ºìÐÓÊÓƵ

Topic

Urchin distribution modelling and grazer control for kelp restoration

Department of Biology

Date & location

• Friday, September 5, 2025
• 11:00 A.M.
• Clearihue Building, Room B007

Examining Committee

Supervisory Committee

• Dr. Julia Baum, Department of Biology, University of Victoria (Supervisor)
• Dr. Francis Juanes, Department of Biology, UVic (Member)
• Dr. Jasmin Schuster, Program Manager, Kelp Rescue Initiative (Outside Member)

External Examiner

• Dr. Daniel Okamoto, Department of Integrative Biology, University of California, Berkeley

Chair of Oral Examination

• Dr. Jason Fisher, School of Environmental Studies, UVic

Abstract

Kelp, brown algae in the Order Laminariales, serve as important biogenic coastal habitat for a multitude of species. Losses of the highly productive kelp forest ecosystems these species form has been driven by climate change and the expansion of sea urchins, the dominant herbivore in kelp forests, leading to large swaths of kelp forests transitioning to urchin barrens from overgrazing. Yet, kelp forest restoration is hindered by the limited understanding of where urchin grazing pressures are highest and the extent to which urchin grazing prevents outplanted kelp from establishing. I address this gap first by developing predictive species distribution models (SDMs) through ensemble approaches that combine parametric and machine learning methods using long-term spatial datasets for three common urchin species in British Columbia (B.C.). Performance of these models, which identify environmental factors influencing urchin occurrence and abundance and generate novel spatial predictions of grazing intensity across B.C., varied with urchin response type and species. While occurrence models performed well (AUC = 0.82-0.87), abundance models showed lower predictive power (R² = 0.41-0.42, F1 = 0.28-0.46). Regardless of species or model type, dissolved oxygen and temperature variables consistently emerged as the most influential environmental predictors. These SDMs effectively captured species-specific spatial patterns aligned with known ecological traits, helping to disentangle urchin spatial patterns and the abiotic factors shaping them. Next, I conducted an in-situ field experiment in Barkley Sound (B.C.), in collaboration with the Kelp Rescue Initiative, testing how urchin exclusion and green gravel outplanting performed across a gradient of urchin densities. I found that while physical exclusion reduced urchins within plots where kelp was outplanted, kelp survival and growth remained generally low, indicating potential limitations of exclusion design or weak kelp response under high herbivory pressures. Larger gravel substrates improved kelp survival but did not affect growth. Overall, sites with lower urchin densities showed better kelp survival and growth. These results suggest that exclusion devices alone may be insufficient in areas with intense grazing. Through this integrated modelling and experimental approach, this thesis delivers new tools and evidence-based insight to guide targeted and effective kelp restoration by identifying the drivers of grazing patterns and evaluating restoration techniques. More broadly, this work supports ecosystem-based management, species recovery, and climate resilience planning in Canada’s marine ecosystems.