So departmental officials decided to take active measures to protect the marine invertebrates, commissioning and funding the university’s Swire Institute of Marine Science to initiate a coral restoration programme.

The programme was launched in 2016 and is managed by Yu. Just as a woodland can be reforested, so can a reef, and the institute’s team began replanting corals.

Initially, the scientists glued living coral fragments directly onto rocks. But corals grow slowly, and left alone it would take them decades to increase to a size large enough to provide new habitats for other sea creatures – a crucial role of corals in the ecosystem.

Why Hong Kong has the toughest coral in the world

So this year, the department expanded the programme. The institute was commissioned to boost the restoration process using a two-tiered approach – placing artificial structures called “reef tiles” on the seabed for corals to settle on and kick-start biodiversity, and at the same time seeding the reef tiles with living coral.

Artificial reefs, made from a variety of materials, have been dropped into the sea from the Maldives to the Mediterranean to boost biodiversity by attracting wildlife. Coral “replanting” has also been tried, but the institute’s programme was the first combining the two methods.

The reef tiles were made by the Robotic Fabrication Lab in the university’s faculty of architecture, where a team led by associate professor Christian Lange 3D-printed them from clay. It was the first use anywhere of clay, chosen for being environmentally friendly, in 3D printing.

The design requirements – to provide a healthy habitat for both coral seedlings and sea creatures, and to prevent sediment build-up on top of the tiles – were translated into a mathematical model that was then coded into an algorithm by Lidia Ratoi, assistant lecturer in the university’s faculty of architecture, and a researcher in robotic fabrication.

During the printing process, Ratoi’s algorithm guided the robotic arm, equipped with what is known as a cold extrusion system that pushes out clay. On the upper surface of the tiles, the algorithm created an intricate pattern of elevated, folded ridges. This pattern is known in mathematics as a Gosper curve. Ratoi’s algorithm tweaked it, smoothing out the angles to make it more suitable for coral growth. The ornamentation is exactly the same on each tile.

“For me, as an architect and robotics researcher, this project was an eye-opening experience,” Ratoi says. “I am grateful I got to use my skills in the service of nature, and I hope that the project will highlight the beauty and power of technology when used in the right way.”

In July 2020, 128 reef tiles were placed on the seabed at Coral Beach, Moon Island, and in a bay near the WWF Marine Life Centre, both in Hai Ho Wan Marine Park. They cover an area of 40 square metres (431 square feet) in total.

Six live coral fragments were seeded in the ridge folds of each tile. Collected by scientists from the seabed, the pieces had been dislodged from local living coral either naturally or by human impact such as boat action, and would almost certainly not have survived without intervention.

The tiles were placed on a sandy bottom where corals cannot grow naturally because they need rocks to attach to. This sand is a consequence of sedimentation caused by construction on land and reclamation in the sea. Sediment settles on corals and buries them.

“Historical evidence suggests that it used to be inhabited by a coral community; this is one of the reasons we are trying to restore corals there,” says Yu, referring to the marine park.

To restore coral communities in Hong Kong, understanding the present state of the region’s corals was not enough. The scientists first had to learn the corals’ past and make a prediction about their future.

Jonathan Cybulski, a historical ecologist at the institute and a doctoral candidate at the university, has been investigating the changes in Hong Kong’s coral community – from 5,000 years ago to the present.

Cybulski’s research found that the region’s coral community has varied greatly with time. This prompted a question: what kind of coral community, composed of which species, should be the result of restoration? Should it be a community like those from 200 years ago, 50 years ago, or just 30 years ago?

The key consideration, says Cybulski, was that the coral community the scientists were building “have a good shot” at coping with present and future environmental conditions – salinity, temperature and the amount of nutrients in the water.

Coral restoration is a relatively new field, and the institute’s programme is an experiment to find effective ways to restore coral communities. There are many potential pitfalls.

For example, not all coral species would make good “roommates” on the same reef tile. Some are aggressive and attack other species in the struggle for space. Others can be even more competitive – a colony would attack not only a different species, but anything that it is not its own genetic clone.

Having taken both coral biology and local coral history into account, the institute’s scientists seeded the tiles with different combinations of three dominant local genuses of coral, Acropora, Platygyra and Pavona, that coexist in Hong Kong’s coral communities. Each genus brings its own contribution to the new colony. Acropora is fast-growing but physically fragile, while Platygyra is slower growing, but robust.

Other coral species, and Hong Kong has 90 of them, are expected to join naturally, as their larvae settle on the safe habitat of the tiles.

The scientists will be specifically monitoring two developments: which coral combinations produce faster regrowth and coverage, and which ones generate the best increase in biodiversity.

The tile design itself facilitates biodiversity with inbuilt “structural complexity – little crevices and niches, attracting invertebrates to settle in”, says Yu.

These small organisms – worms, shrimp, snails, as well as smaller, microscopic creatures – are the foundation of a healthy ecosystem. The more different species, the better. If one is lost, a similar one can step in to fulfil the same function. “Biodiversity creates a safety net for ecosystem collapse,” Yu explains.

In Hoi Ho Wan Marine Park, wildlife moved into the new 3D-printed habitats almost immediately. “An hour after we put the first tile down I could already see fish, hermit crabs and sea cucumbers settling there,” Yu says.

To function well, an ecosystem also requires predators. For example, some ecologists in Britain advocate the reintroduction of wolves to naturally control the population of deer and allow regrowth of forest.

In the waters around Hong Kong, the equivalent of the deer are the long-spined sea urchins of the Diadema genus that graze on living coral. The predators that would help Hong Kong’s corals regrow are spiny lobster and black-spotted tuskfish.

“They used to be a common catch in Hong Kong in the 1990s,” says Yu, adding that both species have been fished out of existence in the region.

Yu says she would like to see both brought back because they would be far better than humans at controlling Diadema. Everything is connected in an ecosystem. Under natural conditions the sea urchins’ diet consists largely of algae – arch-rivals of corals. Only when their numbers are out of control do the sea urchins start to damage coral. Predators would control their numbers, resulting in a better-balanced ecosystem.

Yu calls rebuilding coral communities in Hong Kong “a cutting-edge restoration project”. She says the goal is for the terracotta tiles to eventually become invisible under the mass of coral, but makes it clear that it will take years for the coral communities to grow to that size.

“This is a pioneer study,” she says. “We need the courage to try new things, but there are also uncontrollable factors. We would appreciate people being understanding and patient. Restoration is crucial, bit it is only a part of the larger equation to ensure the future of coral reefs.”