Management implications
Given that local adaptation and phenotypic differentiation in forest trees (Savoleinen et al . 2007) has been closely tied to variation in climate (Alberto et al. 2013), populations may become increasingly maladapted as climate change continues (Shaw & Etterson 2012; Franks et al. 2014, Aitken & Bemmels 2015). Maladaptation due to climate change is expected to be greatest in populations from the warmest extent of their range, while populations at the cold edge may benefit from slightly warmer temperatures (Aitken & Bemmels 2015). Provenance trials have shown this response experimentally with tree productivity declining as the climate distance transferred between home site and garden site increases (O’Neill et al. 2008; Evanset al. 2016; Grady et al. 2015). Transfer functions can help determine how far a population can be moved before growth declines below a specified level; this tool, combined with climate change forecasts, is one of the best ways to implement assisted migration in order to manage for future forest health and productivity. This method has been used to recommend seed transfer zones and distances for economically important conifer species in British Columbia (O’Neillet al. 2008, 2017), and as a caution to move trees at a reasonable, step-wise pace to track climate change (Grady et al.2015). Based on our results of declining tree performance as climate transfer distance increases, we may expect decreases in tree productivity and increasing maladaptation as local conditions become increasingly arid, especially for populations in southern Arizona that are close to the thermal and low elevation edge of their distribution (Ault et al. 2014; see Fig. 1).
In addition to tracking declines in growth metrics with climate transfer distance, it is important to consider how shifts in phenology in forest trees will affect dependent communities (Whitham et al. 2020). Numerous studies have shown trophic-level asynchronies across different ecosystems due to mismatches in phenological changes (Thackeray et al. 2010; Renner & Zohner, 2018). For example, warmer, earlier springs can facilitate earlier phenology of many tree species, which may or may not be synchronized with the emergence and reproductive cycles of important community members (Visser et al. 2006; Kudo & Ida 2013; van Asch et al. 2013). This can cause a phenological mismatch between plants and the species that rely on them for food or habitat (Renner & Zohner 2018), disrupting species interactions and trophic cascades (Bailey et al . 2006; Smith et al. 2011). Since cottonwood trees provide food and habitat for thousands of dependent species (Whitham et al . 2006; Lamit et al.2015), changes in growth, morphology, and phenology with altered climate will likely affect species interactions, community composition, and functionality (Whitham et al . 2020).