Vulnerabilidade ambiental às mudanças climáticas e ao uso e cobertura da terra na região semiárida do nordeste do Brasil

• Autor(a) principal: SILVA, João Paulo Ferreira da
• Data de Publicação: 2019
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UFRPE
• Texto Completo: http://www.tede2.ufrpe.br:8080/tede2/handle/tede2/8218
• topic: Mudança climática; Fragmento florestal; Bacia hidrográfica; Ecossistema; CIENCIAS AGRARIAS::RECURSOS FLORESTAIS E ENGENHARIA FLORESTAL

Landscape design, resilience, and the provision of ecosystem services are associated with occupation and land use. Forest cover plays important roles in environmental balance, and knowledge of its spatial dynamics and distribution has been central to river basin planning and management. The study site was the area of the Brígida river, Northeastern Brazil. The objective of this work was to analyze the pattern of forest fragmentation and its implications for the environmental vulnerability of the Brígida river Basin. To perform the work, remote sensing techniques were used in conjunction with landscape metrics and map algebras. Were performed Image processing, training and supervised classification on the GHG in-earth platform - Google Earth Engine, Code and Workspace mode. We used Landsat Collection 1 orbital images generated by the Landsat 4-5 sensor Thematic Mapper TM, Landsat 7 sensor Enhanced Thematic Mapper Plus (ETM +) and Landsat 8 Operational Land Imager (OLI) for orbit 217 points 65 and 66, in the period between the years 2000 to 2015 and spatial resolution of 30 m. The Random Forests classification algorithm and the accuracy calculation was performed with the API of the acuracy GH based on the matrix of confusion among the classes agriculture, water, Savannah-Southeastern Forest, Savannah / Seasonal Forest and Intensive Use (Rainfed agriculture, pasture native soil, exposed soil, agricultural fallow and urbanized nucleus). Fragmentation analysis was done using the software ArcGis version 10.4 and Fragstats version 4.2. To order the variances of the metrics between the years 2000 and 2015 for each class, a PCA (Principal Components Analysis) factorial analysis was performed. The analysis of the environmental vulnerability of the basin was estimated by the method of Hierarchical Analysis of Processes (AHP), fusification of fourteen subcriteria in a matrix format organized in physiographic, anthropogenic and climatic factors. The estimation of potential and provision of ecosystem services was evaluated from the map of land use / coverage. In the fifteen-year period, there was a 16% reduction in the Savannah-Steep Forested cover - SEF, with a positive increase of 13.93% of the intensive use area, which demonstrates a direct conversion between these two landscape classes. Among the landscape metrics, the connectivity (COHESION) and percentage of the highest class in the landscape (LPI) for SEF coverage showed the greatest variations between the years 2000-2008 with stabilization in subsequent years. The rotations of the main components indicate a strong relationship between the drastic periods of the basin water deficit with the largest variations of the SEF higher class metrics (PLAND) and spot number fragmentation (NP) for the SFE transition cover. The three scenarios of environmental vulnerability indicate that on average 42% of the basin area is between high and very high vulnerability. The variation in the supply potential of ecosystem services in the basin has a high relation of significance with the water storage capacity of the reservoirs and a moderate variation in the period of fifteen years for the other services provided by the land use / cover. The estimated totals in ($ USD 1000) for S.E. of provision are of USD 4,137.37 and Regulation of $ USD 11,442.84. Planning of the use of natural resources should take into account the fragilities of ecosystems and river basins as a unit of the landscape for adopting measures of conservation, management and control of the use of natural resources.