Mapeamento de pastagens com a caracterização espectral dos diferentes niveis de degradação no estado de Rondônia
| Ano de defesa: | 2022 |
|---|---|
| Autor(a) principal: |
Moura, Valdir
 |
| Orientador(a): |
Johann, Jerry Adriani
|
| Banca de defesa: |
Johann, Jerry Adriani
,
Mercante, Erivelto
,
Souza, Ranieli dos Anjos de
,
Azevedo, Emilio Carlos de
,
Aquino, Renato Eleoterio de
|
| Tipo de documento: | Tese |
| Tipo de acesso: | Acesso aberto |
| Idioma: | por |
| Instituição de defesa: |
Universidade Estadual do Oeste do Paraná
Cascavel |
| Programa de Pós-Graduação: |
Programa de Pós-Graduação em Engenharia Agrícola
|
| Departamento: |
Centro de Ciências Exatas e Tecnológicas
|
| País: |
Brasil
|
| Palavras-chave em Português: | |
| Palavras-chave em Inglês: | |
| Área do conhecimento CNPq: | |
| Link de acesso: | https://tede.unioeste.br/handle/tede/6354 |
Resumo: | The major changes that have been occurring in the diet, especially in emerging countries, create a projection of greater consumption of animal and plant foods and, consequently, the need to increase production without generating environmental impacts on forest remnants. Thus, the knowledge and understanding of the spatial-temporal dynamics of agriculture and cattle raising areas is a strategic issue since Brazilian agribusiness accounts for about 21.5% of the national GDP, requiring more specific studies on the areas occupied by agriculture and cattle raising. Concerning agriculture, the monitoring of agricultural production, carried out by the IBGE and CONAB, is based on sample surveys with rural producers, cooperatives, and agricultural financing data, which gives them high cost, time-consuming execution, subjectivity, imprecision, and lack of knowledge about the spatial distribution of agricultural production. Conversely, cattle-raising occupies approximately 220 million hectares, distributed throughout all regions of the national territory; however, around 60% present some level of degradation, generating losses and, consequently, a reduction in the carrying capacity of animals. With this, new deforestation arises so the animals can be housed comfortably. Therefore, it is necessary to propose objective methods and the use of geotechnologies associated with data mining techniques and spatial variability analysis methods through precision agriculture (or precision cattle breeding) to make the current area and crop estimation system more efficient and dynamic. This is because it is possible to work with good quality images available daily, identifying the suitability of areas and indicating their efficient use so that the conversion of exploitation forms can be carried out without degrading the forest. In this context, the general objective of this research was to develop a methodology to map the land use and evaluate the levels of pasture degradation, using different remote sensing and machine learning products, in the state of Rondônia. To this end, the work was divided into three papers: the first, investigates the origin of the current dynamics of land use and occupation in Rondônia, evaluating the consequences and environmental impacts of the colonization models, whose implementation was a factor that accelerated deforestation in the state, resulting in the landscapes over which farming and cattle ranching advances; the second proposes a new intelligent and automated methodological structure for defining desirable land use, mapping and quantifying the areas occupied by agricultural crops and pasture, aiming to delimit productive areas through conversion between use classes based on multicriteria, allowing the reordering of exploitation for livestock and agricultural production to be carried out without advancing over the forests; the third is a proposal for defining new levels of pasture degradation that meets the reality of the field and, furthermore, evaluates the spectral behavior of the different levels of degradation in order to discretize them spectrally. |