The Role of Switchgrass in Avoiding the ‘Food vs. Fuel’ Dilemma

Second-generation bioenergy crops (also known as ‘next generation’ energy crops), often perennial plants as lignocellulosic sources from which biofuel is derived, have been championed as an avenue to avoid the food vs. fuel dilemma (Valentine et al. 2012). The use of lignocellulosic sources from second-generation bioenergy crops for ethanonl production has gained popularity, particularly since recent price increases in grains from first-generation bioenergy crops (e. g., maize) have been attributed, in the popular press and elsewhere (BRDI 2008), to diversions towards biofuels. Indeed, as recently as August 2012, the United Nations has urged the U. S. to reconsider its ethanol mandates. The latest U. S. government figures indicate that approximately 40% of the maize production in the U. S. is dedicated to ethanol production (Wise 2012). The use of lignocellulosic ethanol sources is therefore welcome in this regard, especially if second-generation bioenergy crops can be grown in areas that would not compete with current food and feed crop (e. g., maize, soybean) production (e. g., on "marginal" lands; see Cai et al. 2011), such that the latter could be utilized exclusively for direct and indirect (i. e., as animal forage) human consumption.

Switchgrass, Panicum virgatum L. (Poaceae), is one of a number of second-generation biomass energy crops that can be converted to lignocellulosic ethanol to the transportation sector. Other herbaceous candidates being considered worldwide for agronomic production include, but are not limited to, Miscanthus x giganteus, Miscanthus spp., sorghum, flaccidgrass, Napiergrass, sugarcane bagasse (residue), and maize stover (more commonly referred to as corn stover). Switchgrass is native to North America (Fig. 1), and is a perennial, obligate-outcrossing C4 grass capable

Figure 1. Current habitat suitability map for switchgrass (Panicum virgatum L.) based on the biology of the species rather than exclusively from herbarium specimens (from Barney and DiTomaso 2010). The map corresponds well with the current range of switchgrass, with darker shading indicating higher habitat suitability. Image used by permission of Elsevier.

of producing reliable biomass yields in agronomic production fields (Fig. 2) for approximately 10 yr after planting (U. S. Department of Energy 2011). Two ecotypes (also called cytotypes) have been noted in this species: with a few exceptions, ‘lowland’ ecotypes are predominantly tetraploid (2n = 4x = 36) and tend to comprise southeastern and coastal U. S. populations, while the ‘upland’ ecotypes are mostly octaploid (2n = 8x = 72) and tend to be more interior in their U. S. distribution (Zalapa et al. 2011; Zhang et al.

2011) . Switchgrass occasionally reaches "common" status in certain prairies (Howe et al. 2002; Baer et al. 2005; Haught and Myster 2008), marshes (Ford and Grace 1998), conservation reserve program (CRP) settings (Mulkey et al. 2006; Adler et al. 2009), and along roadsides and waste places (Radford et al. 1968). Elsewhere, it is typically not a large component of natural areas (Grelen and Duvall 1966) and hence is often found in much lower densities than those grown in agronomic settings.

Switchgrass is a leading cellulosic biofuel feedstock candidate owing to its high productivity (Sanderson et al. 1996). Even before President George W. Bush’s specific mention of switchgrass in his 2006 State of the Union address, switchgrass had been the target of extensive development as a bioenergy crop by the U. S. Department of Energy (DOE) and other entities (Sanderson et al. 1996), due in part to its high forage yields (Parrish and Fike 2005), which was one of its original utilizations. Switchgrass’ current favored status as a biomass-based renewable energy crop stems from its high yield and seed production under low-input conditions in monoculture at different regional cultivar testing fields in several states in the U. S. (Sanderson et al. 1996). This is complemented by substantial predicted biomass yields, particularly in the Midsouth (approaching 23 Mg/ha; Wulschleger et al. 2010). In terms of biomass and ethanol production, with

Figure 2. Agronomic switchgrass (cv. Alamo) field in east Tennessee, USA. Currently, fields > 50 ha exist at numerous farm sites in this region, with biomass intended to be utilized for lignocellulosic ethanol production. Photo credit: M. Nageswara-Rao.

Color image of this figure appears in the color plate section at the end of the book.

some exceptions, switchgrass is comparable to other second-generation herbaceous lignocellulosic bioenergy crops and first-generation crops (Table 1). If such yields are feasible in areas where maize and other row crops for human consumption are not being grown, switchgrass cultivation may indeed successfully avoid "food vs. fuel" controversies. However, aspects of associated landscape and land-use change, coupled with concurrent improvements, including multi-use strategies that could lead switchgrass indirectly into the food supply chain (e. g., first-cut for forage), will ultimately dictate the long-term sustainability of switchgrass as a bioenergy crop.

Table 1. Annual biomass and/or ethanol yields of switchgrass (Panicum virgatum L.) compared to herbaceous bioenergy crop alternatives in a sampling of recent studies where direct comparisons have been made. Most empirical work was based in the United States; Ra et al. (2012) was conducted in Japan. Refer to reference for specific growing, post-harvest,

or modeling conditions.

Switchgrass biomass yield

Alternative biomass yield

Reference

20 Mg/ha

Miscanthus x giganteus: 40 Mg/ha

Miguez et al. 2012a

8.6 Mg/ha

Napiergrass: 25 Mg/ha

Knoll et al. 2012b

~10 Mg/ha

Flaccidgrass:

comparable

Aravindhakshan et al. 2011b

~40 t/ha

Miscanthus x giganteus: ~90 t/ha

Dohleman et al. 2012

~9 t/ha

Napiergrass: ~52 t/ha

Ra et al. 2012bc

Switchgrass ethanol yield

Alternative ethanol yield

~4,000 L/ha

Maize: comparable

Varvel et al. 2008

45 gal/t biomass

Sugarcane bagasse: 52 gal/t biomass

Ewanick and Bura 2011c

aModel predictions

bOnly "best" alternative shown

cOnly "best" conditions shown