We don't track the impact of mines in half the world

Modern sustainable societies rely on mining for the supply of critical materials. Batteries and electric vehicles require metals such as lithium and cadmium, wind turbines require rare earth metals, and our cities require aggregates such as gravel and sand for construction.

On the other hand, mining also produces materials such as coal that are detrimental to a low-carbon future.

Tens to hundreds of thousands of landscapes around the world are changing as industries evolve to meet changing demands for raw materials. But can we measure this impact and confidently weigh the pros and cons of mining activities?

As the mining industry remains highly controversial, independent research can play an important role in measuring the human and environmental impacts of mining activities and tracking progress as the industry evolves. Masu.

view from above

Because mines are often easily spotted on satellite images, a growing body of research is looking at the impacts of mining from what can be seen from above.

For example, in 2019 we began examining satellite images of Indonesia's nickel-mining region (one of the world's leading producers of nickel needed for battery technology) and discovered vast areas that were highly affected (above). (see image).

The view is dramatic from space, but we wanted to know more about what these images could tell us about what's happening on the ground.

The site shows a mix of formal and informal (or illegal) mining and evidence of extensive land clearing in areas of high biodiversity. However, some companies operating in this sector do not produce public reports.

Some companies produce reports that include little production or impact data. This made us wonder how much of the world's mining impacts are actually measured.

To answer this, we have spent the past five years manually scanning global satellite imagery to visually identify and delineate mine areas. We have identified approximately 120,000 square kilometers of tailings dams, pits, waste rock dumps, and other features around the world.

These features occupy all kinds of ecosystems and landscapes and take many different forms.

Where is the data?

Our latest research is NatureNow let's compare the satellite imagery to the S&P Capital IQ Pro database. The S&P Capital IQ Pro database is a global mining database used in nearly all global mining impact studies.

Surprisingly, we found that more than half of the regions identified in satellite images do not have corresponding production information in the database.

This lack of data may be due to companies not reporting anything (a large portion of the world's mining activity may be illegal), or abandoned sites with no companies to report. There may be many.

However, it is also clear that a digital divide exists even in countries with strict mining regulations, such as Australia. Approximately 3600 square kilometers, or 39 per cent of Australia's visible mined area, has no production data in the S&P database. This shows that the extensive work of collecting reports produced by mining companies is not yet finished.

As researchers, the task of collecting data on our own is overwhelming, so we often have to rely on these incomplete databases. It can take hours to find information about a single mine.

More transparency needed

We have a huge challenge ahead of us: not only collecting and interpreting the information generated by thousands of mining companies around the world, but also ensuring that we improve transparency in our industry.

At a minimum, knowing what is being produced locally will help you understand the scale of the impact, but ideally, knowing the local water and soil quality, the amount of waste being produced, The idea is to have companies directly report on local biodiversity and their impact on the region. community.

So far, global studies of mining impacts appear to be affected by routine undermeasurements or, at best, rough assumptions. This poses major problems for governments, NGOs, researchers, mining investors, and mining-affected communities, who have a collective task of managing, tolerating, and regulating impacts.

Without transparent data, we cannot have rational conversations about the risks we are taking. In fact, we cannot even do justice to projects that have minimal or diminishing impacts.

Can AI fill that gap?

A shift towards industry-wide, global transparency is essential, but perhaps too late to reduce risk. Therefore, the research community strives to create reliable alternatives to industry-reported data.

We believe remote sensing and machine learning (AI computer vision) are key.

Because mine features are highly variable in shape and appearance (despite each site having associated features), manual, human-driven methods have been the most successful to date .

However, through this manual process, we have now accumulated a large amount of data that can be used to train an AI to intelligently recognize mine areas and potentially track growth within the terrain in real time. .

Despite the enormous scale of the mining industry, the huge profits it generates, and the often irreversible impact it has on landscapes, communities and the critical raw materials it produces, an independent organization that tracks its impact around the world Only a handful of analysts have done so.

Relatively modest investments in funding more independent research could go a long way in improving outcomes for those affected by mining. In the pursuit of a low-carbon future, many lives and livelihoods may depend on a low-carbon future.

Banner: Diversity in mine morphology, structure, and environment around the world. We demonstrate the complexity of automatic detection of mines from satellite images.Image: Provided/Science Direct