Forest fragmentation: another perspective

Abstract

In his recent (and very favorable) review of our edited book, Tropical Forest Remnants[1xLeigh, E.G Jr. Trends Ecol. Evol. 1997; 12: 414Abstract | Full Text PDFSee all References][1], Egbert Leigh raised some issues of general interest to those who study the ecology of fragmented habitats. I would like to respond to two of these points, and propose a differing view.First, I question Leigh's contention that, because of formidable taxonomic constraints, fragmentation researchers should focus on less-diverse, seasonal forests, rather than those of super-diverse regions like the Amazon, as is being attempted in the Biological Dynamics of Forest Fragments Project (BDFFP). The Amazon, in my view, is simply too extensive and important to ignore. Moreover, while it is true that the identification of over a thousand tree species in the BDFFP study area has been a massive challenge, there has been much progress in recent years. Important work on forest dynamics[2xFerreira, L.V and Laurance, W.F. Conserv. Biol. 1997; 11: 797–801Crossref | Scopus (115)See all References, 3xSee all References], biomass[4xLaurance, W.F et al. Science. 1997; 278: 1117–1118Crossref | Scopus (376)See all References][4], regeneration[5xSee all References, 6xSee all References], and phytosociology[7xSee all References][7]in our fragmented landscape has recently appeared, or will soon appear, in major journals.Second, Leigh suggests that the study of ecological interactions and distortions, revolving around plant communities, is needed to provide a `unified, coherent understanding of the effects of forest fragmentation.' Such interactions could occur, for example, if certain pollinators or seed-dispersers declined in fragments, thereby leading to the eventual collapse of their dependent plant species, which then might cause a cascade of ecological changes affecting yet other species.While this sounds useful and scientifically appealing, in practice there have been surprisingly few demonstrations of the importance of such higher-order interactions in fragmented habitats, especially given the prominent role that symbioses are known to play in tropical forests[8xSee all References][8]. Indeed, one of the chapters in our book, in which Australian investigators invested several years in a failed attempt to detect the effects of loss of a keystone seed-disperser (the cassowary) on its dependent tree species, concluded that the study of such interactions was fraught with risk, and often prone to failure[9xSee all References][9]. Given the inherent complexities and nonlinear behavior of many ecological interactions, I suggest that the nature, direction and magnitude of higher-order effects in fragmented habitats will often be extremely difficult to predict. Hence, while the study of higher-order interactions is likely to remain a useful arrow in the fragmentologist's quiver, we are far from ready to view it as a general paradigm for predicting the ecological consequences of fragmentation.