1 Main text

Climate-smart forest management has been proposed as a nature-based solution integrating climate mitigation goals with adaption measures to enhance the resilience of forest resources (Verkerk et al. 2020). Weatherall et al. (2022) defined it as “sustainable adaptive forest management and governance to protect and enhance the potential of forests to adapt to and mitigate climate change.” But with widespread loss in vitality of many world forests related to drought, beetle attacks, and other disturbances (Patacca et al. 2023), it is proposed that foresters’ attention needs a further shift from climate change mitigation to adaptation (Messier et al. 2019; Jandl et al. 2019). A large majority of European and North American forest managers acknowledge human-caused climate change and its negative effects on forest vitality and productivity, but only a small minority have adopted adaptation measures so far due to a lack of clear guidance on implementation (Morris et al. 2016; Sousa-Silva et al. 2018). Even though a range of adaptation options has been proposed (Spathelf et al. 2018; Messier et al. 2019; Bauhus et al. 2021), uncertainties, trade-offs, and context dependence seem to paralyze decision-making (Hessburg et al. 2021).

Climate change adaptation in forests is a complex exercise of building social-ecological resilience (Nikinmaa et al. 2020), which requires balancing a multitude of trade-offs between objectives and scales of operation (Nikinmaa et al. 2023). Climate change adaptation also involves monitoring, experimentation, and learning while resolving a range of possible lock-ins, either technological, socio-economic, cultural, or institutional, which may hamper its implementation (Groen et al. 2023).

Land managers and policy makers implementing climate-smart forest management seem to feel they are caught between a rock and a hard place. We illustrate this by focusing on the two basic structural characteristics of the forest that adaptive management is targeting: forest composition and stand structure.

Regarding forest composition, the science is advancing steadily (e.g., Benomar et al. 2022). On the one hand, there is increasing evidence of climate change-induced stress and tree mortality in forest ecosystems (Forzieri et al. 2021), with little possibility of natural tree species migration due to slow dispersal and strong landscape fragmentation (Collingham and Huntley 2000). On the other hand, assisted migration could bring clear benefits in terms of adaptiveness and tree productivity (Benito-Garzón and Fernández-Manjarrés 2015), but also holds potential risks, mainly related to biological conservation issues like invasiveness (Camenen et al. 2016) and lack of associated biodiversity, and to genetic conservation issues like outbreeding depression, possibly resulting in, e.g., unadapted day-length or frost sensitivity. Science-based decision support tools are being developed to maximize benefits and minimize risks of assisted migration (e.g., Bindewald et al. 2021; Fremout et al. 2021). There are cautious options like capitalizing on the genetic variation in local populations, complemented by slow introgression of more thermophile and drought-resistant provenances (Fady et al. 2016). Despite these opportunities, it remains a decision fraught with questions of ethical or political nature: whether to prioritize conservation of autochthonous material or promote novel species better adapted to providing ecosystem services under climate change. And even where managers see a clear course of action, execution may be hampered by national or regional regulations that, depending on the zoning (Natura 2000 area, for example), dissuade or sometimes prohibit assisted migration of non-native tree species or non-autochthonous provenances. In this context, Aubin et al. (2011) have offered a useful framework for national and local decision-makers to determine for what species and where and when assisted migration might be a desirable option.

Regarding forest structure, the optimal stand density one should strive for under climate change scenarios is still heavily debated in the forest management community. The focus of the discussion is whether to thin forests more intensively in order to allow precipitation to be distributed over fewer trees, as evidenced by, e.g., del Campo et al. (2022), or to thin less intensively to maintain a humid understory microclimate, as brought forward by Zellweger et al. (2020). At first, this appears to be a mere trade-off between management objectives (tree growth versus forest microclimate-dependent biodiversity), but there might be more to it. Although the available (meta-) studies about thinning effects on tree growth under drought stress (Sohn et al. 2016; Navarro-Cerrillo et al. 2019) are consistent in finding better vitality and growth in thinned stands, it is unclear if results would be conclusive for the full range of conditions, considering that few thinning experiments on moist microclimate-requiring late-successional species are available.

The mentioned adaptation strategies related to forest composition and stand structure illustrate that there are definitely management options to meet the challenges, but clearly, climate-smart forestry is currently caught between a rock and a hard place. In the case of species selection, the choice is between keeping less-adapted native tree species or promoting presumably better-adapted, more southernly growing tree species and provenances. In the case of regulating stand structure, the choice is between maintaining the forest microclimate in closed stands or increasing thinning intensity. These dilemmas have deterred many forest managers and decision-makers from moving ahead with adapting our forest to the uncertainties of climate change and other disturbances. The time has come for hard decisions to be made.

While some decisions await more scientific evidence, promoting diversity in both forest composition and stand structure is one of the clear, no-regret options available that could and should be immediately implemented. Boosting the establishment (and maintenance through stand tending) of functionally diverse tree species mixtures (Messier et al. 2019, 2021; Feng et al. 2022) does increase the overall resilience (Jactel et al. 2017), multifunctionality (Van der Plas et al. 2016), and productivity (Feng et al. 2022) of managed forests over a wide range of contexts. It has almost unanimous support from the scientific community and therefore could be readily implemented (Messier et al. 2021). Boosting stand structural diversity also seems a no-regret option but has not yet been largely studied. Apart from variation in diameter and age class distribution, spatial mosaics of dense and open canopies further accentuate variation in the abiotic conditions of landforms, slope, and aspect, creating varied microclimatic conditions and a broad range of habitats to accommodate a large set of species.

In the context of the heated debate on the European Nature Restoration Law, the here proposed no-regret adaptive management interventions might be interpreted as having a productivist view on forests, but this would be a serious misunderstanding. Also, for forests in conservation areas or targeting non-provisioning ecosystem services and conservation of forest-related biodiversity, the continuity of a vital tree cover is becoming paramount. The time has come to make hard decisions to help adapt our forests, whatever their societal function, to the many threats and uncertainties of the future.