AI-assisted ocean intelligence for artisanal fisheries — powered by satellite oceanography and Copernicus Marine data. Operationally validated along the Peruvian coast.
Across the Humboldt Current System, small-scale fishing fleets face high search uncertainty, rising fuel costs and limited access to satellite-derived oceanographic intelligence.
Decisions on where and when to fish rely largely on local experience without scalable, satellite-informed decision support.
Longer trips, increased fuel consumption and underused capacity reduce the economic viability of artisanal fisheries.
Copernicus, NOAA and NASA products contain rich oceanographic signals that rarely reach the artisanal fleet in actionable form.
PredictaMAR is an early-stage ocean intelligence platform that combines satellite constellations, Copernicus Marine Service data and AI-assisted scoring to support artisanal fishing decisions in the Humboldt Current System.
Optimized processing of satellite constellations, integrating Copernicus, NOAA and NASA datasets.
A satellite-based, weighted and species-specific scoring system for artisanal fishing zone prediction.
Validated through pilot deployments with artisanal fleets in Callao, Pucusana and Ancón along the Peruvian coast.
A field-tested methodology under active development, focused on transparent science and scalable deployment across artisanal fisheries.
Mesoscale oceanographic analysis at 1 km resolution to characterize coastal dynamics relevant to artisanal fishing.
Copernicus Marine Service data integrated with NOAA and NASA products through an optimized multi-satellite pipeline.
Currently supporting 10 target species in pilot operational workflows through a weighted scoring system.
First field validation achieved with less than 20 meters of positional error under pilot conditions along the Peruvian coast.
Methodologies and results documented through peer-reviewed venues and open repositories such as Zenodo.
Optimized cloud workflows for a deployable operational prototype designed for artisanal fleets.
Access the operational prototype of PredictaMAR — a cloud-hosted AI-assisted ocean intelligence interface for artisanal fishing zone exploration.
PredictaMAR is in active pilot operations along the Peruvian coast, with measurable scientific and commercial traction.
Figures describe the current early-stage operational pilot of PredictaMAR along the Peruvian coast (Callao, Pucusana, Ancón). The platform is under active development.
A founding team combining ocean intelligence, systems engineering and marine science, with active collaborations across European and Latin American institutions.
Ocean intelligence architect. Leads satellite pipeline development, MacroScore methodology and strategic partnerships with INESC TEC Porto, CTN Spain and IMARPE Peru.
iD0009-0000-6557-5321Systems engineering and operational infrastructure. Leads deployment architecture, data pipeline automation and integration with Google Earth Engine and Copernicus APIs.
iD0000-0002-4004-1793Marine science and field operations. Leads validation methodology, artisanal community engagement and data quality assurance across pilot sites in Callao, Pucusana and Ancón.
iD0009-0004-5997-0551PredictaMAR is being validated through international scientific venues, peer-reviewed journals, open repositories and active field engagement with artisanal communities.
Top recognition at the Universidad Nacional de Ingeniería startup competition.
Accepted for oral presentation at the European Geosciences Union General Assembly 2026.
Scientific paper currently under review in a Q1 peer-reviewed journal.
Built on operational Copernicus oceanographic products.
Pilot deployments and validation in Callao, Pucusana and Ancón.
Working paper deposited in Zenodo with persistent DOI.
PredictaMAR's methodology is documented through international conferences, peer-reviewed journals and open repositories with persistent identifiers.
We collaborate with research institutions, public agencies, NGOs and partners working on artisanal fisheries and ocean sustainability.