Relationship of the Illinois soil nitrogen test to spring wheat yield and response to fertilizer nitrogen

Abstract

The Illinois soil N test (ISNT) has been used to distinguish between soils that are responsive and non-responsive to fertilizer N in Illinois. We examined the suitability of this test, together with more traditional measures of soil fertility, including spring nitrate-N and soil organic carbon (SOC), for predicting yield and N fertilizer response of wheat (Triticum aestivum) on hummocky landscapes in Saskatchewan. The relationship between ISNT-N and wheat yield and fertilizer N response was assessed using data and soils previously collected for a variable-rate fertilizer study. Soils were re-analyzed for ISNT-N. Our goal was to determine if ISNT-N could be used to improve the prediction of crop yields. Although ISNT-N was correlated with both unfertilized wheat yield (r = 0.467, P = 0.01) and fertilizer N response (r = -0.671, P = 0.01) when data from all study sites were combined, correlations varied according to landscape position and site. Stronger correlations between nitrate-N and both unfertilized wheat yield (r = 0.721, P = 0.01) and fertilizer N response (r = -0.690, P = 0.01) indicated that ISNT-N offered no advantage over nitrate-N. Although both tests broadly discriminated between sites with high or low N fertility, few relationships were detected on a point-by-point basis within a field. Stepwise regression equations predicting yield and yield response did not include ISNT-N, due in part to the high degree of collinearity between ISNT-N and other variables such as SOC, suggesting that ISNT-N alone was not a key indicator of soil N supply. Key words: Illinois soil nitrogen test, potentially available N, soil N, fertilizer N recommendations