Trends in ecology: shifts in ecological research themes over the past four decades

E has grown dramatically as a discipline over the past 100 years from a subdiscipline of biology into its own discipline with many subdisciplines (Kingsland 2005; Ayres 2012). The goal of ecology is to understand the relationships between organisms and their environment, and to use these relationships to help address various complex and challenging environmental problems (Figure 1). As with any expanding scientific field, ecology has evolved from a focus on a central set of core principles into multiple lines of subdisciplinespecific inquiry, resulting in a flourishing of ecological literature (WebFigure 1). Ecologists have periodically assessed the important concepts (Cherrett 1989; Reiners et al. 2017), questions (Sutherland et al. 2013), and themes (Thompson et al. 2001) in ecology to understand the trajectory of the field and to evaluate the societal relevance of the discipline. Comprehensive assessments provide an overview of the intellectual structure of the field, present a framework for pedagogical development and curriculum design, offer guidance for new research directions, and can sometimes help identify paradigm shifts in the science. Previous attempts to determine and define these key concepts, questions, and themes have relied on tools such as literature reviews, professional surveys, and synthesized information in textbooks. Surveys of professionals in the field offer insight into what the scientific community perceives as important (Reiners et al. 2017), but may not convey the topics that are actually being addressed in current research (perceived versus realized importance). Literature reviews, on the other hand, can provide useful syntheses of a body of research. However, it is impossible for a single researcher to keep up with the rapidly growing amount of available literature, making it increasingly difficult to see the overarching picture of ecology. Moreover, the selection of studies to be included in such reviews is prone to unintended human biases, even when selection procedures are established ahead of time (Murtaugh 2002). The development of automated content analysis (ACA) has given researchers the means to review vast amounts of literature in a more datadriven, unbiased manner. ACA is an innovative method for qualitative and quantitative text mining that uses text parsing and machine learning to identify the main concepts and themes discussed within a body of literature (NunezMir et al. 2016). ACA methods have been applied for reviews in forestry (NunezMir et al. 2015, 2017) and other fields (Chen and Bouvain 2009; Cretchley et al. 2010; Travaglia et al. 2011), but not to describe the overarching themes across the entire discipline of ecology. We contend that ACA can provide the systematic, datadriven analyses across time that are Trends in ecology: shifts in ecological research themes over the past four decades