On étudie une zone de la Cordillère de la Costa au sud du Chili (41°15'S-41°00'S), habitée par l'ethnie Mapuche. Là, il y a encore des précieuses forêts tempérées, qui ont une haute priorité de conservation au plan mondial. Mais, il est constaté, un processus continu de transformation lié à l'extraction de bois de chauffe, l'élevage extensif, la perturbation de l'habitat du conifère Fitzroya cupressoides, et le déploiement des plantations forestières. Sur la base d'une classification multi-temporelle d'images satellite, une modélisation spatio-temporelle a été mise en œuvre pour comprendre et anticiper la dynamique de la détérioration de la forêt. L'analyse rétrospective révèle qu'entre 1976 et 2008, la forêt a été réduite 12400 ha, soit un taux de déforestation 0,25% par an. 20 modèles de déforestation sont calibrés, 4 d'entre eux sont sélectionnés pour construire des scénarios prospectifs. La simulation prévoit que la diminution de forêt entre 2008 et 2040 pourrait atteindre 13000 ha. Deux autres modèles mis en œuvre simulent l'évolution de la qualité de l'habitat de F. cupressoides. Les scénarios prévoient une augmentation de la surface détériorée ; l’extrapolation à partir de la période d’apprentissage 2001-2008 pendant laquelle la plus grande part de terres indigènes a été légalisée, par rapport à la période précédente à 2001, prédit une augmentation plus significative de la surface détériorée. On fournit un outil pour amplifier la gestion territoriale des systèmes environnementaux toujours non incorporée dans la planification chilienne. Enfin, la thèse met également d'intéressantes comparaisons entre modèles qui diffèrent par rapport aux paramètres de calibration. / SAn area of the Cordillera de la Costa is studied in southern Chile (41°15'S - 41°00'S), inhabited by ethnic Mapuche. There are still valuable temperate forests that have high conservation priority worldwide. But it is found, a continuous process of transformation related to the extraction of firewood, ranching, disruption of habitat conifer Fitzroya cupressoides, and deployment of forest plantations. Based on a multi-temporal satellite images classification, a spatial-temporal modeling was implemented to understand and predict the dynamics of the deterioration of the forest. Retrospective analysis revealed that between 1976 and 2008, the forest has been reduced 12,400 ha, a deforestation rate of 0.25% per annum. 20 models of deforestation are calibrated, 4 of them are selected to construct scenarios. The simulation predicted that reductions of forest between 2008 and 2040 could reach 13,000 ha. Two other models implemented simulate changes in habitat quality F. cupressoides. Scenarios predict an increase in the damaged surface; extrapolating from the learning period 2001-2008 during which the largest share of indigenous lands was legalized, compared with the previous period to 2001, predicts a significant increase in the damaged surface. In this way, it provides a tool to enhance the local management of environmental systems still not incorporated into the Chilean planning. Finally, the thesis also makes interesting comparisons between models that differ with respect to the calibration parameters.

It turns out that for most users of spatial data presents a harmonization of data (as an elimination of heterogeneity) as the greatest problem - a transformation their own data sets to the data specifications produced by the INSPIRE Directive so as not to limit current processes processing, management, sharing and presentation of data. Therefore, they need the widest possible description of spatial data, because it will only eliminate the problems of heterogeneity. The author of this Ph.D. thesis dealt with the opportunities which ontologies bring to the process of data modelling. The aim was to describe using ontologies the selected part (classification systems) of data models of spatial data themes "Land Cover" and "Land Use" which are defined in Annexes II and III of the INSPIRE directive. Developed ontologies serve as a source of information on classification systems. Their evaluation (so-called reasoning) will provide new information (relationships between taxonomy elements) that play an important role in the transformation process of classification systems as part of the harmonization of spatial data. Creating of ontologies consisted of three main steps - selection of appropriate methodology, development of a basic ontology structure and iterative process of populating and specification of...

