Evaluating soy‐dris for predicting manganese deficiency and sufficiency

Abstract

Better diagnostic tools are needed to predict Mn deficiency and sufficiency to ensure maximum yield with the minimum of fertilizer inputs. An alternative to the sufficiency range method (SRM) is to use the balance of nutrients approach as in the DRIS (Diagnosis and Recommendation Integrated System) method. A computer program has been developed as a diagnostic tool with plant nutrient levels as inputs and plant indices as outputs that indicate which element is most deficient. This program, SOY‐DRIS, was evaluated for Mn predictions using plant data from four field experiments. Data for check plots (0 Mn) for each non‐Mn treatment were used as input for the program and the Mn index indicating deficiency or sufficiency was the output. The predicted deficiencies and sufficiencies were tested against yield responses to Mn. Of 74 check plots, SOY‐DRIS correctly diagnosed the Mn nutrient status in 62 cases. There were 32 true positives (Mn deficient) and 30 true negatives (Mn sufficient) indicating that both deficiency and sufficiency were accurately predicted. If borderline cases and cases where insufficient Mn was added to give a yield response were eliminated, only one incorrect diagnosis was made giving a 99% correct diagnosis by the SOY‐DRIS program. The SRM gave results nearly identical tothe SOY‐DRIS progam.