Inconsistent recognition of uncertainty in studies of climate change impacts on forests

Abstract

Background. Uncertainty about climate change impacts on forests can hinder mitigation and adaptation actions. Scientific enquiry typically involves assessments of uncertainties, yet different uncertainty components emerge in different studies. Consequently, inconsistent understanding of uncertainty among different climate impact studies (from the impact analysis to implementing solutions) can be an additional reason for delaying action. In this review we (a) expanded existing uncertainty assessment frameworks into one harmonised framework for characterizing uncertainty, (b) used this framework to identify and classify uncertainties in climate change impacts studies on forests, and (c) summarised the uncertainty assessment methods applied in those studies. Methods. We systematically reviewed climate change impact studies published between 1994 and 2016. We separated these studies into those generating information about climate change impacts on forests using models -'modelling studies', and those that used this information to design management actions-'decision-making studies'. We classified uncertainty across three dimensions: nature, level, and location, which can be further categorised into specific uncertainty types. Results. We found that different uncertainties prevail in modelling versus decision-making studies. Epistemic uncertainty is the most common nature of uncertainty covered by both types of studies, whereas ambiguity plays a pronounced role only in decision-making studies. Modelling studies equally investigate all levels of uncertainty, whereas decision-making studies mainly address scenario uncertainty and recognised ignorance. Finally, the main location of uncertainty for both modelling and decision-making studies is within the driving forces-representing, e.g. socioeconomic or policy changes. The most frequently used methods to assess uncertainty are expert elicitation, sensitivity and scenario analysis, but a full suite of methods exists that seems currently underutilized. Discussion & Synthesis. The misalignment of uncertainty types addressed by modelling and decision-making studies may complicate adaptation actions early in the implementation pathway. Furthermore, these differences can be a potential barrier for communicating research findings to decision-makers.

Will forest dynamics continue to accelerate throughout the 21st century in the Northern Alps?

“…Second, considering only a small number of scenarios can mask potential non‐linearities and turning points, such as the shift in compositional trajectories in RCP 8.5 scenarios documented here. The assessment and communication of scenario uncertainty is, thus, pivotal for the elaboration of robust policy and management recommendations (Petr et al, 2019). Our results, however, also indicate that strong dampening feedbacks exist that make future forest structure and composition considerably less variable than climate change scenarios.…”

“…Our study is among the first to study a complete ensemble of climate change projections (n=22) by means of forest landscape simulation modeling (see also Snell et al, 2018). Addressing scenario uncertainty in simulation modeling can improve our understanding of future patterns in the trajectories of changes (Kalliokoski et al, 2018; Petr et al, 2019). It furthermore allows the identification of potential non‐linearities, such as the, on average, considerably different responses between RCP 8.5 scenarios and the other RCP scenarios regarding late century compositional changes.…”

Socio-hydrology, politicization of water science and implication of the Eyes on Earth Study on the contemporary research dialogue in the Lancang-Mekong Basin

“…To better understand the conflict of ideas among multi-stakeholders (Earle, Jägerskog and Öjendal, 2010;Swann and Bosson, 2008) and examine the quality of research inputs (Michalska-Smith andAllesina, 2017, Sarewitz, 2016), we adapted Baker's model (Baker, 2016) and outlined twelve key challenges in "standard research procedures" (see Table 2). (Baker, 2016), we believe that the root of the problem lies in the outcome-oriented research process and different research mindsets of multi-stakeholders as other studies suggest (Bouleau, 2019;Petr et al, 2019;Bolsen and Druckman, 2015;Lorenz et al, 2013). Traditionally, when the research results are published, there are limited ways how to correct the controversial assessments (Haciyakupoglu et al, 2018;Lim, 2018).…”

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

“…Deep uncertainties pose nontrivial conceptual challenges and require different decisionanalytical approaches compared to a situation under well-characterized uncertainty (e.g., where a single probability density function can be identified) (Lempert et al 2003;Marchau et al 2019;Walker et al 2001). In forest management studies that focus on decision-making, deeply uncertain factors are commonly represented by a limited amount of scenarios (see Petr et al (2019) for a review) and as time-independent rather than dynamic factors (e.g., constant discount rates over time) (e.g., Augustynczik et al (2017)). Improved quantification of these uncertainties can potentially improve the design of forest management strategies.…”