Blending Geoscience Research with Machine Learning

Who we Are

We are the John Lab, a research group led by Professor Cédric John and hosted in the Digital Environment Research Institute (DERI) at Queen Mary University of London. At DERI, Professor John leads the Data Science for the Environment and Sustainability Research Platform, one of four key research directions for the institute. Before joining DERI, the group was based for nearly 16 years in the Department of Earth Science and Engineering at Imperial College London.

Our research approach blends machine learning and AI with cutting edge field and experimental methods in Earth and Planetary Sciences. The Digital Environment Research Institute is the centre of digital, data science, and AI research at Queen Mary and it underpins the university’s vision for its research Strategy 2030. For us, DERI is the perfect home for an interdisciplinary team working on scientific machine learning applied to environmental and Earth Science problems.

DERI and Queen Mary University of London are affiliated with the Alan Turing Institute.

Testimonials

Hear what former members of the lab have to say about their time with us.

Adhipa Herlambang (PhD Student, 2018-2021)

Adhipa Herlambang (PhD Student, 2018-2021)

“It was a great honor to be part of the amazing John Lab. Participating in the clumped isotope lab during my Ph.D. was an immensely satisfying experience in many ways. It allowed me to develop a series of learning experiences in a very friendly atmosphere.”
John MacDonald (Postdoc, 2013-2015)

John MacDonald (Postdoc, 2013-2015)

“I had a great time working in the John Lab. Cedric was a fantastic mentor to me as a postdoc, and he really helped me in getting to my current career stage as a Senior Lecturer in Earth Sciences at the University of Glasgow.”
Tobias Kluge (Postdoc, 2012-2015)

Tobias Kluge (Postdoc, 2012-2015)

Research topics in the John lab are at the cutting edge of the respective field and provided me a fascinating insight in current scientific developments. They were a strong motivation for my own research. A vibrant and motivated research group inspired exciting project ideas and enabled thorough scientific discussions.”
Marta Marchegiano(Postdoc, 2019-2021)

Marta Marchegiano(Postdoc, 2019-2021)

“I did my first postdoc at John lab where I learnt about the clumped isotope technique. Cédric is a very thoughtful and knowledgeable supervisor, he always took the time to teaching me about this exciting thermometer. Thanks to his enthusiasm I became very passionate about carbonate geochemistry and this experience was an important stepping stone for my future career. “
Sarah Robinson (PhD Student, 2019-2022)

Sarah Robinson (PhD Student, 2019-2022)

“Over the course of my time with John’s Lab I have benefited from a strong and diverse group of peers. I have gained knowledge in not only my own field, but the fields of my peers through discussions in bi-weekly lab meetings. I will miss working within John’s lab as I move onto my next chapter.”
Annabel Dale (PhD Student, 2011-2015)

Annabel Dale (PhD Student, 2011-2015)

“Cedric was great to work with as a supervisor for my PhD and afterwards whilst co-supervising a PhD student. His enthusiasm and good instinct for interesting science, means that working in the John lab is an excellent place to learn and develop research skills.”
Niranjana Sundararajan (MSc Student, 2022)

Niranjana Sundararajan (MSc Student, 2022)

“Working on my MSc thesis with Cedric as my supervisor was the most interesting, productive and challenging part of my academic experience at Imperial. Cedric is an excellent supervisor-  providing constant support, direction and the encouragement necessary to reach research goals.”
Qi Adlan (PhD Student, 2019-2022)

Qi Adlan (PhD Student, 2019-2022)

“Working in John Lab makes me feel engaged and valued. Cedric is very knowledgeable and kind — he helped me reach my full potential to become an independent researcher.”

research Highlights

We have two main research themes: Applied Artificial Intelligence for Earth and Space Sciences (we call it Earth-Centric AI) and Carbonate Research. You can also find about our publications and the software and data stemming from our research.

AI Research

Our group applies data-centric machine learning techniques to Earth and planetary sciences, leading to innovative approaches for analyzing and interpreting data in these fields.

Carbonate Research

Our research in carbonates focuses on the processes involved in their formation and alteration, and how these processes impact the geochemistry and stratigraphy of carbonate rocks.

Publications

Our group has published over 100 peer-reviewed papers in a variety of fields, including clumped isotopes, carbonate diagenesis, climate change, stratigraphy, AI and machine learning.

Software and Data

Our research has led to the development of free software tools that have been widely used by researchers in our fields. These tools have made it easier to analyze and interpret data..

