You are bombarded with so many new ways of doing things “better” or “saving money” or “improving efficiency.” Just in the world of crop fertility alone, there are so many machines and techniques and products and rates, it will make your head spin. Most of you are willing to try something new once in a while, but I assume that in general farmers are pretty critical of new ideas. You want to see good strong evidence that these ideas are an improvement before investing in them. After all, you don’t want to make this part of your regular program unless it works better and provides a better return than your old system.
Remember how long it took maltsters and brewers to switch from using Harrington barley, even when new varieties were so much better, agronomically? That’s because Harrington worked really well for them. It was a consistent and predictable performer, and these beermakers were not going to switch without mountains of evidence to prove the switch was worthwhile. I’m not saying you have to be that slow to adapt — the great thing about a family-run business like yours is that you can adapt quickly — but there is a message here. When it comes to adopting new ideas, “quick” has to go hand in hand with “smart.”
That is why doing your own test strips is an important way to evaluate a new product or technique on your own farm. Yet, from what I can gather, farmers rarely do test strips.
Perhaps it’s because those of us in agriculture are so used to the government regulating everything. We’ve been trained to think that if someone is selling it, it must work. Yes, it might work somewhere, but it might not work on your farm, in your system. Or it might “work” on your farm, but it might not increase your profits.
In the name of reducing fuel costs, many of you have switched to one-pass spring operations where you put all your seed and fertilizer down at the same time. So how much fertilizer can you put with the seed and not have yield-reducing seed damage in your soil type? Well, do a test strip. Do cereal seed treatments pay? Should I top dress the crop with micronutrients? Does Pod Ceal reduce shattering losses when straight combining canola? You can’t know the answer to these questions unless you leave test strips.
I know why farmers don’t test strip. It’s a pain in the butt to bugger around with test strips when you just want to get the crop in the ground or in the bin. But at the same time, if you’re spending all that money on a new input product or a new technique, it would make sense to know that it works better than your usual system before you make it part of your regular program.
Get Help With Test Strips
Step one: Don’t work alone. John Heard, fertility specialist with Manitoba Agriculture, has some good advice for farmers who want to do their own test strips. A key first step, he says, is to find a “co-ordinator” who can help you create a proper study and develop useable results.
“On-farm tests are seldom successful unless a grower (even with all the tools of guidance systems, variable rate controllers and yield monitor) teams up with a co-ordinator,” Heard says. Many agribusiness companies have research staff or agronomists who could aid in this role. You might have to pay them, but if a few farmers get together, they could share the cost.
The U. S. is a little more advanced when it comes to on-farm trials. State and university extension services often have co-ordinators on staff, and grower associations, such as Iowa Soybean Association, encourages on-farm trials and helps set them up. “Grower associations in the U. S. have also adopted very rigourous standards for their growers to follow,” Heard says. “If conditions of replication and randomization are not met, the results are ruthlessly discarded into the trash bin of coffee shop talk.”
Heard suggests you visit the on-farm research websites for Iowa Soybean Association and University of Nebraska at Lincoln. Here are the addresses: http://www.isafarmnet.com/andhttp://southeast.unl.edu/agriculture/pdfs/PgmHi-OnFarm. pdf
You might also ask your provincial ag office about courses on on-farm testing. Heard has taught one and he says the International Plant Nutrition Institute (www.ipni.net)has one. “But people have not been breaking down the door to enroll,” Heard says.
Keep It Simple
Here are John Heard’s tips to create an on-farm trial that gives you scientific results.
1. You have to keep your trial to one question — one variable — not four or five. For example, does topdressing 15 pounds of liquid nitrogen improve wheat protein and yield compared to no topdressing?
2. Select a suitable trial location. The suitable location is a uniform piece of land. On-farm trial plots are more reliable as plot length is increased. The more uniform the field, the shorter the acceptable plot length. Tests could be as short as 300 feet but strips of 1,000 feet or greater will help identify differences better, Heard says. When there is known variability in the field, make sure all strips contain the same variability. If there is a slope, make sure all treatments go up-and-down the slope, not across. The grower needs to bring field history to the table when selecting areas that otherwise to the helper appear “flat” and uniform in February. “I have seen strip test results tainted by a salinity gradient across the field and the grower
picked the location!” Heard says.
3. Select the treatments. You want fertilizer rates, for example, that are different enough to create some significant results. The Iowa Soybean Association protocols for on-farm trials require that nitrogen rates differ by 50 pounds.
4. Replicate and randomize the strips. If you are not going to replicate on your farm, then get a lot of your buddies to do the same thing and pool data (and get a co-ordinator to interpret that data.) The number of replications depends on how big and reliable the differences are. If the differences are small, you need more replications. Iowa State Soybean association insists on six reps for fungicide trials on corn and soybeans and four reps for manure and nitrogen-rate trials. Strip randomization is a piece of cake now. With a variable rate applicator, you simply dial down or up the rate for adjacent passes. Without variable rate applicator, you use hands-free guidance systems to plant parallel passes while skipping the in-between treatments. You come back and plant the “vacant” passes with Treatment B later.
5. Determine data to be recorded. You need more than just final yield. You want to also include rainfall and specific observations. For example, disease ratings are important for a fungicide trial. You need to know the severity of the disease before you can assess whether a fungicide worked. For a fertility trial, the initial soil test is critical and in-season tissue testing is valuable, too. These recorded observations help explain the yield differences, or lack of differences.
6. Statistically analyze the trial results. Again, this is where a co-ordinator can help. In many cases, differences of a few bushels don’t mean much.
For More on this…
Read Les Henry’s column in this issue. Also read the article by Donna Fleury on page 12 about the Tuck family’s on-farm trial techniques.
Ralph Clark, a longtime Grainews reader and an old family friend from Lauder, Man., told me that most of the jokes he reads in Grainews these days are the same ones he’s read before. Ralph has a thick scrapbook of jokes he’s been collecting. I asked if he would pick a bunch of his favourites that I can use. This won’t help him in his quest for new jokes, but it helps me keep my joke box topped up. See the box at the bottom left of this page for Ralph’s first installment. Thanks Ralph.