|Brown Brandl, Tami|
|Jones, David - University Of Nebraska|
Submitted to: American Society of Agricultural and Biological Engineers
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/3/2011
Publication Date: 4/20/2011
Citation: Brown Brandl, T.M., Jones, D.D. 2011. Feedlot cattle susceptibility to heat stress: an animal specific model. American Society of Agricultural and Biological Engineers. 54(2):583-598.
Interpretive Summary: Animal response to heat stress originates from a combination of three areas: weather, the individual animal characteristics, and the management of the animals. A model was developed to summarize the individual animal characteristics into a single value of animal susceptibility. The model was developed using novel modeling techniques to summarize 11 different characteristics. The model was tested using a team of experts in the field of cattle heat stress. These experts were asked to assess 10 hypothetical animals for susceptibility to heat stress. This paper provides details on the type of model chosen, the development of the model, and the testing of the model.
Technical Abstract: The extreme effects of heat stress in a feedlot situation can cause losses exceeding 5% of all the cattle on feed in a single feedlot. These losses can be very devastating to a localized area of feedlot producers. Animal stress is a result of the combination of three different components: environmental conditions, animal susceptibility, and management. This paper describes the development of a model to predict individual animal susceptibility to heat stress. The model utilizes a hierarchal knowledge-based fuzzy inference system with 11 animal characteristics (color, sex, species, temperament, hair thickness, previous exposure, age, condition score, previous cases of pneumonia, previous other health issues, and current health) to predict susceptibility to heat stress. In an attempt to validate the model, a team of experts were asked to assess the susceptibility to heat stress of 10 hypothetical animals. The output of the model was tested against the experts’ opinions. The validation equation had a slope of 1.05 ± 0.05 with an intercept of 0.06 ± 0.03 and a R2 of 0.86. The opinions of the experts were also compared to the model output by comparing the class of stress susceptibility of each of the 10 hypothetical animals. The model and the experts agreed perfectly on 6 of the 10 animals. Further, the model prediction and the experts’ opinions deviated no more than one class on the remaining 4 animals. This exercise revealed that there was good agreement between the model output and the experts’ opinions.