Prospective Coral Bleaching Predictions
The NOAA bleaching heat stress projections for Maldives are categorised as “Warning” (April) and “Alert level 1” in May. This is a slight improvement compared to the predictions made earlier in the year, although June remains at “Alert Level 2”.
Monthly Progress
At Landaa, only 4 new coral frames were transplanted this month, partly due to reduced guest numbers during the global lockdown, and partly due to the pause in operations during this period of elevated ocean temperatures (newly transplanted frames are particularly at risk of bleaching).
At Kuda Huraa, 416 frames were monitored (cleaned, repaired, photographed) around the island, completing the Water Villas site. With the help of fellow Resort staff, we have moved all 500 frames at the Water Villas site, to position them in the shade offered by the boardwalks and water villas. This will lower the sunlight intensity during the upcoming period of elevated ocean temperatures. Based on last year’s coral studies, the greatest survival rates are seen on coral frames exposed to some sunlight during the day (“partial shade”). Some coral frames were left in “full sunlight” and “full shade” as comparison baselines.
We continue to experiment with shading structures at the Channel site, where 30% of our frames are showing some paling or bleaching (with a few instances of disease and tissue necrosis), and we have relocated the most vulnerable frames to deeper waters.
The graph (below) plots the temperatures recorded at the Water Villas and Channel sites (February-April), with spikes up to 34°C. The two sites have similar depths, however, the Water Villas site has lower temperatures, likely due to the stronger water currents.
CoralWatch – Monitoring for Signs of Coral Bleaching
At Landaa, we recorded species of both Acropora and Pocillopora showing signs of bleaching and stress-related coral fluorescence (down to depths of 8m, and at temperatures of 31°C). During a guest snorkelling excursion, we were able to monitor the coral bleaching status of the reefs around the inhabited local island of Kudarikilu. We recorded bleaching in various colonies of Pocillopora (5m deep, temperature of 30°C).
At Kuda Huraa’s House Reef site (12m deep), most of the corals transplanted from last month’s mass collection have adjusted well to their new environment. There are no signs of coral bleaching, despite an increase in average ocean temperatures of 1°C since early March (the maximum temperature was 31°C on 8 April). On 19 April, we checked on the healthy wild coral colonies out on the reef flat, which we had identified and tagged with buoys in March. Overall, many of the colonies had started to pale or show signs of coral bleaching (Acropora digitifera, A. hyacinthus, A. tenuis, Pocillopora verrucosa, P. meandrina, Porites species).
Coral Plates in Aquarium One (plates KH01, KH02, KH05) and Aquarium Two (plates KH03, KH04)
- KH01 – 8 dead fragments plus 10 fragments with some bleaching or tissue loss (probably caused by the warmer intake water). The two merged Acropora hyacinthus fragments have started to grow over the digitifera, and are competing with A. cytherea.
- KH02 – All 36 Galaxea fascicularis fragments remain healthy and continue to grow well, with healthy encrusting and no bleaching (one dead fragment from a previous month). There is some fusing between colonies of the same colour morph, but the two colour morphs are not fusing with each other.
- KH03 – 16 fragments have died; the remaining 13 living fragments are bleaching or partially dead. Larger fragments at the top of the plate are healthier, with hyacinthus better suited to higher light levels. The A. valida fragments continue branching, but have small areas of calcification.
- KH04 – Towards the end of April, several healthy fragments died within one week, likely due to the warmer intake water. hyacinthus variant 2 is bleaching, and variant 3 has lost tissue.
- KH05 – Our new fascicularis plate continues to grow well, with 2 rapidly encrusting fragments; 1 fragment is partially bleached.
Voavah Frame Relocation
Over the course of 2 full days, an additional 148 coral frames were relocated to Four Seasons Private Island at Voavah (from Landaa’s Water Villas and Parrot Reef sites). Our Reefscapers staff were kindly assisted by an enthusiastic volunteer task force of Resort staff colleagues (Boat Crew, Recreation, Dive and Launch).
The new artificial reef to minimise Voavah’s beach erosion consists of 4 parallel lines of 261 relocated frames (180 Small, 42 Medium, 13 Large, 19 Hearts, 7 Diamonds). The next step will be to replant these frames with new coral fragments from Ghulifalhu (Malé Atoll).
The plugs and drilled holes in the parent colonies are monitored and photographed every week. The health of the corals is assessed against the Coral Watch colour chart, most recently on 26 April (9 weeks after the start of the experiment). Currently, there are no clear health differences between the tested plug sizes.
R1 | Plugs | All 6 were lost (no epoxy) or have died (4 weeks after transplantation). |
Parent | Quickly recovered from slight bleaching, but now slightly paling again due to warm temperatures. | |
R2 | Plugs | 2 died; 2 plugs are healthily encrusting; 2 plugs are surrounded by CCA (Crustose Coralline Algae). |
Parent | Very healthy, with good recovery around and over the holes. | |
R3 | Plugs | 1 plug died; 5 plugs healthy after 9 weeks (some slight paling is returning). |
Parent | Very good health, with encrusting over the Epoxy for all 6 holes. | |
R4 | Plugs | 2 plugs dead; 4 plugs starting to bleach. |
Parent | Very healthy encrusting for all 6 holes. |
Our Reef-Monitoring AI Catamaran
After more than a year of work on artificial intelligence, we have reached an important milestone… our program successfully analysed several complete frames! The software automatically identified the taxonomic genus of the coral fragments and recorded their sizes, all without any human intervention.
This progress opens the door to ‘big data’ analyses by running the program on the whole database (20-30 days of continuous data processing). To do this, we need to make the deep learning models reliable in a wider variety of situations through additional training. We have already conducted one more batch for fragment detection and two more batches for frame segmentation (our two deep learning models).
Get the full story in our Development Journal at Reefscapers.com.

AI analysis of growth (cm) of the coral colonies on frame #LG3324, over time (2017-2019)