The use of marginal lands for biofuel production has been proposed as a promising solution for meeting biofuel demands while avoiding food-feed-fuel conflicts. However, uncertainty surrounds whether marginal lands can be reliably located, as well as their inherent biofuel potential and the possible environmental impacts. We developed a quantitative approach that integrates high-resolution land cover and land productivity to classify productive croplands and nonarable marginal lands in a nine-county region in southern Michigan. The classified lands were then examined with the spatially explicit modeling framework using the Environmental Policy Integrated Climate (EPIC) model to estimate net energy (NE) and soil organic carbon (SOC) changes associated with the cultivation of different annual and perennial production systems. Simulation results suggest that biofuel production systems underperform on marginal lands when compared to productive croplands. However, we found perennial grasses could perform better than annual crops. Hence, when growing perennial bioenergy crops on marginal lands instead of productive croplands, less additional land (about 0.09 ha per each hectare planted) would be needed to achieve the same NE than if growing annual bioenergy crops (additional 0.17 ha per hectare planted). Miscanthus ( × ) and switchgrass ( L.) can produce 112.43 and 74.61 GJ ha yr NE, respectively, and have the potential to sequester, on average, 0.59 and 0.23 Mg C ha yr SOC, respectively. Notably, simulation results indicate substantial variability of the NE and SOC storage potential across the study region. Thus, although perennial energy crops are promising options for biofuel production on marginal lands, given the large spatial variability, regional- and site-specific management strategies are required for sustainable biofuel production.
Copyright © by the American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America, Inc.