The test result is only as good as the sample. You hear that over and over, for any situation from political polling to assessing grain quality. Taking representative samples is the critical step to understanding the quality of your production. It is impossible to test the “whole” in most cases, so a sample is generally taken and the results of the tests on the sample are presumed to represent the whole.
Holly Gelech, business development manager at BioVision Seed Labs, echoes the need for rigor when creating representative samples on the farm and points to the issues if not. “There are two main problems that could arise from tests on samples that are not representative of the lot,” she explains. “First, a quality issue could be inflated or secondly, a quality issue that exists in the lot could be minimized or not even identified.” Either scenario is very undesirable to growers who produce large volumes of grain and need to have quality results that reliably predict the quality of the lot that is either for sale or will be used as seed in the following year.
“In order to understand the quality of a particular seed or grain lot,” says Gelech, “farmers need to understand how to get a representative sample and be prepared ahead of harvest to gather it. In that way, the sampling process will be seamless and not interfere with or slow down harvesting operations.” It’s one thing to get the grain off the field and into the bin, and another to know exactly what you have in that bin. The quality of grain harvested from a single field can vary quite widely. High spots or drowned out spots, areas in the lee of a shelter belt, salinity, and other issues will impact grain quality.
Stream sampling is the preferred method to create a representative sample. “Harvest is the best time to do this,” explains Gelech, “when the grain is being transferred from the truck or grain cart to the bin.” For the majority of farmers, this will involve taking manual samples/scoops from the flow of grain from the truck to the auger. There are automated stream samplers that are used in more commercial situations, but manual will do the job just as well. The real key to stream sampling is to take a high number of samples from all places in the stream at a regular interval you’ve determined in advance. At the end of the process, the sample is mixed, divided, and reduced to a reasonable amount to facilitate grain grading and other testing as well as having sufficient volume to provide samples to buyers if needed. The equipment required to get a representative sample does not have to be sophisticated. Generally a good number of clean 20-litre pails and a scoop or sampling ladle is all that is required.
“Static sampling from a bin or bag is fairly commonplace today,” says Gelech. “What farmers have to watch here is not taking samples from the bottom of the bin or just from the door.” In both cases, these two spots will not produce a representative sample for different reasons. In the case of the floor, smaller grains and weed seeds tend to accumulate there. In the case of the door, there can be moisture infiltration from outside or other environmental changes that are not happening throughout the rest of the bin. Bin probes are an absolute requirement for getting a representative sample from a bin. “The problem with bins and bin probes today is that bins are getting really big,” says Gelech. “It’s becoming more difficult to get a representative sample out of some of them.”
The Canadian Grain Commission has an excellent guide on their website on how to take a representative sample and later divide and reduce it.
If a field or grain type is binned together and comes from a number of fields or occupies a number of bins, a composite sample can be created to represent that lot or variety or field. The Canadian Grain Commission defines a composite sample as being composed of a number of distinct portions, each obtained in a prescribed manner from consecutive samples. Those portions are then blended to make the composite.