Comparison of DRIS and alternative nutrient diagnostic methods for soybean

The Diagnosis and Recommendation Integrated System (DRIS) provides advantages over the Sufficiency Range (SR) approach to diagnosing the nutrient status of soybean (Glycine max L.). However, a number of modifications to DRIS have been proposed, including the use of only one method for calculating nutrient indices, and incorporating nutrient concentrations. In previous research, the author found that derivation and interpretation of DRIS diagnoses could be simplified by: 1) using a logarithmic transformation of nutrient ratio data; 2) using population parameters rather than high‐yield subpopulation values; 3) using a single index calculation method; and, 4) incorporating a measure of the probability of yield response to a treatment. Diagnoses by the SR approach, DRIS, three revisions of DRIS, and two new concentration‐based diagnostic methods were compared using diagnostic norms derived from a data base of over 4000 soybean tissue analyses and yield observations. Virtually all diagnoses of the most limiting nutrient were the same, except that the SR method did not make diagnoses on 5 week‐old samples. Traditional DRIS diagnoses were the least conservative, indicating only the order in which N, P and K would likely limit yield. Revised methods, particularly the concentration‐based methods, indicated fewer limiting nutrients, and agreed well with the SR method. Yields in the test data set increased with nearly every nutrient application, so the least conservative DRIS showed the greatest yield advantage. No diagnostic method consistently identified the nutrient causing the greatest yield response as most limiting.