Skip to main content
ARS Home » Research » Publications at this Location » Publication #114089


item Eigenberg, Roger
item Nienaber, John - Jack
item Hahn, G
item Brown Brandl, Tami

Submitted to: Energy Metabolism of Farm Animals Symposium Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 7/22/2000
Publication Date: N/A
Citation: N/A

Interpretive Summary: Heat production is an important measure of animal response. However, it has been limited to closely controlled laboratory facilities. Recent measures of body temperature have been shown to closely follow heat production measurements. Body temperature measurements can be easily taken on a frequent basis. This paper discusses the possibility of estimating heat production with body temperature measurements taken at 48-sec intervals. Results from growing/finishing pigs held at optimal and hot conditions were used to make the comparisons. The comparison showed that the body temperature measurements could be used to estimate heat production when adjustments were made for animal weight.

Technical Abstract: This study was conducted to determine the suitability of tympanic temperature (TT) as a surrogate for estimating swine dynamic heat production (HP). The effects of three thermal environments on growing- finishing swine were investigated using indirect calorimetry and TT measures. Six pigs (two groups of three with an approximate beginning weight of 45 kg) were individually penned in a thermoneutral (18C) environment. The pigs were moved individually into three calorimeters, each providing one of three thermal environments: 28C low air speed, 28C high air speed, and 18C air speed. The first group of pigs provided the data to develop the regression equations; the data from the second group was used to test estimates of HP based on TT and animal weight. At the completion of the experiment, the dynamic swine HP averages had ranged from a low of 125 watts/pig to a high of 257 watts/pig, with average TT ranging from 38.4C to 39.7C. All records of HP and TT were significantly correlated (P<0.01) through all treatments. Regression equations were developed to transform the TT record and animal weights into a HP estimate. An independent data-set (group 2) demonstrated mixed results, with average HP errors ranging from 0.6% to 7.7%, and accounting for 54 to 76% of the variability.