Dominica Harrison

University of Victoria
Country flagCanada

Research Interests

ApplyKite TwinLearns as you apply

Make your next move.

Prepare your application

Review drafts before you send them.

Explore related searches

Articles (8)

Habitat complexity plays a critical role in coral reef ecosystems by enhancing habitat availability, increasing ecological resilience, and offering coastal protection. Structure-from-motion (SfM) photogrammetry has become a standard approach for quantifying habitat complexity in reef monitoring programs. However, a major bottleneck remains in the two-dimensional (2D) classification of benthic cover in three-dimensional (3D) models, where experts are required to manually annotate individual colonies and identify coral species or taxonomic groups. With recent advances in deep learning and computer vision, automated classification of benthic habitats is possible. While some semi-automated tools exist, they are often limited in scope or do not provide semantic segmentation. In this investigation, we trained a convolutional neural network with the ResNet101 architecture on three years (2015, 2017, and 2019) of human-annotated 2D orthomosaics from Kiritimati, Kiribati. Our model accuracy ranged from 71% to 95%, with an overall accuracy of 84% and a mean intersection of union of 0.82, despite highly imbalanced training data, and it demonstrated successful generalizability when applied to new, untrained 2023 plots. Successful automation depends on training data that captures local ecological variation. As coral monitoring efforts move toward standardized workflows, locally developed models will be key to achieving fully automated, high-resolution classification of benthic communities across diverse reef environments.

Year:

2025

Coral life cycle dynamics are poorly understood in most reefs, especially at the large geographic scales commensurate with ocean transport, genetic flow, and other synoptic scale processes. We present a spatially explicit, large-scale, and multi-temporal study of coral settlement along a 30 km long reef system in the Southwest portion of Hawaiʻi Island. Here, we focused on interannual variability in coral recruitment from 2021 to 2024, a period without a major marine heatwave. We used stratified random site selection to place 320 coral settlement tiles at 32 sites (10 tiles per site) at 10 ± 3 m water depth annually to monitor recruitment of the three most common coral genera found in the region (Montipora, Pocillopora, Porites). Site-level interannual variability in coral recruitment was high yet the overall geographic distribution of recruits was consistent through time. This occurred despite a decrease in benthic temperature and recruitment rates during the study period. Persistent geographic patterns in coral recruitment strengthen our understanding of mechanisms and conditions that drive reef resilience. They also strongly suggest a need to protect areas of high recruitment while studying drivers of low recruitment in contrasting habitats. This approach will further increase support of coral production in an era of climate- and coastal pollution-driven declines in coral reefs.

Year:

2025

Sponges are remarkably active animals with a range of behaviors that are broadly termed sneezes. Little is known about what triggers sneezes or how sneezes propagate in an animal that lacks either muscular or nervous systems. We explored the triggers of sponge behavior in situ in the demosponge Suberites concinnus . The placement of this tennis ball-sized sponge at the focal point of a camera array on an instrument platform in Barkley Sound, BC, Canada, provided a unique opportunity for our investigation. The observatory has a suite of environmental sensors measuring parameters such as photosynthetically active radiation (PAR), oxygen, turbidity, salinity, pressure, current speed, and temperature. An array of 8 cameras designed for 3D imaging was used to capture hourly images. Our findings revealed both long-term (annual) and short-term (daily) behaviors in S. concinnus . The sponge contracted for 2 mo each winter. In February, it gradually expanded and became very active with short-term full-body contractions each lasting 11 h through the summer. A meta-analysis of contraction behaviors in other species of sponge showed that contraction duration is a factor of sponge size and that all sponges contract faster than they expand. Invertebrates and fish were frequently in contact with the sponge, but no single interaction was found to correlate with patterns of sponge behavior. However, significant correlations between sponge behaviors and chlorophyll and turbidity were found using generalized additive mixed models. Our study underscores the dynamic nature of these sessile filter feeders and their responses to their environment.

Year:

2024

High-resolution water quality maps derived from imaging spectroscopy provide valuable insights for environmental monitoring and management, but the processing of all pixels of large datasets is extremely computationally intensive and limits the speed of map production. We demonstrate a superpixel approach to accelerating water quality parameter inversion on such data to considerably reduce time and resource needs. Neighboring pixels were clustered into spectrally similar superpixels, and bio-optical inversions were performed at the superpixel level before a nearest-neighbor interpolation of the results back to pixel resolution. We tested the approach on five example airborne imaging spectroscopy datasets from Hawaiian coastal waters, comparing outputs to pixel-by-pixel inversions for three water quality parameters: suspended particulate matter, chlorophyll-a, and colored dissolved organic matter. We found significant reduction in computational time, ranging from 38 to 2625 times faster processing for superpixel sizes of 50 to 5000 pixels (200 to 20,000 m2). Using 1000 paired output values from each example image, we found minimal reduction in accuracy (as decrease in R2 or increase in RMSE) of the model results when the superpixel size was less than 750 2 m × 2 m resolution pixels. Such results mean that this methodology could reduce the time needed to produce regional- or global-scale maps and thereby allow environmental managers and other stakeholders to more rapidly understand and respond to changing water quality conditions.

Year:

2024

Year:

2024

Year:

2021

Year:

2021

Year:

2021

Collaborators (5)

Gregory P. Asner

Arizona State University

UNITED STATES
View details

Roberta Martin

Associate Professor

Arizona State University

UNITED STATES
View details

Julia Baum

Professor

University of Victoria

CANADA
View details

Warren Gallin

University of Alberta

CANADA
View details

Kelly Hondula

Arizona State University

UNITED STATES
View details