Abstract
Wildlife corridors can help maintain landscape connectivity but novel methods must be developed to assess regional structural connectivity quickly and cheaply so as to determine where expensive and time-consuming surveys of functional connectivity should occur. We use least-cost methods, the most accurate and up-to-date land conversion dataset for East Africa, and interview data on wildlife corridors, to develop a single, consistent methodology to systematically assess wildlife corridors at a national scale using Tanzania as a case study. Our research aimed to answer the following questions; (i) which corridors may still remain open (i.e. structurally connected) at a national scale, (ii) which have been potentially severed by anthropogenic land conversion (e.g., agriculture and settlements), (iii) where are other remaining potential wildlife corridors located, and (iv) which protected areas with lower forms of protection (e.g., Forest Reserves and Wildlife Management Areas) may act as stepping-stones linking more than one National Park and/or Game Reserve. We identify a total of 52 structural connections between protected areas that are potentially open to wildlife movement, and in so doing add 23 to those initially identified by other methods in Tanzanian Government reports. We find that the vast majority of corridors noted in earlier reports as "likely to be severed" have actually not been cut structurally (21 of 24). Nonetheless, nearly a sixth of all the wildlife corridors identified in Tanzania in 2009 have potentially been separated by land conversion, and a third now pass across lands likely to be converted to human use in the near future. Our study uncovers two reserves with lower forms of protection (Uvinza Forest Reserve in the west and Wami-Mbiki Wildlife Management Area in the east) that act as apparently crucial stepping-stones between National Parks and/or Game Reserves and therefore require far more serious conservation support. Methods used in this study are readily applicable to other nations lacking detailed data on wildlife movements and plagued by inaccurate land cover datasets. Our results are the first step in identifying wildlife corridors at a regional scale and provide a springboard for ground-based follow-up conservation.