'Background noise' is seriously distorting beef-breeding data
My recent item on the Irish Cattle Breeding Federation (ICBF) and the use and ownership of cattle breeding data triggered responses.
Several people said that I underestimated the contribution that current chief executive Sean Coughlan made to steadying the ICBF ship. They made the point that prior to Coughlan's arrival, ICBF was floundering in terms of delivering meaningful figures and indices. The highly qualified Coughlan brought IT skills which identified and removed the glitches from the system.
I also questioned the volatility and inconsistencies in beef indices and the contribution that we can expect from the maternal indices on beef bulls.
Callers agreed with this observation and quoted reasons as to why this is happening. While the Irish system of collecting commercial records and data is unique and potentially the most comprehensive in the world, there is also a lot of what the geneticists call "background noise" in our data. Some of this noise is accidental but some, unfortunately, is not.
For example, late registration of calves will cause major distortion of growth rate records. Breed societies have tried to clamp down on this malpractice, but even when caught fiddling the system, offenders usually get off lightly.
Levels of meal feeding, especially to pedigree cattle, vary enormously. This must distort indexes. Another practice popular with some pedigree breeders is the introduction of the 'Holstein Friesian cow' to rear pedigree calves. As well as giving the calf an artificial boost, being relieved of the calf gives the pedigree beef cow a better chance of going back in calf. How does this affect the data for maternal assessment and indices.
How accurate are farmers at matching a calf to its dam and sire, especially when stock bulls are used? When we use the commercial records, from marts, meat plants, etc, there is huge scope for background noise in the data.
In this, the beef records are fundamentally different from the dairy records which largely come from targeted milk recording. Even the ICBF sheep records using targeted farms are likely to deliver more accuracy than the beef equivalents.