A behind the scenes look at soil testing
I went on a tour of the soils lab at Brookside. Here’s a video, and some notes.
Soil testing is amazing. Many critical parameters can be measured on a small sample. The test results can then be used to improve turfgrass performance, adjust soil pH, apply necessary fertilizers, and much, much more.
On a recent visit to Brookside Labs, I was able to record, in almost one continuous take, the steps a sample goes through—from arrival at the lab, to disposal after all tests are completed. With help from many technicians and soils lab manager Liz Skinner, the video includes demonstrations of lab procedures and machines, along with explanations of what is happening at each stage of a soil sample’s trip through the lab.
I’d like to point out some things that are shown in the video. I’ve written about some of these in ATC’s soil testing newsletter. You may have noticed the drying, grinding, scoop sizes, and so on when you watched the video. If you haven’t watched it yet, you can keep these points in mind as you watch.
Samples are dried at the lab prior to grinding. A lot of things happen when the soils contain water. Microbial activity continues, ion exchange continues, nitrogen mineralization occurs, etc. These processes are already happening in the soil, and then when the sample is removed from the soil and suddenly experiences exposure to a lot of air, there are unpredictable changes in microbial activity, ion exchange, and nitrogen mineralization. To stop these from happening—to essentially freeze the soil in the condition it was in at the time the sample was pulled from the ground—I recommend drying the samples yourself, prior to sending to the lab.
Soil nutrient analyses—the standard tests for soil pH, for soil organic matter by mass loss on ignition, for Mehlich 3 or 1 N ammonium acetate extractable nutrients—these tests are all conducted on soil. Soil in this case means material that has passed through a 2 mm sieve. This means there is no gravel, no big chunks of undecomposed living and dead plant material, and no thatch. What is in this soil is sand, silt, clay, humus, and small pieces of undecomposed plant material that are no larger in size than a grain of sand. This is excellent. I’m glad this is how soils are tested. But this means that if you want to check what’s happening with thatch, or mat, and how coring or topdressing have influenced those, you’ll want to check more than just the soil organic matter. You’ll want to check the total organic material as well.
Did you notice the scoop sizes? A 1 g scoop for Mehlich 3 extraction. A 2 g scoop for soil organic matter. A 7 g scoop for soil pH—and the reason for such a large scoop is simply to get enough solution volume for the pH electrodes to sit. With such small sample sizes, it’s essential that the sample you collect and send to the lab is representative1 of what you want tested. I recommend being extra careful about sample depth (I recommend 10 cm, that’s 4 inches), and I recommend leaving the thatch and mat and base of the grass plants on the sample. Let the lab remove that stuff every time, for consistency. As you see in the video, there’s a machine for that.
It’s impressive to see how the soils lab can process and test up to 4,000 soil samples per day. And those samples may get a scoop for pH, a scoop for nutrient extraction, a scoop of soil for organic matter, and scoops of material for other tests as well: Bray 2 phosphorus, KCl-extractable aluminum, a pH buffer, soil salinity, etc. Those tests are all run with blind quality control samples, a known soil with known test results, so the validity of the test results can be confirmed. And then through the wonders of software, and a lot of care and hard work by the staff, those results all get put together onto one report at the end of the day and are sent to the clients.
It’s usual to consider a “representative” sample to be a composite sample made up of multiple subsamples, mixed together, with the sample that gets sent to the lab being a subsample itself, from the mixed material that was collected. Or it may be that you skip the mixing part, and just take a lot of samples and send them to the lab. Because turf is growing in an unplowed soil with a gradient of organic matter that decreases with soil depth, it’s important to conduct soil tests on material that is not only comprised of material from a known depth, but also that the material itself is equally sourced from the entire depth gradient. I suspect that the very process of collecting multiple subsamples and compositing them, or of sending multiple subsamples to the lab, makes it more likely that the deeper parts of a turfgrass soil profile are overrepresented in the sample, and that the soil closer to the surface may be slightly underrepresented in the scooped material. This is one of the reasons why I currently prefer single core sampling for assessment of soil chemical properties in turfgrass. ↩︎