Quantification of plant-available potassium (K) in a corn-soybean rotation: A long-term evaluation of K rates and crop K removal effects.

摘要

A long-term field study conducted at five Purdue Agricultural Centers located in west-central, north-central, south-eastern, north-eastern, and east-central Indiana was designed to assess the soil's ability to supply potassium (K) in a corn (Zea mays L.) -- soybean (Glycine max Merr.) rotation, and to evaluate the assumptions in the current soil K fertilizer recommendation. The objectives of this study were to 1) evaluate the long-term impact of five K rates, application frequency (annual or biennial), soil depth, and the soil K budget (net K addition/removal) on soil exchangeable K (Kexch) levels; 2) to quantify the effects of K application frequency and K rates on grain yields, and evaluate currently established Kexch critical levels for corn and soybean; and 3) to evaluate the long-term effect of K fertilization rates, net K addition/K removal, and residual K fertilizer on subsoil Kexch. Soil sampling during the growing season (0-20 cm) was synchronized to immediately precede the rapid K accumulation that corresponds to greatest growth rates. Post-harvest sampling was performed with a hydraulic soil probe to obtain intact cores to a depth of 60+ cm. Grain yield and grain K concentrations were measured to calculate crop K removal rates at harvest. Soil Kexch increases with net K additions varied across sites, in which differences were not related to differences in soil cation exchange capacity as indicated in the current recommendation. Within each site, the rate of Kexch decrease (as a function of the soil K budget) in soils with net K removal was smaller in magnitude than the increase observed in soils per unit net K addition. Additionally, annual K additions proved more effective in building-up soil Kexch than biennial applications. However, yield differences across application frequencies were not observed. Yield response to K additions was not observed in soils with Kexch considered limiting according to the recommendation framework. Non-linear models suggested that lower than currently-recommended Kexch levels can be risked at the 0-10 and 0-20 cm depths without significantly affecting grain yields. However, analysis of yield as a function of Kexch could not clearly identify a critical level or range in soil Kexch levels, irrespective of location or soil depth increment. Potassium fertilizer applied remained in the upper 20 cm of the soil profile. Subsoil Kexch depletion was observed at all sites, which resulted from crop K removal exceeding K rates additions. However, for near-maximum net K removals between 700-450 kg ha-1, long-term subsoil K exch declines at the 20-30 cm depth layer were less or equal than ≤16 mg kg-1 across sites. Consequently, the often reported lack of topsoil Kexch change in soils with large net K removals could not be attributed to subsoil Kexch dynamics or cycling.