The scale of the human exposome, which covers all environmental exposures encountered from conception to death, presents major challenges in managing, sharing, and integrating a myriad of relevant data types and available data sets for the benefit of exposomics research and public health. By addressing these challenges, the exposomics research community will be able to greatly expand on its ability to aggregate study data for new discoveries, construct and update novel exposomics data sets for building artificial intelligence and machine learning-based models, rapidly survey emerging issues, and advance the application of data-driven science. The diversity of the field, which spans multiple subfields of science disciplines and different environmental contexts, necessitates adopting data federation approaches to bridge between numerous geographically and administratively separated data resources that have varying usage, privacy, access, analysis, and discoverability capabilities and constraints. This paper presents use cases, challenges, opportunities, and recommendations for the exposomics community to establish and mature a federated exposomics data ecosystem.