The unique canopy structure, leaf morphology, and physiology of Cornus drummondii

Abstract

Dense canopies are a key characteristic of clonal shrubs that enables their encroachment of mesic grasslands. These dense shrub canopies displace shade-intolerant grasses, resulting in reduced fire intensity and a gradual grassland-to-woodland transition. While the importance of dense canopies to clonal woody encroaching shrubs is well documented, the structure of their canopies and the mechanisms enabling clonal shrubs to facilitate dense canopies are not yet understood. To fill this knowledge gap, I investigated the canopy structure of Cornus drummondii (chapter 2) and the growth investment strategy enabling C. drummondii to facilitate dense canopies (chapter 3). In chapter 2, I measured the vertical distribution of leaves and light transmission in canopies of C. drummondii and their response to grazing and simulated browsing. In doing this, I also assessed the accuracy of two indirect methods of measuring leaf area index (LAI; the one-sided area of leaves per ground area). My results indicated that unbrowsed C. drummondii canopies had a mean LAI of ~8, exceeding the LAI of most temperate deciduous forests, and distributed half of their total LAI within a single, vertical 50 cm canopy section. Canopy sections with greater leaf density had lower light extinction rates compared to less dense sections. The evaluation of LAI in C. drummondii canopies with indirect methods revealed that an AccuPAR LP-80 ceptometer could accurately predict LAI in unbrowsed canopies, despite their high densities. However, the ceptometer overestimated the total LAI of browsed canopies by 46%. An Einscan Pro 2X Plus 3D handheld scanner had high precision at estimating the leaf area of individual ramets but became less accurate as leaf area increased. These results indicate that indirect LAI measurements can predict LAI in C. drummondii canopies despite the density of these canopies and varying rates of light extinction. In chapter 3, I investigated the vertical distribution of leaf traits and physiology in relation to light availability across canopies of C. drummondii and the impact of simulated browsing and grazing. My results revealed that leaf mass per area (LMA) and leaf nitrogen per area (Na) varied ~3-fold across canopies, resulting in major differences in leaf physiological functioning. High LMA leaves had high photosynthetic capacity, while low LMA leaves used a novel strategy for maintaining light compensation points below ambient light levels. In response to browsing, C. drummondii modified its vertical allocation of leaf traits by increasing LMA and Na at lower canopy depths, leading to a greater photosynthetic capacity deeper in browsed canopies compared to control canopies. This response, along with greater light availability in browsed canopies, resulted in greater photosynthetic rates and resource-use efficiency deeper in browsed canopies compared to control canopies. My results suggest that the high LAI canopies of C. drummondii and its compensatory growth response to browsing are driven by the capacity of C. drummondii to dramatically alter leaf traits in response to light gradients—both spatially to achieve dense canopies and temporally to achieve compensatory growth. Together, these two studies provide a better understanding of the dense canopy structure of C. drummondii and the morphological and physiological mechanisms enabling C. drummondii to facilitate dense canopies and respond to grassland disturbance by browsing, both of which are key factors contributing to the successful encroachment of grasslands by C. drummondii and other woody encroaching species.