National Institute for Space Research
Biblioteca Digital da Memória Científica do INPENot a member yet
6175 research outputs found
Sort by
Classificação de parques aquícolas na represa de Furnas em função da resposta espectral multitemporal
No full textThe Brazilian government is encouraging increased production in fish in large reservoirs nationals. In Furnas reservoir (MG) were proposed sixteen aquaculture parks for the installation of cages and we intend to implement a framework for monitoring water quality in these parks. Therefore it is important to know the parameters of water quality and its variation over time. This paper presents a method for classifying these aquaculture parks in function of its spectral response and multitemporal. To consider different situations in the analysis were used ten Landsat images, five in the rainy season and five in the dry season. The images were selected from the database of images from the TM sensor of Landsat 5 satellite. We used the bands 1, 2, 3 and 4, forming a database with 40 variables. In the final stage, we used the hierarchical cluster analysis by Ward method to group the parks into similar clusters.Pages: 5435-544
Imagens orbitais como alternativa para restituição do relevo e uso da terra de indicações geográficas vitivinícolas na Serra Gaúcha, RS, Brasil
No full textSome areas in Brazil are being studied to define wine terroirs and improve the quality of wines. The Brazilian Agricultural Research Corporation (Embrapa Grape and Wine) currently uses satellite images and geographic information system (GIS) for studies of geographical appellation for fine wines. Aerial surveys initially used to restore the maps were very expensive and tested alternative in this work to reduce costs were ALOS (Advanced Land Observation Satellite) image to generate digital elevation model (DEM) and to orthorectificate high resolution image such as IKONOS for land use survey and vineyards. Recently in the Serra Gaúcha Wine Region, a study for the geographical appellation of Altos Montes is being developed with partnership of the producers. The ground control points (GCP) was obtained by precision global position system (GPS), a Rational Polynomial Camera (RPC) model was built to obtain the stereoscopic image pair PRISM (Panchromatic Remote-sensing Instrument for Stereo Mapping) and IKONOS. Although ALOS be out of service since April 2011, the PRISM system is still useful for researching the relief features. So it was possible to characterize the area of geographical appellation by multiple criteria such as altitude, slope, exposure, land use and it was still possible to register the vineyards of producers in GIS and its attributes. Planimetric maps, image-maps and land use maps were produced for a detailed scale (1: 5000).Pages: 664-67
A cartografia geoambiental como método de caracterização e localização de lavouras arrozeiras, no município de Mata-RS
No full textRemote sensing and geoprocessing are geotecnologies important in supporting the data of agriculture survey, allowing studies and the understanding of environmental relations in a determined culture. In this work, physical variables were related, such as altimetry, declive and kinds of relief in order to identify and classify rice farms in Mata-RS, in 2011, through the use of geotecnologies. To carry on this study, satellite images from Landsat 5 series (INPE) were used, beyond the use of software Spring 5.2.1 (INPE) and AutoCAD Civil 3D 2012 (Autodesk). To analyze the relief, images SRTM (NASA) were used, which allowed the knowledge of altitude and declive of the area. The identification and classification of rice farms in the city were made through Spring 5.2.1 software. Considering the relief, it was found that most of the rice farms are located between 38 and 128 meters, and only a small area above 144 meters, and that most are located in declive of up to 5% inclination, considered flat surfaces. Considering the geomorphology, it was observed that the majority of rice farms in Mata are grown on the fluvial plain of the Toropi river.Pages: 401-40
Utilização de SIG na caracterização das áreas cultivadas com soja no Oeste do Paraná no ano agrícola 2010/2011
No full textSoybean is one of the most representative crops in Brazil. Paraná is one of the largest states of soybean production at the national level. The study area is located in the state of Paraná, being considered as polo in terms of mechanization and technology in the management of soybean. Studies involving factors of climate, intensity of use, features and physical properties of soil, impediments to mechanization, can provide a broader view of the soil potential, this knowledge is essential to proper use and management. Associating the importance of soybean with proper use and soil handling, the study aims to prepare maps of agricultural suitability of areas cultivated with soybean, using satellite image and information of soil type and declivity of the land, provided by official agencies, integrated into a Geographic Information System - GIS. The land suitability map was generated by the crossing of soil types and slope classes. The Oxisols and Alfisols with a slope of 20% were considered more suitable to handling and development of soybean crops. The soy mask was elaborated through supervised classification of Landsat 5/TM. Subsequent the soy mask was crossed with the suitability map, identifying and quantifying areas of soybeans crops cultivated in areas considered propitious and/or not propitious. As a result it is noteworthy that approximately 91% of the soybean crop areas are in places considered propitious to the development of culture.Pages: 748-75
