Research on the evolution and optimization of ecological networks in arid areas based on machine learning and multi-source remote sensing data

The construction and optimization of ecological networks is a crucial strategy to enhance the stability of ecosystems. In arid and semi-arid regions, where vegetation degradation and water stress are prevalent, improving the connectivity of ecological units is crucial for achieving optimal ecologica...

Full description

Saved in:
Bibliographic Details
Main Authors: Ruyi Pan, Jie Zhang, Wentao Xue, Qianqian Xia, Jiangxiang Liu, Junjie Yan, Hongbo Ling, Xufan Jin
Format: Article
Language:English
Published: Elsevier 2025-09-01
Series:Environmental and Sustainability Indicators
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S266597272500251X
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849240610736701440
author Ruyi Pan
Jie Zhang
Wentao Xue
Qianqian Xia
Jiangxiang Liu
Junjie Yan
Hongbo Ling
Xufan Jin
author_facet Ruyi Pan
Jie Zhang
Wentao Xue
Qianqian Xia
Jiangxiang Liu
Junjie Yan
Hongbo Ling
Xufan Jin
author_sort Ruyi Pan
collection DOAJ
description The construction and optimization of ecological networks is a crucial strategy to enhance the stability of ecosystems. In arid and semi-arid regions, where vegetation degradation and water stress are prevalent, improving the connectivity of ecological units is crucial for achieving optimal ecological network design. This study proposes a new methodological framework that integrates Morphological Spatial Pattern Analysis, circuit theory, and machine learning models to explore the spatiotemporal evolution and optimization of ecological networks in Xinjiang from 1990 to 2020. The study presents a refined classification system for ecological units, focusing on the analysis of spatiotemporal patterns of vegetation degradation and drought stress, and proposes specific strategies for ecological restoration, including optimizing ecological corridors by introducing buffer zones and planting drought-resistant species, restoring key ecological areas like forests and wetlands, establishing desert shelter forests and artificial wetlands in desert regions to prevent desertification, and other similar strategies. The results show: 1) The core ecological source regions (Core) area decreased by 10,300 km2, and the secondary core regions decreased by 23,300 km2. 2) After model optimization, the connectivity of ecological sources significantly improved, with the dynamic patch connectivity and dynamic inter-patch connectivity increasing by 43.84 %–62.86 % and 18.84 %–52.94 %, respectively. 3) The proportion of areas with extraordinarily high and high vegetation cover decreased by 4.7 %, while the area of highly arid regions increased by 2.3 %. 4) Change point analysis revealed that TVDI values in the range of 0.35–0.6 and NDVI values in the range of 0.1–0.35 are critical change intervals, with vegetation showing significant threshold effects under drought stress. 5) The area of high resistance increased by 26,438 km2, the total length of ecological corridors increased by 743 km, and the total area of corridors increased by 14677 km2, with significant growth in biological corridors, facilitating smoother species migration paths. This study provides reasonable references for ecological restoration in arid regions and introduces a new approach to optimizing ecological networks.
format Article
id doaj-art-881dfb9366364bccaa07dae1b2a42c1c
institution Kabale University
issn 2665-9727
language English
publishDate 2025-09-01
publisher Elsevier
record_format Article
series Environmental and Sustainability Indicators
spelling doaj-art-881dfb9366364bccaa07dae1b2a42c1c2025-08-20T04:00:32ZengElsevierEnvironmental and Sustainability Indicators2665-97272025-09-012710083010.1016/j.indic.2025.100830Research on the evolution and optimization of ecological networks in arid areas based on machine learning and multi-source remote sensing dataRuyi Pan0Jie Zhang1Wentao Xue2Qianqian Xia3Jiangxiang Liu4Junjie Yan5Hongbo Ling6Xufan Jin7Institute of Resources and Ecology, Yili Normal University, Yining, 835000, China; College of Resources and Environment, Yili Normal University, Yining, 835000, ChinaInstitute of Resources and Ecology, Yili Normal University, Yining, 835000, China; College of Resources and Environment, Yili Normal University, Yining, 835000, ChinaInstitute of Resources and Ecology, Yili Normal University, Yining, 835000, China; College of Resources and Environment, Yili Normal University, Yining, 835000, ChinaInstitute of Resources and Ecology, Yili Normal University, Yining, 835000, China; College of Resources and Environment, Yili Normal University, Yining, 835000, ChinaInstitute of Resources and Ecology, Yili Normal University, Yining, 835000, China; College of Resources and Environment, Yili Normal University, Yining, 835000, ChinaInstitute of Resources and Ecology, Yili Normal University, Yining, 835000, China; College of Resources and Environment, Yili Normal University, Yining, 835000, China; Corresponding author. Institute of Resources and Ecology, Yili Normal University, Yining, 835000, China.Xinjiang Institute of Ecology and Geography, Chinese Academy of Sciences, Urumqi, 830011, China; State Key Laboratory of Ecological Safety and Sustainable Development in Arid Lands, Xinjiang Institute of Ecology