In violation of this assumption, it is known that trees do accumulate wood differentially along the stem, and that this longitudinal variability can change from year to year. Few efforts have been made to describe annual longitudinal variability, and quantify the error in estimated annual whole-stem wood production related to assuming that constant wood production along the stem. In the present study, I present a stem analysis of 30 Populus grandidentata to address this. Dendrochronological techniques are used to develop three chronologies: a traditional tree-ring width chronology from basal height, a novel chronology developed from tree rings grown in the crown of the trees, and a specific volume increment chronology calculated from measured annual volume increment data. A novel taper chronology is also presented. In Chapter 2, comparisons are made between the chronologies to explore differences in inter-annual variability, and the suitability for using tree-ring data from basal height as a proxy for annual wood production. Both basal and crown tree-ring width chronologies were strongly correlated with the volume chronology (r = 0.96 and 0.88, respectively), suggesting that the basal chronology is a superior proxy for stem volume. However, a chronology of taper along the stem indicates that the reliability of either chronology to represent specific volume increment (SVI) changes over time, resulting in different common signals, especially in the last decade of this dataset. If accurately capturing the relative year-to-year changes in stem wood volume is desired, stem dissection and development of an SVI chronology is required. In Chapter 3, two models that use tree-ring data to estimate annual wood production are compared to volume measurements from the stem analysis. The two models are a site-specific allometric model of biomass, and a simplified conic model of volume. Additionally the conic model is decomposed into the three dimensions of growth along which variability exists (around the circumference, along the length of the stem, and height) to identify which dimension introduces the most error when no variability in that dimension is assumed. Relative error (RE) analysis and regression analysis show that stem analysis is superior in cases where few trees are used and accurate measures of wood increment are needed. At the population level, the allometric and conic models show different strengths. Allometric models are more accurate than the conic model (RE = -16% and -18%, respectively) and are better for carbon budgets, whereas the conic model was more precise than the allometric model (R² = 0.94 and 0.86, respectively; interquartile range = 24% and 41%, respectively) and maintains inter-annual variability, which is necessary in cross-validation efforts. Decomposition of the conic model supports previous findings that height is the second most important parameter, following diameter at breast height, in models of woody tissue growth. In Chapter 4, basal, crown and specific volume chronologies are compared to eddy covariance estimates of carbon dioxide flux between the forest and the atmosphere, including net ecosystem exchange, gross primary production and ecosystem respiration. At the University of Michigan Biological Station (UMBS), crown-grown tree-ring widths from P. grandidentata individuals are good recorders of the inter-annual variability of net ecosystem production. Coupled with other environmental information from UMBS, these records implicate defoliating insects as a previously under-appreciated modifier of stand level respiration and gross primary production. These histories of ring widths, volume and taper have unique potential to improve our understanding of how carbon is stored in and flows through forests within the terrestrial biosphere. In the face of global change, forests will experience new stressors, and changes in frequency of known stressors, that reduce the ability of trees to store carbon in woody tissues. A diversity of tree-ring-based chronologies can describe the sensitivity of carbon stores to these stressors, improving predictions of how forests respond to environmental changes.
Degree ProgramGraduate College