A physiological and biophysical model of coppice willow (Salix spp.) production yields for the contiguous USA in current and future climate scenarios

High-performance computing has facilitated development of biomass production models that capture the key mechanisms underlying production at high spatial and temporal resolution. Direct responses to increasing [CO2] and temperature are important to long-lived emerging woody bioenergy crops. Fast-growing willow (Salix spp.) within short rotation coppice (SRC) has considerable potential as a renewable biomass source, but performance over wider environmental conditions and under climate change is uncertain. We extended the bioenergy crop modeling platform, BioCro, to SRC willow by adding coppicing and C3 photosynthesis subroutines, and modifying subroutines for perennation, allocation, morphology, phenology and development. Parameterization with measurements of leaf photosynthesis, allocation and phenology gave agreement of modeled with measured yield across 23 sites in Europe and North America. Predictions for the continental USA suggest yields of 17Mgha(-1)year(-1) in a 4 year rotation. Rising temperature decreased predicted yields, an effect partially ameliorated by rising [CO2]. This model, based on over 100 equations describing the physiological and biophysical mechanisms underlying production, provides a new framework for utilizing mechanism of plant responses to the environment, including future climates. As an open-source tool, it is made available here as a community resource for further application, improvement and adaptation. Willows in short rotation coppicing are emerging as major sustainable feedstocks for bioenergy. As long-lived C3 crops, they will experience significant global change. Within the BioCro framework we have developed a new model for this crop that captures the key mechanisms of response to rising atmospheric CO2, temperature and precipitation to predict yield and yield stability across the USA. The model, is mechanistically rich and based on over 100 equations describing the physiological and biophysical mechanisms underlying production. It showed a good ability to predict recorded yields at a range of sites in Europe and N. America given soil and climate data. High yields, exceeding those of the perennial C4 grass feedstocks Miscanthus and switchgrass were predicted across much of the north-east USA, reaching >17 Mt ha-1 yr-1 at the best locations. Rising temperatures over this century decreased yields, an effect partially ameliorated by rising [CO2]. Using the BioCro framework, common to other perennial feedstocks, avoids confounding biological differences in species comparisons with differences in model structure and model assumptions. As an open-source tool, it is made available here as a community resource for further application, improvement, and adaptation.