By Dan Kaiser
One thing you need to know about calculating removal-based rates is that the so-called rules of thumb for the amount of P and K in grain may or may not represent what you have in the field. One concern I have is that some growers treat their P and K like balancing a checkbook, trying to calculate what is taken off the field and putting that amount back on in fertilizer. In almost all cases, this type of accuracy is not needed for removal-based guidelines. Here is why:
1. The amount of P and K in grain can vary greatlyWhile rules of thumb or standard factors can provide a good starting point, the actual concentration of P and K in the grain you are removing cannot be known unless you grind a small amount of grain and have it analyzed. The graphs below show the actual range in values in my set of roughly 15,000 samples collected over the past 14 years. Tables in both the corn and soybean fertilizer guidelines for Minnesota list the median P and K removal values as well as the 25% and 75% confidence intervals. The confidence interval data indicates where we are confident the actual mean is. For example, a 200-bushel corn crop should remove anywhere from 50 to 66 lbs P2O5 and 36 to 44 lbs K2O on an annual basis. We tend to treat the removal values as an exact value when it is a range.
2. There is more P and K in the soil than is indicated by the soil test
3. The rate that soil tests change can be unpredictable
4. The probability that fertilizer will increase yield decreases with increasing soil test valuesSoil test categories can be thought of as representing the ability of the soil to supply all of the given nutrients for a crop. As soil test values increase, the ability of the soil to supply all nutrients to the crops increases until a point that the fertilizer applied is primarily replacing what is taken up from the soil by the crop. If you are looking at risk assessment, there is data available as to the probability that fertilizer will increase crop yield using the current soil test classifications in Minnesota and an estimated percentage of yield that should be produced if fertilizer is not applied. This data is only available right now for phosphorus in Minnesota as the potassium data has been much more variable and harder to predict.
5. It is a good time to take soil samples if you have not done so for a whileIt seems like some producers have started to hit the easy button and blindly apply P and K based on removal without knowing what their current soil test values are. One area of concern I have is building soil tests too high. While we do not think of soluble phosphorus being an issue in fields as soil tests increase, the risk for P loss to tile lines will increase. There is no agronomic reason to build soil tests too high, so knowing where you are and adjusting rates down can reduce P loss to the environment without sacrificing yield. Many treat their soil like a bank for nutrients, but this bank can be a bit leaky over time should the concentration of some nutrients get too high.
In closing, I want to reiterate that there is no way to apply exactly the amount of nutrients removed on an annual basis from fields. Being close is typically sufficient and you will not find that soil test values decrease if your application rates of P and K and what was removed by the crop are within plus-or-minus 10 to 15 lbs. Soil testing is critical to look at trends in your fields, providing you with important insights on your management practices.