The loss and degradation of koala habitat is one of the most significant threats to koalas (Phascolarctos cinereus) in New South Wales (NSW). Addressing this challenge is a key priority of the NSW Koala Strategy. However, substantial knowledge gaps remain regarding where the risks of habitat loss and degradation are highest. This dataset provides spatially explicit predictions of (1) the risk of woody vegetation loss and (2) the risk of koala habitat loss across NSW. Predictions are generated separately for three major land-clearing drivers: (a) agriculture, (b) forestry, and (c) infrastructure.

The predictions were derived from statistical models fitted to deforestation patterns observed in cadastral units from 2011 to 2021, using data from the NSW Statewide Land and Trees Study (SLATS). Model development was informed by expert-identified variables, refined through stepwise selection. To account for regional differences in deforestation rates and drivers, NSW was divided into Koala Modelling Regions (KMRs), with independent risk models constructed for each land-clearing driver. Woody vegetation clearing data and a wide range of predictor variables were initially processed at a 25 m × 25 m resolution before being aggregated to cadastral land units.

This dataset is linked to the NSW Koala Research Plan project 2022_05 Understanding drivers and identifying solutions to koala habitat loss and degradation.

More details about the study can be found on: Chung, Y. F., Maron, M., Drielsma, M., & Rhodes, J. R. (2025). Do private land conservation policies and programs adequately consider climate change? Biological Conservation, 308, 111187. https://doi.org/10.1016/j.biocon.2025.111187