Mapeamento litológico do Vale do Curaçá (BA) a partir da fusão de componentes principais de dados dos satélites ALOS-PALSAR e TM-Landsat
No full textThe purpose of this work was the lithological mapping of the Curaçá Valley, northern State of Bahia, using satellite data from ALOS-PALSAR (L-band and HH-HV polarization) and TM-Landsat. The processing chain was divided in four steps: (i) pre-processing; (ii) processing; (iii) RGB-IHS fusion; and (iv) MaxVer classification of hybrid image. The first step consisted of the coregistration of the reflected and radar images (coherence -  and amplitude - A). The second step consisted of the principal component transformation of the reflected and radar sets of images independently. The third step involved the composite of principal components of radar and reflected data using the RGB-IHS fusion method, in which the I channel (intensity) was replaced by PC1 obtained from the radar set of images. The fourth and last step corresponds to the MaxVer classification of the hybrid image. The results showed that the hybrid image represents an excellent product for photogeological mapping, allowing the individualization of the main lithological units. This result may be related, among other things, to the peculiar physiographic features of the study area, as scarce vegetation and shallow soil profile. On the other hand, the MaxVer classification from hybrid image showed a moderate result (kappa = 0,45), which allowed an incomplete discrimination of the lithological units, even considering a substantial number of homogeneously distributed training samples.Pages: 3635-364
Respostas polarimétricas de florestas degradadas pelo fogo na Amazônia a partir de imagens ALOS/PALSAR
No full textFire is one of the main factors leading to forest degradation in the Amazon by changing species composition, biomass and structure. Considering the wide geographical extension of forest fires, remote sensing provides essential data for mapping, monitoring and even modeling fire. Despite radar sensors provide information on forest structure, no study based on such data mainly focused on fire affected areas in the Brazilian Amazon. The hypotheses of the present study are that wildfires affect forest structure and the radar signal is sensitive to these changes. Given this scenario, the objective of this work was to evaluate the sensitivity of L-band SAR data to shifts in forest structure caused by wildfires through the polarimetric responses. The present study area was situated in the Arc of Fire, in the State of Roraima (Northern Amazonia), where 39 plots (0.25 ha each) were set out and inventoried, across five fire disturbance classes. Polarimetric responses of the whole set of plots were acquired from two single look complex SAR images (L band). The results indicated the dominance of the VV polarization scattering in primary and lightly burned forests, whereas a dominance of the scattering in HH polarization was noted in heavily and frequent burned forests. In addition, the degree of depolarization of the incident wave (pedestal height) was higher in primary and lightly burned forest plots. We concluded that the L-band SAR signal was sensitive to structural changes of the forest due to wildfires. This indicated significant shifts in the geometry of these targets.Pages: 8429-843
Imagens CBERS para o estudo de manguezais do Brasil: aplicações, potencialidades e limitações
No full textBrazil shows the third largest world mangrove area, but the increase of anthropogenic pressures has been leading to high loss of these forests. This highlights the need for synoptic, multiscale and multitemporal studies of Brazilian mangroves by the use of remote sensing tools. For this purpose, the CBERS (China-Brazil Earth Resources Satellite) images are valuable tools. In this paper we presented and discussed features, applications, potentialities and limitations of the CBERS-2 and 2B images from sensors IRMSS, CCD and HRC for the study of Brazilian mangroves. For a case study, images of these three sensors of the coastal zone of Sergipe State (Northeast Brazil) were processed in GIS. The use of CBERS images allows the identification (IRMSS), mapping, area quantification, change detection, identification of types of mangroves stands, analysis of canopy closure and biomass estimate (CCD) and extraction of detailed information about mangrove coverage (HRC). The selection of the most appropriate sensor for a particular mangrove study should mainly occur by the scale of analysis. The joint use of the three sensors is suitable for micro-region studies, since this approach offers a range of supplementary information, which is not possible to obtain using images of a single sensor.Pages: 2226-223
