National Institute for Space Research
Biblioteca Digital da Memória Científica do INPENot a member yet
6175 research outputs found
Sort by
Análise multitemporal da expansão turística e o seus reflexos nas mudanças da cobertura do solo do município de Bonito-MS por meio de imagens de satélite
No full textThe touristic expansion in the city of Bonito, in the State of Mato Grosso do Sul - Brazil has been changing the economy in the region due to the recent precautions to the environment which raises the necessity for ecotourism activities. Firstly it was tried to use a classification based on the digital numbers from NDVI to discriminate the land cover. The accuracy of this classification was calculated by the kappa statistic with a hybrid classification as reference. Also through the hybrid classification it was attempted to relate the land cover to the touristic expansion measured by the ICMS (tax on consumption of goods and services) of the selling activity. For this we used 3 TM/Landsat 5 images dated from 25 July, 1987; 23 July, 1998 and 23 July, 2004. The classification based on NDVI values was not accurate since the k values were from each date as respectively 0.1950, 0.2056, and 0.2442. The correlation between land cover classes and ICMS values have the best direct correlation (r = 0.9993) between urban area and commerce economical activity. It showed that the touristic expansion have changed the land cover pattern in the city of Bonito MS mainly in the urban area. The article presented a methodology based on remote sensing and GIS technology to understand the dynamic of the spatial organization which can be a powerful tool for policy makers.Pages: 7148-715
Interpretação de cicatrizes de movimento gravitacional de massa em imagens de alta resolução do sensor ADS 80 na geração de mapa de risco de deslizamentos em trecho do projeto do trem de alta velocidade brasileiro
No full textThe design of high speed train to be built in Brazil will connect the Campinas city in São Paulo to Rio de Janeiro city. It is a linear work unprecedented in the Country, which will go through various types of land experiencing different conditions. This job presents a risk mapping of landslides in the region of the fourth section of the planning route, on a region near the city of Itatiaia in Rio de Janeiro. Landslides were inventoried by meas of photointerpretation using ADS 80 sensor images. Thematic maps of the region with data from pedology, geomorphology, land use and land cover, lithology and hypsometry were used for the map algebra which generates the risk and hazard maps. The results indicate the existence of high risk to the occurrence of landslides in almost section four showing the need of a very thorough geotechnical evaluation of the slopes geological conditions that will suffer interventions during engineering work. This article was based on a publication relative to the conclusion job of the authors pos graduation Landslide hazards analysis of Itatiaia massif talus associated to the project of high speed brazilian train presented in march of 2012, on the University of Brasília, in cooperation whit the SIPAM (Amazon Protect System) and Brazilian Air Force.Pages: 4294-430
Identificación y mapeo de minerales de alteración hidrotermal con imágenes ASTER en el Distrito Minero San José. Macizo del Deseado, provincia de Santa Cruz
No full textIn the northwestern part of the Deseado Massif, within the San Jose Mining District, an intermediate sulfidation style epithermal deposit is actually in operation. It hosts a number of siliceous veins carrying gold and silver. Within this district, they have been identified several areas with hydrothermal activity evidences, temporally and spatially associated to effusive episodes of the Jurassic volcanism. In this area, ASTER images were used for the identification and mapping of hydrothermal alteration zones, by the application of different digital analysis techniques. The preliminary geologic recognition was made by the interpretation of RGB color compositions from VNIR and SWIR bands. For the detection of alteration areas and minerals associations, ratio images and spectral indexes were generated. For a more specific mineralogical determination a spectral classification by SAM technique (Spectral Angle Mapper) was performed, using the USGS (United States Geological Survey) spectral library signatures as reference. The results obtained by the different methodologies were compared among them and with the alterations maps obtained from field and laboratory works. Digital processing techniques application on ASTER images allowed us to obtain very useful information from the prospective point of view (hydrothermal alteration areas distribution, alteration types, mineral associations, relative abundance, etc.). Given the results achieved, the methodology used in the San Jose Mining District, could serve as reference for the implementation of future exploratory works in the whole region.Pages: 3643-365
Uso de imagens orbitais de radar como ferramenta para sistemas de alerta a inundações
No full textThe aim of this study is to conceive a methodology for flood alert. The study area, situated in southern Gulf of Mexico, embraces the surroundings of Villahermosa city, Mexico, where a major flood took place in 2008. RADARSAT-1 and RADARSAT-2 images were interpreted in conjunction with altimetric and pluviometric data. In a previous study (Silva et al., 2011), the USTC algorithm (Unsupervised Semivariogram Textural Classifier) was used to identify cover types representative of water, dry forest and flooded forest/buildings. Furthermore, in order to quantify the extension of inundation, precipitation data coming from climatological stations were associated with the evolution in time of the flood area distribution. Such a procedure allowed the identification of the inundation duration. In addition, the morphometric analysis permitted to establish areas within the basin that are prone to inundation, which was confirmed with the generation of flooded scenarios and RADARSAT image analysis. Based on these results, it was possible to determine a lower limit of alert for precipitation and corresponding inundated areas in 2008. The fact that flood-related natural disasters are possible to happen in Villahermosa enhances the importance of the proposed methodology. If the necessary data sets are available, the same approach can be applied in other coastal areas of the tropics.Pages: 6665-667
Comparação de métodos estatísticos para avaliação espectral de atributos de solo
