Jump to Main Content
Application of the Smith-Hazel Selection Index for Improving Biomass Yield and Quality of Switchgrass
- M. Z. Z. Jahufer, M. D. Casler
- Crop science 2015 v.55 no.3 pp. 1212-1222
- Panicum virgatum, biofuels, biomass production, combustion, data collection, dry matter accumulation, economic impact, energy crops, ethanol, ethanol production, feedstocks, fermentation, field experimentation, forage, genetic correlation, genetic techniques and protocols, genetic variation, heat, heritability, lignin, selection index, selection response, Wisconsin
- Apart from the importance of switchgrass (Panicum virgatum L.) as forage for livestock, it is useful as a high-value cellulosic biofuel feedstock. Breeding for a biofuel crop is complicated by the existence of multiple platforms for conversion of biomass to energy. Our main objective was to investigate the relative merits of single-trait selection, correlated response to selection, and Smith-Hazel-index-based selection for the following traits: biomass dry matter yield (YLD), ethanol (ETOH), Klason lignin (KL), and high heating value (HHV). The genetic analysis was based on a 2-yr data set generated from evaluation of 144 half-sib (HS) families in sward-plot trials at Arlington and Marshfield, WI. There was significant (P < 0.05) additive genetic variation among the families for all traits and for family × site interaction for the traits HHV and YLD. The estimates of narrow sense heritability on a HS family mean basis ranged from 0.37 for YLD to 0.51 for KL. Genetic correlation of YLD with ETOH, HHV, and KL were 0.38, 0.27, and 0.01, respectively. The index constructed to increase YLD and ETOH and reduce KL was most successful for a fermentation platform. This index enabled identification of families for enhancing ethanol production that would have been missed if selection was based solely on YLD. The index weighted to increase YLD and KL best suited a combustion platform. Both of these two indices had economic impacts superior to any other selection index evaluated.