Modeling Climate Refugia for Chengiodendron marginatum: Insights for Future Conservation Planning

Zhe Yu, Qin Yan, Jianping Li, Yan Zheng, Chenlong Fu, Bo Jiang, Lin ChenPlants 2025 · DOI

TL;DR

This study models the potential habitat distribution of Chengiodendron marginatum, an ecologically significant plant, under past, present, and future climate scenarios. It identifies key climatic drivers and pinpoints critical "climate refugia" in southern Hainan, Taiwan, and northeastern Guangxi, offering crucial insights for conservation planning in the face of climate change.

What Problem Does This Paper Address?

Climate change poses a significant threat to biodiversity, altering species distributions and potentially leading to extinction. Understanding how species will respond to these changes is crucial for effective conservation. Chengiodendron marginatum, a valuable germplasm resource for breeding, is facing habitat fragmentation and decline due to human activities. However, there's a lack of research on its distribution and how it might be affected by future climate change. This paper aims to fill this knowledge gap by predicting the species' suitable habitats across different time periods and climate scenarios, thereby informing conservation strategies.

Methods

The researchers employed a robust approach combining the Maximum Entropy (Maxent) modeling algorithm with GIS spatial analysis. Maxent is a widely used and effective method for predicting species distribution based on occurrence data and environmental variables. The study utilized a comprehensive set of environmental predictors, including bioclimatic, soil, and topographic variables. To account for long-term climate shifts, they incorporated paleoclimatic data (mid-Holocene) alongside contemporary and future climate projections derived from CMIP6 models under two contrasting Shared Socioeconomic Pathways (SSPs): SSP126 (sustainable development) and SSP585 (fossil-fueled development). The model's accuracy was rigorously evaluated using the Area Under the Curve (AUC) metric, and the relative importance of environmental variables was assessed through percentage contribution and permutation importance, further validated by a jackknife test. Finally, centroid migration analysis was used to track the shift in the center of suitable habitat over time.

Datasets

  • Species Occurrence Data: 248 validated occurrence points for Chengiodendron marginatum, collected from field surveys and databases like GBIF, NSII, and CVH.
  • Environmental Variables:
    • Bioclimatic Data: 19 variables from the WorldClim global climate database (mid-Holocene and current data from WorldClim V1.4 and V2.1, respectively).
  • Future Climate Data: Projections from the Beijing Climate Center Climate System Model (BCC-CSM2-MR) under SSP126 and SSP585 scenarios for two future periods: 2041-2060 (2050s) and 2061-2080 (2070s).
  • Soil Data: 21 variables from the Harmonized World Soil Database (HWSD2.0).
  • Topographic Data: Elevation, slope, and aspect from the Shuttle Radar Topography Mission (SRTM).

Key Insights and Ideas

  • Precipitation is King: The study strongly highlights that precipitation, specifically the precipitation of the driest month (bio14) and annual precipitation (bio12), are the overwhelmingly dominant factors shaping the distribution of C. marginatum. This suggests that water availability, particularly during dry periods, is the primary limiting factor for this species.
  • Climate Refugia Identified: The research pinpoints southern Hainan, Taiwan, and northeastern Guangxi as crucial climate refugia. These areas are projected to maintain high-suitability habitats for C. marginatum across all analyzed climate scenarios and time periods, making them vital for the species' long-term survival.
  • Habitat Dynamics: Expansion and Contraction: While the species experienced habitat expansion from the mid-Holocene to the present, future projections under both SSP126 and SSP585 scenarios indicate a general trend of habitat contraction, particularly for high-suitability areas. This underscores the vulnerability of the species to ongoing climate change.
  • Anthropogenic vs. Climatic Drivers: The study acknowledges that while climate is a major driver, anthropogenic factors like habitat fragmentation (e.g., due to land development) can create discrepancies between predicted and actual suitable habitats, especially in human-dominated landscapes.

Key Results

  • Model Accuracy: The Maxent model demonstrated excellent predictive accuracy with an AUC value of 0.942.
  • Current Suitable Habitat: Approximately 98.38 × 10⁴ km² of suitable habitat exists currently, primarily in East, Central, and South China. High-suitability areas are concentrated in southern Hainan, Taiwan, and northeastern Guangxi.
  • Future Habitat Projections:
    • Under SSP126, total suitable habitat is projected to decrease by 14.33% (2050s) and 12.21% (2070s) compared to the present. High-suitability areas could decrease by 35.19% to 44.44%.
  • Under SSP585, total suitable habitat is projected to decrease by 12.57% (2050s) and 16.47% (2070s). High-suitability areas could decrease by 19.75% to 30.25%.
  • Centroid Shift: The center of suitable habitat is projected to shift, with some scenarios showing a westward movement towards Guangxi and others a northeastward shift towards Hunan, ultimately converging in northeastern Guangxi.

Limitations and Future Work

  • Anthropogenic Factors: The study acknowledges that anthropogenic factors, such as habitat fragmentation and land development, were not explicitly quantified. This could lead to discrepancies between predicted and actual species distributions, particularly in human-modified landscapes. Future work should integrate socio-ecological variables to improve model realism.
  • Data Limitations: While the study used multiple databases and field surveys, the accuracy of species distribution models is inherently dependent on the quality and completeness of occurrence data.
  • Future Work Suggestions: The authors suggest integrating socio-ecological variables, conducting assisted migration and native vegetation recovery in projected expansion areas, implementing emission mitigation strategies, and leveraging the species' horticultural value for ex situ conservation and artificial propagation in identified refugia.

Summary updated

Abstract

<i>Chengiodendron marginatum</i>, an evergreen tree or shrub belonging to the Oleaceae family, represents a critical germplasm resource with considerable potential for novel cultivar breeding. To elucidate the adaptive responses of <i>C. marginatum</i> to climate change and provide strategic guidance for its conservation, this study investigates the changing patterns in its potential suitable habitats under various climate scenarios. We employed an integrated approach combining maximum entropy (Maxent) modeling with GIS spatial analysis, utilizing current occurrence records and paleoclimatic data spanning from the mid-Holocene to future projections (2041-2060 [2050s] and 2061-2080 [2070s]). Climate scenarios SSP126 and SSP585 were selected to represent contrasting emission pathways. The model demonstrated excellent predictive accuracy with an AUC value of 0.942, identifying precipitation-related variables (particularly the precipitation of driest month and annual precipitation) as the primary environmental factors shaping the geographical distribution of <i>C. marginatum</i>. Current suitable habitats encompass approximately 98.38 × 10<sup>4</sup> km<sup>2</sup>, primarily located in East, Central, and South China, with high-suitability habitats restricted to southern Hainan, Taiwan, and northeastern Guangxi. Since the mid-Holocene, an expansion of suitable habitats occurred despite localized contractions in Southwest China. Future projections revealed moderate habitat reduction under both scenarios, and high-suitability areas decreased substantially. Importantly, under both scenarios, persistent high-suitability habitats were maintained in southern Hainan, Taiwan, and northeastern Guangxi, which are identified as essential climate refugia for the species. These findings provide a basis for understanding the response of the species to climate change and offer valuable guidance for its conservation.
BibTeX
@inproceedings{yu2025modeling,
  author = {Zhe Yu and Qin Yan and Jianping Li and Yan Zheng and Chenlong Fu and Bo Jiang and Lin Chen},
  title = {{Modeling Climate Refugia for Chengiodendron marginatum: Insights for Future Conservation Planning}},
  booktitle = {Plants},
  year = {2025},
  doi = {10.3390/plants14131961},
}
Figure