Environmental Data Science (EDS) is an open-access transdisciplinary journal dedicated to advances in data-driven methods to understand and predict environmental processes and impacts, and their patterns in space and time. The methodological scope is defined broadly to encompass artificial intelligence (including machine learning, deep learning, and computer vision), statistics, data mining, and econometrics, as well as hybrid approaches that combine data-driven methods with physical process-based modeling.

EDS is a venue for topics relating to the geosphere (the solid earth and its processes), cryosphere (e.g. ice, snow, permafrost and tundra), biosphere (biodiversity and ecosystems), hydrosphere (oceans and fresh water, including the water cycle) and atmosphere (e.g. meteorology, climatology). It also welcomes work that shows how data science can inform societal responses to environmental problems, such as climate change, air quality, energy, natural resources and land use. Papers in EDS can address a range of data types from in-situ observations, to remote sensing, as well as data simulated by physical models, and reanalysis products.

EDS promotes open research by encouraging authors to share their data and code and by publishing reviews alongside accepted articles. For more details about the types of article published in the journal, see the Instructions for Authors.