The amazing diversity of Poaceae: trait variation across space, time, and lineage

Abstract

The grass family Poaceae is one of the most successful plant families on Earth. Poaceae is comprised of over 11,500 species, making it the fifth-largest plant family in current existence. It is also one of the most dominant – grasslands and savannas are found on every continent except Antarctica and cover over a quarter of the planet’s terrestrial surface, impacting biogeochemical cycles on a global scale. Finally, grasses play fundamental roles in the development of human civilizations (forage, fiber, and fuels), ecosystem regulation of the water cycle, and regulation of biodiversity within plant and animal communities. Given the global significance of grass species, it is important to understand the mechanisms of success within this plant family. Plant traits are typically used as ecological tools for assessing species adaptations to varying environments. Species that have a range of traits (morphological and physiological) are more likely to persist in environments that have complex spatial and temporal variability. Trait diversity has facilitated the great success of grasses. How can the traits of this vastly diverse group of plants be used to make predictions of grassland change in the future? In Chapter 2, I explored three methods of organizing grass species to better understand how trait diversity varies among photosynthetic pathway (C₃ or C₄), life history (annual or perennial), or evolutionary history. I accomplished this task by measuring 11 structural and physiological traits of 75 naturally-occurring species of grass on the Konza Prairie Biological Station (northeastern Kansas, USA). My results show that the traits of grasses are best represented by their evolutionary history. Photosynthetic pathway only revealed significant differences among physiological traits while structural traits varied by life history. Evolutionary history, on the other hand, was found to significantly explain differences found in both structural and physiological traits when species were grouped by either Tribe or C₄ lineage. These findings indicate that models utilizing photosynthetic pathway to group grasses, a commonly used practice, likely minimize the existing variability and oversimplify landscape predictions of grassland change. In Chapter 3, I examined how grass traits may vary temporally and spatially using two Panicoid species, Dichanthelium oligosanthes subsp. scribnerianum (C₃) and Panicum virgatum (C₄). To assess temporal variability, I measured leaf stomatal and isotopic/elemental composition traits from specimens dating back to 1887 at the Kansas State University Herbarium (KSC) and the McGregor Herbarium at the University of Kansas (KANU). To assess spatial variability, I measured a suite of traits from eight different grasslands across the Great Plains of North America representing six unique ecoregions. While differences in traits were found across space and time for both species, my results show that Δ13C has been increasing in Dichanthelium oligosanthes and decreasing in Panicum virgatum over time, illustrating differential responses to water stress for these species. Results from both chapters demonstrate that there is substantial inter- and intraspecific trait variation in Poaceae. My research suggests that incorporating aspects of evolutionary history, as well as spatial and temporal trait variability, better characterizes the natural variability in grass traits in the Great Plains and allows greater mechanistic insight into how these species respond to climate variability.