Tools are under development to assess biomass yield and composition and their related traits in Miscanthus.

© Inra — E. Rosiau

For biomass production, Zub et al. [26] showed that biomass yield during the second year could be used to predict the biomass yield of the third year, whatever the harvest date. This correlation requires further investigation over a longer period and on a wider sample of genotypes to determine whether the yield difference between harvest dates is the same for juvenile and mature phases of the crop.

In addition, Zub et al. [150] found that the aboveground volume including the stem number, the stem diameter and the plant height was a good predictor of plant biomass yield. Within genotypes, strong positive relationships were observed between biomass yield and the aboveground volume regardless of crop year (equal to 0.70 and 0.82 for autumn and winter harvests during the second year, respectively).

For biomass production, Hodgson et al. [59] developed Near-Infrared Reflectance Spec­troscopy (NIRS) calibration models for biomass quality to determine acid detergent lignin (ADL), acid detergent fiber (ADF), and neutral detergent fiber (NDF) from sample spectra of M. x giganteus, M. sacchariflorus and M. sinensis. The corresponding concentrations were predicted with a good degree of accuracy based on the coefficient of determination (values of R2 being higher than 0.80), standard error of calibration, and standard error of cross-validation values.

Regarding the statistical analysis, the residual model error in the analysis of variance model needs to be small to enable the comparisons between genotypes for quantitative traits. In Miscanthus, the residual term was high during the second and third years of the crop (Zub et al. [150]) and could hamper inter-genotypic comparisons for traits such as aboveground biomass yield or related traits. Without more plots or samples (it is indeed important for the
breeder and the producer to save place and cost), one way for reducing the residual term is to take into account intra-genotypic competition effect in the statistical model [150]. As it implies observations at the plant level, intragenotypic effect assessment requires easy — to-measure variables, such as the stand volume as a predictor of the aboveground biomass [150,151].

Modeling of emergence and plant growth using three and four-parameter logistic func­tions and the Gompertz function were tested to best describe the dynamics of crop emer­gence and of plant growth. The Gompertz function was found to be the best to estimate emergence dynamics while four-parameter logistic to estimate growth dynamics [151].

In contrast to the genetic tools, where intensive research has been conducted during the last decade, the phenotypic evaluation represents a bottleneck for Miscanthus, since high throughput tools are still required today.

4.4 Conclusion

The development of renewable energy sources is being investigated across the world and there is a growing demand for bioenergy feedstock that does not compete with food production and which has a low environmental impact. Miscanthus is developing as a serious player in the renewable energy sector. To realize its potential, new varieties are needed with the productivity and processing traits required for bioenergy production. This will require a full exploration of the genetic resource base of Miscanthus and its related species and the development of appropriate genetic tools. Many such projects are in progress throughout the world and the next decade is likely to deliver exciting new developments for Miscanthus as a renewable energy source.