Distribuição espacial de massa seca de plantas de cobertura de outono/inverno

Abstract

The use of soil cover plants in crop rotation plans has been shown to be an effective way of promoting the quality and stability of the no-tillage system (SPD). The choice of species is a very important factor, as these can positively or negatively influence the later crop, and the distribution of the biomass on the surface is more important than its quantity or volume. The use of optical sensors can bring important benefits by speed, agility and field scale in surveys involving the production of dry mass of cover crops, carbon sequestration, nutrient cycling, SPD quality and crop yields. The objective of this study was to determine the spatial variability of dry mass (Chapter I) of different fall / winter cover crops, through NDVI obtained by optical sensor (chapter II), and to verify its influence on soybean crop productivity (Chapter III). The study was carried out in three experimental areas of a property located in Boa Vista das Missões - RS. The dry matter (DM) and NDVI sampling was georeferenced through the sampling grid of 0,5 ha (70,71m x 70,71 m). For the determination of the spatial variability of MS in the experimental areas, weekly green mass sampling was carried out at the sampling points, with the aid of a sample frame of known area (0,25 m²). For the determination of the Vegetative Index (NDVI), the GreenSeeker manual was used, and the readings were carried out on the same dates of collection of the phytomass and at the same sampling points. For Chapter III, soybean productivity was determined by a productivity map obtained from the harvester. Data were submitted to descriptive and geostatistical statistical analysis. The normality hypothesis was tested by the W test (p < 0,05). The similarity between the maps was verified by the relative deviation coefficient (RDC) and the correlations through Pearson's simple correlation (p < 0,05). It was possible to establish a preferential time for the diagnosis of the spatial variability of dry mass for the crops of white oats and rye, but not of turnip. The NDVI obtained by means of the GreenSeeker ground sensor was effective in determining the variability of the dry mass production of the white oat crop, however, it was not effective when applied to rye and forage turnip. The RDC was an efficient parameter allowing to evaluate the similarity between dry mass and NDVI thematic maps. The variability of the winter cover crop of forage turnip did not influence soybean yield in the studied period.