Sea Ventures Marine Response Unit
Monitoring our coral projects
At Sea Ventures MRU, we recognize that monitoring is key to celebrating success and overcoming challenges. We use a variety of monitoring methods and approaches to evaluate quantitative outcomes in our restoration projects that require systematic measurements rather than relative observations or opinions. However, monitoring has been a challenge in coral restoration, as funding limitations often hinder the capacity to assess long-term restoration success indicators such as coral growth, survivorship, and ecological drivers of change (predation, diseases, water quality, mechanical damage, etc.). Given these limitations, in recent years we have harnessed the utility of photomosaics to carry out our coral outplanting monitoring. Our approach harnesses the vanguard technology to evaluate the growth, survival, and other vital metrics of coral outplanting projects.
In recent years we started to integrate photogrammetry, visual computing, and GIS tools to deliver clear and detailed products to our stakeholders.​
3D Models
Tracking coral bleaching recovery using 3D Models
Monitoring is key to gathering evidence of the resilience of corals. In coral restoration this is a key trait that managers seek in the corals they propagate. Using monitoring techniques and local expert knowledge we have been able to record the recovery process of century-old Orbicella faveolata colonies after the [worst in-record] 2023 bleaching event in Puerto Rico.
October 2023 December 2023 January 2024
Coral microfragments monitoring
In addition to acroporids, we monitor the growth and survivorship of massive coral species propagated via micro-fragmentation in collaboration with ISER Caribe’s Centro de Investigación y Restauración de Organismos Marinos (CIROM).
Machine Learning
Extracting data from orthomosaics can be a time-consuming process in the lab. We are harnessing machine-learning algorithms to achieve automated Acropora palmata segmentation for fast and accurate data extraction.
Want to track all outplants in restored areas to obtain robust demographic performance data? The conspicuous pixel signature of A. palmata provides an ideal scenario to train AI tools and achieve similar results compared to human manual segmentation, exponentially reducing the time and effort to extract Acropora data from benthic imagery.
One goal of coral restoration might be to increase the structural complexity of a coral reef area. Using photogrammetry products such as 3D models, Digital Elevation Models (DEMs), and GIS processing we have been able to measure changes in structural complexity indicators within the short-term context of rubble stabilization.
