Regular monitoring is essential to understand the health of our coral reefs and observe the impacts of climate change. High-quality underwater cameras enable faster data collection at reef sites, however, analysing photos has remained a manual, painstaking process. ReefCloud is designed to help scientists to retrieve highly accurate insights within minutes, as opposed to weeks or months. Using this machine learning algorithm for automated image analysis, ReefCloud gains more knowledge with every image added. With over 1.3M images from across 24 countries, the algorithm has created over 62 million data points on the health of the world's reefs to date. Interactive dashboards, report cards and data access tools enable rapid interpretation, reporting and communication on coral reef conditions across geographies.

If climate change is outpacing science, can we use data and A.I. to help us outpace climate change? We developed ReefCloud to allow the world’s coral reef monitoring community to work together to improve reef monitoring. Through automated image analysis, ReefCloud provides comprehensive, up-to-date and robust insights into coral reef conditions with 85-95% accuracy and 700 times faster than previous methods allowed. ReefCloud provides statistical analysis on the impact of disturbances derived from a number of external environmental datasets, such as information on sea surface temperatures and tropical cyclones, to provide scientists with additional context on changing coral conditions. We make data sharable, technologies compatible and information more widely available.

ReefCloud integrates international coral reef monitoring efforts and supports scientists to manage, analyse, and share data, fast-track the recording of reef condition, and communicate where and how reefs are changing. ReefCloud caters for the needs of multiple agencies across geographies and has been extensively user-tested to ensure each interaction is fit-for-purpose. Scientists and managers from Australia, Fiji, Palau, Vanuatu and the Maldives have been closely collaborating with NGOs, universities and government organisations to ensure individual needs are considered, from local stakeholders (Indigenous communities, marine parks, and state managers) to national and international bodies (national ministries and international science advisory networks).

The project originates from 2019, when AIMS and the Australian government co-funded an initiative to explore opportunities to address coral reef health. ReefCloud harnesses advances in image analysis and statistical modelling to create an integrated approach to support reef monitoring organisations by revolutionising the speed and accuracy of data analysis. During an initial innovation sprint in 2019, we developed machine learning algorithms and computer vision approaches to automatically tag images of coral and generate insights on its health. A second innovation sprint refined the vision for ReefCloud, identified user needs, tested concepts, developed a UI framework and defined a partner engagement strategy. Collaboration on cloud-infrastructure and data pipeline enabled a robust, scalable, and cost-effective cloud solution. ReefCloud was officially launched at the Our Ocean conference on April 12th 2022 in Palau. By then, ReefCloud had already analysed over 1.3 million images across 24 countries from over 200 contributing users.

Through automated image analysis, ReefCloud estimates coral reef composition 700 times faster than manual analysis and collects 10X more information than previous methods. ReefCloud also replicates expert observations with 85-95% confidence. The tool has already supported over 200 users across 24 countries to analyse over 1.3 million coral images. In Palau, ReefCloud automatically analysed 1.4 million data records from over 20,000 images collected by the Palau International Coral Reef Centre (PICRC) in less than four hours. This work regularly takes the team several months. ReefCloud democratizes data and enables real time reporting on ecosystem-specific indicators. It overcomes the challenge of limited capacities for monitoring, especially in developing countries. ReefCloud also directly supports the tracking of global targets and initiatives for SDG 14 and the Convention on Biological Diversity by providing insights at local, national, and global scales and suggesting course corrections to meet desired outcomes.