McGaughran, AngelaHale, Phoenix2026-05-142026-05-142026https://hdl.handle.net/10289/18278Biodiversity loss is more prominent in freshwater versus terrestrial or marine ecosystems, with temperate estuaries believed to be the most degraded of all ecosystems globally. This is due, in part, to the proliferation of invasive non-native species and the inadequacies of current biomonitoring practices for monitoring population trends over sufficient spatiotemporal scales to inform management decisions. Environmental DNA (eDNA; genetic material shed to the surrounding environment by organisms) is an increasingly popular, non-invasive biomonitoring method that provides the opportunity to increase the scale at which population trends can be monitored, while also having the potential as a sampling method for population genetic studies. My thesis aimed to use a variety of methods to understand source-sink population dynamics at varying spatial and temporal scales and explore the value genetic techniques can provide to species management, using the invasive pest fish, Gambuis affinis, as a test case. Chapter 2 used traditional netting techniques to survey occupancy patterns of G. affinis to understand the species’ physical dispersal drivers within a large coastal ecosystem in the South Island of New Zealand. I demonstrated a decrease in abundance and likelihood of G. affinis presence as tide height increased, and a decrease in abundance of G. affinis at a specific site as rainfall increased. These findings show that monitoring of occupancy patterns of freshwater fish is often done at too small a spatiotemporal scale and that future work should incorporate additional methods to increase the scale at which monitoring is performed. Chapter 3 analysed mitochondrial DNA collected across New Zealand from both tissue and water samples to explore the efficacy of eDNA as a sampling tool for determining mitochondrial variation at large spatiotemporal scales. I showed that >99% of the G. affinis mitochondrial assemblage was shared by both tissue and water samples, but the remaining detected genetic variation was unique to either tissue or water only. These findings show that, while that eDNA holds great promise for biomonitoring in freshwater environments, careful consideration of study design is required for its use in assessing the drivers of biodiversity differentiation. Together, my thesis demonstrates the utility of eDNA metabarcoding techniques as a complement to traditional survey methods to increase the spatiotemporal scale at which biomonitoring is undertaken. Continued application of a combination of these methods will facilitate improved knowledge of occupancy and dispersal patterns for G. affinis and, as such, greater protection of our native ecosystems and species.enAll items in Research Commons are provided for private study and research purposes and are protected by copyright with all rights reserved unless otherwise indicated.invasivegeneticsecologyeDNApest speciesmetapopulationsource-sinkThesis