The first objective of this study was to validate the feeding behavior measurements from a radio frequency electronic system (GrowSafe (TM) System Ltd., Airdrie, AB, Canada) and examine the software sensitivity to different parameter settings (MPS) to quantify feeding behavior traits. Data was continuously recorded 24 h per day using the GrowSafe™ system for 32 heifers over 81-d. Ten animals were randomly selected and evaluated over 6-d using time-lapse video recordings. Different parameter settings (MPS) from the electronic system (GrowSafe (TM)) used to record feeding behavior data, bunk visits (BV) frequency and BV duration, were compared with the observed (video) values.
The second objective of this study was to quantify meal criterion; examine the associations between feeding behavior traits, performance, and feed efficiency; and the effects of breed type on feed efficiency (residual feed intake - RFI) and feeding behavior traits in heifers fed high-grain diets.
Results from study one demonstrated that the GrowSafe (TM) system 4000E could accurately predict BV and meal data compared to observed data. The 100 s, used for the maximum duration between consecutive EID recordings to end an uninterrupted BV, was the appropriate MPS to predict BV frequency and duration, and meal frequency and duration compared to observed data using the GrowSafe (TM) 4000E system. The system’s ability to detect the animal’s presence or absence at the feed bunk was 86.4 and 99.6 percent, respectively.
Results from the second study demonstrated that the meal criterion for heifers fed high-grain diets was 13.8 min. The 4 methods to calculate meal criterion demonstrated no differences in results of frequencies and durations of meal and the number of bunk visits per meal. Similar phenotypic correlations were found between the feeding behavior traits with RFI derived from the base model or with adjustments for final back fat. The adjustment of RFI to final back fat changed the RFI rank between breeds. The addition of feeding behavior traits to the RFI base model could accounted up to 40.4 percent of the variation in DMI not explained by ADG or MBW.
| Identifer | oai:union.ndltd.org:tamu.edu/oai:repository.tamu.edu:1969.1/ETD-TAMU-2010-08-8157 |
| Date | 2010 August 1900 |
| Creators | Mendes, Egleu Diomedes Marinho |
| Contributors | Carstens, Gordon E., Tedeschi, Luis O. |
| Source Sets | Texas A and M University |
| Language | en_US |
| Detected Language | English |
| Type | thesis, text |
| Format | application/pdf |
Page generated in 0.0015 seconds