Jim, removal rates are exactly what the term implies. It's replacing the amount of nutrients removed with the harvested portion of the crop, therefor, in order to maintain a soils fertility at a desired level, the amount of nutrient applied needs to equal the amount that is removed or else the fertility level will decrease. Neither banding or broadcasting the nutrients changes this fact.

When thinking in terms of cutting rates, it really shouldn't be a question of; can I cut rates? Rather, it should be a question of; when can I cut rates? To answer this question, one needs to understand the basis of the "buildup" portion of a fertility recommendation.

Through research, we know that in order to insure that a given soil can supply a sufficient amount of nutrient so as to not limit yield, we have to build that soil to a certain level. That level of build varies by state and soil type, but in the case of P in my area of Illinois, that level has been determined to be 50 lbs. P as reported by a soil test. This is the level where the soil can supply all of the plants P needs. In actuality, one can maintain a soil at a lower level and not limit yield if you are making annual fertilizer nutrient applications. This is where your understanding of recommendations being based upon "broadcast" application comes into play.

Let's say for argument sake that your corn crop removes 70 lbs. P2O5. If you apply enough phosphate to replace the 70 lbs., we know that approximately 20% (14 lbs.) of the crops P needs will be supplied by the broadcast fertilizer. The rest (56 lbs.) will be supplied by the soil. Because you are supplying the 20% through the applied fertilizer, there is really no need to build to a level of 50. In reality, a level of 45 would suffice.

In the case of banded fertilizer, because we are increasing plant use efficiency to a level of approximately 50% of the applied fertilizer nutrient being available to the crop (instead of 20% available with broadcast), the soil itself does not need to supply as much nutrient to the plant. The implication here is that we no longer need to build the soil to as high a level as we would if we were broadcasting our fertilizer.  At what level we transition from buildup to maintenance in a banding situation I do not have a firm answer, but I'm increasingly becoming confident that it is somewhere around the 35 lb. level. Time will ultimately provide the answer, but in the meantime that's what I'm shooting for.

The overriding point that I want to make is that we are doing ourselves a serious disservice when we talk about using "half rates", or "reduced rates". Instead we should be talking about when we can start to transition from a "build" program to a "maintenance" program. If one is using a banding program, I believe we can start that transition much earlier than we can with a broadcast program. This is where the fertilizer savings is going to come from.

I will however point out that Tony Vyn from Purdue makes a good argument for a band/ broadcast program. He says his data indicates that a combination of the two is the best approach, but I'll leave that argument for another time.