Water, nitrogen and carbon balance of bioenergy crops: impact of crop species and cropping practices

Second generation biofuels could provide renewable energy to the transportation sector while mitigating climate change. However, their greenhouse gas, energy and environmental balances will probably depend on the feedstock used for their production. Bioenergy crops that could be used for second generation biofuels will have to fulfil several requirements, including high productivity, low input requirements, and low environmental impacts. The objective of this work was to assess the water, N and C balances at the plot scale for various bioenergy crops with different management. The study is based on a long term field experiment, called 'Biomass and Environment', established at the INRA experimental station in Estrees-Mons, northern France. This experiment includes two perennial C4 crops (Miscanthus x giganteus and switchgrass), two semi-perennial forage C3 crops (fescue and alfalfa) and two annual C4/C3 crops (fibre sorghum and triticale). It compares two nitrogen treatments and two dates of harvest of perennial crops: early (October) or late harvest (February). Measurements have been carried out on: i) biomass production; ii) soil water stocks, monitored continuously during 7 years; iii) root depth and density; iv) drainage and nitrate concentration in drained water, assessed from soil water and mineral N content measurements (in mid-autumn and late winter) and using the STICS model; v) soil organic carbon (SOC) stocks in 2006 and 2011-2012; vi) the fate of ¹⁵N-labelled fertiliser applied during 4 or 5 successive years. Thanks to their deep rooting system, perennial and semi-perennial crops consumed more water than annual crops. The amount of drained water was lower under semi-perennial than annual crops (64 vs. 133 mm yr^-1 average over 7 years), despite an equivalent biomass production. It was intermediate under perennial crops (56-137 mm yr^-1) and negatively correlated to biomass production, itself depending on crop species and N rate. Nitrate concentration in drained water varied between 2 and 23 mg l^-1. It was generally lower under perennial than other crops, except for miscanthus on the first year of measurement. SOC stocks increased markedly over time under semi-perennial crops (+0.93 t C ha^-1 yr^-1), whereas no significant change occurred under perennial and annual crops. The ¹⁵N recovery in the harvested biomass was lower for perennial than other crops, particularly when harvested late, but compensated by a higher ¹⁵N recovery in belowground organs and soil. The overall ¹⁵N recovery in the soil-plant system was 69% in perennials, 61% in semi-perennials and 56% in annual crops, suggesting that important fertiliser losses occurred through volatilisation and denitrification. In our pedo-climatic conditions, the C4 perennial crops performed best in terms of production, water and nitrogen use efficiency, and nitrogen losses towards the groundwater and the atmosphere. However, only semi-perennial crops yielded in SOC sequestration. (author)