Optimization of landscape connectivity: A directed dynamic ecological network topology perspective
- 1Field Scientific Observation and Research Station of China's Ministry of Education for Shanxi Alpine grassland ecosystem, Shanxi, 030006, China
- 2Institute of Loess Plateau, Shanxi University, Taiyuan, Shanxi 030006, China
- 3Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing 100085, China
Abstract
The intensification of landscape fragmentation severely disrupts the directional fluxes of information, energy, and matter within ecosystems, thereby compromising ecological networks (ENs) functionality and process integrity. However, traditional ENs connectivity optimization predominantly relies on undirected topologies, neglecting the inherent asymmetry and weight heterogeneity of ecological flows—a methodological oversight that creates a critical disconnect between theoretical optimization strategies and actual ecological dynamics. To address this gap, we developed an integrated framework coupling static structural analysis with dynamic directed-flow optimization and applied it to the Sanchuan River Watershed (SRW), a typical ecologically fragile catchment on the Chinese Loess Plateau (CLP). This framework was designed to minimize ecological resistance costs while precisely identifying key nodes and priority corridors. Our results identified 35 ecological sources, 80 potential corridors, 19 pinch points, and 16 barrier points within the SRW. Comparative analysis of static versus dynamic topologies revealed three distinct categories of critical nodes: key origin-destination hubs, pivotal transit nodes, and nodes with previously underestimated importance. Notably, neglecting these “underestimated” nodes in conservation planning scenarios reduced global network efficiency by over 50%. Furthermore, by employing an improved Dijkstra algorithm, we accurately calculated the least-resistance paths between node pairs, significantly optimizing landscape connectivity. This study demonstrates that the explicit incorporation of flow directionality into ENs analysis yields a more realistic representation of ecological processes, effectively rectifying the limitations of traditional static models. The proposed framework offers a robust, operational paradigm for regional ENs planning and landscape connectivity optimization in ecologically fragile landscapes, with broad applicability to similar hilly-gully catchments in the CLP and other regions under intensive anthropogenic disturbance. • Connectivity planning of undirected ENs constrains regional sustainability. • Developed a framework for analyzing the structure and dynamics of ENs. • Integrated directional flow properties of ENs to enhance landscape connectivity. • Used the improved Dijkstra algorithm to find the shortest resistance path. • Quantified ecological impact of nodes underestimated by static analysis.
Related
- Mapping tsetse fly connectivity in Uganda with machine learning landscape genetics
- A non‐Euclidean movement model to quantify landscape connectivity with open spatial capture–recapture models
- Ecological connectivity modelling with WebAssembly
- Ecological role of emergent properties in the chemodiversity landscape
- Ecological networks balance connectivity with flexibility to perturbations
BibTeX
@article{li2026optimization,
title = {{Optimization of landscape connectivity: A directed dynamic ecological network topology perspective}},
author = {Ruobin Li and Ziqiang Du and Yuying Yan and Shurong Wu and Hong Zhang and Keming Ma},
journal = {Ecological Informatics},
year = {2026},
month = may,
doi = {10.1016/j.ecoinf.2026.103818},
url = {https://doi.org/10.1016/j.ecoinf.2026.103818},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Optimization of landscape connectivity: A directed dynamic ecological network topology perspective
AU - Ruobin Li
AU - Ziqiang Du
AU - Yuying Yan
AU - Shurong Wu
AU - Hong Zhang
AU - Keming Ma
PY - 2026
DA - 2026/05/19
JO - Ecological Informatics
DO - 10.1016/j.ecoinf.2026.103818
UR - https://doi.org/10.1016/j.ecoinf.2026.103818
AB - The intensification of landscape fragmentation severely disrupts the directional fluxes of information, energy, and matter within ecosystems, thereby compromising ecological networks (ENs) functionality and process integrity. However, traditional ENs connectivity optimization predominantly relies on undirected topologies, neglecting the inherent asymmetry and weight heterogeneity of ecological flows—a methodological oversight that creates a critical disconnect between theoretical optimization strategies and actual ecological dynamics. To address this gap, we developed an integrated framework coupling static structural analysis with dynamic directed-flow optimization and applied it to the Sanchuan River Watershed (SRW), a typical ecologically fragile catchment on the Chinese Loess Plateau (CLP). This framework was designed to minimize ecological resistance costs while precisely identifying key nodes and priority corridors. Our results identified 35 ecological sources, 80 potential corridors, 19 pinch points, and 16 barrier points within the SRW. Comparative analysis of static versus dynamic topologies revealed three distinct categories of critical nodes: key origin-destination hubs, pivotal transit nodes, and nodes with previously underestimated importance. Notably, neglecting these “underestimated” nodes in conservation planning scenarios reduced global network efficiency by over 50%. Furthermore, by employing an improved Dijkstra algorithm, we accurately calculated the least-resistance paths between node pairs, significantly optimizing landscape connectivity. This study demonstrates that the explicit incorporation of flow directionality into ENs analysis yields a more realistic representation of ecological processes, effectively rectifying the limitations of traditional static models. The proposed framework offers a robust, operational paradigm for regional ENs planning and landscape connectivity optimization in ecologically fragile landscapes, with broad applicability to similar hilly-gully catchments in the CLP and other regions under intensive anthropogenic disturbance. • Connectivity planning of undirected ENs constrains regional sustainability. • Developed a framework for analyzing the structure and dynamics of ENs. • Integrated directional flow properties of ENs to enhance landscape connectivity. • Used the improved Dijkstra algorithm to find the shortest resistance path. • Quantified ecological impact of nodes underestimated by static analysis.
SN - 1574-9541
SN - 1878-0512
ER -