SPATIAL VARIABILITY PATTERNS OF PHOSPHORUS AND POTASSIUM IN NO-TILLEDSOILS FOR 2 SAMPLING SCALES

The increasing use of grid soil sampling methods and variable-rate fer tilization requires better understanding of patterns and causes of lat eral variability of nutrients. This study assessed patterns of spatial variability of P and K for two scales of sampling on eight no-tilled corn (Zea mays L.) and soybean [Glycine max (L.) Merr.] fields. The fi elds had received no P or K fertilization since the last harvest but h ad varied histories of fertilization. Fifty 10-core (0- to 15.2-cm dep th) composite samples were collected from 2.2-m(2) areas separated 3.0 5 m from each other along each of two 150-m intersecting transects lai d out across and along old crop rows. Sixty-seven single-core samples spaced 7.6 or 15.2 cm were collected from 7.85-m segments of the trans ects across rows. Variography revealed high random and spatially struc tured variability for most fields. The structure of the variability di ffered among fields and directions within a field. Plots of soil-test P and soil-test K values, fast Fourier transform analyses, and semivar iograms showed clusters and periodic patterns of variable regularity a nd size across fields. The cyclic patterns were more defined across cr op rows. Small-scale cyclic patterns with a period of approximate to 1 m revealed for some fields probably resulted from repeated banded fer tilizer applications. Large-scale cyclic trends with periods of 15 to 18 m or a multiple of this distance were probably the result of broadc ast fertilization with commercial bulk spreaders. Sampling methods sho uld be developed for fields with high spatial variability of cyclic st ructure at various scales. Producers should pay more attention to unif orm application of fertilizers and manure.