Predicting diet quality of donkeys via fecal-NIRS calibrations
Academic Article
Overview
Identity
Additional Document Info
Other
View All
Overview
abstract
Successful applications of fecal-near infrared reflectance spectroscopy (fecal-NIRS) techniques have been reported for ruminant animals. Information on the ability of fecal-NIRS to characterize diet quality in equines is lacking. The objective of these studies was to determine the potential of fecal-NIRS to predict diet quality of free-grazing equines. Two independent in vivo feeding trials, one in Texas (United States) and one in Kenya, were conducted to generate paired samples of diet chemistry:fecal spectrum (D:F). Using 20 female donkeys (Equus asinus), 14 (10 US, 4 Kenya) in vivo pen feeding trials were conducted to generate 140 (100 US, 40 Kenya) D:F paired samples. Over 25 species of forage and crop residues ranging from 3.3% to 21.4% crude protein (CP) were used to blend unique diets. Three CP predictive equations based on paired samples from US alone, Kenya alone, US + Kenya combined, and one predictive equation for digestible organic matter (DOM) from US alone were developed. The standard errors of calibration (SEC) and R2 values were 0.77 and 0.97, 0.97 and 0.95, and 0.88 and 0.90, respectively, for the US, US + Kenya, and Kenya CP equations. The US DOM equation resulted in an SEC of 2.58 with a corresponding R2 of 0.60. Validation of the US CP equation using an independent dataset resulted in standard error of prediction (SEP) and R 2 of 1.79 and 0.82, respectively, indicating acceptable predictive ability. The validation results (SEP = 15.56) for the US DOM equation were not satisfactory. We calibrated and validated fecal-NIRS equations to predict the DOM and CP contents of diets for donkeys. Crude protein content of diets was predicted with acceptable levels of accuracy, but prediction of diet digestibility was less successful. The degree of accuracy obtained for CP equations indicated that fecal-NIRS can be considered as a tool for routine nutritional management of donkeys.
