But the Kielder Forest is much wider than a conventional corridor. Few studies have looked for gene flow in genuine corridors; even fewer have found it. One study investigated the genetic diversity of small marsupials in a narrow forest corridor traversing 4.5 kilometers of grazed grasslands in Queensland, Australia. It found that genetically distinct populations had persisted at either end. Mixing was a myth.
Other studies have shown that conservation corridors work. But most have looked at short corridors of 100 meters or so through largely natural landscape. "Just because species can use and travel along short corridors in a natural setting does not mean that they will be successful dispersing along much longer corridors embedded in a large, heavily impacted landscape," says Gregory. "Still less that such movements occur frequently enough to allow enough gene flow to occur so that the connected habitat blocks function as one population."
Many corridors are useless. Developers in the Brazilian Amazon are required by the national Forest Code to leave 60-meter-wide forest strips along riverbanks. But a study of birds and animals in those corridors found that anything with a width less than 400 meters had no benefit on the numbers of species along the banks, let alone on promoting gene transfer. Yet, far from widening the corridors, the Brazilian parliament recently voted to relax the Forest Code.
This all matters because big claims are made for conservation corridors. The suggestion is that they can minimize the impact of development by substituting for wild habitat. If that is wrong, much of our conservation efforts will come to naught. Gregory and Beier have set up a project, Do Corridors Work?, to work out what makes a successful corridor, and have issued a plea for help finding ones to investigate.
Perhaps we should not make the perfect the enemy of the good. Surely any corridor is better than none? But consider this. The edges of wild areas are known danger zones for wildlife, where predators and diseases may invade. Linking two existing protected areas with a long narrow corridor may expose it to greater danger along these edges. Unless the benefit exceeds the threat, then there is serious potential to do harm.
Gregory and Beier do not dismiss corridors altogether. They believe they can still work, if designed properly. The problem is, we don't really know what that means.
This article originally appeared in New Scientist.