Rates and Residual Effects of Potassium Fertilization in a Brazilian Soil

Potassium (K) fertilizers are a common necessity in terms of plant nutrition in acid soils of the tropics, including Brazil. In many areas, farmers cut back on fertilizer use due to expenses, which could compromise good yields, profit, and food safety in the future. Studying the impacts of K fertilizer cut back on Brazilian soils is important to demonstrate the effects in the medium to long run and this was the main objective of this study. The field trial has been carried out for seven years in a clay Oxisol in Itiquira, MT, growing soybeans every summer and maize second crop or Brachiaria grass during the fall/winter. Treatments involved: (1) four rates of K₂O (22.5, 45, 90, and 135 kg K₂O/ha), plus a control with no K₂O added, in interaction with suppression or not of K after the third year, (2) three levels of base saturation (40%, 55%, and 70%), (3) three rates of phosphogypsum application (0, 2, and 4 t/ha), (4) two times of K₂O application, and (5) two localities of application. Responses to K₂O were frequent along the years, but relatively small up to the sixth year. The average negative effect on grain yield of both crops from the suppression of K₂O up to the sixth year was 385 kg/ha for soybeans and 617 kg/ha for maize second crop, yearly, comparing the K recommended rate of 90 kg/ha and the control. In the seventh year the same comparisons rendered the same negative effects of 951 kg/ha and 862 kg/ha for soybean and maize, respectively, showing a trend for higher response to K from now on. Cutting back the K rate after the third year still did not have much of an effect on soybean and maize second crop. In certain years, yields of both crops were positively affected by phosphogypsum application, while maize second crop yield was also positively affected by liming.

The experiment has shown the soil to have a high buffering capacity in K. It is very possible that recycling of K from deeper soil layers is still helping to maintain yields even when applying relatively low rates of K₂O. It is intended to continue with this experiments for several more years to find out about the residual K effect in the long run.