Fecal-FT-NIRS as a Noninvasive Tool for Assessing Diet Quality of Mediterranean Deer

Abstract

In order to assess the diet quality of two Mediterranean deer species we developed and validated a Fourier transform near-infrared diffuse reflectance spectroscopy methodology on feces (Fecal-FT-NIRS) for the determination of acid detergent fiber (ADF), neutral detergent fiber (NDF), lignin, C:N ratio, and enzymatic digestibility of organic matter (EDOM). We used rumen contents and fecal samples from 149 red deer (Cervus elaphus hispanicus) and 111 fallow deer (Dama dama) from southeast Spain (n = 520 observations). Spectra from the feces were related with rumen conventional chemical analysis through chemometric regression with partial least-squares (PLS). Specific predictive equations from red and fallow deer data separately were generated, as well as merged equations after pooling all deer samples. All the predictive equations had a high linearity with high correlation coefficients (r = 0.8 – 0.99). The selected equations had a reliable accuracy considering the root-mean-square errors of prediction (RMSEP), calibration (RMSEC), and cross-validation (RMSECV) in relation to the range of values for which the NIRS calibration was set for each parameter. Broad-based equations from combined samples were demonstrated as being useful for all nutritional parameters determination in red and fallow deer simultaneously. Equations obtained for the red deer data were also successfully applied to fallow deer and vice versa for NDF, ADF, C:N, and lignin determination, while for EDOM assessment the specific equations for each species were more accurately applied. Once validated, the Fecal-FT-NIRS technique can be considered as a suitable noninvasive tool for monitoring deer diet quality variations in Mediterranean environment. This method has the possibility to overcome interspecific barriers of direct fecal analysis by using rumen digesta as its reference method.