SOIL SAMPLING TECHNIQUE FOR BAND-FERTILIZED, NO-TILL FIELDS WITH MONTE-CARLO SIMULATIONS

Band application of P fertilizer in no-till fields results in extremely variable soil test P (STP) along lines perpendicular to the bands. This variation may be represented by two interlaced populations, each with a unique STP mean and standard deviation. The objectives of this research were to: (i) characterize STP populations in a band-fertilized field; (ii) use these population characteristics to calibrate a Monte Carlo based model to simulate STP populations; and (iii) use simulated STP populations to evaluate different soil sampling procedures. A corn (Zea mays L.) field was band fertilized with P, and 1 yr later STP (mg P kg(-1)) was measured on individual soil cores with the bicarbonate extraction procedure. Out-band STP mean and standard deviation (SD) were 13.9 +/- 4.75, in-band STP mean and SD were 127.4 +/- 29.7, and composite STP mean and SD mere 23.3 +/- 28.69. A computer simulation model based on the Monte Carlo technique reproduced the observed populations very closely. The results show that the number of completely random soil cores needed to represent a band-fertilized field is prohibitive. Rapid collection of a single representative composite sample with equipment capable of taking a slice of soil across the fertilizer band and accurately weight in-band and out-band STP by volume is recommended.