Spatio-Temporal Evolution and Driving Forces of Habitat Quality in China’s Arid and Semi-Arid Regions: An Interpretable Machine Learning Perspective for Ecological Management

Shihao Liu and Jinchuan Huang ()
Additional contact information
Shihao Liu: Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China
Jinchuan Huang: Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China

Land, 2025, vol. 14, issue 10, 1-24

Abstract: Against the global biodiversity crisis, arid and semi-arid regions are sensitive indicators of terrestrial ecosystems. However, research on their habitat quality (HQ) evolution mechanism faces dual challenges: insufficient multi-scale dynamic simulation and fragmented driving mechanism analysis. To address these gaps, this study takes northern China’s arid and semi-arid regions as the object, innovatively constructing a “pat-tern-process-mechanism” multi-dimensional integration framework. Breaking through single-model/discrete-method limitations in existing studies, it realizes full-process integrated research on regional HQ spatiotemporal dynamics. Based on 1990–2020 Land Use and Land Cover Change (LUCC) data, the framework integrates the InVEST and PLUS models, solving poor continuity between historical assessment and future projection in traditional research. It also pioneers combining the XGBoost-SHAP model and Geographically and Temporally Weighted Regression (GTWR): XGBoost-SHAP quantifies nonlinear interactive effects of natural, socioeconomic, and landscape drivers, while GTWR explores spatiotemporal heterogeneous mechanisms of landscape pattern evolution on HQ, effectively addressing the dual challenges. Results show the following: (1) In 1990–2020, cultivated and construction land expanded, with grassland declining most notably; (2) Overall HQ decreased by 0.82%, with high-value areas stable in the west and northeast, low-value areas concentrated in the central region, and 2030 HQ optimal under the Ecological Protection (EP) scenario; (3) Natural factors contribute most to HQ change, followed by socioeconomic factors, with landscape indices being least impactful; (4) Under future scenarios, landscape Patch Density (PD) has the most prominent negative effect—its increase intensifies fragmentation and reduces connectivity. This study’s method integration breakthrough provides a quantitative basis for landscape pattern optimization and ecosystem management in arid and semi-arid regions, with important scientific value for promoting integration of landscape ecology theory and sustainable development practice.

Keywords: habitat quality; landscape scenic index; machine leaning; PLUS model; arid and semi-arid regions (search for similar items in EconPapers)
JEL-codes: Q15 Q2 Q24 Q28 Q5 R14 R52 (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2073-445X/14/10/1937/pdf (application/pdf)
https://www.mdpi.com/2073-445X/14/10/1937/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jlands:v:14:y:2025:i:10:p:1937-:d:1757614

Access Statistics for this article

Land is currently edited by Ms. Carol Ma

More articles in Land from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().