Reefs are shallow water areas where carbonate producers such as corals, coraline algae and halimeda, live. They are one of the world’s most diverse environment, making this an ideal location to understand the complex ecology of marine organisms. The sediments produced by reefs and other organisms with a carbonate skeleton are known as ‘carbonate sediments’, or simply ‘carbonates’.

Carbonates have a fundamental importance in Earth Sciences: they are bio-chemical sediments that contains rich information about the past conditions of our Earth, and about transformations occuring in the subsurface. Carbonates often form the skeleton of marine organisms, and therefore offer traces of life in the deep-past. Because they require liquid water to precipitate, carbonates are also important to trace the presence of water in our solar system.

The range of scientific problems we can address using carbonate minerals range from understanding the evolution of life in the deep past, constrain past climate and sea-level changes, reconstruct diagenetic transformations in the subsurface, understand the thermal history of basins, and determine the 3D spatial geometry and sequence stratigraphy of coral reef deposits.

Sadly, modern reefs are under threat from climate change and human activity. Warming sea surface temperatures can lead to corals expelling their zooxanthela algae, a process known as “bleaching” since the coral patches become all white – this can lead to the death of the reef. In addition, plastic pollution has become a major problem, as we explain in the video below.

Carbonate reefs have existed for at least 700 million years, if not more. During this time, reefs have left behind vast quantities of sediments, forming carbonate rocks.

Our research extends to studying ancient reefs in the field. Stratigraphy is the study of how rocks are organised in space and time. Hence, stratigraphers are interested not just in what the rocks are made of, but also in what processes led to the depositon of the rocks. One of the best way to study the stacking pattern of rocks at the small scale remains to look at outcrops. Therefore, we have studied outcrop and core stratigraphy to reconstruct sea-level changes, past climate change, and the heterogeneities of carbonate rocks at a range of scale.

For instance, we studied the relationship between fractures, fluids and diagenesis extensively in the salt dome of Jebel Madar in Oman. We also tried to understand how fractures control dolomitization (the transformation from limestone to an Mg Ca carbonate – dolomite) in the Neoproterozoic to Mesozoic carbonate successions of the Central Oman Mountains. For both of these projects, we collected samples and measurements at the outcrop, and combined geochemistry (including clumped isotopes) with petrographe and mineralogy of the samples to understand past fluid flow in the rocks.

But fieldwork is not limited to the dry land: we also work on cores recovered from drilling vessel such as the IODP Research Vessels, and we periodically sail as well. You can read more about our upcoming our CARAPACE Project here.

We use AI in both modern reefs, and in our work on ancient carbonate sediments. For modern reefs, we can use Earth Observations to monitor the health of coral reefs, both from satellite images or from remote operated vehicles during dives. You can read more about our work on remote sensing with machine learning for coral reefs in this post. And you can find out all about our work on using AI to classify organisms living around coral reefs using here. We also use satellite data to track coral reefs heath.

We have also worked for years on the topic of computer vision for carbonate description. This subfield of deep learning allows to analyse images and use this information to derive knowledge. We use images of rocks, such as cores or thin section, to automate the classification of the rocks, as well as the counting and identification of individual carbonate grains. This allows us to describe field and lab data at an unprecedented speed, and accuracy.