A collaboration between scientists from the U.S. Forest Service, the University of Oxford’s Wildlife Conservation Research Unit (WildCRU program), and the University of Montana has drawn attention to the risks of deforestation, providing conservationists with the tools to predict and plan for future forest loss. The study, “Multiple-scale Prediction of Forest Loss Risk across Borneo,” recently came out in the journal Landscape Ecology. Co-lead author and research landscape ecologist Samuel Cushman, Rocky Mountain Research Station, and his team focused their research on Borneo, a rugged island shared by three nations in the Malay Archipelago of Southeast Asia. Borneo forests have among the highest biodiversity and the highest forest loss rates on the planet, having lost a staggering 30 percent of its forest since the 1970s. The loss of Bornean forests releases significant amounts of carbon dioxide into the atmosphere and threatens species such as the orangutan, Sumatran rhino, and the Sunda clouded leopard.
The research team’s innovative approach to conducting data analysis, described in a recent news release, included using existing maps of the area and the machine learning algorithm ‘Random Forests’ to build a multi-scale model of deforestation on the island from 2000- 2010. After calculating the historic links between landscape variables and deforestation, the team used this information to predict the future deforestation risk facing Borneo’s remaining forests. The goal of this study was to provide national authorities with a tool that would help them recognize potential deforestation threats in the future.
The research findings confirm that this multiple-scale modelling approach offers a powerful method for analyzing land use change. In addition, the results highlight the immense and imminent deforestation risk to Borneo’s forest biodiversity, with risk patterns related to the differing topographies and landscape structures of the three nations. However, despite its focus on Borneo, the study findings are useful to all forest conservationists, and could help analyze risk in tropical forests around the world.