Estrus detection in dairy cattle is one of the biggest issues in the dairy industry, contributing to an annual loss of $300 million in the dairy industry for failed estrus detection due to prolonged calving intervals and days open, losses in milk production, increased veterinary costs, etc.

Madureira augusto
University of Guelph / Ridgetown Campus
Madureira was previously a graduate student at the University of British Columbia.
University of Guelph
University of British Columbia

Standing to be mounted has been the gold standard for heat detection, but the frequency of standing events has been declining over time. Thus, technology has brought us automated activity monitors (AAMs) to better identify estrus behaviour, with the goal of getting cows pregnant efficiently.

Our laboratory at the University of British Columbia is studying how this data can be best used to detect estrus for the best opportunity for pregnancy.

The evolution of automated activity monitors

Increased physical activity is considered a secondary feature of estrous expression in dairy cattle, and nowadays AAMs have been used to reliably identify cows in estrus. There are many different types of monitors on the market.

Accelerometers and pedometers are the most common AAMs used in the dairy industry. In general, most AAMs follow the same concepts in order to create estrus or health alerts.


For example, when using accelerometers, the monitor will use an algorithm to calculate an index of physical activity by combing the data measured by the 3D accelerometer. A rolling average in physical activity is calculated to use as the baseline for each animal, which in turn becomes a reference value to calculate the relative increase or decrease in activity at a given period of the day.

In the case of alerts for estrus, the relative increase in physical activity has to reach a threshold set for each AAM system. Pedometers use similar concepts to create alerts; however, it will consider step counts or lying time as a measure of physical activity.

Research using many different types of sensors within different management systems have reported the positive predicted value (total of true events versus total estrus events detected by the AAM system) to vary between 80% to 95%.

Using more than one measurement obtained by activity monitors can improve the detection of true events. For instance, studies have shown the association between rumination and physical activity, as rumination times decrease on the day of estrus.

What factors impact AAM readings?

The AAMs are indeed able to effectively detect estrus events. Our laboratory became interested in understanding what animal factors may impact the readings of the AAM.

  • Parity: Our studies have shown that parity has a marked effect on estrus measurements as multiparous cows expressed shorter duration and a lower change in increased physical activity when compared with primiparous cows. Other research has also found that parity was significantly associated with walking activity and reported that for each additional parity, walking activity at an estrus episode was reduced by 21.4%.

  • Body condition score (BCS): Body condition score has also been shown as a major factor affecting physical activity at estrus, as animals with lower BCS (less than or equal to 2.5) have less intense estrus measurements. Other research has found similar results, where they reported a 0.25 increase in BCS was correlated with an increase in physical activity at estrus.

  • What about milk production? Although commonly reported to impact mounting behaviour, milk production is weakly correlated with estrous expression using AAM. The apparent contradiction between the results of our studies regarding milk production and estrous expression may lie on the fact that AAMs measure overall physical activity, not only mounting behaviour. Overall, greater estrous expression was only weakly associated with milk production, emphasizing the influence of other factors, such as BCS and parity, and potentially factors such as group size, health status and lameness.

How does estrus activity correlate to pregnancy per A.I.?

Interestingly, we have found a strong correlation between the relative increase in activity at estrus from the AAM systems and pregnancy per A.I. This relationship has been confirmed using multiple different AAM systems in multiple countries.

We found animals that had a greater intensity of activity at estrus had around 12 percentage points (or around 30% relative improvement) greater pregnancy per A.I. than animals that had lower estrous expression.

In that study, we included only animals that spontaneously came in estrus. But even when using a timed A.I. protocol that induced ovulation using estradiol (allowed in Brazil) and that consequently induce estrus, we found a similar effect of estrous expression intensity (Figure 1).

Distribution of pregnancy per A.I. (%) of all insemination events according to relative increase in activity at timed A.I. detected by an automated activity monitor

Even more interesting, we also found a correlation with pregnancy losses from 31 to 60 days post-A.I. (Figure 2) and estrous expression.

Distribution of pregnancy losses (%) according to relative increase in physical activity at timed A.I. detected by an automated activity monitor

Cows that had a lower increase in physical activity were more likely to have reduced pregnancy per A.I. and increased pregnancy losses.

Multiple ovulations and embryo transfer protocols have been important breeding technologies with great value to the dairy industry, as they can accelerate gains in genetic progress.

We also found similar positive associations between estrous expression of donor and recipient cows and the respective collection, development and transfer of embryos. We conducted an experiment evaluating the viability of embryos produced in vivo in heifers enrolled to a superovulation protocol.

Greater intensity of estrous expression detected by the AAMs was associated with donor heifers having greater number of embryos collected (3 more) and greater percentage (8% more) of viable embryos per collection.

On a different study, we evaluated the impact of estrous expression of recipient cows on the success of embryo transfers. The animals with greater estrous expression had higher pregnancy per embryo transfer than those with low estrous expression (41.5% versus 30.6%).

Can we fix the decreased fertility of animals with low estrous expression?

Previous data from our group has indicated that lower-intensity estrus episodes are more likely to be followed by ovulation failure, shorter intervals from the beginning of estrus to ovulation and lower fertility.

A recent study from our lab evaluated the administration of a single dose of GnRH at the moment of A.I., with the idea that GnRH could benefit animals with lower estrous expression by reducing the proportion of animals with ovulation failure.

Cows were divided into four groups based on the intensity of estrous expression and on their GnRH treatment: high estrous expression with no GnRH injection, low estrous expression with no GnRH injection, high estrous expression with a GnRH injection, and low estrous expression with a GnRH injection.

We found that administration of GnRH at the time of A.I. was able to increase the pregnancy per A.I. of animals with low estrous expression to be comparable to those with high estrous expression (Figure 3).

Pregnancy per A.I. by GnRH and estrous expression

Our research was the first to report the association of intensity of estrous expression with fertility, ovulation, pregnancy losses and the viability of embryos.

The measurements of estrus events captured by AAM are correlated with BCS and parity but have found to be weakly associated with milk production as initially hypothesized.

Animals that had greater estrous expression had higher pregnancy per A.I. and reduced pregnancy losses.

These results provide further evidence that measurements of estrous expression (i.e., on spontaneous estrus, timed A.I. and superovulation programs) might be a reliable predictor of fertility and could be used as a tool to assist in the decision-making of reproduction strategies at the farm level.

Future research is needed to better understand and interpret the data from the AAMs to optimize breeding decisions.

Take-home messages

  • More information from the AAM can be useful.
  • Intensity of estrus measured by activity monitors is associated with fertility and can be used for much more than only alerts.
  • Artificial insemination (spontaneous or timed A.I.) and embryo transfer can both be affected by expression of estrus and its intensity.
  • Reproductive programs with strong reliance on estrus detection are highly efficient.
  • The intensity of estrus expression can be useful for breeding decisions at the end of timed A.I. protocols.
  • An injection of GnRH at A.I. can improve fertility, particularly among those animals with low-intensity estrus expression.
  • Expect variability among farms.