Department of Agricultural and Biological Engineering, Purdue University, West Lafayette, IN 47907, USA.
Department of Agricultural and Biological Engineering, Purdue University, West Lafayette, IN 47907, USA; Department of Civil Engineering, Indian Institute of Technology Madras, Chennai, 600036, India.
J Environ Manage. 2018 Apr 15;212:198-209. doi: 10.1016/j.jenvman.2018.01.060. Epub 2018 Feb 22.
Biofuel has emerged as a substantial source of energy in many countries. In order to avoid the 'food versus fuel competition', arising from grain-based ethanol production, the United States has passed regulations that require second generation or cellulosic biofeedstocks to be used for majority of the biofuel production by 2022. Agricultural residue, such as corn stover, is currently the largest source of cellulosic feedstock. However, increased harvesting of crops residue may lead to increased application of fertilizers in order to recover the soil nutrients lost from the residue removal. Alternatively, introduction of less-fertilizer intensive perennial grasses such as switchgrass (Panicum virgatum L.) and Miscanthus (Miscanthus x giganteus Greef et Deu.) can be a viable source for biofuel production. Even though these grasses are shown to reduce nutrient loads to a great extent, high production cost have constrained their wide adoptability to be used as a viable feedstock. Nonetheless, there is an opportunity to optimize feedstock production to meet bioenergy demand while improving water quality. This study presents a multi-objective simulation optimization framework using Soil and Water Assessment Tool (SWAT) and Multi Algorithm Genetically Adaptive Method (AMALGAM) to develop optimal cropping pattern with minimum nutrient delivery and minimum biomass production cost. Computational time required for optimization was significantly reduced by loose coupling SWAT with an external in-stream solute transport model. Optimization was constrained by food security and biofuel production targets that ensured not more than 10% reduction in grain yield and at least 100 million gallons of ethanol production. A case study was carried out in St. Joseph River Watershed that covers 280,000 ha area in the Midwest U.S. Results of the study indicated that introduction of corn stover removal and perennial grass production reduce nitrate and total phosphorus loads without compromising on food and biofuel production. Optimization runs yielded an optimal cropping pattern with 32% of watershed area in stover removal, 15% in switchgrass and 2% in Miscanthus. The optimal scenario resulted in 14% reduction in nitrate and 22% reduction in total phosphorus from the baseline. This framework can be used as an effective tool to take decisions regarding environmentally and economically sustainable strategies to minimize the nutrient delivery at minimal biomass production cost, while simultaneously meeting food and biofuel production targets.
生物燃料已成为许多国家的主要能源来源。为了避免因谷物乙醇生产而导致的“粮食与燃料之争”，美国已通过相关规定，要求到 2022 年，生物燃料生产主要使用第二代或纤维素生物原料。农业残留物，如玉米秸秆，目前是纤维素原料的最大来源。然而，为了补充因清除残留物而损失的土壤养分，增加农作物残留物的收割可能会导致更多肥料的使用。或者，可以引入较少依赖肥料的多年生草类，如柳枝稷（Panicum virgatum L.）和芒属（Miscanthus x giganteus Greef et Deu.），作为生物燃料生产的可行原料。尽管这些草类在很大程度上减少了养分负荷，但高生产成本限制了它们作为可行原料的广泛采用。尽管如此，仍有机会优化原料生产，在满足生物能源需求的同时改善水质。本研究提出了一种基于土壤和水评估工具（SWAT）和多算法遗传自适应方法（AMALGAM）的多目标模拟优化框架，以开发具有最低养分输送和最低生物质生产成本的最优种植模式。通过松散耦合 SWAT 与外部溪流溶质输运模型，大大减少了优化所需的计算时间。优化受到粮食安全和生物燃料生产目标的限制，以确保粮食产量不减少 10%，乙醇产量至少达到 1 亿加仑。在美国中西部的圣约瑟夫河流域进行了案例研究，该流域占地 28 万公顷。研究结果表明，引入玉米秸秆清除和多年生草类生产可在不影响粮食和生物燃料生产的情况下减少硝酸盐和总磷负荷。优化运行产生了一个最优的种植模式，其中流域面积的 32%用于清除秸秆，15%用于种植柳枝稷，2%用于种植芒属。与基线相比，最优方案可使硝酸盐减少 14%，总磷减少 22%。该框架可作为一种有效工具，用于制定环境和经济可持续战略的决策，以在最小生物质生产成本下最大限度地减少养分输送，同时满足粮食和生物燃料生产目标。
