The first work-package of Logist’EC focused on feedstock production, the first step of the logistics chain, which is of prime importance since it contributes a major fraction of environmental impacts and economic costs. It therefore impacts the sustainability of the supply chain, but also the rest of the logistics since it determines the amounts, density and quality of available biomass around the biomass processing units. The type of energy crops, their management, as well as insertion in cropping systems determines yield potentials and biomass quality, possible options for biomass densification or transportation, and the profile of production costs, among others. Lastly, some aspects of environmental impacts of energy crops were still unclear at the outset of the projet: there were relatively few data on the effects of perennial grasses or woody species on soil organic carbon, especially after the destruction of the stands.
A this first task within this work-package aimed at synthesizing current knowledge about biomass yields, quality and the environmental impacts of various candidate biomass crops. A database including 858 yield data of 36 biomass crops from 28 scientific papers was elaborated. A meta-analysis aiming at ranking the yields of energy crops was carried out.It is detailed in the following deliverable while a summary is given below.
A statistical analysis based on direct and indirect comparisons was performed to compare the mean yield values of the species included in the database. Miscanthus x giganteus was significantly more productive than most of the other energy crops included in the database. Giant reed (Arundo donax) and Pennisetum purpureum were significantly more productive than Miscanthus x giganteus but both were studied at a limited number of sites. By contrast, Erianthus, Phragmites australis, Phalaris arundinacea, Miscanthus sacchariflorus and Miscanthus sinensis were the least productive species. This database and on-farm data of Miscanthus x giganteus gathered in the Bourgogne Pellets supply area were also used to develop a Bayesian statistical model to predict the yields of energy crops using site-specific measurements. Two applications were developed to predict (i) the future yields of Miscanthus x giganteus using past yield data, (ii) the yield of an energy crop in a given area from yield data collected in the same area, but for a different crop species. The first application was used to predict the future yields of Miscanthus x giganteus in the supply area of Bourgogne Pellets.