You are here: Home / Research Topics / Research Highlights / Individual Highlight

Research Highlights

Individual Highlight

Improved predictions of deforestation in Borneo

Photo of Map of Borneo showing areas of forest loss between 2000 and 2010 in yellow, areas of forest persistence from 2000 to 2010 in green, and areas that were not forest in 2000 in black.Map of Borneo showing areas of forest loss between 2000 and 2010 in yellow, areas of forest persistence from 2000 to 2010 in green, and areas that were not forest in 2000 in black.Snapshot : A collaborative team led by Forest Service research ecologist Samuel Cushman has produced a substantial breakthrough in advancing predictive modeling of drivers and patterns of deforestation. The method combines multi-scale optimization with machine-learning predictive modeling to identify the drivers of deforestation and map relative future deforestation risk.

Principal Investigators(s) :
Cushman, Samuel A.  
Research Station : Rocky Mountain Research Station (RMRS)
Year : 2017
Highlight ID : 1350

Summary

Deforestation is the leading driver of biodiversity loss in the world today. Improved understanding of the factors that drive the rate and pattern of forest loss are critical to guide proactive conservation of forest ecosystems and the biodiversity they support. The island of Borneo in Southeast Asia has some the highest levels of biodiversity in the world and the highest deforestation rates. Samuel Cushman, a scientist at the Forest Service’s Rocky Mountain Research Station, developed a novel application of multi-scale machine-learning predictive modeling to identify the drivers and predict the rates and patterns of forest loss across the island of Borneo. The model produced much stronger and more accurate predictions than traditional approaches and has provided detailed and specific guidance for conservation prioritization and management efforts in Borneo. The project improved predictive modeling of deforestation with a novel application of multi-scale machine learning predictive modeling. The project provided highly accurate predictions of future deforestation risk across the full extent of Borneo, which provides critical guidance for proactive conservation and management efforts.

Forest Service Partners

External Partners

 
  • University of Oxford Wildlife Conservation Research Unit