No full textIn this paper we tested the possibility of using a spectral evaluation methodology who understood the whole spectral curve of some soil classes. For this purpose were collected 103 soil samples on a grid with spacing of 500 meters among plots. These samples were subjected to analysis of chemical, particle size and spectral attributes. Statistical evaluation of the spectral data, the main focus of this work was performed by simple linear regression using SAS software in all wavelengths collected by the spectroradiometer and also by regression with the method of partial least squares (PLS) in Parles software. The models for the attributes of soil H++ Al3+, Ca2+, SB, CTC, total sand content, clay content proved high potential correlation, since all had R2 greater than 0.80. The calibration models for Mg2+ and C show only the possibility to discrimination of low and high concentrations. The calibration models generated for the attributes pH, Al3+, K+, P and silt content exhibited R2 less than 0.5 and are considered low in comparison to other work and therefore not entered the cross validation method. In the validation phase of the calibrated models, to estimate the attributes C and Mg 2+ did not show satisfactory performance, unlike other attributes evaluated showed that R2 greater than 0.80. Thus, it was verified that the method is valid for most attributes, with exceptions made for pH, Al3+, K+, P and Silt content.Pages: 9277-928
Uso de técnicas de processamento de imagens Landsat-TM para o mapeamento de florestas exóticas e nativas em regiões montanhosas
No full textMost of the remaining forests in the southeast region of the country are located in mountainous areas. To manage and monitor these forest resources is necessary periodic mapping. His mapping can be done through satellite images of medium spatial resolution, because your area has big extensions. However, automatic methods of classification may have their performance reduced because of topographic effects. This article demonstrates the application of image processing techniques to mitigate these effects in images TM. After these processes, we investigated the performance of the unsupervised classification algorithm ISOSEG in TM with and without such processing. The results show the processing applied reduced the lighting effect and highlighted the subtle differences between targets with spectral similarity, such as native and exotic forest. These characteristics allowed the segmentation of images and improved the accuracy of the mapping.Pages: 3428-343
Zoneamento e Análise das Consequências Socioambientais das Áreas Inundáveis no Baixo Jucu em Vila Velha (ES) com emprego de Imagens Temporais CBERS-2B/CCD
No full textThis article aims to main zoning areas often flooded plain of the lower river Jucu in Vila Velha (ES) and surroundings, with the aid of GIS tool integrated products and Remote Sensing techniques using free digital satellite images from CBERS-2B/CCD INPE. Then, we attempted to analyze the environmental impacts associated with natural flooding of these regions and the use and occupation undue soil. Based on the integration and analysis of the physical and environmental aspects along with the socioeconomic, showed that the methodology of delimitation of seasonally flooded area using temporal satellite images shown satisfactory and can be applied in regions with similar morphologies. This methodology is also important information in making decisions, as the most appropriate ordering of these landscapes, preventing irreversible environmental and social damage or difficult and costly solution occur.Pages: 1758-176
Mapeamento de unidades ambientais do município de Caucaia (CE) através de técnicas de geoprocessamento: Subsídios para o ordenamento territorial
No full textIn this study was performed the analysis of geoenvironmental aspects of the city of Caucaia, state of Ceará. Caucaia has shown in recent years a growing population, triggering a pressure on natural resources, justifying the need for studies to identify areas most vulnerable and the most environmentally suitable to population density, generating elements for territorial planning. To perform the mapping, we used the field work and the use of ortophotos scale 1:20,000. We identified and mapped four natural regions, seven environmental systems and eleven environmental subsystems, verifying that depression sertaneja is the environmental system of higher occurrence in the city, being followed by the accumulation glacis. Thus, territorial planning, based on the aspects geoenvironmental, emerges as an important delineator of organizational conditions and functionality environments, given that always considers its environmental and socio-spatial dynamics integrated.Pages: 4032-403
Ortorretificação e classificação de imagens dos satélites de alta resolução World View2 e GeoEye
No full textIn the processing of high-resolution images there are still difficulties. And, due to the fact of specific software for treating high resolution images presents high cost, it is necessary to test different types of classifiers and parameters available in softwares low cost or free market to classify high-resolution images. Based on these considerations, the present study was developed to orthorectify and to test different algorithms to classify high resolution images, in this case images of satellite Word View2 and GeoEye. There were tested the following supervised classifiers available in ENVI 4.8 software: Spectral Angle Mapper, Maximum Likelihood, Mahalanobis Distance, Minimum Distance and Parallelepiped, applying them to sampled areas of two watersheds included in the Atlantic Forest in the state of Rio de Janeiro. The results showed that the best classifier for most classes was the Maximum Likelihood (MAXVER) to both images and watersheds. In general, the greatest possibility of confusion was found between the classes: Initial Vegetation and Advanced to Medium Vegetation (0.20), Advanced to Medium Vegetation and Banana (0.28), Initial Vegetation and Agriculture (0.29), Initial Vegetation and Banana (0.38), Banana and Agriculture (0.50), Pasture and Agriculture (0.80). The MAXVER was the classifier that presented best Overall Accuracy (65,38% and 63,165 to each sampled area) and Kappa Coefficient (0,60 and 0,57). But, these values were considered unsatisfactory.Pages: 1379-138
Delimitação automática da microbacia hidrográfica do Rio das Lontras, através de dados SRTM
No full textKnowing the limits of territorial units is essential for planning and management of water resources. Thus, the automatization of the delimitation of watersheds and its characterization are fundamental to the appropriate execution of the National Policy of Water Resources. Therefore, the aim of this work was to implement a methodology of automatic delimitation of watersheds, using data from de project Shuttle Radar Topographic Mission (SRTM), processed in a Geographic Information System (GIS). The use of SRTM data integrated in GIS enabled a quick and efficient delimitation of the watershed, and for contributing to the characterization of the delimitation, it can be applied in the management of water resources, contributing to the process of making decisions.Pages: 5515-552