| | Introduction | Materials and methods | Results
Introduction
Plant breeding is a long-term commitment that puts together the agronomic, disease resistance and quality characteristics into an economically viable cultivar. However, for malt quality, the cost and time required for testing is both expensive and limiting so only a small part of the breeding program can be screened. The malting tests are also destructive and require larger seed samples that are not available in the early generations of a breeding program.
During the commercial malting process the maltster also combines grain lots into the malt house and increases the variability of the grain. A rapid test of the whole grain will allow the variability of the grain to be predicted and controlled before the barley even enters the malt house. This will improve overall quality of the final malt.
The Field Crop Development Centre (FCDC) began a joint project with Canada Malting in Calgary in 1996. The objective was to develop NIRS calibrations for the primary quality factors used by the malting and brewing industry to measure malting quality. These included Grain Protein, Fine Extract, Diastatic Power, Alpha-Amylase, Total Malt Protein, Soluble Malt Protein, Wort B-Glucan, Malt Friability, Homogeneity, and Viscosity.
NIRS is an excellent tool to screen large numbers of whole grain samples in a short period of time. For over 20 years the FCDC has used NIRS to screen for feed quality and presently screens over 35,000 samples every year. NIRS is a non-destructive test, requiring as little as 25 grams of seed, allowing the breeder to screen material at a very early stage in the breeding program.
Materials and Methods
The development of the malting quality equations were done on breeding samples representing everything from feed barley to the best malting quality available. Samples were selected at the FCDC and scanned on a FOSS 6500 Spectrophotometer. The samples were then sent to Canada Malting. Canada Malting malted these samples in a Phoenix micro malt plant using 150 g of seed. They used a standard cycle in the phoenix plant and the samples were fully modified. Data was sent back to the FCDC to be used for calibration development.
The final research calibration set consisted of approximately two hundred samples per year for five years beginning in 1996. After 2001, samples were added each year in order to strengthen the calibration. In order to insure that we had maximum variation both genetically and environmentally we eliminated redundant samples; therefore, not all samples were used to build the calibration. All the equations developed are based on whole grain, unmalted barley.
The FCDC has successfully used these calibrations to predict malting quality on whole grain samples from the breeding program since 1998. In 2002 we began the process of adding commercial malt samples into the equations. The commercial samples came from three malt houses at Canada Malting in Calgary and from Rahr Malt in Alix, Alberta. The barley samples were taken from lots before malting and the malt analysis came from the same malting company that produced the finished malt.
Results
The commercial malt samples represented a narrow range of variability in the original calibrations. This was expected as these were all malt varieties selected by the malt house. In general the commercial samples contained much more variability within the sample compared to the original research samples. Figures 1 to 4 show the relationship of the commercial malts in the overall calibration sets. By adding them into the calibration set we reduced the accuracy slightly but improved the consistency (Table 1). This translated into slightly different RSQ and Standard Error of Calibration (SEC).
Table 1. Relation Between the Research Calibration Based on Micro Malt Data (MM) and the Commercial Calibration (CM) for the Characteristics Measured.
It is evident to us that the success of the calibrations has been the wide segregating variation we had in the genetic research samples, which was built upon with the commercial samples. Because of the variability in commercial malts, which is also increased due to the lot size in the malt house and the variability introduced by both the steeping and killing processes, we will see final commercial malts have a greater variance and would expect them to differ from research malts from a pure source. This should not reduce the usefulness of this technology to the malting industry but can only help them to determine how to blend lots to meet optimum quality through their malting process. It also allows breeding programs to screen for malt quality cheaply and quickly at early stages in the breeding program.
James H. Helm, Lori Oatway and Patricia Juskiw
Alberta Agriculture Food and Rural Development
Field Crop Development Centre, 5030-50 Street, Lacombe, Alberta, T4L 1W8 Canada
Presented at the 18th North American Barley Researchers Workshop, July 17-20, 2005 |
|