(187d) Analysis of Marginal Land Definition on Integrated Landscape Design and Biofuel Supply Chain Network Design Optimization

Land available for lignocellulosic feedstock production is unevenly distributed across the landscape and depends heavily on the criteria used to classify it. Additionally, the landscape-scale decisions of where to plant bioenergy crops and how to manage them may have a large effect on the efficiency of the integrated biofuel supply system. Therefore, a detailed and integrated analysis of the landscape design and biofuel supply chain network design is needed to better understand the trade-offs among different types of marginal land and how the supply chain can be designed alongside the landscape to find holistically sustainable and economically beneficial outcomes for the large-scale production of biofuels.

In this study, we use the SALUS (systems approach to land-use sustainability) process-based model to generate spatially explicit, high-resolution estimates of biomass yields and soil organic carbon (C) changes under switchgrass (Panicum virgatum). We then use an integrated mixed-integer linear programming approach to simultaneously optimize the landscape design and biofuel supply chain network design for switchgrass grown on alternative definitions of marginal land. These include non-arable lands (i.e., Land Capability Classifications V-VIII) and parcels identified as historically abandoned and recently abandoned in previous studies.

We find that the differences in yield distribution, soil organic C sequestration distribution, and overall available land area, lead to very different optimal supply chain configurations and policies for landscape design (establishment and management), and that C market valuation ($/MgCO₂e) significantly influences what lands are considered favorable to establish bioenergy crops. With a higher cost of C, biomass may be planted on lands better suited to sequestering C located farther from a biorefinery despite the increase in transportation costs. Furthermore, by designing the landscape and supply chain simultaneously and at a high resolution, the model finds that integrated ‘win-win’ solutions are possible such that a significant amount of C negative biofuel can be produced for only a modest increase in biofuel price when compared to coarse resolution models or independently optimized landscapes and supply chains.