Heat stress on dairy cows impacts more than just the quantity of milk produced – warming temperatures also reduce the fat and protein content of the milk, new research finds.
The study, accepted and published online May 29 in Environmental Research Letters, identifies a new impact of climate change on milk production, with researchers finding that the economic losses from a decrease in fat and protein in milk equal those of well-established, heat-related losses in yield. Farmers in the U.S. are compensated based on milk’s fat and protein content.
“The heat-induced dilution of these valuable milk components is happening a bit under the radar,” said senior author Ariel Ortiz-Bobea, associate professor in the Dyson School of Applied Economics and Management, in the Cornell SC Johnson College of Business. “When you account for the deterioration in milk composition, the economic loss ends up being of the same order of magnitude as the yield effect, so it just basically doubles the damage.”
Researchers, including first author and doctoral student Jeisson Prieto, used data from 2007 through 2016 on the milk production of around 6.5 million cows and compared that with weather data from the same period, with precision down to 2.5-mile grids across 43 states. The team found, aligning with previous research, that the amount of milk the cows produced dropped sharply at a certain threshold of heat and humidity. But the negative effects on milk composition began at lower temperatures and increased as temperatures rose.
“If it’s a day in the 60s or 70s, you don’t see any effect on yield, but the milk starts to get diluted gradually,” Ortiz-Bobea said. “Unlike the yield effect, which only happens in the summer, this is happening all the time.”
The team calculated farmers’ potential loss of revenue, finding that an average 10-point increase on the temperature-humidity index results in a 1.2% reduction in milk yield but a 2.8% reduction in revenue over a year, a $1.65 billion loss for an industry that produces 20% of U.S. animal products.
“It’s another headwind for dairy farmers,” Ortiz-Bobea said. “Milk prices are low, and farmers are struggling, and that usually leads to more consolidation, which changes the landscape in rural areas, literally and economically.”
The researchers also found little evidence of cows growing more heat-resilient – they saw almost no variation in heat-response across cows of different ages, farm sizes or regions; the only adaptation to heat was structural, or where in the country the industry chooses to farm, Prieto said.
“This is why we have a lot of cows in the northern, cooler regions, like New York and Wisconsin,” he said. “But for farmers these thresholds for quality and for quantity are always relevant – and they may become more important in the future.”
Ortiz-Bobea said the study could help broaden the scope of research and development in the dairy industry.
“A lot of the innovation in the industry has been focused on a few indicators around increasing output,” Ortiz-Bobea said. “We see productivity rising but sensitivity to climate might be rising as well, and we’re not selecting for those traits that might make cows more resilient.”
Prieto spent six months finding the right methodology to analyze the massive amount of data – with 120 million data points on cows’ milk production alone, all provided by the U.S. Council on Dairy Cattle Breeding. Combined with the daily weather data, Prieto had to rely on the Cornell Center for Social Sciences, which provided access to a server large enough to run the models he created.
The team hopes to work with even more granular data that will help them determine the cows’ response to heat with more precision.
“If we can have daily data and benchmarking for specific cows, we might be able to identify which ones are more heat-resilient,” Ortiz-Bobea said. “We can also start to look at if there’s a trade-off between higher yields or higher protein and fat and heat resilience. The goal is to share these insights that have broader implications for the industry.”
Co-authors of the paper include Christopher Wolf, the E.V. Baker Professor of Agricultural Economics in the Dyson School (SC Johnson); Kristan Reed ’07, adjunct professor of animal science in the College of Agriculture and Life Sciences; and Ph.D. student Ziyi Lin.
Funding was provided by the USDA’s National Institute of Food and Agriculture and the National Science Foundation’s AI Research Institutes program.
