Ferrugem da folha de trigo (Puccinia triticina Erikss.) taxa de formação e distribuição das lesões

Abstract

Leaf rust of wheat (Triticum aestivum L. Thell) is the one of the most important diseases affecting this crop world-wide. The causal agent of wheat leaf rust is the fungus Puccina triticina Erikss, which belongs to the Pucciniaceae family. The main disease symptoms, are normally shown in the upper leaf surface, but under severe epidemics they can be observed in the leaf sheaths. The uredium is brown in color and generally in the circular form. In Brazil, under favorable ambient conditions, the disease causes significant damage for the majority of cultivars under cultivation in the country. Genetic resistance and the use of chemicals are the most common control practices under way. The objectives of this work are: 1) to determine the frequency distribution of leaf rust lesions within wheat leaves; 2) to fit distribution models to the observed data; 3) to examine the standard point pattern for the lesion locations within a leaf; 4) to model the density area for leaf rust lesions in a single wheat leaf. The present work was justified considering economic importance of wheat leaf rust as result of frequent genetic resistance break down and the heavy dependence on fungicide use to disease control. Therefore, this study should contribute to add epidemiological knowledge by examining the increment on lesion density as well the distribution pattern for lesion location. A sampling area of 225 square meters, in commercial farming of five hectares of wheat BRS loro, was used for collecting wheat leaf samples. From leaf rust onset four samples were carried out in the field. A total of fifty main tiller were collected every time. The leaves were detached from the main tiller and a digital image was made for each sampled leaf. A software, especially developed, was used to assist counting and pointing the exact location of leaf rust lesions directly from the image. The quantification of the number of lesions was made in each one of the leaves. The data set representing the epidemic progress was adjusted to a Logistic model. The results showed that the values found for lesion rate density were of 0,38 and 0,37 day-1 in the flag leaf and the leaf immediately below, respectively. The results indicate that leaf rust lesions localization followed a random pattern at the beginning of the epidemic shifting to an aggregate pattern as the epidemic advances in time. Finally, a parametric model representing the density area for rust lesions was adjusted to the data from randomly chosen leaves from the sample. The importance of these findings is discussed in this thesis