With the increasing cost of feed and social awareness of climate change, dairy farmers are under greater pressure to produce more milk with less resources. There is a need to explore long-term solutions to reduce livestock emissions and improve overall on-farm efficiency. Researchers at Agriculture Victoria are investigating the opportunity to use genetics to breed for cows that produce less methane and efficiently convert feed to milk.

A cow produces 70-120kg of methane per year. The environmental equivalent of driving a car 11,951km. About 90 per cent of that a by-product of feed fermentation.The cow rumen is filled with a variety of microbes that breakdown plant material into particles the cow can use as energy. Through this process, the microbes generate methane which is then burped out.

The amount of methane created greatly depends on the population of microbes in the rumen and that is determined by the cow's genetics.

Methane production in cattle also has a strong link to feed efficiency.

The feed-energy that goes towards generating methane could be used more economically for milk production.

In a group of cows that eat the same amount of feed and produce the same amount of milk, there will be some cows that release more methane and some cows that release less.

The challenge is to identify animals that produce less methane and use them to breed the next generation. Selecting cows for lower methane is also selecting for improved efficiency.

Dairy farmers have already begun to breed for more efficient cows by choosing to breed animals with high breeding values for FeedSaved, a trait released by DataGene in 2015 that identifies the most feed-efficient cows.

Farmers currently use breeding values to rank animals in their herds on a genetic basis. To generate breeding values, the animal's DNA is associated with their physical characteristics, in this case methane. This genetic tool predicts which bulls or cows will breed the best offspring.

However, breeding values for methane are not currently available. The gold standard of measuring methane is expensive and labour intensive, which leads to small datasets.

Although measurements in these small groups are extremely accurate, the breeding values produced are unreliable and cannot be used in industry application.

Generating methane breeding values is still a work in progress, but the results are promising for a cost-effective and reliable approach. To combat the data size problem, researchers are looking at new ways to collect methane data that is cheaper and easily obtained.

One strategy uses mid-infrared (MIR) spectroscopy to analyse milk samples collected at routine herd testing. A light is beamed through the milk sample and based on how the light interacts with the molecules in the milk a pattern is produced that is unique to each cow, similar to a signature.

The individualised pattern can then be used to predict a cow's methane production. By taking advantage of this technique, researchers can obtain large datasets with many animals.

Results show that breeding values generated using the large MIR methane dataset are three times more reliable than those calculated using the precise gold standard dataset, which has much fewer animals.

The real test will come when breeding values combine the gold standard and MIR methane measurements, which is standard industry practice. If the increase in reliability is high enough, a trait will be available that would have otherwise been too expensive to be included in routine genetic evaluations.

MIR is a tool that can be used to scale up datasets size, but it is still a rough prediction. To ensure the most efficient animals are consistently selected, it is imperative that gold standard methane data is collected to prove that it works.

MIR is an inexpensive way to make breeding values more reliable and is used for routine traits like fat and protein per cent. Obtaining measurements for methane is a simple recycling of the collected samples. However, the measurements are crude compared to the gold standard method.

But how crude can measurements be and still provide valuable information for farmers? Hopefully with the help of funding agencies, researchers and farmers, this is a question we will be able to answer in the near future.