Soil nitrogen and phosphorus behavior in a long-term fertilization experiment

Accurate fertilizer recommendations rely on quantitative estimation of nutrients supplied by soil and fertilizer nutrients immobilized by soil. Understanding variation in these processes over space and time is critical for site-specific nutrient management. Our objective was to characterize spatial variability in N and P cycling for a corn [Zea mays L.]-soybean [Glycine max (L.) Merr.] rotation in southern Minnesota glacial-till soils. Soil samples and grain yield measurements were taken annually from 0.014-ha cells within two 16-ha fields. We determined effects of fertilizer P additions and crop P removal on soil test phosphorus (STP) and determined relationships between STP changes and soil variables. We also determined temporal stability of soil mineralizable N and inorganic N. The spatial patterns of mineralizable N were consistent over time. The spatial pattern of soil NO3-N was consistent with mineralizable N at a well-drained site, but not at a poorly-drained site. Change in STP per kg P net addition or removal exhibited spatial autocorrelation. Declines in STP under net P removal were directly related to initial STP values. Increases in STP under net P addition were significantly related to pH at both sites and mineralizable N at one site. Temporal stability in mineralizable N suggests that predictive approaches to site-specific N management may succeed when the environment for mineralization is uniform. Within-field variability in the relationship between STP and net P addition may substantially affect fertilizer P rates required to attain critical STP values and should be accounted for in variable-rate P applications.