MOVING AWAY FROM A DEFAULT
Reframing provenancing strategies on eco-evolutionary risks highlights that all strategies come with risks, even local provenancing. We recommend moving the discussion beyond the local vs. non-local debate and towards context-dependant choices that enhance resilience and future adaptability. Growing evidence suggests that no single strategy will be appropriate for all contexts. Rather, the most appropriate strategy in any situation will be influenced by a range of context-dependant factors.
To create resilient plantings, provenancing must consider the dynamic conditions imposed by environmental change (Aitken & Bemmels 2016). Genetic variation is a key factor in long-term evolutionary resilience and adaptability (Sgrò et al. 2011; Kardos et al. 2021). Provenancing strategies that mix local and non-local provenances aim to boost resilience by increasing genetic diversity, including adaptive variation. Considering eco-evolutionary genetics can therefore help guide whether local, non-local or a combination of both may be most appropriate to enhance genetic variation and thus adaptability in any given situation (Hoffmann et al. 2020). While we focussed on the eco-evolutionary risks associated with provenance choice, other factors may influence provenance choice, including ecological (e.g., biotic interactions; (Bucharova et al. 2021), logistical (e.g., availability of seed; (Broadhurst et al. 2016) and socio-cultural (e.g., significance of local genetic identity of plants; Mauriceet al. 2013). However, these factors need to be weighed against the eco-evolutionary risks associated with not changing provenancing strategy, especially in changing environments.
Moving forward, it will be important to address barriers and knowledge gaps that currently inhibit context-dependant provenancing decisions. These include better understanding of the long-term performance, and risks associated with, both local and non-local provenances for different species in different restoration contexts and deriving practical guidance on provenancing choice (Breed et al. 2018). Greater direct evidence on the long-term performance of not only different provenances (e.g., local and non-local provenances) but also different provenancing strategies will also assist decision-making (Bailey et al. 2021). Determining whether generalized recommendations can be derived based on plant life history, functional traits and/or genomic information across species will also help guide recommendations for species without empirical data.
Several initiatives are working to address these knowledge gaps. For example, there are already several practitioner guides that incorporate climate uncertainty, evolutionary genetics, and landscape fragmentation into seed sourcing decisions with the aim of increasing genetic diversity and adaptability (Harrison et al. 2017; Rossettoet al. 2019). Establishing long-term provenance trials (as per example 1 above), especially those that are embedded within mixed species revegetation plantings (e.g. Bailey et al. 2021), will help generate long-term fitness data on provenances in a relevant revegetation context and thus help enable adaptive seed sourcing decisions into the future.
CONCLUDING REMARKS
With a focus on eco-evolutionary factors, our viewpoint is that no provenancing strategy is risk free. The risk of changing from the default local provenancing strategy needs to be weighed against the risk of not changing. We do not suggest local provenancing is no longer appropriate, nor that provenancing strategies must change from local. There are clear situations where local will be most appropriate and where factors not discussed in depth here (e.g., logistics and cultural) may influence provenance choice more than eco-evolutionary factors. Rather, the choice of provenancing strategy should consider all risks when determining the most appropriate provenancing strategy for a given species and context. As such, the intention of our paper is to move the discussion of provenancing away from a default strategy and whether local provenances are more-or-less appropriate than non-local provenances, and towards conscious and context-dependent provenancing decisions.