Ecology, once a subdiscipline of biology, has rapidly expanded to become its own unique field with many subdisciplines and research areas of its own. Ecologists need to identify key research themes and how these themes change over time to determine what has already been done in the field and what the future directions of ecology are. However, with the rapid growth of ecological literature, traditional review methods are no longer feasible to cover the breadth of ecology topics. Here, we utilized a newly developed text mining and machine learning technique called Automated Content Analysis (ACA) to identify and describe 46 key themes in 84,841 abstracts from 33 of the top journals in ecology. We also used ACA to evaluate how these themes have shifted in frequency over time and how they are distributed across the 33 journals.
Results/Conclusions
Our ACA showed that at the frontiers of ecology themes such as microbial ecology, genetics, biogeochemistry, and management and policy have all increased in relative frequency over the last four decades. These shifts have led to a decline in classical local-scale research topics such as disturbance, life history, and plant reproduction. Recent advancements in technology, data availability, and analytical techniques have led to a divergence in the scales ecology is studied at to address ecological problems at both micro- and macro-scales. In addition, human-related themes have increased in frequency over time, indicating that ecologists are embracing the human dimensions of ecological systems. The results of our ACA also provide a roadmap for ecologists to navigate the vast amounts of ecological literature. New ecologists such as graduate students can utilize this roadmap to find journals that align with their research goals or identify potential target journals for their manuscripts. Moving forward, these results can help ecologists develop new educational and research frameworks that align with emerging themes in ecology.