Geração de modelos e índices digitais para para diagnóstico da erosão de área agrícola

Abstract

Erosion is one of the main problems of soil erosion, this degradation causes the loss of the most fertile soil layers causing damage and damage to the environment. Thus, the objective of this work is to generate topographic digital models and indexes for the diagnosis of erosive process in the agricultural area, using images acquired by the Remotely Piloted Aircraft (RPA) of the School Capão da Onça Farm (FESCON) of UEPG. The models were generated from the images taken by RPA with support points georeferenced by the Global Navigation Satellite System (GNSS) and planialtimetric survey using a total station. The indices were generated using remote sensing, photogrammetry and aerophotogrammetry techniques, using the Geographic Information System (GIS). First, the area with erosion was defined, after the definition of the flight planning in the laboratory, in the field were demarcated the support points soon after taking photographs with the RPA, the support points were collected with GPS which were 10 At 15 minutes at each point, another fieldwork was the planialtimetric survey using the regular mesh methodology. The work resulted from tests performed on the number of control points used to generate each Model Digital Elevation (MDE) which is a product generated through aerial photography. None, 4, 9 and 11 control points were used to generate different MDEs, the product generated from the MDE is the Model Digital Terrain (MDT) which has only the elevation of the ground surface discarding everything above, like trees, buildings, cars, etc. MDTs have generated new products, with interpolated MDT data it is possible to generate terrain slope map, runoff flow, water flow direction among other products which indicates and assist in detecting erosion or possible soil erosion. As mentioned earlier these indicators were made from the same area with the number of different support points, and each generated a similar but different product, one with some features more visible than others. In addition to the different models with the photographs, he obtained a terrain MDT using the planialtimetric survey with the regular terrain grid. And it obtained products with a significant difference from the MDT generated by aerial photography. Thus, with the knowledge acquired in the field, the model that best represents the study area is the MDT generated with aerial photographs with 11 control points. The indices were generated from the model that was most suitable being the 11 control points.