Biometria e índices de vegetação por sensoriamento remoto em área de pastagem no município de Cachoeiras de Macacu, RJ

Abstract

Agricultural and livestock production is influenced by different climatic factors and international demand. Livestock currently plays an important role in the country's economic activity and, as a result of this influence, there is a need to rethink the management stages based on the application of new techniques and technological tools in the management of these areas. The present work presents an evaluation of the biometric characteristics and the processing of orbital images to obtain the vegetation indices of a commercial pasture area located in the municipality of Cachoeiras de Macacu, RJ. For the biometric characterization, destructive and non-destructive methods on the surface were considered, as well as the method based on orbital images from the Landsat-8 (L8) and Sentinel-2 (S2) satellites. The leaf area index (LAI) was determined in the campaigns, using a ceptometer and the height for two areas, defined as: i) outdoor (30 ha) with 160 animals grazing every 15 days, and ii) indoor (25 m²), fenced inside the outdoor area and without grazing. In the destructive method adopted in three campaigns, the vegetation in the external area was collected at ground level and taken to the laboratory. The orbital images were selected, considering the lowest cloud interference, pre-processed and algebraically manipulated with geoprocessing tools. The meteorological variables for characterization of the environmental conditions of the place were obtained by a micrometeorological station installed in the area and the precipitation was measured by a meteorological station located close to the study area. The IAF presented an average of 2.10 m2.m-2, with a minimum value of 0.68 m2.m-2, and a maximum value of 5.00 m2.m-2, while in the internal area the average was 2. 72 m2.m-2, with a minimum of 1.31 m2.m-2; and a maximum of 5.31 m2.m-2. The height of the vegetation in the external area averaged 29.19 cm, with a maximum value of 75 cm and a minimum of 8 cm, for the internal area it was found an average of 48.41 cm, with a maximum value of 90 cm and a minimum of 12 cm. For Landsat-8 images, the NDVI (Normalized Difference Vegetation Index) ranged between 0.61 and 0.83; SAVI (Soil-Adjusted Vegetation Index) between 0.32 and 0.68; and IAF with values between 0.52 and 4.56, for Sentinel-2 the NDVI variation was between 0.58 and 0.79, SAVI between 0.29 and 0.61 and IAF between 0.42 and 2.17. The IAF values estimated by the SAVI index with the observed IAF (ceptometer) showed accuracy, with a Willmott coefficient of agreement (d) of 0.87 (Sentinel-2) and 0.86 (Landsat-8) for L=0. 25. The study showed that the results show seasonal behavior, which indicated periods with greater availability of water and solar energy in the environment favor the growth of vegetation. Monitoring becomes an important tool to help understand seasonal dynamics and provide information for projects and management of pasture areas.