On the Space Filling Nature of Trees: Clarifying and Validating a Model of Plant Architecture With Laser Scans of Tree Crowns

Abstract

The scaling of mass and energy in the biosphere is most dramatically illustrated in forest ecosystems. Individual plants can span up to 12 orders of magnitude in size as they climb through a canopy. During growth, trees allocate carbon and water dynamically to meet a variety of physiological and ecological constraints via the deployment of leaf area, the history of which is inscribed in branching architecture. Terrestrial Laser Scanning (TLS) provides new opportunities to map out the geometric complexity of branching architecture in tree crowns, and infer the physiological functioning of individuals using allometric theory for the scaling of size and performance. This dissertation is focused on clarifying and validating the West, Brown and Enquist (WBE) model in order to mechanistically predict allometric relationships in tree crowns. We advance WBE by testing its core assumptions for the first time with Terrestrial Laser Scanning (TLS)–a technique for laser imaging large tree crowns. These measurements establish a diverse dataset of the largest range of plant vascular networks analyzed to-date, a key test for scaling theory that purports to describe biological function across an arbitrary range of sizes. This work emerges from a controversial disconnect between the allometric predictions of WBE and various empirical evaluations of those predictions. More specifically, measurements of fine-scale branching plus allometric observations of tree architecture from forest plots have been discordant with WBE. The field is advanced in a movement toward resolving this conflict by i) clarifying the core assumptions of the theory and proposing new empirical approaches to testing them, and ii) demonstrating that geometric patterns broadly match clarified predictions. In particular, the studies contained herein clarify theoretical predictions by consistently emphasizing the driving force of leaf area in plant network development. By proposing novel proxies of leaf area and branch development, we validate core predictions of the theory and outline key deviations, proposing extensions to the WBE model where needed. We extensively validate these results with destructively harvested data, to account for bias in remote sensing techniques which are pervasive in studies of TLS. The most critical gap we address is the role of branch extension and light foraging in affecting the geometry of tree branching networks. We measure light-foraging in small (terminal twigs and branches) and show they adhere to broad theoretical assumptions from WBE, namely space-filling and the preservation of metabolic service volume, which produces the core proportionality in metabolic scaling. I use this mechanism to explain a long-standing inconsistency between theoretical predictions and tree branching data, namely the presence of curvature on log-log plots of power laws. The novel methods and results presented here point the way toward linking tree geometry to physiological functioning (e.g. water use, respiration/photosynthetic rates, growth rates) in order to further the applicability of allometric relationships and broad-spectrum remote sensing to ecological and evolutionary studies of tree architecture.