New approaches for understanding plant invasions
Hovick, Stephen Matthew
MetadataShow full item record
Despite the research attention invasion biology has received over the past half-century, the field is in many ways still searching for general principles. This dissertation attempts to fill some of that gap by using new approaches to address invasive species dynamics and traits that may be associated with invasiveness in plants. In my first study, I compared colonizer suppression by the wetland invasive purple loosestrife (Lythrum salicaria) with suppression by the native dominant broad-leaved cattail (Typha latifolia). Loosestrife suppressed colonizers significantly more than cattail based on nearly every measure of colonizer success. In my second study, I tested the hypothesis that an invasive species (loosestrife) should have a stronger response to enemy release in nutrient-enriched conditions than a dominant native species (cattail). The enemies in this study were specialist insect herbivores on each species and included Galerucella calmariensis, a biocontrol agent widely used to control loosestrife populations in North America. Contrary to expectations, herbivores suppressed loosestrife and cattail biomass equally regardless of nutrient enrichment. However, despite this proportionally similar response, when nitrogen was abundant loosestrife produced as much biomass in the presence of herbivory as cattail did when protected from insect herbivores. This suggests that when nutrients are plentiful biocontrol may not depress loosestrife biomass enough to permit increases in native species abundance. The findings from my third study were consistent with this result. I conducted surveys of 46 wetlands that had releases of G. calmariensis to assess whether biocontrol success on loosestrife was dependent on nutrient availability. The likelihood of biocontrol success increased with increasing biocontrol efforts, but this relationship only held in low-nitrogen sites. In my fourth study I expanded the scope of my research, using plant traits from 20 wetland species to quantify and predict invasiveness. I aligned species along an axis of invasiveness based on trait data and then tested that axis by regressing it against invasiveness scores derived from comprehensive invasive species compendia. Trait-based invasiveness scores explained nearly 40% of the variance in literature-based invasiveness scores, suggesting this method has promise as a screening protocol for assessing invasive potential in plants prior to their introduction.