Prediction of Soil Nitrogen Supply in Corn Production using Soil Chemical and Biological Indices

Abstract

Assessment of the soil N supply capacity is essential to optimize N fertilizer use. The soil N supply capacity of 102 soil samples (0–15 cm) from 25 sites collected from 2004 to 2007 across four Canadian provinces was evaluated by comparing a group of chemical N availability indices with soil mineralizable N pools and a field‐based measure of soil N supply. Soil N supply was estimated by corn (Zea mays L.) N uptake corrected for starter fertilizer N. Two subgroups were created based on the soil texture and were compared to the whole data set. Grouping soils provided limited benefits in predicting soil potentially mineralizable nitrogen (N0), but improved the prediction of soil N supply. The N0 was weakly related to soil N supply for the whole data set (r = 0.09) and in fine‐textured soils (r = 0.37) but the relationship was improved (r = 0.68) in medium‐ to coarse‐textured soils. The N0 was not necessarily a good predictor of soil N supply under field conditions which emphasizes the need to also consider environmental conditions. The UV absorbance of a 0.01 M NaHCO3 extract at 205 nm (NaHCO3–205), the hot KCl extractable NH4–N (HotKCl–N) and Pool I (a labile mineralizable N pool) plus NO3–N were the most promising N availability indices because they are easy to perform and they were positively and significantly related to soil N supply in the whole data set as well as the soil texture subgroups (0.28 ≤ r ≤ 0.62). This study demonstrated that grouping soils based on texture can increase the proportion of variation in soil N supply explained by N availability indices when data from contrasting environmental conditions, soil types, and years are used.