Oceans of Data: Teaching Climate Change through Data Science
Abstract
This article explains how an IT engineer, a statistician, and two thematic researchers built a course on teaching climate change and data science at the same time. Over almost 10 years, this course has been offered to approximately 300 students, enrolled in two different masters programs, in physical oceanography and data science. For the physical oceanography students, this course provides the opportunity to learn how to manipulate and extract information from large datasets. For the data science students, not familiar with climate sciences, they learn about the current state of the climate and its evolution. The primary learning objective of the class is to promote interdisciplinary collaboration through project-based learning, helping these two distinct groups of students to bridge the gap between data science and climate dynamics. These projects aim to manipulate observations and/or model data that are indexed in time, space and, sometimes, ensemble dimensions, in order to observe changes in the ocean. Students are enthusiastic about reproducing emblematic examples such as sea level rise, marine heat waves, ocean heat content change and the evolution of sea ice concentration. Through these examples, they learn to study climate change using appropriate tools. To help them achieve this goal, students are taught fundamentals concepts from a range of disciplines: cloud computing, descriptive and predictive statistics, machine learning, and also oceanography. We find that such a multidisciplinary course, in front of a varied audience of students, is extremely relevant to tackle the increasing challenge of big data in the environmental sciences.
Related
- Climate Change Finance and Carbon Offsets: Addressing Global Climate Challenges through Innovative Funding and Emissions Reduction Mechanisms
- Guide to Teaching with Data in Statistics and Data Science Classrooms
- A Framework for Teaching the Responsible and Effective Use of Generative AI in Statistics and Data Science
- Nonparametric Regression on the Circle with Applications to Climate and Environmental Science
- Spatio-Temporal Stochastic Interventions for Causal Inference in Climate Science
BibTeX
@misc{tandeo2026oceans,
title = {{Oceans of Data: Teaching Climate Change through Data Science}},
author = {Pierre Tandeo and Sally Close and Guillaume Maze and Frédéric Paul},
howpublished = {HAL},
year = {2026},
month = oct,
url = {https://imt-atlantique.hal.science/hal-05793654v1},
note = {CC BY-NC-SA 4.0},
}
RIS
TY - UNPB
TI - Oceans of Data: Teaching Climate Change through Data Science
AU - Pierre Tandeo
AU - Sally Close
AU - Guillaume Maze
AU - Frédéric Paul
PY - 2026
DA - 2026/10/12
PB - HAL
UR - https://imt-atlantique.hal.science/hal-05793654v1
AB - This article explains how an IT engineer, a statistician, and two thematic researchers built a course on teaching climate change and data science at the same time. Over almost 10 years, this course has been offered to approximately 300 students, enrolled in two different masters programs, in physical oceanography and data science. For the physical oceanography students, this course provides the opportunity to learn how to manipulate and extract information from large datasets. For the data science students, not familiar with climate sciences, they learn about the current state of the climate and its evolution. The primary learning objective of the class is to promote interdisciplinary collaboration through project-based learning, helping these two distinct groups of students to bridge the gap between data science and climate dynamics. These projects aim to manipulate observations and/or model data that are indexed in time, space and, sometimes, ensemble dimensions, in order to observe changes in the ocean. Students are enthusiastic about reproducing emblematic examples such as sea level rise, marine heat waves, ocean heat content change and the evolution of sea ice concentration. Through these examples, they learn to study climate change using appropriate tools. To help them achieve this goal, students are taught fundamentals concepts from a range of disciplines: cloud computing, descriptive and predictive statistics, machine learning, and also oceanography. We find that such a multidisciplinary course, in front of a varied audience of students, is extremely relevant to tackle the increasing challenge of big data in the environmental sciences.
ER -