Leaf trait variation and decomposition in short-rotation woody biomass crops under agroforestry management

The integration of trees into agricultural landscapes has a strong influence on plant-soil interactions, with major impacts on rates of nutrient cycling across multiple spatial scales. Shifts in abiotic conditions owing to agro-forestry practices, such as changes in soil nutrient and microclimatic conditions, are commonly identified as primary mechanisms for these changes. Yet agroforestry systems are also likely to influence litter decomposition rates through their influence on morphological and chemical traits of crop leaves and residues. However, explicit evaluations of how agroforestry management results in intraspecific leaf trait plasticity, and how this variation then links with litter decomposition, remain sparse. Using a 30-year-old agroforestry system intercropped with short-rotation willow (Salix dasyclados), we i) examined the effects of agroforestry management (i.e. N-2-fixing trees vs. non N-2-fixing trees vs. monoculture) and distance from tree row on key crop (willow) leaf morphological (i.e. leaf area, specific-leaf area [SLA]) and chemical (leaf carbon (C) and nitrogen (N) concentration) traits, and then assessed ii) how this intraspecific leaf trait variability influenced decomposition rates. Willow leaves within agroforestry treatments exhibited significantly greater leaf area (p = 0.01), SLA (p = 0.002), and leaf N (p = 0.001) than willow leaves in monoculture. Following a 56-day incubation, willow leaves from agroforestry treatments did not decompose more quickly than those in the monoculture. Our results indicate high intraspecific variation in willow leaf traits owing to agroforestry management. However, intraspecific variation in willow leaf traits did not correspond to statistically significant differences in decomposition rates. We show that variation in willow traits near tree rows in agroforestry systems translates into limited variability in agroeco-systems processes such as litter decomposition (as compared to monoculture arrangements). Our results have implications for optimal nutrient management in agroforestry systems.