Blending Indigenous Knowledge and artificial intelligence to enable adaptation

In 2021, PolArctic completed a pilot project in Sanikiluaq, an Inuit community in Nunavut, Canada to develop the area’s potential for commercial inshore fishing. As LESLIE CANAVERA explains, the resulting artificial intelligence (AI) model integrated traditional Indigenous Knowledge, scientific data and remote sensing techniques to locate previously undiscovered fishing locations.

This innovative project was the first to combine Indigenous Knowledge, satellite data, scientific research, and AI to pioneer a response to a problem caused by the climate crisis—in this case, changing climate patterns that affect the availability of fish. Together, we were able to create a new framework for how Arctic Indigenous communities can create sustainable, adaptable mariculture (marine farming) solutions. The project was jointly funded by the Nunavut Fishery Association, Qikiqtaaluk Corporation and WWF.

Adaptation requires knowledge

Knowing about and being able to predict the changes in environmental conditions that affect habitat growth and harvest are critical to our ability to cultivate and farm the ocean. These factors are often site-specific and change with the environment. Although AI is an incredibly powerful tool, it is limited by the data that inform it. This gap is especially relevant in areas like the Arctic, where satellite data is often limited by cloud cover and extended darkness.

Our project explored various modelling architectures and data sources that could overcome these limitations and highlighted the value of including and valuing traditional knowledge from Indigenous People, who have a deep understanding of their land and ocean. PolArctic was able to solve the most pressing data issues by conducting interviews and research and by consulting reports, local maps and the Nunavut Coastal Resource Inventory (NCRI). The project was also supported by C-CORE, a Canadian innovation organization that develops technology solutions.

Merging two knowledge systems

The resulting AI developments identified areas of the sea near Sanikiluaq, Nunavut, that were likely to contain higher concentrations of scallops, clams and kelp, providing a boon to the local mariculture industry. Most significantly, the project was the first AI model of its kind to treat Indigenous Knowledge and western science as equals, training with and validating both knowledge systems. Merging these two knowledge bases provided tangible benefits to the local mariculture industry in its fight to adapt to the worsening effects of climate change.

PolArctic used NCRI information from Sanikiluaq, with permission from the community, to digitize geospatially referenced maps as a data layer. Historical trawl sampling and remote satellite sources were then added. This approach eliminated both the risk of travel during the COVID-19 pandemic and the need for expensive hardware.

 

Expanding the benefits

Future projects could focus on continuing to identify, define and refine biophysical factors to integrate into the AI framework—such as shellfish growth rates, protein yields, nutrient requirements and optimized kelp farm placements. The resulting knowledge will help local and regional mariculture operations to adapt and grow sustainably in the face of climate change.

Actively revitalizing the traditional technologies connected to agriculture, mariculture and natural resource management are key to addressing climate change in the Arctic. These technologies can reduce greenhouse gas emissions by shortening the distances travelled to gather foods while also providing economic benefits (like food security) and social benefits (like preserving traditional knowledge of hunting and gathering).

AI and Indigenous culture are often positioned as if in conflict with each other, but this project achieved a level of success that would not have been possible without the benefit of both. It is a testament to how merging human knowledge and advanced AI techniques can begin to address some of the most pressing issues facing our people.

Given that Indigenous Knowledge spans the Arctic, this programme could be replicated across Nunavut and beyond as new Indigenous data sources are identified and ethically incorporated.