Determinação da folha diagnóstico e avaliação do estado nutricional de canola (Brassica napus L.)

Abstract

The objective of this work was to evaluate the position of the leaf with better relation with the productivity and the nutritional status of the plant by means of the Diagnosis of Nutritional Composition (CND) method for canola cultivation. The experiment started in 2015, the Hyola 433 hybrid was sown mechanically in no-tillage system. For the validation of the standards, a 5 x 4 factorial experiment was developed with four replications, in randomized blocks, where sulfur doses (0, 30, 60, 90 and 120 kg ha-1 of elemental sulfur) and nitrogen ( 0, 40, 80, 120 kg ha-1 of urea) in Dourados, at the Experimental Farm of the UFGD, harvest 2015/16. 240 leaf samples were collected (3rd, 4th and 5th pair, from the apex of the stem). The database was divided into three reference populations, using the leaf pairs as criteria and the transformation of the data by the centered logarithmic relationship (CND-clr). The subpopulations were separated by the Cate-Nelson method taking into account the productivity (kg ha-1) and the nutritional balance, so it was classified as TN (True negative); TP (True positive); FN (False Negative) and FP (False Positive). The classification was tested, indicating the Negative Predictive Value (NPV); Positive predictive value (PPV); Accuracy (Acc); Sensitivity and Specificity. CND standards and indices were calculated by the mean and coefficient of variation centered on nutrients as well as the index of nutritional imbalance (CND-r2). A normality test was performed on the data and then the discrepant values (outliers) were detected and removed. Multiple discriminant analysis (MDA), characteristic rectifier operation (ROC) analysis and principal component analysis (PCA) were performed to obtain information on which variable discriminates the predicted groups better, the nutritional state of the plant and the interaction study between nutrients. Through the CND method and statistical analysis proposed in this work, the fourth pair of leaves can be indicated as diagnostic leaf, in the collection of samples for foliar diagnosis and evaluation of the nutritional status of canola, in a crop rotation system in the state of Mato Grosso do Sul.