Modelagem dinâmica do desmatamento na província de Manica - Moçambique

• Ano de defesa: 2020
• Autor(a) principal: Castigo David Augusto Machava
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; Brasil; IGC - DEPARTAMENTO DE CARTOGRAFIA; Programa de Pós-Graduação em Análise e Modelagem de Sistemas Ambientais; UFMG
• Palavras-chave em Português: Modelagem Dinâmica; Desmatamento; Simulação de Cenários; Manica - Moçambique; Modelagem de dados – Aspectos ambientais –; Desmatamento – Manica (Moçambique)
• Link de acesso: http://hdl.handle.net/1843/33558

Mozambique is a Southern African Country with high and diversified plant cover. This natural wealth is being threatened by a set of related factors being the expansion of agricultural areas most cited as the main cause of deforestation. In order to stop deforestation the intensification of agricultural production has been pointed as a solution. The continuous anthropic pressure on forest resources has led to decreased patterns of vegetation cover. This research aims to analyze the deforestation phenomenon in order to understand its dynamics in Manica Province - Mozambique and to explore the possible contributions of the intensification of agriculture in the reduction of deforestation. It was also intended to simulate future scenarios for deforestation from past trends. To achieve the proposed objectives, the mapping of land use and land cover was first carried out through the classification of satellite images Landsat 5 (TM) from the year 2007 and Landsat 8 (OLI) from the years 2015 and 2018. Then a questionnaire was applied to 227 heads of families between December 2018 and March 2019 in the central region of Manica province to characterize the production system of family farmers and identify spatial variables related to deforestation. After coding and processing the questionnaire data, Principal Component Analysis (PCA) and Cluster Analysis (CA) were performed to create a typology of farmers in the region. The agricultural suitability of the lands was also estimated by applying the multicriteria analysis method in order to define and delimit more favorable areas for agricultural intensification. Finally, the dynamic deforestation modeling methodology was applied in the DINAMICA EGO software. In the analysis of the questionnaires and by using the PCA it was possible to reduce 24 initial variables in 7 components that explain 76% of the total variance. The Clusters Analysis (CA) based on the 7 main components identified 3 clusters of family farmers with specific characteristics of agricultural production modes and their associations with deforestation. Cluster 1 (N = 86) is the majority, consisting of immigrant farmers for whom the decrease in forest area is mainly caused by the expansion of agricultural areas and exploitation of biomass energy (firewood and coal). Cluster 2 (N = 59) consists of farmers living in the area since birth, mostly community leaders. For these farmers the increase in production area is motivated by the need to produce more to sell surpluses. Cluster 3 (N = 82) is young farmers and producers of biomass fuel. This analysis led to the conclusion that the causes of deforestation are multiple and complex and their resolution must go far beyond the current proposal for intensifying agriculture. The assessment of the agricultural suitability of the land identified about 1268400 hectares with high aptitude for agricultural intensification without the suppression of remaining forest areas. Areas of high suitability are propitious for expanding and intensifying agricultural production and thus preventing their expansion over the forest. Finally, with dynamic modeling it was noted that deforestation occurs near of major and minor roads, in the buffer zone of protection areas, in the smallest slopes and in the areas near the main towns and cities of the province. Future deforestation scenarios based on the historical trend of the analyzed period estimate a decrease in forest cover area by about 1592,000 hectares by the year 2025, in which the province will remain only with 5% of forest area. This research can contribute to assist the formulation of Public Policies in Mozambique to combine socioeconomic development with nature conservation.