As mudanças de uso e cobertura da terra ocasionam alterações no balanço de energia, na temperatura do ar, na precipitação, na umidade do ar e na circulação regional e global. Consequentemente, mapas de uso e cobertura da terra e suas respectivas características físico-químicas e biológicas constituem-se em uma importante variável na modelagem numérica de sistemas terrestres. Entretanto, na maioria dos modelos regionais de previsão do tempo e clima, o mapa de uso e cobertura da terra não é atualizado com frequência e encontra-se defasado, o que influencia os resultados das simulações. Desta forma, o presente trabalho tem como objetivo principal verificar o impacto nas simulações numéricas do RegCM4 oriundas da atualização do mapa de uso e cobertura da terra e dos parâmetros físicos como, por exemplo, o índice de área foliar (IAF), a reflectância no visível ( 0,7m) e no infravermelho próximo e médio (>0,7m) utilizados pelo Biosphere Atmosphere Transfer Scheme (BATS), além da validação dos dados provenientes do Tropical Rainfall Measuring Mission (TRMM). Consequentemente, a comparação entre os dados de precipitação mensal estimada pelo TRMM e 183 estações meteorológicas espalhadas por todo o Brasil apresentam uma concordância de aproximadamente 97%. Ainda, em relação à precipitação, temperatura máxima e temperatura mínima, obtém-se um melhor ajuste do modelo RegCM4 quando os dados de entrada do modelo são compostos pelas reanálises do SST-ERA-Interim e ERA-Interim e com a parametrização cúmulos proposta por Emmanuel. A partir da atualização do uso e cobertura da terra utilizado no modelo de superfície BATS para o ano 2007 obteve-se uma melhora de 10% na simulação da precipitação, aumentando de 0,84 para 0,92 o coeficiente de correlação (significante a p<0,05, teste t-student). Do mesmo modo, a simulação realizada com a alteração dos valores de fração máxima de cobertura vegetal apresentou valores de precipitação 18% acima dos observados. Ainda, as alterações da reflectância no visível e no infravermelho próximo e do índice de área foliar superestimaram a precipitação em 19% e 23%. Ressalta-se que todas as simulações apresentaram boa concordância no que diz respeito à temperatura máxima e mínima, apresentando valores muito próximos ao esperado. A variação trimestral dos parâmetros físicos utilizados pelo modelo de superfície BATS reduziram para 3% as superestimativas de precipitação, provendo uma correlação de 92% (significante a p<0,05, teste t-student). Em relação às variáveis meteorológicas, as principais diferenças encontradas na evapotranspiração, precipitação, umidade relativa do ar e temperatura a 2 metros concentram-se na região noroeste do Estado do Mato Grosso, nas divisas dos Estados de Mato Grosso do Sul e Mato Grosso com a Bolívia (áreas alagadas do Pantanal brasileiro), região noroeste do Paraguai e para a região da Bacia do Rio da Prata na Argentina, Região Nordeste do Brasil entre outras, devido principalmente à alteração da classe Floresta Ombrófila Densa para áreas de pastagem e agricultura e à substituição de áreas de agricultura por áreas de gramíneas, pastagem, vegetação arbustiva e Floresta Estacional Decidual. / The land use and land cover changes modify the air temperature, precipitation, air moisture, energy balance and regional and global circulations. Consequently, the land use and land cover maps and its physical-chemical and biological properties are important variables for numerical modeling of terrestrial systems. However, in most of regional weather and climate models the land use and land cover maps are not frequently updated, being out-of-date and, influencing the results of simulations. Therefore, this work has as main objective to analyze the impacts in numerical simulations by RegCM4 of land use and land cover maps updating, as well as the alteration of physical parameters, such as the leaf area index (LAI), the visible (<0.7 m) and near infrared and medium (> 0.7 m) reflectance used by the Biosphere Atmosphere Transfer Scheme (BATS). The validation of simulated precipitation by comparison to the Tropical Rainfall Measuring Mission (TRMM) data is also an objective. The comparison between monthly precipitation data estimated by TRMM and the 183 weather stations distributed throughout Brazil presented a concordance of approximately 97%. Also related to rainfall, maximum and minimum temperature simulations, better assessments of RegCM4 were found when SST-ERA-Interim, ERA-Interim reanalysis and cumulus parameterization proposed by Emmanuel were used. In addition, the update of the South