Across the world, Indigenous people manage more than 80 per cent of its vital ecosystems and threatened species. In Australia, about four million square kilometres of combined land and sea Country are Indigenous-titled lands. That's over half of Australia's total land area.
At the same time, Indigenous people face many challenges in land management. Climate change, for one, is driving up species extinctions and ecosystem losses. Such threats are also increasing in scale and urgency.
In the face of complex environmental management problems, Indigenous people in Australia have looked to enlist the help of innovative technologies such as drones and artificial intelligence (AI).
But it's critical to first ensure that these technologies can be paired with Indigenous knowledge (IK) and governance.
AI for Healthy Country
Over the past four years, the Healthy Country AI program has been empowering Indigenous people to respond to environmental challenges. Co-developed with Traditional Owners, the program is a world-first and award-recognised. Its overall aim is to survey species and habitats of cultural and ecological significance to Indigenous people. By mixing ethical AI and modern science with Indigenous knowledge, it's delivering practical solutions for conserving precious ecosystems on Indigenous lands.
Over the program's lifetime, its list of projects have expanded, and so has its long list of list of collaborators.
Today, Healthy Country AI is proudly tackling a range of complex environmental issues. These solutions are being delivered for Indigenous rangers and on-country enterprises.
Though it continues to expand, it has held fast to its key principles around ethical data collection and data analysis. This ensures that the program's efforts:
- are governed by Traditional Owners,
- reflect the priority areas of concern for local Indigenous communities, and
- support on-ground adaptive management efforts.
Measuring impact and monitoring recovery
One of the program's many highlights includes work with Bininj Traditional Owners in Kakadu National Park. The Traditional Owners have worked with CSIRO, Microsoft and scientists to use data collected from Healthy Country AI. The initiative has been supported by the Australian government's National Environmental Science Program (NESP).
Over 2018-19, Bininj Traditional Owners used the data to guide effective weed management efforts in Kakadu’s floodplains. The resident magpie geese populations were threatened by invasive para grass, which limits their ability to create nesting sites. The infestation of para grass was choking the floodplains across a huge area.
Using the data produced by ethical drone monitoring, the rangers were able to get accurate estimates of magpie geese populations, and of para grass sites. It meant that rangers could check how effective their weed management was. Within just nine months, the count of magpie geese in one wetland jumped from 50 to 1,800.
A couple of years later, in Cape York, Wik Elders and Aak Puul Ngantam (APN) Cape York rangers worked with CSIRO, Microsoft and other scientists, funded by NESP, to adapt Healthy Country AI to protect turtle nests in the area. They were able to efficiently analyse tens of thousands of helicopter and drone images to monitor turtle nesting sites and target management of feral pig populations in the area which led to pig predation on turtle nests in key areas falling by up to 90 per cent.
Supporting decision-making from above
Healthy Country AI has also been harnessed in the recently announced SpaceCows project. It's tapping into data collected from satellites and animal tags to manage large herds using AI (and space technology). And along the way, it's helping to create evidence-backed ‘best practice’ for similar projects.
The satellite-linked ear tags generate animal movement data. These data are used to create distribution patterns, and are then fed into a models, where they are combined with other environmental data. The models provide decision-support tools used to support efficient mustering of feral herds that also account for environmental, cultural and economic values.
Supported by the Smart Farming Partnership, the SpaceCows project involves CSIRO, Microsoft the North Australian Indigenous Land and Sea Management Alliance (NAILSMA), satellite IoT company Kinéis, James Cook University, Charles Darwin University, Mimal Land Management Aboriginal Corporation, APN Rangers, Normanby Land Management.
Empowering Indigenous rangers to use Healthy Country AI tools on country
Whether it's protecting turtle nests, combating invasive weeds, or tracking herd populations, co-design is at the core of Healthy Country AI. Tackling such complex environmental challenges critically requires close collaboration with Indigenous leaders.
Now, a newly announced $2.6 million dollar boost will help take the Healthy Country AI project to the next level. The funding will provide on-ground training for local Indigenous communities and land managers. This training will enable Indigenous rangers to drive and develop AI and digital tools themselves. These are the same tools that will support evidence-based decision-making on country. The Indigenous-led program involves NAILSMA and CSIRO. And it's in collaboration with The Telstra Foundation, Microsoft, NESP and the Women in STEM and Entrepreneurship program.
Not only will the partnership help advance the important work that Indigenous rangers are already doing - it's helping to fill an important piece of the narrative. It's recognition that providing Indigenous rangers with the skills to design, drive and use AI and digital support tools is a critical component in supporting on-ground action that responds to local healthy country priorities. These digital skills can also offer Indigenous people valuable job and on-country enterprise opportunities.
The Healthy Country AI program, partnership and impact continues to grow. At the same time, it remains focused on creating pathways for digital inclusion for Indigenous rangers. These pathways will ultimately inform their adaptive management of country.