The gradient (vertical variability) of soa fertility indexes makes difficult to define lhe depth of soil sampling m the no tillage system to represem the status of soil fertility. The objectives of this study were initially to compare the crop yield in conventional and no tillage systems with some fertilization and crop sequences si different soil and cultivation periods, and as they were similar in yield, to define me depth of soil sampling in the no tillage system for phosphorus and potassium recomendation. Four long-term experiments (10 to 19 years) in conventional and no tillage systems from regions of the state of Rio Grande do Sul - Brazil, were used: EEA-UFRGS, in Eldorado do Sul in a Rhodic Paleudult soil, FUNDACEP-FECOTRIGO, in Cruz Alta, CNPT-EMBRAPA, in Passo Fundo, soils and COTRISA. in Santo Ângelo, all in Rhodic Hapludox soils. Since the crop yields were similar in both tillage systems. soil samples were collected in different layer thickess from lhe sou surface in the no tillage system and in the 0-20cm layer in lhe conventional tillage. The phosphorus and potassium content were somewhat below to the criticai leveis for me Paleudult soil and much higher for the Hapludox soils. The adjustement for lhe soil depth is about 7 cm for lhe Paleudult soil, and in the other soils, with high nutrient content, any soil layer from the soil surface up to around 20cm does not affect the fertilizer recommendation.
calibration adjustment; soil sampling; no tillage