America 2007 land use and land cover map used by BATS has improved the simulation of precipitation in 10%, increasing the correlation coefficient from 0.84 to 0.92 (significant at p <0.05, t-test student). Moreover, the results show good agreement between simulated and observed maximum and minimum temperature, with correlation coefficients near to 0.95 for both variables. Correspondingly, the simulations performed with adjustments on maximum fractional of vegetation cover showed precipitation mean value 18% above the observed data. Changes in visible and infrared reflectance and in the leaf area index overestimated the precipitation in 19% and 23%, respectively. It is noteworthy that all simulations showed good agreement for maximum and minimum temperature, with values very close to those observed. The three-monthly (seasonal) change of the physical parameters, which characterize the surface model BATS, provided a reduction of 3% in the mean simulated precipitation if compared to the results obtained without seasonal variation of surface parameters. The linear correlation between simulated and observed monthly precipitation obtained by considering seasonal changes in the surface parameter was equal to 92% (significant at p <0.05, Student\'s t-test). Regarding to atmospheric variables, the more significant differences observed in evapotranspiration, precipitation, relative humidity and temperature above two meters from surface are located in the northwest of Mato Grosso state, at the Mato Grosso do Sul, Mato Grosso states and the Bolivia bordering (placed on the Brazilian Pantanal wetlands), over Northeastern Brazil region, northwestern Paraguay, and River Plate Basin, among others. These important differences are due to the substitution of Tropical Rain Forest by pasture and agriculture and, the replacement of agricultural areas by areas of grass pasture, shrub and Deciduous Forest.

O estudo avaliou a influência das árvores urbanas na manutenção da qualidade do ar e na melhoria da saúde pública, na cidade de São Paulo/SP. Para isso, essa dissertação apresenta três artigos, sendo o primeiro um artigo de revisão sistemática da literatura científica dos últimos 10 anos, contendo pesquisas que envolvem ao menos dois dos três temas: florestas urbanas, poluição do ar e saúde pública, de forma integrada. O segundo trabalho demonstra como a distribuição do solo urbano e as árvores urbanas, levantadas através de um Inventário Florestal, influenciam na qualidade do ar. O terceiro artigo discute o processo de urbanização e como o uso e ocupação do solo podem influenciar nos casos de Câncer de pulmão em idosos. As pricipais conclusões observadas são que as publicações da área vêm crescendo com o passar dos anos em revistas consideradas de alta qualidade científica, porém há uma carência de estudos que englobem os três grandes temas de forma integrada; Copa arbórea/arbustiva e Relvado podem ser utilizados como ferramentas eficientes na gestão pública, no combate à poluição por Material Particulado de 10&mu;m; áreas abertas de relvado, como praças, campos e jardins, assim como a presença de árvores, podem ser considerados como fatores que favorecem a saúde da população de idosos paulistanos, e são encontradas quando mais distantes do centro da cidade. / The study evaluated the influence of urban trees in maintaining air quality and improving public health in São Paulo. For this, this dissertation presents three articles, the first one is a systematic review of the scientific literature of the last 10 years, containing research involving at least two of the three themes: urban forests, air pollution and public health, in an integrated way. The second work demonstrates how the distribution of land use and urban trees, raised through a Forest Inventory, influence the quality of the air. The third article discusses the process of urbanization and how land use and land cover can influence the cases of lung cancer in old people. The main conclusions observed are that the publications have been growing over the years in journals considered of high scientific quality, but there is a lack of studies that encompass the three major matters in an integrated way; Arboreal / Shrub Cup and Grass can be used as efficient tools in the public management, in reduction of pollution by particulate matter of 10&mu;m; Open areas such as squares, fields and gardens, as well as the presence of trees, can be considered as factors that favor the health of the elderly population of São Paulo, and these environments are found when further away from the city center.

