BUILDING FRAMEWORKS FOR UNDERSTANDING INVASIONS AND EXTINCTIONS FOR BIODIVERSITY SCIENCE

Abstract

Scientists have long been interested in mechanisms that increase and decrease biodiversity on Earth and the effects they have on organisms’ interactions and functions. Global biodiversity loss is now outstripping accumulation and far exceeds expected background levels and has drawn comparisons to previous mass extinctions. The ongoing Holocene–Anthropocene extinctions differ from prior biodiversity loss, because humans have been directly implicated as major drivers of current loss—overexploitation, habitat modification and destruction, climate variability, spread of pathogens, and invasion by exotic species. Biodiversity change and especially loss can alter ecological assemblages irreversibly, which consequently can change the direction and magnitude of key ecosystem functions that organisms, including humans, rely on. Therefore, it is increasingly important to develop methods for quantifying and understanding phenomena linked to biodiversity change. In my dissertation, I develop methods to: (1) assess risk of a global-scale invasion of a pest species, (2) predict host associations for a generalist pest species, and (3) develop models to understand extinction dynamics within a clade of conservation interest. In my first chapter, I developed a framework and used it to assess a rapidly spreading regional U.S. grape pest, the spotted lanternfly planthopper (Lycorma delicatula; SLF), to spread and disrupt the global wine market. I found that SLF invasion potentials are aligned globally because important viticultural regions with suitable environments for SLF also heavily trade with invaded U.S. states. For my second chapter, I estimated host plant associations for SLF with phylogenetic imputation and predicted SLF host associations for the U.S. Many known and predicted high association host species are found in the uninvaded Midwest, Southeast, and West Coast as well as the Mid-Atlantic and Northeast, where SLF is present. Should SLF spread further, these regions are likely to experience impacts to resident trees. For my third chapter, I proposed a method for detecting three non-random extinction models and used the imperiled Caribbean lizard genus Leiocephalus as a case study to test it. Past extinctions showed directional loss of larger Leiocephalus species. However, future predicted extinctions are random for body size but show stabilizing extinction of species with either smaller or larger limb and tail lengths. Shifting extinction for Leiocephalus may occur because of changing pressures that now include anthropogenic habitat loss. Altogether, these studies attest to the value of developing and evaluating approaches to describe biodiversity dynamics in the Anthropocene.