Near Infrared Spectrometry May Help Separate U.S. Wool from Foreign Wool

Near Infrared Spectrometry May Help Separate U.S. Wool From Foreign Wool

MACKENZIE FLETCHER
Sheep Industry news Contributor

Near Infrared Spectrometry (NIRS) may be a useful tool for sheep producers to use as an estimate for yield and fiber density and also for companies to determine the country of origin for wool.

That was one takeway from a presentation to the Wool Council during the ASI convention in Charleston.

John Walker, professor and director of research at Texas A&M University’s AgriLife Research and Extension Center, gave the presentation.

NIRS is a spectroscopic method that uses the near-infrared region of the electromagnetic spectrum (from about 800 nm to 2500 nm). Typical applications include pharmaceutical, medical diagnostics, food and agrochemical quality control and combustion research.

Using NRIS to measure wool has been an ongoing project, originally funded by a grant from the U. S. Department of Agriculture under the direction of Chris Lupton of Texas A &M. However, that grant is no longer available so Lupton is no longer overseeing the work. The research is currently being conducted by Walker.

The project’s initial objective was to determine if improved instruments and calibration algorithms would improve the precision and accuracy of NIRS for commercial determinations of wool base. The project obtained bale core samples and lab measurements from Yocom McColl and used as few samples as possible to make the predictions, a different approach from Lupton’s research that maximized the number of samples used for calibration.

Both approaches gave similar results. To determine the accuracy of the tests, a math model was used to separate the calibration error from the standard lab error. The 95 percent confidence interval for NIRS predictions of AFD and Yield is about ±1.5 ì and ±3.4%, respectively. However about half of the error is a result of sampling error and error of the standard lab test.

However, Walker said the bottom line is that nothing done in the U.S. is going to set standards for the commercial trading of wool.

Australia and New Zealand have tested and rejected using NIRS to determine wool base.

However, Walker said he thinks it has a lot of potential for individual animals. If 100 samples are submitted, lab scoured yield and OFDA could be conducted on 10 of them and NIRS could be used for the remainder, which is an improvement and less expensive than the lab. Walker said it is important to keep updating the calibration of the NIRS machine to maintain testing accuracy.

Walker said what would be ideal is to use a side sample staple to predict the average fiber diameter, and yield with a single NIRS determination. He said such yield determination would be based on skirted fleeces with bellies out.

Future plans for NIRS research include developing calibrations for individual animals based on a 12-month fleece in order to aid in individual animal selection for fine wool sheep and also developing calibrations for hair contamination.