Lorem ipsum dolor sit amet, consectetur adipiscing elit. Curabitur eleifend tortor nec augue pretium
Biodiversity loss and ecosystem collapse has been ranked as the second most severe risk to economies and societies over the next decade, according to the World Economic Forum’s Global Risks Report 2025.
It’s not surprising given that the World Wide Fund for Nature’s latest Living Planet Report found that wildlife populations have declined by an average of 73% over the past 50 years, with the biggest decrease in freshwater (85%), terrestrial (69%) and marine (56%) populations.
The report, which tracked almost 35,000 vertebrate populations of 5,495 species from 1970 to 2020, warns that parts of the planet are approaching dangerous tipping points – driven by nature loss and climate change. This includes South-East Asia. Home to more coral reefs than anywhere else in the world, almost half of the planet’s mangroves, and a third of the world’s coastal and marine habitats, the region is a hotspot for biodiversity.
However, deforestation, rising temperatures and extreme weather events pose a real threat. Indeed, up to 42% of the region’s species could be lost by the end of the century if organisations, countries and governments don’t intervene.
Nature-based Solutions (NbS) are the answer, according to some experts. The Southeast Asia Climate and Nature-based Solutions (SCeNe) Coalition is a network of nine non-profit organisations working to advance conservation and climate action in South-East Asia. Its goal is to bridge the gap in investment, technical expertise and resources in NbS and accelerate the delivery of such conservation projects.
“Data and technology are essential tools in scaling and monitoring NbS,” says Edwin Seah, NbS partnerships lead at SCeNe Coalition. These include geospatial mapping, remote sensing and drones that provide real-time insights into ecosystem health, carbon sequestration and biodiversity changes.
“In regions such as South-East Asia, where ecosystems are vast and diverse, these tools are crucial in addressing the scale and complexity of conservation efforts,” he adds.
SCeNe Coalition is developing tools and resources to support organisations to deliver high-quality NbS through three key workstreams:
“Through these workstreams, we aim to direct carbon and climate finance towards just and equitable climate mitigation and adaptation work, and to the conservation of biodiversity and critical habitat,” Seah says.
To date, more than 4,500 users from 279 organisations have used the NbS Tool, where almost 100 organisations have identified over 280 NbS projects. “We want to ensure that high-quality NbS projects don’t just take root but thrive – delivering triple benefits for nature, people and climate across the region,” he adds.
Technology can also help to scale up coral restoration efforts. Working with non-profit coral restoration organisation Our Blue Spaces, Malika Meghjani, assistant professor in computer science and design at Singapore University of Technology and Design, developed software to help them analyse coral types and their growth.
The technology can clean up images in low-visibility waters, measure coral areas and identify corals using machine learning, saving time and resources. “Marine biologists take pictures, put a scale against them, measure how much the coral has grown and that’s their record,” Meghjani explains. “We wanted to enhance these images so they can see them better, label them and figure out the size of them in one tool.”
Talking about the challenges she faces, Sam Shu Qin, marine biologist and co-founder of Our Blue Spaces, says: “Low visibility in Singapore’s waters is our biggest pain point, so sometimes it’s hard to get good pictures or videos to document coral growth and survey marine life. The process is also manpower-heavy – divers go out for the whole day to measure the data before we can actually get something we can analyse.”
She adds: “We’re at a trial stage where we have a coral nursery at ONE°15 Marina in Sentosa that we use as a test area to optimise the software before going out to open water. We’ve done a few runs to test the manoeuvrability, and whether we need to adjust the parameters or the drone structure to optimise documentation.”
The long-term goal is to have a system that can memorise where transplanted corals are so that growth can be monitored over six months or a year. “There are still things that need to be looked at, like how you train the machine to understand the difference between corals and rocks.”
-----------------------------------------------------------------------------------------------------------------------------------------------
Corals from the Sentosa coral nursery are planted on adjacent seawalls – the coral garden. Malika Meghjani has created 3D maps of the garden to track progress. She says: “They’re interactive maps, so you can see close-up images. You can see the species name, segmented area for it and how the map looks overall. It’s essentially Google Maps for underwater environments.”
Meghjani and Sam Shu Qin are also looking at how to optimise maintenance of the reefs. “We’re trying to identify those hotspots where this is required,” Meghjani says. “If everything looks okay then they don’t need to dive.”
Shu Qin adds: “Different species have different resilience to thermal strength, so that’s where technology can help us identify which corals are better adapted to future conditions. In Singapore, we’re finding corals that can tolerate higher sedimentation and temperature stress.”
Meghjani is also exploring whether certain properties on the water surface indicate what’s happening below.
“We have small autonomous buoys that can sense the water quality – the temperature, pH level and salinity – and tell us about the underwater domain as well,” she says.
Meghjani has worked on a project that uses artificial intelligence to streamline underwater data collection and real-time decision-making for marine biologists. Project MERLION (Marine ExploRation with Language-guIded Online iNformative Visual Sampling and Enhancement) addresses challenges such as limited data storage and communication bandwidth by identifying and reporting key events in real time.
“If objectives change from time to time and marine biologists are interested in looking at puffer fish and cloud fish, but then want to see turtles, they can communicate this in natural language in real time,” she says. “MERLION can align itself with the user’s requirement and give only those unique events that are meaningful.”
Visual data is enhanced in real time and monitored long term. Previously, the robot would be left underwater for three or four hours if it was powered autonomously by batteries, Meghjani says, and the marine biologist would then spend another four hours looking at the entire video sequence.
“We found a scoring metric where we could score the automated summaries based on what humans think,” she adds. “We showed users the entire video sequence or visual experience and they would select the summaries from it. We benchmarked against that, and our MERLION approach came close enough to the human score.”
Katie Smith is a freelance writer and editor.
SCeNe Coalition: www.scenecoalition.org
Project MERLION: www.youtube.com/watch?v=o1hDcecdX5g
