糖心视频

Illegal bomb fishing is causing extensive damage to some of the world’s most biodiverse coral reefs, according to a new pilot study involving Lancaster Environment Centre coral reef researcher Dr Tim Lamont. Using an AI-powered underwater listening system, the team found that bombs are being detonated in a global marine biodiversity hotspot at a rate of almost one per hour, revealing the true scale of one of the world's most destructive fishing practices and providing vital evidence to support conservation efforts.

Bomb fishing is recognised as one of the most destructive fishing practices in our oceans. The practice involves using improvised explosives to stun or kill large numbers of fish for harvest. This efficiency comes at a significant ecological cost, with explosions causing widespread destruction. Bomb fishing is particularly prevalent on highly threatened tropical coral reefs, where concentrated fish stocks are often found in shallow waters. Globally, , and bomb fishing represents an acute additional stressor. Despite being universally outlawed, bomb fishing has been documented .

Each blast can indiscriminately kill fish tens of metres away, while simultaneously causing physical destruction to the coral reef framework. Repeated blasts can leave entire reefs degraded into rubble fields which are unable to recover without costly interventions. These ecological impacts also have significant socio-economic consequences for communities that depend on coral reefs. Declining fish populations can reduce food security and income, while reef degradation can also weaken natural coastal protection and reduce the appeal of reefs for tourism.

A range of factors can incentivise bomb fishing, including poverty and a lack of alternative livelihoods. Weak enforcement can also allow the illegal supply chains for explosive components - as well as the practice itself - to continue.

However, the true extent of bomb fishing across impacted regions remains unknown, making it difficult to identify the hotspots and patterns of activity needed to inform effective management.

In their paper, , published in Remote Sensing in Ecology and Conservation, a team of researchers, involving Dr Lamont and from universities in Indonesia and the UK, set out to determine the intensity of bomb fishing, estimate the maximum extent of reef degradation caused by the practice, and document patterns of activity over an extended period.

The team focused on the Spermonde Archipelago, in Southwest Sulawesi, Indonesia - a suspected hotspot for bomb fishing. The archipelago sits within the Coral Triangle region of Southeast Asia, the most biodiverse marine biogeographic realm in the world. While bomb fishing has been identified as a primary driver of reef decline in this region, its full extent has not previously been quantified.

To address this gap, the research team used underwater microphones with an estimated detection distance of around 10 miles, reflecting the fact that noise from the bomb blasts can travel vast distances underwater.

Shockingly, the researchers found that the coral reef was bombed once every 62 minutes on average. Across 16 months of underwater audio, the microphones recorded almost 3,570 cases of bomb fishing from 3,650 hours of recordings. This equates to around 8,500 incidents per year and an estimated 160,000m? of reef degradation. To process the thousands of hours of audio generated by the study, the team developed an AI model capable of rapidly identifying each bomb blast, allowing them to process a month’s worth of audio in just four hours.

Dr Lamont said:

“It’s horrifying to discover that bomb fishing is still so prevalent in places like this – it causes so much destruction in globally important biodiversity hotspots. But the technology developed in making this discovery is also an encouraging proof of concept.”

The findings offer a glimmer of hope for conservation efforts, demonstrating the potential of this low-cost approach to quantify an underreported and highly destructive fishing pressure at scale. It offers the potential to identify bomb fishing hotspots and peaks in activity to support targeted patrols or rapid-response restoration efforts, while carefully designed intervention campaigns that combine AI-accelerated data analysis with local community engagement could provide new opportunities to reduce the practice in effective and just ways.

“Now it’s possible to automatically detect bombs with low-cost equipment, it opens the door for strides forward in monitoring and enforcement,” said Dr Lamont. “This technology could really help in the fight to end bomb fishing around the world.”