Evaluation of NDF, ADF and Acid Detergent Insoluble Ash to Predict Forage Selection Among Heifers Grazing Pastures Containing a Binary Mixture of Brassicas and Grass

Diet selection is impacted by a myriad of factors among grazing cattle. A greater understanding of diet selection can allow improved estimates of performance, digestion and nutrient balance. However, few data are available related to diet selection by cattle grazing cool season annual forages. Often, estimates of diet selection among cattle grazing mixed pastures require complex laboratory analyses and can be variable. However, innate markers (e.g., NDF, ADF, acid detergent insoluble ash) may allow for prediction of diet selection in cattle grazing pastures that contain limited amounts of plant species. Diet samples were collected from ruminally cannulated heifers grazing a binary mixture of annual ryegrass (Lolium perenne) and brassica (Raphinus sativus, forage radish and Brassica rapa, purple top turnip) at 3 stocking rates (1.7, 2.3 and 2.9 AUM × ha^-1). Forage samples were clipped 11 d prior to grazing and masticate was collected 2-, 24-, and 46-d after cattle began grazing pastures. Subsequently, forage samples and masticate were analyzed for NDF, ADF and acid detergent insoluble ash (ADIA). Estimates of brassica or grass intake did not differ in response to increased stocking rate when intake was estimated with NDF (P ≥ 0.52), ADF (P ≥ 0.48), or ADIA (P ≥ 0.34). However, intake predictions using NDF and ADIA resulted in differences in time (quadratic; P < 0.01) but ADF did not (P ≥ 0.61). Interestingly, estimates of brassica and grass intake were 52% greater and 32% less when predicted using ADF in comparison to NDF. Similarly, estimates of brassica and grass intake using ADF were 35% greater and 38% less in comparison to ADIA. The overall pooled coefficient of variation for estimates of grass and brassica intake was 55%, 60% and 65% for NDF, ADF and ADIA, respectively. It seems logical that NDF may allow for improved estimates of brassica and grass intake because true fiber has no potential for contamination by endogenous sources. Improvement among non-destructive technologies that allow rapid analysis of samples (e.g., near-infrared reflectance spectroscopy) may allow further improvements in estimates of diet selection among cattle grazing binary mixtures of annual forages.