Scholarlink

Context <p>The Kavango Zambezi Transfrontier Conservation Area (KAZA-TFCA) hosts numerous protected areas and a high diversity of fauna, including emblematic species. Because animals move between protected areas through landscapes used by rural communities, preserving landscape connectivity is essential for long-term biodiversity conservation. Most existing connectivity studies in dry forest landscapes relying on expert knowledge characterize habitat preferences through poorly detailed global land use/land cover (LULC) maps. In this work we leverage recent advances in remote sensing that enable more detailed and cost-effective LULC mapping and environmental characterization to improve potential connectivity models.</p> Objective <p>The aim of this study was to develop a framework to assess how remote sensing data can enhance rule-based resistance models of potential landscape connectivity, using the African savanna elephant (<i>Loxodonta africana</i>) as focal species.</p> Methods <p>Landscape resistance maps of elephant movement were generated from two LULC classifications using ecological information on habitat preferences derived from the literature. These LULC products consisted of a global classification and a more detailed classification distinguishing forest types and forest structure. Additional ecological and geographical variables known to influence elephant movement were also incorporated. A graph-based potential connectivity index was modeled, and the models were evaluated qualitatively against corridors identified by the stakeholders in a local district in Zimbabwe.</p> Results <p>As hypothesized, distinguishing between the type and structure of the forest in LULC improved potential connectivity precision. Furthermore, including additional environmental variables further highlighted the potential connectivity contrasts between inside participatory mapped corridors and outside.</p> Conclusions <p>This study provides a resource-efficient, operational framework for testing the effects of detailed LULC in dry forest landscapes and environmental variables on potential connectivity models and corridor identification. The approach offers valuable insights for landscape-level conservation planning and can support more informed, evidence-based management of potential connectivity for elephants movement in fragmented landscapes.</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

Enhancing rule-based potential connectivity models for African savanna elephants (Loxodonta africana) using remote sensing in Binga District, Zimbabwe

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月19日

查看全文
  • Colas Guillon,
  • Lilian Blanc,
  • Hadrien Vanthomme,
  • Sébastien Le Bel,
  • Jacques Baudry,
  • Camille Deforceville,
  • Hélène Dessard,
  • Maxwell Phiri,
  • Julie Betbeder

摘要

Context

The Kavango Zambezi Transfrontier Conservation Area (KAZA-TFCA) hosts numerous protected areas and a high diversity of fauna, including emblematic species. Because animals move between protected areas through landscapes used by rural communities, preserving landscape connectivity is essential for long-term biodiversity conservation. Most existing connectivity studies in dry forest landscapes relying on expert knowledge characterize habitat preferences through poorly detailed global land use/land cover (LULC) maps. In this work we leverage recent advances in remote sensing that enable more detailed and cost-effective LULC mapping and environmental characterization to improve potential connectivity models.

Objective

The aim of this study was to develop a framework to assess how remote sensing data can enhance rule-based resistance models of potential landscape connectivity, using the African savanna elephant (Loxodonta africana) as focal species.

Methods

Landscape resistance maps of elephant movement were generated from two LULC classifications using ecological information on habitat preferences derived from the literature. These LULC products consisted of a global classification and a more detailed classification distinguishing forest types and forest structure. Additional ecological and geographical variables known to influence elephant movement were also incorporated. A graph-based potential connectivity index was modeled, and the models were evaluated qualitatively against corridors identified by the stakeholders in a local district in Zimbabwe.

Results

As hypothesized, distinguishing between the type and structure of the forest in LULC improved potential connectivity precision. Furthermore, including additional environmental variables further highlighted the potential connectivity contrasts between inside participatory mapped corridors and outside.

Conclusions

This study provides a resource-efficient, operational framework for testing the effects of detailed LULC in dry forest landscapes and environmental variables on potential connectivity models and corridor identification. The approach offers valuable insights for landscape-level conservation planning and can support more informed, evidence-based management of potential connectivity for elephants movement in fragmented landscapes.