Towards a better mechanistic understanding of edge effects

Context

Predicting and managing edge effects requires an understanding of the mechanisms that drive them. However, analytical methods that dominate edge effects research are not well suited to discriminating mechanisms, because they do not measure ‘indirect’ edge effects: effects that are mediated by covariates in statistical models.

Objective

To discuss the value of indirect effects for improving mechanistic understanding of edge effects.

Methods

We explain how measuring indirect effects improves mechanistic understanding, and provide guidance on how to do so. We also conduct a literature review to examine awareness of indirect effects in empirical studies of mechanisms underpinning edge effects. Finally, we use a recent paper in Villaseñor et al. (Landscape Ecol 30:229–245, 2015) as a case study to discuss how failure to measure indirect edge effects may limit mechanistic understanding.

Results

Indirect effects provide a means to translate conceptual models of edge effects into mechanistic pathways that are testable and quantifiable. Moreover, failure to measure indirect edge effects can result in impacts of habitat edges being underestimated. However, few studies that we identified in our literature review quantified indirect effects (7 %, n = 72). Worryingly, 11 % of studies did not account for indirect effects despite using statistical models that potentially contained them, possibly resulting in incorrect inference.

Conclusions

A better awareness of indirect effects will help researchers to understand the mechanisms that underpin edge effects, while ensuring that impacts of habitat edges are not underestimated.