Entender como o balanço hídrico de uma bacia hidrográfica é afetado pela conversão de florestas em cultivos agrícolas e pastagens permite avaliar os possíveis impactos da mudança de uso e cobertura do solo na sustentabilidade regional. Com o modelo hidrológico de grande escala VIC (Variable Infiltration Capacity) calibrado para as condições locais simulou-se os possíveis efeitos em três cenários de substituição da cobertura vegetal na bacia do rio Ji-Paraná, Rondônia. Um cenário imagina a substituição total da floresta pela cultura da soja, e nessas condições a vazão do rio aumentou em 28% durante o período chuvoso, e em até 70% na época seca, quando comparados às vazões atuais. Quando a substituição foi por pastagens esse aumento foi um pouco menor, 11% no período chuvoso e 16% no período seco. Esses dois cenários mutuamente exclusivos foram selecionados por representar condições extremas de uso do solo. Um terceiro cenário simulou a possível condição original da bacia coberta totalmente pela floresta. Esse último cenário permite inferir sobre as possíveis conseqüências das mudanças já ocorridas visto que as medidas de vazão são recentes (menos de 30 anos). Nessa condição a vazão sofreu redução de 9% durante o período chuvoso e de 20% no período seco. O modelo indicou ainda redução de 30% na evapotranspiração dessas culturas, durante o período seco, quando comparado ao cenário de floresta. / The understanding of how the water balance of a hydrographic basin is affected by the conversion of forests into crops and pasture allow to evaluate the possible impacts of the land use and land cover changes in the regional sustainability. Using a macro-scale hydrologic VIC (Variable Infiltration Capacity) model calibrated for local conditions we simulated the possible effects on the hydrology under three scenarios of land cover substitution of the Ji-Paraná basin, state of Rondônia (western region of Brazil). One scenario simulated the total replacement of forest by soybeans crops, under this condition the outflow of the river increased in 28% during the rainy period, and up to 70% at the dry period, when comparative to the current outflows. When forest was replaced by pastures, this increase was a little lower, 11% in rainy season and 16% in the dry period. These two exclusive scenarios have been selected by representing extreme conditions of land use. One third scenario simulated the possible original condition of the basin covered by forest. This last scenario allows to infer on the possible consequences of the actual changes since recent measures of outflow (less than 30 years). In this condition the outflow suffered a reduction of 9% during the rainy period and 20% in the dry period. The model indicated a decrease of 30% in evapotranspiration for these crops in dry period compared to the forest scenario.

Entender como o balanço hídrico de uma bacia hidrográfica é afetado pela conversão de florestas em cultivos agrícolas e pastagens permite avaliar os possíveis impactos da mudança de uso e cobertura do solo na sustentabilidade regional. Com o modelo hidrológico de grande escala VIC (Variable Infiltration Capacity) calibrado para as condições locais simulou-se os possíveis efeitos em três cenários de substituição da cobertura vegetal na bacia do rio Ji-Paraná, Rondônia. Um cenário imagina a substituição total da floresta pela cultura da soja, e nessas condições a vazão do rio aumentou em 28% durante o período chuvoso, e em até 70% na época seca, quando comparados às vazões atuais. Quando a substituição foi por pastagens esse aumento foi um pouco menor, 11% no período chuvoso e 16% no período seco. Esses dois cenários mutuamente exclusivos foram selecionados por representar condições extremas de uso do solo. Um terceiro cenário simulou a possível condição original da bacia coberta totalmente pela floresta. Esse último cenário permite inferir sobre as possíveis conseqüências das mudanças já ocorridas visto que as medidas de vazão são recentes (menos de 30 anos). Nessa condição a vazão sofreu redução de 9% durante o período chuvoso e de 20% no período seco. O modelo indicou ainda redução de 30% na evapotranspiração dessas culturas, durante o período seco, quando comparado ao cenário de floresta. / The understanding of how the water balance of a hydrographic basin is affected by the conversion of forests into crops and pasture allow to evaluate the possible impacts of the land use and land cover changes in the regional sustainability. Using a macro-scale hydrologic VIC (Variable Infiltration Capacity) model calibrated for local conditions we simulated the possible effects on the hydrology under three scenarios of land cover substitution of the Ji-Paraná basin, state of Rondônia (western region of Brazil). One scenario simulated the total replacement of forest by soybeans crops, under this condition the outflow of the river increased in 28% during the rainy period, and up to 70% at the dry period, when comparative to the current outflows. When forest was replaced by pastures, this increase was a little lower, 11% in rainy season and 16% in the dry period. These two exclusive scenarios have been selected by representing extreme conditions of land use. One third scenario simulated the possible original condition of the basin covered by forest. This last scenario allows to infer on the possible consequences of the actual changes since recent measures of outflow (less than 30 years). In this condition the outflow suffered a reduction of 9% during the rainy period and 20% in the dry period. The model indicated a decrease of 30% in evapotranspiration for these crops in dry period compared to the forest scenario.

CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / presente pesquisa teve como objetivo principal a elaboraÃÃo do diagnÃstico georreferenciado do uso e ocupaÃÃo do solo nas Ãreas de influÃncia direta da Sub-Bacia B1 da grande bacia do rio CocÃ, em Fortaleza, CearÃ, a partir da avaliaÃÃo das condiÃÃes de ocupaÃÃo das APP das oito microbacias constituintes, todas inseridas no meio urbano e muito impactadas pelo avanÃo da urbanizaÃÃo ocorrido nas Ãltimas dÃcadas em Fortaleza. Neste sentido, foi realizada uma busca documental em ÃrgÃos pÃblicos e por intermÃdio da internet, envolvendo legislaÃÃes, relatÃrios, depoimentos e registros cartogrÃficos sobre a Ãrea de estudo, alÃm de diversas visitas de campo, para consolidar as informaÃÃes. Entretanto, a escassez de dados gerou a necessidade de elaboraÃÃo de mapas temÃticos que tornassem possÃvel o diagnÃstico. Assim sendo, com base nas deliberaÃÃes do CÃdigo Florestal, nas ResoluÃÃes n 303/02 e 369/06 do Conama e no Decreto Municipal n 12.450/08, foram elaborados mapas georreferenciados delimitando as APP, utilizando como referÃncia ortofotos dos voos de 2004 e 2010, na coordenada UTM â zona 24S, South American Datum1969 (SAD 69), utilizando o programa ArcGis, versÃo 9.3. Foi verificado que a legislaÃÃo municipal apresenta-se mais permissiva que as deliberaÃÃes federais, diminuindo, na maioria dos casos, as APP desta sub-bacia. Em consequÃncia, observa-se excessiva ocupaÃÃo regular e irregular do solo das APP, ocasionando grandes vazios em relaÃÃo à vegetaÃÃo ciliar. Destaque-se que, mesmo atendendo parcialmente a ResoluÃÃo n 369/06 do Conama, hà intensa impermeabilizaÃÃo da superfÃcie, o que potencializa a vulnerabilidade à degradaÃÃo ambiental da Ãrea. Diante deste cenÃrio, evidencia-se o uso predatÃrio destes recursos naturais, restringindo os usos mÃltiplos e as aÃÃes que fomentem o desenvolvimento sustentado. Desta forma, propÃem-se aÃÃes de recuperaÃÃo e proteÃÃo destes ecossistemas, buscando reequilibrÃ-los, uma vez que tÃm grande importÃncia no contexto paisagÃstico e da melhoria da qualidade de vida da populaÃÃo. Por fim, sugere-se que haja maior compromisso dos ÃrgÃos pÃblicos, por intermÃdio de seus gestores, no sentido de facilitar a acessibilidade Ãs informaÃÃes sobre a problemÃtica ambiental do municÃpio, a fim de que haja maior interaÃÃo entre a academia e a gestÃo pÃblica, gerando informaÃÃes que possam subsidiar as tomadas de decisÃo / This research was mainly aimed at making the georeferenced diagnosis of the use and occupation of the lands directly influenced for the Sub-Basin B1 of the great basin of the Coco River, in Fortaleza, CearÃ, from the appreciation of the occupancy conditions of the APP constituents of the eight watersheds, all of that inserted in the urban environment and too much impacted by urbanization that has occurred in recent decades in Fortaleza. In this sense, it was realized a documentary search in public agencies and through the internet, involving legislation, reports, statements and cartographic records of the studied area, besides several field visits, to consolidate the information. However, the scarcity of data has generated the need to prepare thematic maps that make possible the diagnosis. This way, based on the Forest Code deliberations, in Resolutions n 303/02 and 369/06 from Conama, and the Municipal Decree n 12.450/08, it was elaborated georeferenced maps delimiting the APP, using as reference orthophotos ,of flights 2004 and 2010, at UTM â zone 24S, South American Datum1969 (SAD 69) coordinate, using ArcGis, version 9.3 program. It Was verified that the municipal legislation presents more permissive than federal determinations reducing, in the most of cases, the APP of this sub-basin. In consequence of that, there is too much regular and irregular occupation of APP lands, that causes large gaps in relation to riparian vegetation. It is noteworthy that, even taking part of the resolution n 369/06 from Conama, there is an intense surface sealing that potentiates the vulnerability of environmental degradation in this area. In this scenario, it becomes evident the predatory use of these natural resources restricting the multiple uses and actions that promote sustainable development. This way, it is proposed recovery actions and protection of these ecosystems trying to rebalance them, since they have great importance in the landscape and improving the quality of life. Finally, it is suggested a greater commitment of public agencies to facilitate the accessibility of information about environmental issues in the city, so that, there is greater interaction between academia and public administration, generating information that can support decision making.

