« Back

Abstract:

Land cover and land cover changes affect the environment in such ways that they can influence the human health by changing the climate, weather, water, air, biodiversity, wildlife, disease risk and food security. Comparing aerial or satellite images from the same region recorded at different times, known as post-classification comparison, is an approach that aims to identify changes in land coverage and understand forest spatial pattern change over time. These patterns are extremely complex, therefore, for their description and understanding a better approach would be multifractal geometry. In this paper, by using the multifractal analysis the evolution of the forest spatial model in southern Moldova over a decade has been investigated. We explored the potential of aerial image processing, lacunarity and textural analysis to detect a correlation between lacunarity, fractal dimensions and coverage crowns areas. This analysis leads to the determination of the percentage of land coverage, the dynamics of the natural reforestation process, the location of mature trees and the determination of areas where the landscape has changed drastically due to deforestation and / or reforestation. Two opposite forest spatial patterns were identified. For four studied forests, the rate of deforestation was continuously increased between 2005 and 2010. For only one forest, the Valeni forest, the coverage crown area has increased over the analyzed time period.