Ligno-cellulose composting : Case study on monitoring oil palm residuals

Oil palm empty fruit bunches (EFB) composting is increasingly being considered by tropical farmers as a worthwhile alternative to direct spreading in the field. EFB are ligno-cellulose residues comprising 46% cellulose and 16.5% lignin with a high C:N ratio ranging from 45 to 70. In hot regions, it is essential to control variations in the water content of the substrate. This case study set out to optimize the EFB composting procedure by monitoring two key factors, namely the nutritional balance of the biomass and its water content. The trial was conducted on a composting platform belonging to the PT SMART Tbk company (Indonesia) in windrows comprising shredded EFB watered weekly with oil mill effluents. The quantity of effluent to be added was calculated so as to systematically top up to 60% humidity. The initial C:N ratio of the EFB was reduced by urea and/or ripe compost applications (seeding). Application times and rates were studied on two windrows, one in an open area, the other in. an area sheltered by a root to prevent over-rapid drying. After ten weeks the compost could be considered ripe for all the treatments; at that stage, composting had reduced the volume and initial weight of the fresh EFB by 85% and 50% respectively. The combined action of a temperature rise and. windrow turning led to substantial water losses through evaporation, resulting in a sudden drop in windrow humidity at the start of composting. When rainfall was insufficient, effluent applications became essential. The total amount of effluent to be applied during composting (without rainfall) was around 3 m3 per initial tonne of EFB. This amount approximately corresponded to the standard effluents: EFB ratio in the oil mill. Inoculation of the mass of EFB to be composted with ripe compost had a significant effect on the speed with which the mixture was reduced. A urea application a fortnight after the start of fermentation seemed to be beneficial. The resulting compost had a good agronomic value. However, the mineral balance was considerably affected, as the nutrients provided by the effluents were poorly retained by the substrate, and partially lost in percolation water following the weekly watering operations. For instance, almost 50% of the phosphorus, 70% of the potassium, 45% of the magnesium and between 10% and. 20% of the calcium theoretically applied were lost ten weeks into the trial. Better distribution of the effluent applications, combined with a system to recover the leachi.ngs, should substantially reduce these losses, while maintaining suitable humidity for microbial degradation. (Résumé d'auteur)