Table 1.
Species and sources of faecal samples used in NIRS calibration equations to predict faecal nitrogen.
Table 2.
Summary of nitrogen content in faecal samples used in calibration and validation sets.
The results are expressed in % of dry matter.
Fig 1.
Near infrared reflectance spectra of faecal samples from five herbivorous mammalian species used to build a multispecies calibration for faecal nitrogen.
Table 3.
Calibration and validation statistics of monospecies and multispecies calibrations used to determine the faecal nitrogen content (%) in faecal samples by NIRS analysis.
Table 4.
Validation statistics using the multispecies equation to predict the faecal nitrogen content (% of dry matter) in faecal samples of each species.
Table 5.
NIRS predicted faecal nitrogen contents (% of dry matter) in faecal samples and validation statistics using the multispecies equation.
Fig 2.
Relationships between faecal nitrogen (FN) predicted by NIRS and FN estimated by the Dumas dry combustion method in five herbivorous mammal species.
Table 6.
Model selection to explore whether the relationships between faecal nitrogen (FN) predicted by NIRS and faecal nitrogen estimated by the Dumas dry combustion method (reference method) varied among species.
Fig 3.
Bland and Altman plot of the difference between mono- and multispecies NIRS calibrations to predict the faecal nitrogen (FN expressed on a % dry matter basis) in five herbivorous mammal species.
The mean of differences (d) between mono- and multispecies calibration equations are represented by a dotted dark-blue line, whereas limits of agreement (d ± 1.96sd) are represented by dashed lines. Confidence intervals at 95% are shown by shaded areas.