Comparative evaluation of some diagnostic techniques for determining the nutrient requirement of maize grown on hydric dystrandepts
| dc.contributor.author | EscanĚo, Crisanto R. | |
| dc.date.accessioned | 2009-07-15T17:06:31Z | |
| dc.date.available | 2009-07-15T17:06:31Z | |
| dc.date.issued | 1980 | |
| dc.description | Typescript. | |
| dc.description | Thesis (Ph. D.)--University of Hawaii at Manoa, 1980. | |
| dc.description | Bibliography: leaves [207]-216. | |
| dc.description | Microfiche. | |
| dc.description | xvi, 216 leaves, bound ill. 28 cm | |
| dc.description.abstract | Two fundamentally different approaches namely, the conventional critical concentration and "diagnosis and recommendation integrated system" (DRIS) were evaluated to determine their applicability in assessing the nutrient requirements of maize grown on a thixotropic, isothermic soil family of Hydric Dystrandept. Results of 15 fertilizer experiments (eight N x P and seven CaC03 x P) were used in this evaluation. In the conventional critical concentration approach, five methods were evaluated to determine the critical or optimum tissue concentration or range for each nutrient element. The methods were (1) graphical determination of the range of tissue concentrations associated with high yields, (2) determination of the mean nutrient concentration of the highest-yielding 15% of the plots, (3) determination of the nutrient concentration which, when used as the critical concentration, resulted in the highest percentage of correct diagnoses of nutrient deficiency, (4) determination of the nutrient concentration at which the mean yield response to additional fertilizer approaches 0.0, and (5) determination of the nutrient concentration at which yield is maximized (calculated using a quadratic regression model). Methods (1) and (2) were used to determine optimum concentrations of N, P, K, Ca, Mg, S, A1, Mn, Fe, Cu, and Zn. Method (3) was used for N, P, and Ca while methods (4) and (5) were used for N and P. Both methods (1) and (2) gave critical or optimum values as a range of tissue nutrient concentrations. For most nutrient elements, however, method (1) gave a comparatively wider range of tissue concentrations, and estimates of optimum ranges from this technique appear to be highly subjective. Except for N, Mg, and S the optimum nutrient concentrations determined with these two methods are within the sufficiency ranges suggested by either Jones (1967) or Neubert et al. (1969). For N, Mg, and S, high maize grain yields were associated with slightly lower concentrations than those previously reported. The critical or optimum nutrient concentrations obtained with the other three methods are similar and fall well within the published values. In most N x P experiments, highly significant positive correlation was found between tissue N and P concentrations as well as with grain yield. Likewise, the concentrations of tissue K, S, Cu, and Zn were positively correlated with tissue N and P and also with grain yield. Only in the case of Mn was the correlation negative. Since application of fertilizer N and P gave significant yield responses it is unclear whether the optimum ranges of these covariant elements are accurate, or whether they are artifacts of tissue N and P. In the DRIS approach, the percentage of correct diagnoses was the criterion used to determine the diagnostic accuracy for N, P, and Ca. Several variations in the various steps of the DRIS approach were evaluated to determine their effects on diagnostic accuracy. The diagnostic accuracy for N was improved (1) by judicious selection of the high-yielding subpopulation, (2) by selecting the "critical" DRIS index based on its effect on diagnostic accuracy, (3) by using DRIS indices calculated by experiment rather than by replicate, and (4) by using 10 rather than 3 or 36 ratios to calculate DRIS indices. For P and Ca, however, no marked difference in diagnostic accuracy was observed among the several DRIS approaches. Using the percentage of correct diagnoses technique, a critical DRIS index of 0.5 was chosen for N while a value of 0.0 was found for P and Ca. Similar values were obtained when maize grain yield was plotted against DRIS indices for these elements. For most other elements, the plot showed that 0.0 was the critical DRIS index. DRIS norms (optimum values of nutrient ratios and their corresponding coefficients of variation) independently derived from this study gave greater diagnostic accuracy for N, P, and Ca than when using published norms. Both the conventional critical level and DRIS approach were more accurate when the critical or optimum values of nutrient concentrations or nutrient concentration ratios were independently derived from our own experimental data. In the final analysis, the DRIS approach was judged to be better than the conventional critical level approach for diagnosing the N requirements of maize. | |
| dc.identifier.uri | http://hdl.handle.net/10125/9240 | |
| dc.language.iso | en-US | |
| dc.relation | Theses for the degree of Doctor of Philosophy (University of Hawaii at Manoa). Agronomy and Soil Science; no. 1398 | |
| dc.rights | All UHM dissertations and theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission from the copyright owner. | |
| dc.subject | Corn -- Fertilizers | |
| dc.subject | Plants -- Nutrition -- Mathematical models | |
| dc.title | Comparative evaluation of some diagnostic techniques for determining the nutrient requirement of maize grown on hydric dystrandepts | |
| dc.title | Nutrient requirement of maize | |
| dc.type | Thesis | |
| dc.type.dcmi | Text |