NIR calibration models to estimate the methane production potential of fresh grass and maize

The estimation of the methane production potential of different plant substrates helps to increase the efficiency of the process management in biogas production of energy crops. . Therefore, a NIRS based method for a rapid estimation of the methane production potential of grasses and silage maize was developed.

The spectrometer NIRS5000 (FOSS, ground samples) and PSS 1720 (Polytec, fresh samples) were used and calibrations were developed using partial least squares regression.

The methane production potential of dry ground samples of grasses and silage maize can be estimated with an error of about 4 LN/kg dry matter and a coefficient of determination of 0.97. Using fresh chopped material of grasses that has been analyzed stationary with the atline equipment the estimation was possible with an error of 8 LN/kg dry matter and a coefficient of determination of 0.80 being sufficient for screening purposes. The calibration models of online spectra taken on harvesters allow the estimation of the methane production potential with a precision comparable with the stationary measurement.

For silage maize, no applicable calibration model could be developed with sufficient accuracy since the available variance of the silage maize methane production potential of 7 – 9 LN/kg dry matter was too low.

The methane production potential can be estimated more precisely with the developed NIR based method than with any existing batch tests. To further improve the precision and accuracy of the calibrations, the existing database may need to be expanded by material with a higher variability in methane production potential in maize.