METHODOLOGICAL INFORMATION

1. Description of methods used for collection-generation of data: Two main sources of data have been used for this analysis: TotalEnergies' Universal Registration Documents from 1999 to 2022 and the Global Atlas of Environmental Justice (EJAtlas).8 The EJAtlas provides a dataform to systematically characterise and codify the main dimensions of a conflictive project [21,22]. At the end of 2023, the EJAtlas is reaching 4000 entries. The EJAtlas data collects information using a predetermined case entry sheet that combines both free text descriptions of various aspects of the conflict with a set of categorical variables for the contributor to check when applicable. EJAtlas cases include qualitative and quantitative variables such as a case description, geolocation, the main sources of conflict, project details and actors, data on the conflict stage and on forms of mobilisation, impacts, outcomes, and a reference list. This combination of open text and categorisation allows describing the specificity of the conflicts in their local contexts and at the same time conduct large comparative studies across regional, global or thematic scales. Through a global network of collaborators, the EJAtlas gathers information from local sources, always trying to be as faithful as possible to the local narratives and concerns, and referencing all data and significant claims with secondary data. All data submitted by collaborators is then moderated by a central team following a standard set of rules and format. This allows harnessing the geographic coverage and local specificity of a widespread and diverse network of contributors, while systematising information across the data sample. We found the EJAtlas well suited for this study since the repository already contained a significant amount of information on TotalEnergies environmental conflicts and its methodology facilitated the investigation of global patterns of impacts and resistances to TotalEnergies operations. We also draw on complementary information from conflicts not yet documented in the EJAtlas.

2. Methods for processing the data: All Universal Registration Documents for the company from 1999 to 2022 were screened to extract names of specific projects in seven categories: Coal (8), Oil and Gas exploration and production (368), LNG Terminals (47), Pipelines (64), Refineries (49), Power plants & Renewables (78) and Offsets, Recycling and CCS (15). The project names are reported in Supplementary Table 1. This database of TotalEnergies projects was used to understand the global geographic and operational scope of TotalEnergies, as well as to identify environmental conflicts related to specific projects using specific search strings in Google and Google Scholar (see search strings in Supplementary Methods). The identified conflicts were studied in greater depth and documented in the EJAtlas. We selected all EJAtlas cases involving TotalEnergies, its predecessors Total, Fina and ELF or its subsidiaries during the time of development of a conflict, notably CEPSA until 2011 and Novatek until 2022. At the start of this research, 43 cases involving TotalEnergies were already present in the EJAtlas database. We have subsequently added 7 more conflicts and updated some existing cases to include more detail, and contemporary information on the conflicts: we obtained a final list of 50 cases. We codified the cases by type of operation following the energy flux from extraction to consumption including exploration (11), extraction (17), pipeline (5), LNG (7), oil spills (3), refinery (5) petrochemical (1), and climate litigation (1). A full list of the conflicts is shown in Supplementary Table 2 and an EJAtlas featured map9 allows the reader to browse the different cases.