Predicting Miscanthus (M. x giganteus) and Switchgrass (P. virgatum) Productivity for the US Using a Semi-Mechanistic Crop Model.

Introduction: Developing domestic renewable sources of energy can be accomplished, in part, by the agricultural sector by the use of existing (maize and soybean), readily accessible (crop residue) or next generation biomass crops (e.g. Miscanthus, switchgrass, sorghum, etc.). Assessing the potential of next generation perennial biomass crops for delivering the needed feedstock can be greatly enhanced by the use of models which consider a solid connection between the underlying crop physiological process and the interaction with the environment. Objective: develop regional predictions for the potential contribution of perennial grasses as feedstock for the emerging bioeconomy using a semi-mechanistic model named BioCro. Methods: A novel semi-mechanistic model was developed based on WIMOVAC (http://www.life.uiuc.edu/plantbio/wimovac/homepage.htm). The new model, named BioCro, was implemented as an R package and it incorporated optimization, graphics and model diagnostics tools to the physiological, semi-mechanistic WIMOVAC. The NOAA gridded monolevel dataset was used which provides sub-daily meteorological variables for the last 28 years with a 32 by 32 km coverage for the U.S. The model was calibrated for Miscanthus using mainly field experiments carried out in the European Union. The model was then applied and validated with field experiments from Illinois. Some of the variables successfully simulated by the model were CO2 uptake, stomatal conductance, leaf area index, volumetric soil moisture and dry biomass. BioCro was also used to investigate the response of biomass productivity to different environments (year and locations). Results: Based on field experiments conducted in Illinois and the European Union, Miscanthus seems more promising in areas were the total annual precipitation averages are higher than 600mm. Mainly, due to its lower transpiration switchgrass is a more promising crop in dryer areas. Results II: In addition, sensitivity analysis showed that productivity is highly dependent on the underlying physiological parameters. In particular, while increasing the maximum rate of carboxylation has little impact on final yields, the initial response of CO2 uptake to light (i.e. quantum yield) can have a significant impact on biomass productivity.