Abstract

Abstract Coral restoration has emerged globally as a form of life support for coral reefs, awaiting urgent mitigation of anthropogenic pressures. Yet its efficiency is difficult to assess, as sizeable transplantation programmes handle hundreds of thousands of fragments, with survival rates inherently time intensive to monitor. Owing to limited available data, the influence of most environmental and methodological factors is still unknown. To address this issue, machine learning and computer vision were used to track individual colonies' survival, in a world first. Fragments from several species of Acropora and Pocillopora were transplanted over 12 sites across two Maldivian atolls. These colonies grew on coral frames, placed between 1 and 30 m deep. Analysis of monitoring pictures provided health and growth data on 77,574 individual coral colonies to inform the influence of genus, depth, initial fragment size, and substrate on their survival. Among 77,574 fragments, individual survival rate was 31% after 2 years (21% after 4 years), which is much lower than most reported results. Deeper placement was an important success factor for Acropora transplants, but not for Pocillopora. In both genera, smaller initial fragment size was key to increased survival rates. Pocillopora fragments survived better than Acropora fragments at shallow depths (≤7 m), regardless of initial fragment size. Deeper, both genera had similar survival rates, which were influenced by initial fragment size and depth with comparable importance. During the mid‐2019 heat wave, previously transplanted Acropora fragments were 38% more likely to die than Pocillopora fragments. Overall, the total volume of live coral steadily increased over time, by more than 3.7 × 106 cm3 per year, as the volume increase in surviving fragments more than compensated for the volume loss due to mortality. This finding supports the use of targeted coral restoration to accelerate reef recovery after mass bleaching events.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call