Correction Line Ag

Fertilizer Over-Application is Associated with Increased Grid Size There was a study done by the University of Georgia CAES in 2022 where they looked at how grid sizes affected fertilizer application accuracy. They looked at 1, 2.5, 5, 7.5, and 10 acre grids. Variable-rate prescription maps were made according to the grid sizes and soil sample results from those grids. The paper states that, “the Rx maps were spatially analyzed and compared to assess the application accuracy (%) of different grid sizes compared to reference Rx maps.” The study found that the 1 acre grids had an application accuracy of 80%+ for Lime, P, and K. The 2.5 acre grid fields were all under 80% for Lime, P, and K. 5 acre grids were lower than that, and 7.5 were even lower. At 10 acre grids, accuracy was just above 50% across the board. The lower the percentage, the lower the effectiveness in representing spatial nutrient availability. It is understandable to think that larger grids allow for a reduced cost associated with soil samples and the sampling cost. However, this doesn’t factor in how grid size and application accuracy affects the cost of Lime, P, and K. What looked to be cheaper on the front end was actually more expensive when things were all said and done. There tends to be an increase in overapplication with an increase in grid size. Now you might be thinking, “that is in Georgia, so what?” That is beside the point. The point is that there are costs associated with grids (even 2.5 acre grids) that are not accounted for. There are costs associated with only using standard soil tests on those grids as well. This will affect recommendations and variable rate scripts drastically.

A Quick History of Grid Sampling Soil sampling and recommendations based on those samples have come a long way since the early 1900s in some ways. But in other ways, they have not. Consider the first recorded field that got recommendations based on soil samples. This was published by Linsley and Bauer in 1929. It was: • A 30.89 acre field that was broken down into ≈1 acre grids. • There were soil samples taken at roughly 6 inch and 12 inch depths. • Based on those soil samples, the recommendation for Lime was about a half ton per acre. The lime would’ve been shoveled onto the field by hand (Franzen & Mulla, 2015). • The areas that didn’t need lime based off of the soil tests saved both labor and money. All of this was a huge advancement in agriculture at the time. But it is hard to believe that we are still using grids (especially larger than the ones used in 1929) in the year 2026. We have learned so much about soil since then, but our soil mapping, testing, and sampling techniques haven’t always kept up with the pace of learning. So What? Everything in agriculture is variable. Fields aren’t always square, soil is dynamic, and weather throws a wrinkle into everything. This makes measuring anything difficult. What we do measure isn’t always black and white. However, we try to treat something as dynamic as the relationship between plants and soils as static. This is what grid sampling and industry standard soil tests do. Grids treat the field as if it works in 2.5 acre (or more) squares. And coupled with standard chemical tests, we shoot for ppm values within those perfect squares. Seems simple right? But soil is not simple. We might know 0.1% about it, actually. Yet we do know that it is alive and that it cycles. Grids and chemical extractions alone do not really account for that. Consider this justification for 2.5 acre grids: “Although most researchers that have sampled whole fields at small spatial scale (less than 0.4 ha (≈1 acre) sampling density) have found that 0.4 ha is a minimum density to sample if nutrient variability is in a range where variable-rate fertilizer application would make a positive difference to the farmer, most grid soil sampling today is conducted in a 1 sample ha grid (2.5 acre grid). The reason that this is acceptable in the central Corn Belt of the United States today is that most fields and areas within those fields have P and K levels in the high range. Although there is considerable variability of P and K levels within the field, most of the field, regardless of soil test level, will result in the same P and K fertilizer recommendation. Therefore, a less dense grid is acceptable, because failure to represent the P and K nutrient level still results in the proper recommendation as provided by university soil fertility experts (Bullock, 2002; Wittry and Mallarino, 2004).” This is from A History of Precision Agriculture (Franzen & Mulla, 2015). That quote works for a system that is static and that has perfect edges. The reason why the 1 acre or less grids were preferred is because they picked up variability that the 2.5+ acre grids could not. Often times, you don’t have the same elevation, water flow accumulation, or topography within that 2.5 acre grid. This leads to differences in soil structure, carbon, biology, and nutrition within that one grid (reference the picture attached that overlays 2.5 acre grids over Baseline Rx zones). And while the quote above references P and K levels to justify 2.5 acre grids, is that enough? This considers chemical P (Bray, Mehlich, Olsen) and chemical K (AA) and that’s it. This is especially concerning if that P and K will have recommendations to spread in the fall which allows for tie-ups of these nutrients. What about Nitrogen, Calcium, or Carbon?