A Brief History of Edge Effects

100 years of research in four short paragraphs

The ecological importance of edges has been noted and studied for over a century, beginning with Clements’ description of gradients between plant communities that he dubbed “ecotones” (Clements 1907).  Later,  it was noted that diversity and abundance were often higher near edges, both for birds (Lay 1938) and for game species, in general (Leopold 1933), leading to a general conception that edges were “good” and managers should manage for habitat heterogeneity.  An almost complete paradigm shift started in the late 1970’s when studies showed that, although abundance and diversity may increase near edges for some taxa, predation and parasitism often increase as well, leading to the concept that edges may be an “ecological trap” with dire consequences for some species (Gates and Gysel 1978).  This dynamic was even identified as a leading cause of a perceived decline in songbird abundance in eastern deciduous forests in the United States (Brittingham and Temple 1983, Wilcove 1985).  In addition to these trap dynamics, habitat sensitive species were often found to avoid edges (e.g., Burke and Nol 1998), and so habitat quality near edges was considered to be of little value for the most sensitive species.  These factors all led to a fundamental shift in how edges were viewed, especially by the conservation community.  Instead of positive features on the landscape, edges were viewed as largely negative and the preservation of "core" habitat, far removed from the negative influences of habitat edges, was emphasized (e.g., Harris 1998).

This paradigm shift was also accompanied by an explosion of research that began to challenge the new “edges are bad” paradigm with an avalanche of often contradictory results, and as the century turned, a new paradigm emerged that seemed to suggest that edge effects may not be a particularly useful concept because the responses of organisms to edges were idiosyncratic and unpredictable (Ehrlich 1997, Cadenasso et al. 2003, Ewers and Didham 2006).  Many studies emerged that questioned some long-standing assumptions from the edge literature, such as uniformly generally predation rates (Chalfoun et al. 2002) and the consistency of certain species responses (Schlossberg and King 2008). 

Despite the conflicting results that have emerged from the edge literature, our  10 years of research suggests that edge responses are more consistent than generally believed (Ries et al. 2004), and much of the confusion comes from an historical focus on the focal habitat fragment, and not the type of habitat adjacent to it (in other words the habitat that forms the edge). A recent study (Kennedy et al. 2010) even showed that birds in forest patches can have profoundly different responses to different types of open edges (agriculture, mining, and residential). However, we have found that when edge type (both focal and adjacent habitat) is held constant, edge responses are more predictable than previously thought (Ries and Sisk 2004, Ries and Sisk 2008).  Further, even the most basic ecological information describing a species' habitat associations and resource use can be used to make predictions about the direction of edge responses.  However, when considering the landscape as a complex mosaic of habitat of different qualities, rather than just a dichotomous landscape of “habitat” and “non-habitat”, it becomes more difficult to consider how edge responses might scale up to the landscape (Sisk et al. 1997), making  large-scale consideration of edges more difficult.

Despite long-standing research showing the ecological and conservation consequences of being near habitat edges, as well as the ubiquity of edges (both natural and anthropogenic) in all landscapes, practical models and tools to study or consider edges have lagged. The material on this website offers some practical tools that we have developed over the past 10 years to predict edge responses for any species at any edge type, measure edge responses using field data, and to take known or hypothetical edge responses and extrapolate them over landscapes. These tools were developed using terrestrial systems, and considering edges at the scale of “patches” of different cover classes (or “habitat”) within landscapes. But recent applications show their usefulness in aquatic systems (Macreadie et al. 2010) as well. Although the tools we present here come largely from our own research program, we have tried to be as complete as possible in our selected literature citations to provide information on other approaches and tools that have appeared in the literature.

If you have any questions, comments, feedback, or suggestions for other material that would be helpful to present here, please email Leslie Ries at lries@umd.edu.