Mehlich-3 Extractable Phosphorus and Potassium As Affected By Previous Crop, Sample Time, and Fertilization Rate.

The adoption of precision agriculture technologies has increased the number of soil samples submitted for analysis and changed the times that soil samples are submitted in Arkansas. A large proportion of soil samples are now collected shortly after harvest and submitted for analysis in October and November. Eleven fertilization trials, six following rice (Oryza sativa) and five following soybean (Glycine max), were soil sampled periodically from fall to spring in soils fertilized with no P and K or 29 kg P and 112 kg K ha^-1.  Following soybean, soil-test K did not change across time as the linear coefficient was not significant (P = 0.2369). Soil receiving 112 kg K ha^-1 maintained soil-test K 3 mg K kg^-1 above the preplant value; whereas application of no fertilizer-K resulted in a net decrease of 18 mg K kg^-1 at soybean harvest. Following rice, soil-test K increased at a uniform quadratic rate with time peaking at 128 DAH. Shortly after rice harvest, soil-test K was 19 and 26 mg K kg^-1 less than the initial soil-test K for soil fertilized with 112 and 0 kg K ha^-1, respectively. The predicted peaks show that soil-test K increased by 22 mg K kg^-1 following rice harvest with the peak in late December to early January. The 22 mg K kg^-1 change in soil-test K following rice represents 28% of the initial average (across all rice site-years) soil-test K (80 mg K kg^-1). Sample time does not appear to influence soil-test P and K results for samples collected following soybean.