Using post‐season tissue nitrogen concentrations to predict adequacy of in‐season nitrogen management

Abstract

High‐yielding mechanized rice production requires large amounts of nitrogen (N) fertilizer, which can be particularly costly to producers when applied in excess. The development of a tissue test to detect overapplication of N to rice is needed. Nitrogen response trials arranged in a randomized complete block design were established to identify a critical concentration threshold for a tissue test to detect excess N availability. Rice stalk samples were collected from 0 to 20 cm above the soil within 5 days of harvest and analyzed for NO3−‐N concentration. Two models were fit to the data to represent rice grain yield as a function of N concentration: the linear plateau model and the quadratic plateau model. The data were well represented by both models; however, the linear plateau model more accurately described the data with a normalized root mean square error of 0.128 for linear plateau and 0.131 for quadratic plateau and a significant F‐test (p < 0.0001). The join point of the linear response region and the plateau region of the model serves as a critical value that separates the responsive (N deficient) and non‐responsive (N optimum or excessive) regions. The 95% confidence interval of the join point (2.1 mg NO3−‐N kg−1) was selected as a practical, agronomic critical concentration threshold to reduce false positive errors. The data presented here further supports the use of a post‐season tissue test in rice to detect overfertilization and establish a critical concentration threshold for this tissue test.