A Comparison of the Performance of On-Farm OFDA2000 and In-Store FLEECESCAN - Part 2: Sheep Selection and Fleece Classing

Abstract

Fleece testing has traditionally been conducted within a laboratory. Using sub-samples often sourced from the midside, the Airflow, SIROLAN-LASERSCAN or OFDA100 instruments are generally used to estimate the fibre characteristics. More recently, a new generation of testing technologies has allowed testing to be conducted on-farm, in a classing race (OFDA2000) or during shearing (SIROLAN-FLEECESCAN); the latter may also be operated in a central wool store. Commercial usage has determined the costs involved as well as practical advantages and disadvantages of each system, but little information is available that evaluates their precision limits or performance.

Based on a fine-medium Merino flock, this paper aims to compare the performance of the OFDA2000 operated in a race prior to shearing and the SIROLAN-FLEECESCAN operated in a central store. A third comparison is made on snippets sampled by FLEECESCAN but tested under standard laboratory conditions (referred to as LAB-LASERSCAN).

OFDA2000 and FLEECESCAN were used to allocate fleeces to four Mean Fibre Diameter (MFD) categories and two “Elite” classes. It was stipulated that Elite fleeces be ≤19.0 µm and Semi-Elite fleeces be ≤20.0 µm. The MFD calculated for all sale lots was subsequently compared with Coretest results. Both OFDA2000 and FLEECESCAN were able to class fleeces in the four distinctive diameter categories. However in this trial, the OFDA2000 results exhibited a coarse bias that resulted in the allocation of fewer fleeces to the Elite classing lines. Thus, the accuracy of measurement is important in order that specific clip preparation strategies involving micron cut-offs can be achieved.

While the association between fibre characteristics measured by FLEECESCAN and LAB-LASERSCAN was better than that between OFDA2000 and FLEECESCAN or LAB-LASERSCAN, the R2 values were poor for all comparisons. Thus, it is not recommended that measurements derived from different test systems be compared or used in combination. The ranking of a single trait, MFD, was used to compare the testing technologies. When selecting a large proportion of a flock, for example retain 65% of ewes, the three systems had 85-90% of sheep in common. However, when selecting from a tail of the distribution, for example retain 5% of rams, the systems only had 40 - 50% of sheep in common.

This illustrates the importance of precision. Based on a sub-set of this flock, Marler et al. (2002) estimated 95% confidence limits for MFD of ±1.02 µm for FLEECESCAN and ±1.41 µm for OFDA2000. Particularly when testing animals of high commercial value it is recommended that the test precision be at its optimum. In these circumstances, it is advisable to use the test system with the best precision (FLEECESCAN). Further improvements to precision can also be achieved by increasing the number of samples tested by FLEECESCAN or the number of sites tested by OFDA2000. Good test precision is also necessary if tighter MFD categories are to be achieved in clip preparation. 

Citation

"A Comparison of the Performance of On-Farm OFDA2000 and In-Store FLEECESCAN - Part 2: Sheep Selection and Fleece Classing", K.A. Hansford, J.W. Marler and I.M. McLachlan, Report CTF09, Commercial Technology Forum, IWTO Barcelona Meeting, May 2002