Achieving conservation and restoration targets requires not only deciding how much land to act on but also identifying where those actions will have the greatest ecological effect. Traditional prioritization often relies on metrics of habitat condition, opportunity, or broad ecological value. The authors build on sensitivity-based conservation prioritization to provide a complementary approach that explicitly links local habitat-condition change to landscape-scale outcomes relevant to species persistence. The approach aims to identify where local degradation would cause large connected-habitat loss (loss potential) and where local improvement would generate substantial connected-habitat benefits (gain potential).
The extended framework differentiates a persistence-relevant connectivity metric with respect to local habitat condition. In practice, this means calculating the sensitivity of the chosen connectivity measure—one that is relevant to persistence—to incremental changes in local habitat condition. Sensitivity highlights locations where small local changes yield disproportionately large effects on connected habitat at the landscape scale.
By combining sensitivity with current habitat condition, the method separates two distinct conservation-relevant quantities:
This differentiation allows conservation planners to prioritize sites for protection to avoid large-scale losses and to prioritize sites for restoration where gains would be maximal.
The authors applied this sensitivity-based extension to national forest-naturalness data in Norway. Using these national-scale data, they calculated spatially explicit measures of loss and gain potential based on the derivative of the persistence-relevant connectivity metric with respect to local forest-naturalness condition. The study thereby evaluates where within Norway small local changes in forest condition would have the largest connected-habitat consequences at the landscape level.
When applied to Norway's forest-naturalness layers, loss and gain potential were spatially related but far from redundant. That is, areas identified as high loss potential were not simply the same places identified as high gain potential. Importantly, the authors report that gain potential is not simply concentrated in the most degraded forests, indicating that restoration opportunities with large landscape-scale benefits can occur outside areas of lowest condition. The spatial distinction between loss and gain potential suggests different priorities for conservation versus restoration action.
The analysis shows that both high-loss and high-gain areas were poorly represented within current protected areas in Norway. Across the dataset, loss potential was consistently better represented than gain potential in the existing protected-area network, but neither category was well covered overall. This result indicates gaps in current protection relative to locations that would most affect landscape connectivity if degraded or improved.
Extending sensitivity-based prioritization to derive loss and gain potential provides a general and scalable tool for linking local habitat-condition changes to persistence-relevant landscape outcomes. Key implications include:
The manuscript reports results from Norway using national forest-naturalness data and declares funding from the Research Council of Norway. Specific methodological parameters, maps, numerical summaries, and supplementary materials were provided in the original preprint; those detailed data and any additional caveats are reported in the source and are not reproduced in full here. The authors conclude that the sensitivity-based extension is a broadly applicable way to connect local habitat-condition change to persistence-relevant connectivity outcomes and to inform distinct conservation and restoration decisions.