Habitat fragmentation represents the conversion of continuous habitat into a series of disjunct fragments, often surrounded by a modified agricultural matrix. I have conducted numerous studies into the effects of habitat fragmentation, both in relatively recently-modified landscapes, but also in older systems where the biological responses have stabilised.
When ecologists first began studying the consequences of habitat fragmentation, many comparisons were made with islands. As more data emerged from recently cleared landscapes, it became clear that communities associated with habitat fragments and islands differed in many ways, and the fragments as islands analogy was abandoned.
Having completed a study of a series of large forest fragments isolated thousands of years ago my regional climate change, I synthesized previous work on the long-term effects of fragmentation. Unlike those disparate findings emerging from recently-modified landscapes, this work from older systems was remarkably consistent, suggesting there were generalizable consequences of habitat fragmentation, it just took hundreds to thousnads of generations for them to emerge.
I then revisited fragment versus island comparisons, demonstrating that, despite initial differences, fragmented and insular systems converged in many community properties over time. So, although a forest fragment surrounded by crops and a hilltop island in a hydroelectric dam display many differences soon after becoming isolated, as ecological interactions stablize and communities readjust, these two kinds of habitat patch become increasing congruent over time.
In addition to improving our understanding of how fragmented landscapes work, this research is also of genuine applied relevance, informing and prioritising on-ground actions to safeguard biodiversity in fragmented landscapes. This emerging discipline—known as connectivity conservation—is the focus of ongoing research, working closely with landholders and various Natural Resource Management groups to look after existing habitats and maximise functional linkages between them.