Uso do sensoriamento remoto para detectar neossolos quartzarênicos: estudo de caso em Cianorte – PR

Abstract

This course conclusion work presents a study in the area of remote sensing aiming to identify the Quartzipsamments Neosols in northwestern Paraná state. These soils present high susceptibility to erosion which can cause problems such as loss of soil, productivity and silting of water bodies. On the Google Earth® program 5 points were selected inside the city of Cianorte, which supposedly ocurred Quartzipsamments Neosols. After this selection, it was used one RapidEye® image covering part of the Cianorte municipaliy and the verification points. This image was used to calculate the indexes MSBI (Misra Soil Brightness Index, SBIMSS (tasselled Cap - Soil Brightness Index MSS) and SBL (Soil Background Line), and then these indexes calculated were imported to the GIS SPRING and transformed into image, being possible then, to associate them with the bands of RapidEye® scene. Thus, one can perform color compositions and take samples of pixels of the areas identified as Quartzipsamments Neosols. The results that highlighted with more efficiently the Quartzipsamments Neosols were found in the second slicing with an interval from 14583.33 to 17500 for the MSBI index and from 12437.5 to 14312.5 for the SBIMSS index, and for the SBL index no satisfactory results were found for the Quartzipsamments Neosols. However, besides highlighting the Quartzipsamments Neosols the slicing performed also managed to highlight other elements from the image such as the roofs.Thus, it was performed the statistical test Principal Component Cnalysis (PCA), which received a 78.32% score of difference among the samples of Quartzipsamments Neosols and roofs for MSBI index and 77.35% for SBIMSS index. This result is mainly due to the composition of Quartzarenic Neosols which are basically formed of sand, which provides a uniform reflectance on these samples, since the roofs are made of different materials which results in different reflectance percentage for each type of material. Thus, MSBI and SBIMSS indexes were the ones that highlighted more effectively the Quartzarenic Neosols, and for the SBL index no results were found in the three slicing performed. It Is worth mentioning that in this study the results were in function of the indexes MSBI SBIMSS and not directly from the reflectance of roofs and Quartzarenic Neosols, furthermore this work can act as a starting point for future research in the area, since there were no similar studies found in Brazil.