and Geography, Chinese Academy of Sciences, Urumqi, 830011, China; Corresponding author. Xinjiang Institute of Ecology and Geography, Chinese Academy of Sciences, Urumqi, 830011, China.Institute of Resources and Ecology, Yili Normal University, Yining, 835000, China; College of Resources and Environment, Yili Normal University, Yining, 835000, ChinaThe construction and optimization of ecological networks is a crucial strategy to enhance the stability of ecosystems. In arid and semi-arid regions, where vegetation degradation and water stress are prevalent, improving the connectivity of ecological units is crucial for achieving optimal ecological network design. This study proposes a new methodological framework that integrates Morphological Spatial Pattern Analysis, circuit theory, and machine learning models to explore the spatiotemporal evolution and optimization of ecological networks in Xinjiang from 1990 to 2020. The study presents a refined classification system for ecological units, focusing on the analysis of spatiotemporal patterns of vegetation degradation and drought stress, and proposes specific strategies for ecological restoration, including optimizing ecological corridors by introducing buffer zones and planting drought-resistant species, restoring key ecological areas like forests and wetlands, establishing desert shelter forests and artificial wetlands in desert regions to prevent desertification, and other similar strategies. The results show: 1) The core ecological source regions (Core) area decreased by 10,300 km2, and the secondary core regions decreased by 23,300 km2. 2) After model optimization, the connectivity of ecological sources significantly improved, with the dynamic patch connectivity and dynamic inter-patch connectivity increasing by 43.84 %–62.86 % and 18.84 %–52.94 %, respectively. 3) The proportion of areas with extraordinarily high and high vegetation cover decreased by 4.7 %, while the area of highly arid regions increased by 2.3 %. 4) Change point analysis revealed that TVDI values in the range of 0.35–0.6 and NDVI values in the range of 0.1–0.35 are critical change intervals, with vegetation showing significant threshold effects under drought stress. 5) The area of high resistance increased by 26,438 km2, the total length of ecological corridors increased by 743 km, and the total area of corridors increased by 14677 km2, with significant growth in biological corridors, facilitating smoother species migration paths. This study provides reasonable references for ecological restoration in arid regions and introduces a new approach to optimizing ecological networks.http://www.sciencedirect.com/science/article/pii/S266597272500251XEcological networkEcological sourceEcological corridorMachine learningConnectivity
spellingShingle Ruyi Pan
Jie Zhang
Wentao Xue
Qianqian Xia
Jiangxiang Liu
Junjie Yan
Hongbo Ling
Xufan Jin
Research on the evolution and optimization of ecological networks in arid areas based on machine learning and multi-source remote sensing data
Environmental and Sustainability Indicators
Ecological network
Ecological source
Ecological corridor
Machine learning
Connectivity
title Research on the evolution and optimization of ecological networks in arid areas based on machine learning and multi-source remote sensing data
title_full Research on the evolution and optimization of ecological networks in arid areas based on machine learning and multi-source remote sensing data
title_fullStr Research on the evolution and optimization of ecological networks in arid areas based on machine learning and multi-source remote sensing data
title_full_unstemmed Research on the evolution and optimization of ecological networks in arid areas based on machine learning and multi-source remote sensing data
title_short Research on the evolution and optimization of ecological networks in arid areas based on machine learning and multi-source remote sensing data
title_sort research on the evolution and optimization of ecological networks in arid areas based on machine learning and multi source remote sensing data
topic Ecological network
Ecological source
Ecological corridor
Machine learning
Connectivity
url http://www.sciencedirect.com/science/article/pii/S266597272500251X
work_keys_str_mv AT ruyipan researchontheevolutionandoptimizationofecologicalnetworksinaridareasbasedonmachinelearningandmultisourceremotesensingdata
AT jiezhang researchontheevolutionandoptimizationofecologicalnetworksinaridareasbasedonmachinelearningandmultisourceremotesensingdata
AT wentaoxue researchontheevolutionandoptimizationofecologicalnetworksinaridareasbasedonmachinelearningandmultisourceremotesensingdata
AT qianqianxia researchontheevolutionandoptimizationofecologicalnetworksinaridareasbasedonmachinelearningandmultisourceremotesensingdata
AT jiangxiangliu researchontheevolutionandoptimizationofecologicalnetworksinaridareasbasedonmachinelearningandmultisourceremotesensingdata
AT junjieyan researchontheevolutionandoptimizationofecologicalnetworksinaridareasbasedonmachinelearningandmultisourceremotesensingdata
AT hongboling researchontheevolutionandoptimizationofecologicalnetworksinaridareasbasedonmachinelearningandmultisourceremotesensingdata
AT xufanjin researchontheevolutionandoptimizationofecologicalnetworksinaridareasbasedonmachinelearningandmultisourceremotesensingdata