Prediction of rectal temperature in Holstein heifers using infrared thermography, respiration frequency, and climatic variables.

The objective of this study was to develop an equation to predict rectal temperature (RT) using body surface temperatures (BSTs), physiological and climatic variables in pubertal Holstein heifers in an arid region. Two hundred Holstein heifers were used from July to September during two consecutive summers (2019 and 2020). Respiratory frequency (RF) was used as a physiological variable and ambient temperature, relative humidity and temperature-humidity index as climatic variables. For the body surface temperatures, infrared thermography was used considering the following anatomical regions: shoulder, belly, rump, leg, neck, head, forehead, nose, loin, leg, vulva, eye, flank, and lateral area (right side). Initially, a Pearson correlation analysis examined the relationship among variables, and then multiple linear regression analysis was used to develop the prediction equation. Physiological parameters RT and RF were highly correlated with each other (r = 0.73; P˂0.0001), while all BST presented from low to moderate correlations with RT and RF. BST forehead temperature (FH) showed the highest (r = 0.58) correlation with RT. The equation RT = 35.55 + 0.033 (RF) + 0.030 (FH) + ei is considered the best regression equation model to predict RT in Holstein heifers in arid zones. This decision was made on the indicators R2 = 60%, RMSE = 0.25, and AIC = 0.25, which were considered adequate variability indicators.