Spotlight on our Papers:

Interplay between depositional facies, diagenesis and early fractures in the Early Cretaceous Habshan Formation, Jebel Madar, Oman
Interplay between depositional facies, diagenesis and early fractures in the Early Cretaceous Habshan Formation, Jebel Madar, Oman

Diagenesis and fracturing can significantly alter petrophysical properties of subsurface carbonate reservoirs, but the impacts of these processes at the inter-well scale are hard to predict. However, the initial distribution of sedimentary facies is easier to predict, and could template…

Deciphering the state of the late Miocene to early Pliocene equatorial Pacific
Deciphering the state of the late Miocene to early Pliocene equatorial Pacific

The late Miocene‐early Pliocene was a time of global cooling and the development of modern meridional thermal gradients. Equatorial Pacific sea surface conditions potentially played an important role in this global climate transition, but their evolution is poorly understood. Here…

Towards a new understanding of the genesis of chalk: Diagenetic origin of micarbs confirmed by clumped isotope analysis
Towards a new understanding of the genesis of chalk: Diagenetic origin of micarbs confirmed by clumped isotope analysis

Chalk is usually thought to be a homogeneous sediment with a relatively simple early diagenetic history. Here, clumped isotope analyses of samples from a core of Campanian Maastrichtian chalk are presented, indicating that material smaller than 5 µm has a different…

Chemostratigraphy in Miocene heterozoan carbonate settings: applications, limitations and perspectives
Chemostratigraphy in Miocene heterozoan carbonate settings: applications, limitations and perspectives

The temporal variability of geochemical proxies can be used in time intervals characterized by global changes in marine chemistry to achieve improved stratigraphic correlation. The application of this approach in rocks lithified by cementation requires particular attention, as the original…

Predicting marine organic-rich deposits using forward stratigraphic modelling: The Jurassic Najmah source rock–Case study
Predicting marine organic-rich deposits using forward stratigraphic modelling: The Jurassic Najmah source rock–Case study

Predicting the distribution and heterogeneity of marine Mesozoic organic-rich rocks is a challenging task that requires multi-disciplinary data integration supported by innovative numerical modelling. This study aims at investigating the factors controlling marine organic matter production, accumulation, and preservation along…

Combining clumped isotope and trace element analysis to constrain potential kinetic effects in calcite
Combining clumped isotope and trace element analysis to constrain potential kinetic effects in calcite

The field of clumped isotope paleothermometry is over a decade old, but the influence of precipitation rate on the fractionation of clumped isotopes between natural carbonates and their environmental solutions remains unclear. Here we apply two different proxies, carbonate clumped…

Assessment of Factors Controlling Clumped Isotopes and δ18O Values of Hydrothermal Vent Calcites
Assessment of Factors Controlling Clumped Isotopes and δ18O Values of Hydrothermal Vent Calcites

The clumped isotope composition of CaCO3 (Δ47) is a geochemical proxy that can provide mineral formation temperatures and, together with measured carbonate δ18O, inferred fluid δ18O values. Under natural conditions, carbonates form within a relatively wide pH range and varying…

Diagenetic Implications of Stylolitization In Pelagic Carbonates, Canterbury Basin, Offshore New ZealandDIAGENETIC IMPLICATIONS OF STYLOLITIZATION IN PELAGIC CARBONATES
Diagenetic Implications of Stylolitization In Pelagic Carbonates, Canterbury Basin, Offshore New ZealandDIAGENETIC IMPLICATIONS OF STYLOLITIZATION IN PELAGIC CARBONATES

Stylolites are irregular discontinuity surfaces that are thought to result from localized stress-induced dissolution during burial or tectonic compression. The genesis of stylolites and the controls on stylolitization are still debated, and the interplay between stylolitization, generation of carbonate-rich fluids,…

The Palaeocene–Eocene carbon isotope excursion: constraints from individual shell planktonic foraminifer records
The Palaeocene–Eocene carbon isotope excursion: constraints from individual shell planktonic foraminifer records

The Palaeocene–Eocene thermal maximum (PETM) is characterized by a global negative carbon isotope excursion (CIE) and widespread dissolution of seafloor carbonate sediments. The latter feature supports the hypothesis that the PETM and CIE were caused by the rapid release of…

Mental health in the field
Mental health in the field

Field work is an important and valued part of geoscience research, but can also serve as a source of stress. Careful planning can help support the mental health and wellness of participants at all career stages.

Clumped isotope record of individual limestone fabrics: A potential method to constrain the timing of oil migration
Clumped isotope record of individual limestone fabrics: A potential method to constrain the timing of oil migration

This study applied clumped isotope analyses to investigate how different limestone components (larger skeletal grains and enclosing matrix) and cements may have varying degrees of susceptibility to recrystallization during progressive burial. The results also provide new constraints on the temperatures…

Relative performance of support vector machine, decision trees, and random forest classifiers for predicting production success in US unconventional shale plays
Relative performance of support vector machine, decision trees, and random forest classifiers for predicting production success in US unconventional shale plays

Unconventional shale reservoirs have revolutionized the energy industry. However, the prediction of production based on reservoir geology characterization has largely focused on sweet spot definition rather than on over-arching production trends across multiple plays. This study uses machine learning (ML) techniques…