Premium
Soybean Phosphorus and Potassium Deficiency Detection as Influenced by Plant Growth Stage
Author(s) -
Hallmark W. B.,
DeMooy C. J.,
Mooris H. F.,
Pesek John,
Shao K. P.,
Fontenot J. D.
Publication year - 1988
Publication title -
agronomy journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.752
H-Index - 131
eISSN - 1435-0645
pISSN - 0002-1962
DOI - 10.2134/agronj1988.00021962008000040009x
Subject(s) - nutrient , medical diagnosis , phosphorus , potassium , chemistry , zoology , mathematics , medicine , biology , organic chemistry , pathology
Recent data suggests that ignoring the effect of physiological age on nutrient status of soybean ( Glycine max L.) can result in biased P, K, and Zn deficiency diagnoses by the diagnosis and recommendation integrated system (DRIS). To overcome this problem, the use of nutrient multiples in place of some nutrient ratios for calculating nutrient function values has been advocated. Also, accurate P, Mn, and Zn deficiency diagnoses have been made using a small R1 (initial‐bloom) DRIS data base. This study was conducted to determine: (i) if small R1 and R5 (initial pod‐fill) DRIS data bases can detect soybean P and K deficiencies from R2 (full‐bloom) plants, and (ii) if the use of nutrient multiples in the R5 data base improves P and K diagnoses. The R1 data base detected P and K deficiencies, but overemphasized P and underemphasized K diagnoses. The R5 data base overemphasized Ca and N and underemphasized P and K diagnoses. Differences in diagnoses were attributed to high ca (data base concentrations are expressed in lower‐case letters) and low p values in the R5 versus the R1 data base. To alleviate incorrect P diagnoses nutrient multiples involving ca were substituted for nutrient ratios of ca in the R5 data base. This improved ( P <0.01) P diagnoses; however, diagnoses were not as accurate ( P <0.l0) as P diagnoses by the Rl data base. These data indicate that the use of nutrient multiples in DRIS can improve nutrient diagnoses, but that more accurate predictions occur when plants are sampled at the same growth stage as the data base.