<p>Current spatial analyses of deforestation predominantly focus on quantification, often overlooking the geometric and topological configurations that are essential for formulating spatially concrete remedial actions. Skeletons and local connected fractal dimension (LCFD) are established techniques that have been used to summarise geometric features and capture connectivity patterns. The present study analysed deforested areas in the Sumaco Biosphere Reserve at three time points from 1990 to 2018. The skeleton captured 62%, 44%, and 40% of the original deforested patches, respectively, and the complexity of connectivity patterns increased over time. A spatially explicit characterisation of the deforested patches was conducted by combining the LCFD and topological descriptors, which enabled the definition of five prioritisation levels for informed decision-making. In addition, we observed an increase in the complexity of pixel neighbourhood relationships over time. In conclusion, the spatial characterisation of the deforestation skeleton serves to further understand the dynamics of deforestation expansion from the local to the regional scale by highlighting complex connections that are significant for forest protection and mitigation efforts.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Understanding local connectivity and complexity in the skeleton of deforestation

  • Andrea Urgilez-Clavijo,
  • David Andrés Rivas-Tabares,
  • Anne Gobin,
  • Ana María Tarquis Alfonso,
  • Juan de la Riva Fernández

摘要

Current spatial analyses of deforestation predominantly focus on quantification, often overlooking the geometric and topological configurations that are essential for formulating spatially concrete remedial actions. Skeletons and local connected fractal dimension (LCFD) are established techniques that have been used to summarise geometric features and capture connectivity patterns. The present study analysed deforested areas in the Sumaco Biosphere Reserve at three time points from 1990 to 2018. The skeleton captured 62%, 44%, and 40% of the original deforested patches, respectively, and the complexity of connectivity patterns increased over time. A spatially explicit characterisation of the deforested patches was conducted by combining the LCFD and topological descriptors, which enabled the definition of five prioritisation levels for informed decision-making. In addition, we observed an increase in the complexity of pixel neighbourhood relationships over time. In conclusion, the spatial characterisation of the deforestation skeleton serves to further understand the dynamics of deforestation expansion from the local to the regional scale by highlighting complex connections that are significant for forest protection and mitigation efforts.