abstract: Use of off-highway vehicles (OHV) in natural landscapes is a popular outdoor activity around the world. Rapid-growing OHV activity causes impacts on vegetation and land cover within these landscapes and can be an important factor in land degradation and ecosystem change. The Algodones Dunes in southeastern California is one of the largest inland sand dune complexes in the United States and hosts many endangered species. This study examines changes in land cover and OHV activity within two OHV active sites in comparison to an adjoined protected area. The study also investigates potential associations between land cover changes, climate trends, and OHV activity over recent decades. Time-series analysis was used to investigate the spatial-temporal changes and trends in the land cover in the Algodones Dunes from 2001 to 2016. In addition, high-resolution aerial photographs were analyzed to determine spatial patterns of OHV usage in comparison to visitor estimation collected by the Bureau of Land Management and observed changes in land cover composition between the control site and OHVs areas. A decreasing trend in Normalized Difference Vegetation Index over time indicates a decline in the amount of vegetation cover, which corresponds with an increasing trend in albedo and land surface temperature. Results also show a substantial difference in land cover between the control site and OHVs areas, which typically have a lower amount of vegetation cover, higher exposed sand surface, and increased anthropogenic features. Both climatic variations and OHV activity are statistically associated with land cover change in the dune field, although distinct causal mechanisms for the observed declines in vegetation cover could not be separated. The persistence of drought could inhibit vegetation growth and germination that, in turn, would hinder vegetation recovery in OHV areas. Meanwhile, repeated OHV driving has direct physical impacts on vegetation and landscape morphology, such as canopy destruction, root exposure, and increased aeolian sand transport. Active ecosystem protection and restoration is recommended to mitigate the response of declining vegetation cover and habitat loss to the impacts of OHV activity and climatic variability and allow natural recovery of re-establishement of nebkha dune ecosystems in the Algodones Dunes. / Dissertation/Thesis / Masters Thesis Geography 2018

Land cover change over time can be a useful indicator of variations in a watershed, such as the patterns of drought in an area. I present a case study using remotely sensed images from Landsat satellites for over a 30-year period to generate classifications representing land cover categories, which I use to quantify land cover change in the watershed areas that contribute to Malheur, Mud, and Harney Lakes. I selected images, about every 4 to 6 years from late June to late July, in an attempt to capture the peak vegetation growth and to avoid cloud cover. Complete coverage of the watershed required that I selected an image that included the lakes, an image to the North, and an image to the West of the lakes to capture the watershed areas for each chosen year. I used the watershed areas defined by the HUC-8 shapefiles. The relevant watersheds are called: Harney-Malheur Lakes, Donner und Blitzen, Silver, and Silvies watershed. To summarize the land cover classes that could be discriminated from the Landsat images in the area, I used an unsupervised classification algorithm called Iterative Self-Organizing Data Analysis Technique (ISODATA) to identify different classes from the pixels. I then used the ISODATA results and visual inspection of calibrated Landsat images and Google Earth imagery, to create Regions of Interest (ROI) with the following land cover classes: Water, Shallow