Effects of measurement technique and sample preparation on NIR spectroscopy analysis of livestock slurry and digestates

Near infrared spectroscopy (NIRS) was used to analyse livestock slurries and digestates. The purpose was to evaluate the influence of sample preparation, reading set-up and sample temperature during NIR scanning on digestates and livestock slurry samples. To obtain accurate and reproducible values of total solids (TS), total kjeldahl nitrogen (TKN), total ammoniacal nitrogen (TAN) and volatile fatty acids (VFA) contained in dairy and pig slurry and digestate, a total of 36 samples were analysed from different farms in Lombardy, Italy. Three sample preparations (filtration, homogenisation and a raw control), two reading set-ups (petri dish and optical fibre) and three sample temperatures (10, 25 and 35 °C) were tested during NIR scanning. Results showed that analysis of livestock slurries and digestates by NIR spectrophotometry is influenced by sample preparation. Both filtered and homogenised samples generally showed higher correlations (r2) and ratio of standard error of performance to standard deviation (RPD) (0.79 < r2 > 0.98 and 2.26 < RPD > 6.99 for filtered samples; 0.30 < r2 > 0.97 and 1.24 < RPD > 6.31 for homogenised samples) than raw samples (0.03 < r2 > 0.95 and 1.05 < RPD > 4.73), but the better sample preparation was filtration. Spectral acquisition through petri dishes was slightly more accurate than through optical fibre. No observable effects on spectral analysis were caused by altering temperature in the range of 10-35 °C.