Recent findings from a collaborative research project will contribute to developing improved tools for breeding dairy cattle with better heat tolerance.
The current Australian Breeding Value (ABV) for Heat Tolerance is derived from genomics only – the DNA testing of animals. Adding more data sources, such as relevant animal performance records – referred to as phenotypes – to the ABV model should improve its reliability and accuracy.
Dairy UP researchers have developed phenotypic indicators of heat tolerance that can be added to the ABV model.
Published by DataGene, the Heat Tolerance ABV allows farmers to identify and breed animals with greater ability to tolerate hot, humid conditions with less impact on milk production.
DairyUP researcher, Dr Anna Chlingaryanm from the University of Sydney, said managing dairy cattle in hot, humid conditions was an increasing issue for the Australian dairy industry, with climate change and the trend towards intensification.
“Improving heat tolerance in dairy cattle has benefits for the business in maintaining production and reproductive performance in hot months, and animal welfare with improved cow comfort in hot conditions,” she said.
Dairy UP, in collaboration with DairyBio, DataGene and Charles Sturt University, has a suite of projects exploring ways to improve the Heat Tolerance ABV.
As part of this work, the team analysed on-farm data from innovative sensor technologies to better understand which animals are more susceptible to heat.
Data was collected from cows fitted with smaXtec rumen sensors to monitor core body temperatures on three pasture-based dairy farms. A total of 1429 animals were involved in this research, plus an additional 28 heifers from the University of Sydney farm in NSW.
To interpret the data accurately, researchers developed a water threshold model to account for water intake, allowing them to isolate the impact of drinking events on core body temperature. They found that drinking behaviour can serve as a reliable indicator of heat stress. These findings provide new phenotypic information that can complement genomic data and strengthen future calculation of the Heat Tolerance ABV.
The Dairy UP team also developed a hybrid artificial intelligence (AI)-based model (HAIM) to improve assessment of heat tolerance in dairy cattle, combining the predictive capabilities of machine learning algorithms with established statistical models. The model revealed patterns that might remain hidden when using traditional models alone, enhancing the understanding of heat tolerance in dairy cattle and the identification of more heat-tolerant animals.
Another project combines the genetic and performance data from animals with a variety of approaches including the sensor technologies from the Dairy UP project.
