Grain agriculture and the end of the fossil fuel era

Abstract

Listen

Translate article

First paragraphs: A great deal of attention is currently focused on how agriculture in highly industrialized coun­tries contributes to greenhouse gas (GHG) emis­sions, and how certain farming innovations might curb the emissions of nitrous oxide and methane and draw down carbon dioxide from the atmos­phere. What is not being discussed is how agricul­ture in general, and grain agriculture in particular, will need to change as society phases out its dependency on fossil fuels in order to achieve carbon (C) neutrality. Over the last century in the U.S., the number of farmers on the land declined by about 66%, in close proportion to the increase in average farm size (U.S. Department of Agriculture Economic Research Service [USDA ERS], n.d.). Integral to these trends has been the simplification of farming systems in which practices like fertility-generating rotations have been replaced with lower-diversity monocultures maintained by applications of fossil fuel–based fertilizers and pesticides (Crews & Peoples, 2004). Between fossil fuel–powered mechanization and fossil fuel–based input intensi­fication, the energy used by farmers to grow maize in the state of Nebraska is 99.7% from fossil fuels and 0.3% human labor (Grassini & Cassman, 2012; Pimentel & Pimentel, 2008). Even organic farming systems often require prodigious fossil fuel inputs with intensive tractor tillage, manure hauling, and mechanical harvesting (Smith et al., 2015). In con­trast to our modern grain-producing agroeecosys­tems, ancestral agroecosystems and natural ecosys­tems of all types—forests, grasslands, deserts, tundra—have remained productive for millennia with no fossil fuel inputs. In this commentary I explore the dependency of grain agriculture on fossil fuel use in the U.S. set in a global context, and approaches for reducing this dependency, including a shift to perennial polycultures that rely on ecological intensification in place of energy-intensive inputs. . . .