Water, Vegetation, Dark Vegetation, Salty Area, and Bare Earth. The ROIs were used in the following supervised classification algorithms: maximum likelihood, minimum distance, and Mahalanobis distance, to classify land cover for the area. Using ArcGIS, I removed most of the misclassified area from the classified images by the use of the Landsat CDR, combined the main, north, and west images and then extracted the watersheds from the combined image. The area in acres for each land cover class and watershed was computed and stored in graphs and tables.After comparing the three supervised classifications using the amount of area classified into each category, normalized area in each category, and the raster datasets, I determined that the minimum distance classification algorithm produced the most accurate land cover classification. I investigated the correlation of the land cover classes with the average precipitation, average discharge, average summer high temperature, and drought indicators. For the most part, the land cover changes correlate with the weather. However, land use changes, groundwater, and error in the land cover classes may have accounted for the instances of discrepancy. The correlation of land cover classes, except Dark Vegetation and Bare Earth, are statistically significant with weather data. This study shows that Landsat imagery has the necessary components to create and track land cover changes over time. These results can be useful in hydrological studies and can be applied to models.

Changes in land use and land cover are associated with many environmental issues observed on the earth’s surface. In the last decades, these changes were unprece-dented, mainly in tropical forest areas. The Brazilian Amazon, the world’s largest tropical forest, lost around 200.000 km² of primary forest in the last ten years (INPE, 2005). Considering this, and the consequences caused by this deforestation, it is important to know and define correctly the responsible agents, aiming at better pub-lic policies that can help preserve the forest. Searching for indicators that could help to identify the deforestation agents, some studies, such as Mertens and Lambin (1997), suggest that every deforestation process shapes the forest land in a specific way, producing a spatial pattern that can be interpreted as indicative of the agents with specific economic activities. Based on this hypothesis, the objective of this study was to contribute to a better understanding of land change processes in the Amazon forest, investigating the linkages between spatial patterns of deforestation, as visualized in satellite images, and different agents and their specific economic activities. To reach this objective, our methodological approach was based on socio-economic data acquired at a household level combined with data from satellite im-ages. First, different spatial patterns of deforestation were identified on the satellite images, based on the typologies proposed by Husson et al. (1995). Then, some of the identified spatial patterns were isolated and analyzed for specific aspects, such as, the deforestation rate calculated through satellite images. socio-economic character-istics based on household survey data and evolution of land use and land cover based on thematic maps derived from satellite images. In addition, cluster analysis was applied using the socio-economic data (household survey) and land use and land cover data (satellite images) in a search for homogeneous groups related to the spa-tial pattern. In the end, an Analysis of Variance (ANOVA) was applied to confirm the differences between spatial patterns. The results suggested that the different spatial patterns of deforestation found in the study area can be related to specific economic activities. Nevertheless, the re-sults have indicated that the spatial configuration is not a consequence of its main economic activity. They suggest that the spatial configuration is linked to the settle-ment project, and the main economic activity in the spatial patterns is a consequence of a set of factors such as: size of property, location and disposition of the property, presence or absence of infrastructure (road, market, transportation, economic and technical).