Conventional niche overlap measurements are not effective for assessing interspecific competition

Interspecific competition is notoriously difficult to detect and quantify, especially in species that are wide-ranging or otherwise difficult to track in the wild. Research investigating interspecific competition usually relies on niche overlap measurements despite that this approach alone does not yield rigorous inference. As an illustration, we review published research assessing interspecific competition in mid-sized carnivores in North America (bobcat - Lynx rufus; Canada lynx - Lynx canadensis; coyote - Canis latrans), and report on shortcomings associated with commonly used study designs and types of inference. Niche overlap measurements typically focus on one or two resources (e.g., food, space, habitat), often using non-independent sampling units and inadequate replication. Few studies measure overlap variation through space, time, or resource variability, which is crucial for robust assessment. Niche overlap (or lack thereof) is used as evidence both for and against interspecific competition, reflecting a weak link between competition theory, predicted responses, and observations. Overall, challenges associated with conducting competition research in the field promote over-reliance on simple measurements, flawed study designs and weak inference. Minimally, niche overlap studies should include assessment across multiple niche dimensions and spatial or temporal variation in competitor density or resource availability. Dynamic investigative approaches should include new technologies for tracking inter-individual interactions, study designs that leverage quasi-experiments (e.g., decline in shared resources, biological control of one competitor), and synthetic analyses (e.g., meta-regression). Ultimately, better understanding of competition theory vis-a-vis study design and data needs will promote improved understanding of the role of interspecific competition in nature.