Developing a heat load index for lactating dairy cows

Abstract

The temperature humidity index (THI) has been extensively used in the Australian dairy industry as an indicator of heat load conditions. However, there are limitations to the THI, where it does not account for solar radiation or wind speed. In addition, the THI has not been formulated in conjunction with physiological data. Thus, it is not apparent whether the THI provides the best prediction for impact of heat load on lactating dairy cattle. The aim of the present study was to develop a dairy heat load index (DHLI), based on the physiological responses of lactating dairy cows to environmental conditions. The study was undertaken at The University of Queensland, Gatton Campus, Australia, over three summers and two winters. Observations were conducted four times daily at 0800 hours, 1200 hours, 1400 hours and 1700 hours. Weather data were obtained every 10 min from an onsite, automated weather station. Panting score data were used to calculate a mean panting score of the herd. Developing the DHLI encompassed three different modelling techniques, including (1) linear regression, (2) broken stick regression and (3) non-linear logistic regression. Comparisons between the models indicated there was an increase in accuracy when using non-linear logistic regression (R2 = 0.542). The DHLI model developed was based on mean panting score, black globe temperature (°C) and relative humidity. By utilising the black globe temperature, the DHLI is able to incorporate the combined effects of ambient temperature, relative humidity, solar radiation and wind speed in a single unit measure. By combining climatic conditions and a physiological response, i.e. panting score, the DHLI model may become a more accurate prediction of heat load for lactating dairy cows. The DHLI produces a unit value between 0 and 100. When DHLI is 0, there would be no panting and, when DHLI is 100, all cows within the herd would be panting.