Monitoramento das Mudanças Anuais e Inter-Anuais das Margens dos rios São Francisco na Região de Pirapora
No full textSatellite images have shown great utility in water resource research. In this paper, we describe the process of making a Local Model to evaluate the inter-annual and annual dynamic of the São Francisco river banks. This study was carried out within the mid-watershed of the São Francisco river, downstream of the Três Marias dam. River width and area were measured from 21 Landsat-5 TM images covering the November 2009 to September 2011 period. The delineation of the water edges was performed using a sequence of algorithms from the eCognition software package. ArcInfo was used for managing and spatially analysing the data which were then compared with in situ measurements of flow, water level and width in regression analysis. The regression yielded poor results mainly because the drop of the river banks is very steep in the section analysed and there is little relation between water level and width. Still, the results show that determining the area and width of medium size rivers using Landsat-5 TM is quick and relatively precise. Another important information obtained from this study is in regard of the difficulty in comparing the data, that depends on: the shape of the river cross section, the frequency of sampling of the rivers width and the difficulty of obtaining images of good quality (cloudless) during the rainy season.Pages: 5745-575
Estimativa da temperatura superficial da bacia hidrográfica do rio Abiaí com a utilização de imagem do Landsat 5 TM
No full textThe human being stands out compared to other animals like modifier space with greater intensity in the area they inhabit. Thus, the relationships between man and nature have the inherent characteristic of human disturbance. Depending on the degree of action of human front to the natural reflexes is enhanced and the evidence of human activity is the modification of the surface coverage and the effect of variation in terrestrial temperatures. With the object of assisting in this study, Remote Sensing configures itself as an alternative to capture the information about the dynamics present on the surface of the earth, and the geoprocessing helper in organizing the data collected. This study aimed to analyze the relationship between Surface Temperature Estimation of river basin Abiaí and the use and occupation of land by correlating the results. To narrow the search were adopted as instruments: a) description and location of the study area, b) digital image processing of satellite, c) classification of the image and mount the final layout and d) analysis of the results regarding surface temperature, use and occupation of the land and the relationship between them. It was found that there was correlation between the degree of anthropization and increased surface temperatures of the surface, where the higher the interference human modifying the original area and coverage of the earth surface temperature is higher estimated.Pages: 8690-869
Análise multitemporal do Uso e Cobertura da Terra no município de Caçapava do Sul - RS nos anos de 1991, 2001 e 2011
No full textThis study aims to analyze the evolution of the use and land cover in the city of South Caçapava through images of the sensor TM, Landsat 5 satellite in 1991, 2001 and 2011. For the study were used remote sensing techniques, digital processing of images which were found at INPE. The study area, municipality of Caçapava South, lies in the micro saws Southeast and is located between the geographical coordinates: 30° 00' and 31° 00' south latitude and 54\ub0 00' and 53° 00' west longitude and altitude approximately 400 meters above sea level. The digital classification algorithm used was supervised Maximum Likelihood. Held digital supervised classification using the Maximum Likelihood algorithm. The thematic classes were chosen: Planted Forest, Native Forest, Field, Bare Soil, Water and Crops of Winter. Analyzing classes in some years it can be said that the significant changes that had been the Planted For0est where it increased by more than 50% in 2011 compared to 1991, however its area does not correspond to 1% of the territory the municipality. The class with the highest expansion was the Bare Soil, for the period 1991-2011 grew 117.40 km² 58% in the period 1991 to 2011. Classes Native Forest and Field, have coverage of 90% and other uses, distributed in the remaining 10% of the municipal area.Pages: 7540-754