Genetic variation within foundation species is a major source of diversity in functional traits, which in turn shape community structure and ecosystem processes. However, this genetic variation in tree traits can also be modified by environmental conditions, impacting both tree performance and ecosystem function. This means that predicting the role of particular genotypes in changing or restored ecosystems depends on their plasticity in key phenotypes. In this study, we measured the genetic, environmental, and g x e effects of growing environment on leaf traits and decomposition rates using 24 genotypes of Populus fremontii from populations throughout Arizona, planted in three common gardens along at 12°C temperature gradient. As an important foundation species across the Southwest, Fremont cottonwood links terrestrial and aquatic ecosystems, providing large litter inputs to the base of aquatic food webs. We collected leaves after natural senescence, measured their size, specific leaf area (SLA), and carbon to nitrogen (C:N) ratios, and then placed them in litter packs in a central Arizona stream to measure decomposition and colonization by aquatic macroinvertebrates. Finally, we compared decomposition rates in the field to rates of shredding by a key detritivore in the lab.
Results/Conclusions
We found strong interactions between genetic origin (population or genotype) and growing environment in all litter traits. This plasticity in litter traits translated into effects of both genetics and growing environment on litter decomposition and aquatic invertebrate communities. These effects of environment and interactions between environment and genetic variation challenge our ability to predict the ecological roles that particular genotypes or source populations will play as environmental conditions change. However, some potential for predicting phenotypes also emerged. First, for environmental effects, we found that the growing environment in the warmest garden substantially compressed the variance in trait values expressed across cottonwood populations. Second, for g x e effects, we found that the amount of plasticity across growing environments in C:N, a key characteristic of litter impacting its decomposition rate, could be predicted by the climate of the source population. Further research on the effects of growing environment on trait diversity and the evolution of plasticity could aid efforts to make strategic decisions about what to plant in riparian restoration projects.