Cultural Momentum
A 3.5-hour reef stewardship experience, co-designed with Indigenous partners. Participants will plant coral, extend the sensor network, and fund the next year of operations.
Restoration that doesn't depend on the next grant.
Most reef restoration programs operate grant cycle to grant cycle. EPOCH is structured to break that pattern. Year 3 is structured to reach operational break-even, the point at which biorefinery products and ecotourism participation cover the full operating cost of the three-site program. From that year forward the system funds itself indefinitely.
Three hours and thirty minutes, from learning, to action, to record.
Cultural education co-designed with Indigenous partners will introduce the ahupuaʻa system, the watershed-to-reef stewardship framework that has governed these waters for centuries. Participants will meet the live reef visualization platform, see the AI-driven robot survey demo, and understand the specific reef they will dive on later that day.
Guided dive or snorkel alongside the BlueROV2 survey platform. Each participant will plant 5 to 20 coral microfragments at GPS-tagged locations. Wrist-mounted data loggers will record water-quality parameters along the participant's depth and route, extending the fixed YSI EXO3 sensor network at zero marginal cost.
Cinematic footage from participant cameras and the robot will be automatically assembled, paired with the participant's measured contribution: GPS-located coral plantings, species observed, and water-quality data captured during their route. Each participant will leave with a personal Contribution Profile, a permanent record of the restoration they performed.
Every guest extends the sensor network.
Cultural Momentum is structured under the dual-purpose principle that governs every part of EPOCH. By design, participant cameras and wrist data loggers generate a continuous stream of georeferenced benthic imagery and water-quality readings, processed through the same EfficientNet-V2-M species classifier and SegFormer-B2 segmentation model the program robots use. The result is high-volume training data that ecotourism revenue pays to collect.
Participant cameras feed the species classifier and segmentation model at zero marginal cost.
Across three sites, launching Month 19, three sites operational by Month 31.
Each fragment is GPS-tagged and incorporated into the long-term ground-truth dataset.
Knowledge held in partnership.
The program is structured around the CARE Principles framework and Two-Eyed Seeing (Etuaptmumk), a methodology that treats Indigenous knowledge as a parallel epistemological system rather than a content layer. The governance model positions Indigenous research partners as co-investigators rather than informants.
Under the model, Indigenous knowledge is encoded directly into the AI systems as spatial priors, into robot survey scheduling, into the computer-vision training taxonomy, and into prediction-model validation. Cultural data sovereignty agreements are established with each partner organization before data collection begins.
Cultural interpretation materials require approval from Indigenous knowledge partners before delivery.
The reef removed becomes the object retained.
Invasive algae harvested from the bioremediation filters and direct-removal operations will enter the modular biorefinery, where it becomes a bio-based FDM filament suitable for 3D printing. The same filament is designed to print structural components used across all four pillars: sensor housings, biorefinery couplings, ecotourism artifacts. The retail program targets ASTM D6400 or EN 13432 certification for genuine compostability. Consumer-product design files are licensed under CC-BY 4.0.