Identifying decision-relevant uncertainties for dynamic adaptive forest management under climate change

The decision on how to manage a forest under climate change is subject to deep and dynamic uncertainties. The classic approach to analyze this decision adopts a predefined strategy, tests its robustness to uncertainties, but neglects their dynamic nature (i.e., that decision-makers can learn and adjust the strategy). Accounting for learning through dynamic adaptive strategies (DAS) can drastically improve expected performance and robustness to deep uncertainties. The benefits of considering DAS hinge on identifying critical uncertainties and translating them to detectable signposts to signal when to change course. This study advances the DAS approach to forest management as a novel application domain by showcasing methods to identify potential signposts for adaptation on a case study of a classic European beech management strategy in South-West Germany. We analyze the strategy’s robustness to uncertainties about model forcings and parameters. We then identify uncertainties that critically impact its economic and ecological performance by confronting a forest growth model with a large sample of time-varying scenarios. The case study results illustrate the potential of designing DAS for forest management and provide insights on key uncertainties and potential signposts. Specifically, economic uncertainties are the main driver of the strategy’s robustness and impact the strategy’s performance more critically than climate uncertainty. Besides economic metrics, the forest stand’s past volume growth is a promising signpost metric. It mirrors the effect of both climatic and model parameter uncertainty. The regular forest inventory and planning cycle provides an ideal basis for adapting a strategy in response to these signposts.

National Science Foundation http://dx.doi.org/10.13039/100000001