부산대학교 도서관

A field experiment with 15 Miscanthus genotypes including M. × giganteus, M. sacchariflorus, M. sinensis and M. sinensis hybrids was conducted for 14 years at the experimental Ihinger Hof station of the University of Hohenheim in southwest Germany to evaluate interannual yield performance stability over 14 years of harvests of the different genotypes. In this article, a simple formula is presented which could be used to forecast late winter yields using morphological traits in autumn. The data obtained indicated a shorter establishment period to reach a yield plateau of M. × giganteus and M. sacchariflorus than M. sinensis hybrids and M. sinensis genotypes. The best performing genotype was M. × giganteus ( Gig-2, No. 16.21) with a mean spring harvestable yield of 14.1 t DM ha−1 year−1. A correlation analysis with climatic parameters revealed precipitation during the growing period as the key factor for high yields at this site. Likewise, but to a lesser degree, heat sum during the growing period was positively correlated with yields. It could be shown that precipitation/snow during the winter correlated with yield losses, until the harvest date in February/March. Phenological measurements indicated that a high yield potential of the tested genotypes is associated with either an absence of flowering or late flowering. Also, height of the plants and shoot diameter were indicators for high yield potential. Shoot density and plant height at senescence were found to be solid parameters to estimate harvestable biomass in late winter. Yield approximations with a mean accuracy of 80.9% for M.× giganteus genotypes were obtained using the newly developed equation. Yields of M. sinensis hybrid ( Sin-H7) were projected most accurately with this simple formula, resulting in a mean accuracy of 84.5%. [ABSTRACT FROM AUTHOR]