Harnessing leaf optical properties to assess plant traits, genetics and environmental interactions

Motyka Corbin, Jaclyn Parker (2022) Harnessing leaf optical properties to assess plant traits, genetics and environmental interactions. Doctoral thesis, Northern Arizona University.

Text Motyka_Corbin_2022_harnessing_leaf_optical_properties_assess_plant_tra.pdf - Published Version Restricted to Repository staff only until 4 January 2025. Download (3MB) | Request a copy

Abstract

Plant trait expression is at the core of ecosystem functioning. Plants produce sources of bioavailable energy, influence nutrient cycling, modulate energy flow, and impact decomposition which has immense impacts on local, regional, and global processes. As climate change intensifies, identifying which plant traits are adaptive in particular environments is imperative for making effective land management decisions. Global climate patterns are predicted to become more extreme and as a result, plant phenotypes which were once adaptive in certain environments may become maladaptive. Developing high throughput phenotyping strategies which can efficiently and cost-effectively survey these dynamics across geographic scales remains an area of high interest to researchers and land managers alike. Hyperspectral technology is one of the most promising avenues for accomplishing this task. We focus on two foundation species, Fremont cottonwood (Populus fremontii), narrowleaf cottonwood (P. angustifolia) and their hybrid (P. × hinckleyana) which have been shown to support diverse communities and regulate ecosystem processes. These obligate riparian species and their associated communities are under increased threat due to aridification, rising temperatures and anthropogenic water resource use. We collected ground-based hyperspectral leaf reflectance from wild and experimental trees across a steep elevational gradient and integrated molecular, morphological, biochemical, and environmental data to test the limits of using leaf spectra to predict ecologically important traits in the wild and climatically dissimilar areas. We addressed three major topics. First, to understand whether leaf spectra vary between genotypes and populations in different environments, we analyzed samples from wild Fremont cottonwood at home sites as well as reciprocally planted clones from three common gardens planted at low, mid, and high elevation sites. Classification models showed that leaf spectra from the visible, near infrared and shortwave infrared wavelengths (500 nm – 2400 nm) possess sufficient variation to correctly predict genotype, population, and environment identity. Our findings also suggest that leaf spectral phenotypes display plasticity in some spectral regions when introduced to novel environments. Second, we assessed whether leaf reflectance can be used to predict leaf biochemical, morphological and tree architectural traits from reciprocally planted cottonwood clones grown in different environments. We conducted partial least squares analyses to identify which traits and spectral regions are important for predictive modeling. We found that 1) wet leaf reflectance can predict leaf morphological and biochemical traits across common gardens, but not tree architecture, 2) the importance of wavelengths for predictive models varies by trait, and 3) leaf spectra are better predictors of population and environment identity than some traditional leaf traits. Our findings suggest that some (but not all) ecologically important leaf traits may be detectable at the canopy level across climatically different environments. Third, hybrid vigor has been hypothesized to confer a potential advantage for individuals in the face of climate change. As such, monitoring functional trait differences in hybrid systems as well as range expansion or contraction using remote sensing is of particular interest. To determine whether genetic and morphological differences may be detectible between canopies, we surveyed the spectra of fresh leaves from common garden trees of known ancestral identity. We predicted that 1) Fremont and narrowleaf spectra would be distinguishable and 2) leaf spectra could discriminate hybrids from parental species. Partial least squares discriminant analyses using leaf reflectance as a function of Fremont molecular markers as well as tree identity and leaf morphological traits supported these hypotheses. Our findings suggest that leaf reflectance and foliar trait differences between parental and hybrid cottonwood may be detectable at the canopy level and could be tractable for determining whether hybrid swarms are expanding into parental territory or contracting as global temperatures increase. The goal of this dissertation was to assess how leaf optical properties can inform conservation and land management practices as well as the ecology and evolution of plant traits in different environmental scenarios. Our exploration into the various scales and facets of this topic provides a proof-of-concept for integrating cottonwood foliar spectra into a diverse range of applications. Based on our findings, we posit that leaf spectra are useful for exploring gene-by-environment interactions and phenotypic variation at the genotype, population, and interspecies levels across multiple geographic and climatic scales.

Item Type:	Thesis (Doctoral)
Publisher’s Statement:	© Copyright is held by the author. Digital access to this material is made possible by the Cline Library, Northern Arizona University. Further transmission, reproduction or presentation of protected items is prohibited except with permission of the author.
Keywords:	common garden; Fremontcottonwood; Narrow-leaf cottonwood; gene-by-environment interaction; hybrid; hyperspectral reflectance; phenotypic plasticity; Leaf optical properties; Clinate change; Southwest, New
Subjects:	Q Science > QK Botany
NAU Depositing Author Academic Status:	Student
Department/Unit:	Graduate College > Theses and Dissertations College of the Environment, Forestry, and Natural Sciences > Biological Sciences
Date Deposited:	13 Jun 2023 17:02
Last Modified:	13 Jun 2023 17:02
URI:	https://openknowledge.nau.edu/id/eprint/6008

Actions (login required)

IR Staff Record View

Downloads