By Tim Kalinowski on July 6, 2021.
Crop insect guru Dan Johnson spoke about the importance of good sampling techniques at the Farming Smarter summer Field School recently.
Johnson, who is a pioneer in the field of crop insect forecasting and a world respected specialist in grasshoppers and potato psyllids in particular, said the key to good assessment within a farmer’s field is understanding the difference between “accuracy” and “precision.”
Johnson provided two diagrams of a testing target circle to explain this point. The first diagram reflected accuracy with samples taken all throughout the target circle area, and even outside, with counts from the sweeps varying across the different sample points. However, the complete summary of all the sweeps gives a pretty accurate picture of how many insects might be in that field. The other diagram Johnson produced showed a narrow sample area to one corner of the field which precisely charted how many insects were in that sample. Johnson said both types of procedures have their place in research, but only the “accurate” target gives the complete picture of the field, and thus better enables farmers to decide whether an expensive chemical intervention is warranted or not.
Another problem with the way sampling is sometimes done, said Johnson, is when multiple sweeps of different areas of a ditch, pasture or field might turn up, for example, a lot of grasshoppers and thus may seem to indicate a problem. However, he explained, there are really only a few species of grasshoppers that are of particular concern to local crops– the rest pose no threat at all despite their intensity in one field or another. He said the farmer also has to crosscheck his sampling against how much crop damage he can see. Just because an insect species might be present in large numbers does not mean it is actually a significant threat to a crop, and sometimes, he acknowledged, a relatively small number of insects can cause significant crop damage. That’s why crop damage observation is very important.
Johnson also spoke about two of the more critical errors which can occur in data analysis after the fact: Assuming two things are related just because they happen to be occurring in the same field at the same time. And being so afraid to fall into the trap of the first error that the observer misses the connection.
“We have to be careful before we determine it comes from that particular pest on that particular day,” he explained. “There are two kinds of errors regarding correlation. Assuming a correlation, just because things are happening together, indicates a cause. And the other is ignoring correlation because you are worried about that first problem. The second is becoming more common because everybody is repeating number one like it’s a mantra.”
Johnson also spoke about the errors which arise from wishful thinking or blind assumptions. If a farmer has made up his mind he has a grasshopper problem, for example, and he finds a few samples in his field which show higher numbers of grasshoppers in some of his sample sizes he does not interrogate the data any further. Wishful thinking comes when a farmer does not want to accept the data before his eyes because of what it might mean to his workload or bottom line if action must be taken.
Johnson said another factor which hasn’t been much considered in the past, but has become much more important now is the presence of natural enemies, beneficial insects, in fields. He isn’t aware of too many wide-ranging studies in Canada at present which have actually tried to survey how many of these natural enemies are present in ratio with known crop pests, and what that might mean to natural controls of these pest species.
“For 20 or 30 years we have recommended it, and done research on it, but it hasn’t been a formalized survey very often,” he said. “Field Heroes are working on at least drawing attention to them.”
Johnson said it is a much different story in New Zealand where they count both the number of pests present in a given field and the number of natural enemies present at the same time.
“They do kind of a ratio calculation, and use that as their threshold (before spraying),” he explained. “We could do that here. We can do it especially in crops like potatoes.”
He said a relative lack of spraying in local potato fields in recent years has led to a “complex and very healthy community of natural enemies” being built up.
“They are there,” said Johnson,” and are essentially guarding those potato fields.”
He warned use of pesticides in large quantities wipes out both the pest species and their natural enemies, and it is something farmers should carefully consider before making a spraying decision.
Johnson was asked if he could see any potential plans to bring the New Zealand model to Canada.
“That’s a good question for the provincial government agencies,” he responded. “I know they do care about this, and I do know we wrote a recommendation at the end of 2017. That was one thing we did recommend was monitoring those natural enemies. And we are fortunate in that, although there are a lot of species in each group, for example we found almost 20 species of ladybugs in potatoes, they as a group do one thing.”
They all eat pests, explained Johnson, so can be counted together as one group in their impact as natural enemies.
Follow @TimKalHerald on Twitter