approved: .__ Dr. gimothy L. ^Righetti This study evaluates Beufil's Diagnosis and Recommendation Integrated System (DRIS) on hazelnuts. Although sufficiency ranges are currently the main approach for fertilizer recommendations in Oregon, DRIS enhanced nutritional interpretations, and might alter recommendations. Reference values (DRIS norms) were developed using 15 years of published and unpublished field data. DRIS indices and Nutritional Imbalance Indices (Nil) were then calculated for a previously, published experiment. Nil was a good indicator of nutrition limitations to maximizing yield. Obtaining high yield was only possible with low Nil trees. There was a good agreement between DRIS and sufficiency ranges, especially if the relative deficiency or excess associated with corresponding relative excesses or deficiencies are evaluated. DRIS indices and Nil more clearly present both beneficial and harmful effects of altered plant mineral composition. The data indicate that using these two parameters will help in better evaluations and more valuable recommendations. AN EVALUATION OF THE DIAGNOSIS AND RECOMMENDATION INTEGRATED SYSTEM (DRIS) ON HAZELNUTS fCORYLITS AVELLANA) IN OREGON by Othman Alkoshab A THESIS submitted to Oregon State University in partial fulfillment of the requirements for the degree of Master of Science Completed October 2, 1986 Commencement June, 1987

The Diagnosis and Recommendation Integrated System (DRIS), which uses nutrient element concentration ratios as indicators of nutrient deficiency, was used to evaluate current sufficiency ranges for hazelnut trees. Reference values that were derived from published and unpublished field data were used to calculate DRIS indices for N, P, K, Ca, Mg, Mn, Fe, Cu, B, and Zn. A nutritional imbalance index (NII) was computed as the sum of DRIS indices irrespective of sign, and a threshold NII value (mean NII + 1 SD), above which severe imbalances are expected, was established. DRIS diagnoses were then compared with the sufficiency range approach to determine if relative deficiencies or excesses associated with severely imbalanced trees would have been routinely detected in 624 mineral analyses of hazelnut leaves. A previously published field trial was also reevaluated. DRIS diagnosis generally agreed with the diagnoses made by the sufficiency range method, especially if sufficiency ranges for some elements were made more narrow. However, some nutrients were never identified by DRIS as a major relative deficiency or excess in any of the trees judged severely imbalanced, based on the sum of DRIS indices. Nitrogen and Mg deficiencies were not detected unless lower NII thresholds were used. Unfortunately, lowering NII thresholds enough to detect N and Mg deficiencies identified some high-yielding trees as severely imbalanced. DRIS will not detect all deficiencies or excesses. Therefore, DRIS is best viewed as a supplement to sufficiency range diagnoses that provides additional information when severe imbalances are detected.

Previously published DRIS norms for sweet cherry and hazelnut were used to calculate DRIS indices for N, K, P, Ca, Mg, Mn, Fe, Cu, B, and Zn on over a thousand leaf sample analyses for each crop. A nutritional imbalance index (NII) was obtained by adding the values of DRIS indices irrespective of sign. The sample with the lowest NII for each crop was assumed to have "ideal"; levels of elements and these "ideal"; levels were used to create an artificial data base to determine which critical values would be most consistent with DRIS evaluations. By maintaining all but one element concentration at "ideal"; levels and artificially varying another, the concentration of a given element that was associated with the onset of severe imbalances could be identified. In addition to providing ratio‐based diagnoses, DRIS norms provide a means of independently evaluating current sufficiency ranges for elements that DRIS diagnoses as relatively deficient or excessive. Key words: Corvlus avellana filbert Prunus avium leaf analysis

Five hundred leaf analyses of sweet cherry and hazelnut leaves were evaluated to determine if biases in DRIS formulas could prevent the detection of some deficiencies or excesses. Both one‐equation and two‐equation calculation procedures were evaluated. The one‐equation approach has several disadvantages. The data imply that there are limits on minimum or maximum values that an index can obtain for some elements, but not for others. The range of index values obtainable varies considerably among elements. If the sum of DRIS indices, regardless of sign, is used as a criterion to identify imbalances, relative deficiencies or excesses for some elements are masked. These difficulties are lessened but not entirely eliminated with the original two‐equation procedure. Regardless of the calculation procedure the relationship between DRIS indices and the respective concentrations varied for the elements N, K, P, Ca, Mg, Mn, Fe, Cu, B, and Zn. Some indices are so dependent on the concentration of the element involved that a ratio based diagnosis would have little advantage over a sufficiency range diagnosis. Other elements have indices that are so dependent on the concentrations of other elements that the index is of little use in evaluating nutritional status. Although DRIS is a very useful diagnostic approach, it will not detect all deficiencies or excesses. DRIS serves best as a supplement to sufficiency range based interpretation; providing additional information when severe imbalances exist.