The carbonate group at Imperial College London Dr. Cédric John's research team, was founded in 2008 to study carbonate sediments and minerals at a range of scales. 

From the begining, research within the carbonate group has been multi-disciplinary, and has included fieldwork, clumped isotopes analysis, and numerical methods. The dual approach mixing experimental/field with numerical methods is very powerful, and can be used to solve problems in both the applied field (notably for the oil and gas industry) and academia.

The principal motivation for the group's research focus is that the heterogeneous carbonate rocks capture a wealth of geological information in the form of geochemical records and rock textures. The range of scientific problems addressed by our research includes Clumped isotope paleothermometry to reconstruct diagenesis, climate and the thermal history of sedimentary basins, machine learning ("Big Data") and numerical methods applied to a range of problems in carbonate geology, as well as stratigraphy and sedimentology coupled with forward numerical modelling to understand the 3D architecture across scales. 

Clumped isotope paleothermometry

The carbonate research group was a pioneer in adopting clumped isotope for diagenetic work back in 2009. At the time, only one laboratory in the world was running clumped isotopes (John Eiler at CalTech). In the last 10 years, the carbonate research group has established itself as a leader in this technique. Some of our major achievements in this field are the have developed the first calibration for clumped isotopes beyond 90˚C, to have written and released free community software (Easotope) to radically simplify and improve data reduction for clumped isotopes, and to have created in collaboration with the mass spectrometry industry the first fully automated line for clumped isotope analysis (the IBEX, now a commercial product operating in labs located on three different continents).

Clumped isotopes are based on the grouping of heavy isotopes within the lattice of carbonate minerals, and they offer a more robust and accurate proxy to determine temperature of formation (or recrystallization) of carbonates. The group has applied this proxy to many case studies including amongst others early to late dolomites, modern echinoids, paleoclimate of the Cretaceous, and various applications to structural diagenesis and tectonics. More details about the technique and our research on the Carbonate Clumped Isotope Lab page.

Machine learning and numerical methods in carbonate geology

Current machine learning algorithms are great at solving a range of non-linear classification and regression problems. Geology has traditionally been a descriptive science, with rock facies (the way rocks 'look') being an important non-structured data that is used for further interpretations. Thus, careful and accurate description of facies directly impacts the quality and related uncertainty of geological models.

A large part of the group's current research focuses on using artificial neural networks to automatically recognise and quantify rock features, such as facies, diagenetic fabric, and others. In additiona to machine learning, the group explores how numerical methods can help solve non-linear problems in geochemistry and carbonate seimentology. This includes developing new code to model geological processes using programming languages such as Python, Matlab, Scala and Java. More information on our research and achievements can be found on our Carbonate Digital Lab page.

Sedimentology, stratigraphy and forward modeling

Sedimentary  processes occur at a range of time and dimensional scales: basin opening and closing are controlled by tectonic, global sea-level changes, sediment supply by rivers  and in-situ production of (carbonate) sediments proceed to fill accomodation space in the basin. The mathematics controlling the shape and distribution of sediment bodies is complex and non-linear, and although the stratigraphic record gives us constraints on the architecture of carbonate sequences, the incomplete nature of this archive impairs quantitative visualization of the geological processes and products.

To overcome this limitation, the carbonate research group combines traditional field methods to gather information on the sediment and stratigraphy of carbonates with a numerical technique called "forward stratigraphic modeling". The principle is to model forward in time how sediments within a basin are being deposited, eroded and transported, thus allowing to predict the nature of the sediment and the stratigraphic architecture of the rocks. Importantly, uncertainties in the stratigraphic models can be determined by repeating the modeling numerous times, allowing the establishement of uncertainty maps. More information on this type of research and the main achievements can be found on the Carbonate Digital Lab page.