Reefscapers Diary September 2020

Monthly Progress

At Landaa this month, we transplanted 12 new coral frames, and monitored a further 394 frames (mainly at our Blu site).

At Kuda Huraa, we monitored 62 coral frames, and we also revisited our sample of wild colonies and confirmed they are now fully recovered from bleaching. At the shallow Water Villas site, some of the coral colonies on our frames are recovering from bleaching last month, whereas at the Channel site, the corals remain healthy and are growing well. At the Blue Hole site, the new rubble substrate continues to be successful in preventing sinking into the sandy seabed. All the 42 frames at this site were monitored, and their photographs uploaded to our online database.

Coral Plates in Aquarium One (plates KH01, KH02, KH05) and Aquarium Two (plates KH03, KH04)

• KH01 – the newly replenished fragments of Acropora digitifera and Acropora millepora have regained a healthy colouration and have now started to encrust onto the plate.
• KH02 – the Galaxea fascicularis fragments remain healthy overall, and the fragments are fusing with one another (although seemingly at a slower rate).
• KH03 – thoroughly cleaned of algae. The calcified areas are growing and about to fuse. The fragments of Acropora valida remain healthy and continue to grow, although they haven’t yet calcified onto the plate.
• KH04 – also cleaned of algae. Only two original fragments remain (due to the loss from algal overgrowth) and of the 11 newly transplanted fragments, 4 have started to calcify onto the plate (photos, below).
• KH05 – 1 fragment died after bleaching, but the others have now regained a healthy colouration and continue to grow (2 fragments are encrusting).

Coral Reproduction Spawning Experiment

In our coral reproduction study, competition from algae remains an issue, so we performed extra cleaning to prevent smothering. Larger coral colonies were unaffected, but our smaller specimens showed reduced growth.

‘Coral Core’ Experiment

All the parent colonies have displayed zero negative effects from the sampling, which is an encouraging result.
One more coral plug died, giving a 21% success rate with the current methodology. Only 1 plug (R3L1) is encrusting on the surrounding substrate and growing slowly in size; the other surviving plugs are yet to demonstrate outward growth.

In future experiments, we plan to:

• Hold the freshly sampled plugs in a nursery to minimise damage and ensure growth before outplanting.
• To facilitate outward growth, outplant the plugs directly onto the substrate (not in a cavity).
• Outplant plugs from the same parent colony close to one another (shared genetics could lead to better acceptance and larger growth).

Time Lapse Experiment to Record Coral Bleaching

This month, we again captured the coral bleaching process (in Acropora humilis) via our time-lapse camera (video here). The process took approximately 26 days, with the addition of hot water once a day (3.2 litres). Whilst the fragment was bleached by the end of the experiment, no tissue loss was observed. It was therefore decided to allow the temperature to return to normal by providing the fragment with “open flow” from the ocean to record the potential recovery process. (This is ongoing.)

Our video footage again displayed the release of a gas. This time it was localised at a site of tissue damage (perhaps caused by handling), indicating that gas release and tissue damage may be related. Once the tissue is compromised, the internal liquid concentration is able to form equilibrium with the surrounding seawater and therefore the gas is able to escape from the tissue easily, as opposed to becoming trapped, as observed in our previous video.

We also observed the fragment “swelling” during the bleaching process, which was in direct correlation to a reduction in polyp activity. This would suggest a “mechanic” aspect, which needs to be identified within the bleaching process. Similar results were observed by a team of Australian scientists when subjecting Fungia sp. to elevated temperatures. The coral used in their experiment nearly doubled in size before expelling its symbionts (you can view this video here).

Featured on brinno.com >> Developing strategies to mitigate coral bleaching with the help of time lapse in the Maldives

Artificial Intelligence and Coral Identification

During the summer, we improved our deep-learning models to achieve better accuracy in our detections of coral fragments. Since then, we have run our entire analysis on a total of 570 different coral frames, which allows us to see trends in coral growth, and additionally helps to identify areas requiring analysis improvements.

During September, we have made improvements in our computer-vision algorithms, which resulted in increased accuracy. The first batch of analysed pictures provides us with data on around 11,000 fragments. We can now start analysing the effects of various data parameters, such as fragment size at transplantation, or choice between Pocillopora and Acropora species.

Going forward, we need to decide if our models’ error margins are acceptable or if we need to continue improving them. Every method of data analysis can be refined, but time is sometimes better spent by starting to analyse the results, rather than perfecting the models.

Read the full project history of our autonomous catamaran and AI coral identification project over at Reefscapers AI-4Corals.