Next Article in Journal
Paternal Cocaine Exposure and Its Testicular Legacy: Epigenetic, Physiological, and Intergenerational Consequences
Previous Article in Journal
Effects of Decapitation on Chlorophyll Metabolism, Endogenous Hormones, and Tillering Ability in Pinus yunnanensis Seedlings of Different Ages
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Biology
Volume 14
Issue 8
10.3390/biology14081071
biology-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Assessing the Potential Distribution of the Traditional Chinese Medicinal Plant Spatholobus suberectus in China Under Climate Change: A Biomod2 Ensemble Model-Based Study

by
Yijun Lin
†,
Quanwei Liu
†,
Shan Lv
,
Xiaoyu Huang
,
Chaoyang Wei
,
Jun Li
,
Yijie Guan
,
Yaxuan Pan
,
Yijia Mi
,
Yanshu Cheng
,
Xiangyu Yang
and
Danping Xu
*
College of Life Science, China West Normal University, Nanchong 637002, China
*
Author to whom correspondence should be addressed.
†
These authors contributed equally to this work and should be regarded as co-first authors.
Biology 2025, 14(8), 1071; https://doi.org/10.3390/biology14081071
Submission received: 11 May 2025 / Revised: 30 July 2025 / Accepted: 15 August 2025 / Published: 17 August 2025
Browse Figures
Versions Notes

Simple Summary

In order to make better use of the medicinal value of Spatholobus suberectus and further investigate it, the Biomod2 ensemble model was used to predict the future trend of the suitable area of S. suberectus. At present, the suitable area of S. suberectus is mainly distributed in southern China, and with the change in various climatic factors, the suitable areas of S. suberectus will expand as a whole in the future. This study can provide a reference for the planting of S. suberectus in China, such as giving priority to semi-shady slope or forest edge areas for artificial planting.

Abstract

Spatholobus suberectus, a valuable Chinese medicinal plant, faces habitat shifts under climate change. In order to better utilize the medicinal properties of S. suberectus and conduct further investigations, this study utilized the Biomod2 ensemble model to predict and analyze the potential expansion and contraction of suitable habitat areas for S. suberectus in China under changing climatic and environmental conditions. The results showed that, compared to the pre-screened models, the ensemble model significantly improved the prediction accuracy. Currently, S. suberectus is primarily distributed in southern China. Under the projected scenarios of SSP1-2.6, SSP2-4.5, and SSP5-8.5, its suitable habitat is expected to expand overall, with the increased areas concentrated mainly in southwestern, central, and eastern China. As climatic factors shift, the high-suitability center of S. suberectus is predicted to shift slightly southward under the SSP1-2.6 scenario, while under the SSP2-4.5 and SSP5-8.5 scenarios, it is projected to move northwestward. In the future, it will be necessary to optimize the warm and humid growth environment for cultivated S. suberectus in China. Meanwhile, wild S. suberectus populations should be closely monitored for the impact of climate change to prevent potential partial reductions in suitable habitats, thereby ensuring ecological balance and sustainable development.

1. Introduction

Spatholobus suberectus (Leguminosae), “Ji-xue-teng” in China, is a climbing vine plant [1]. The flower of this plant is conical inflorescence, axillary or at the top of small branches, with short yellow-brown hairs on the inflorescence and peduncle. The corolla is white, the flag is flattened and round, the wing petals are slightly longer than the keel petals, the anthers are spherical in shape and uniform in size, and the ovary is sessile and covered with rough hairs. The fruit is pod-shaped, knife-shaped, and densely covered with short brown hairs and has a fruit neck. The seeds are oblong and flattened [2]. S. suberectus is mainly distributed in southern China (such as Yunnan, Guangxi, and Guangdong), and mostly grows in sparse forests, dense forests, valleys, or shrubs at an altitude of 800–1700 m [3].
S. suberectus is a widely used traditional Chinese medicine. It has a bitter taste, a sweet flavor, and primarily acts on the liver and kidney meridians. Research has shown that S. suberectus has significant clinical effects, such as inhibiting the growth of human breast cancer cells [4] and human lung adenocarcinoma cells (anticancer) [5]. Its flavonoid extracts also exhibit antioxidant activity, with certain compounds capable of scavenging free radicals. Additionally, it has protective effects on hematopoietic function impaired by radiation, contributing to the improvement of the hematopoietic system. Furthermore, it has immunomodulatory, antiviral [6,7,8], blood-enriching, and meridian-activating properties, making it useful in the treatment of conditions such as menstrual irregularities, blood deficiency, numbness, paralysis, and rheumatic pain [9,10,11]. The effective chemical components are closely related to the pharmacological effects, with modern pharmacological studies identifying several bioactive compounds, such as isorhamnetin, lupeol, and dandelion sesquiterpene [12]. Among these, isorhamnetin has been shown to inhibit the proliferation of various tumor cells in vitro and significantly suppress tumor growth in vivo [13]. Lupeol alleviates postprandial hyperglycemia [14], and dandelion sesquiterpene inhibits cancer cell growth by inducing apoptosis [15]. Therefore, S. suberectus exhibits both anticancer and antidiabetic activities. At present, research on S. suberectus mainly focuses on its chemical composition and clinical efficacy, while studies regarding its geographic distribution are scarce. There is significant research potential in predicting the future potential distribution of S. suberectus in China under the influence of climate and other environmental data changes over the coming decades.
Global climate change is progressing rapidly [16], and environmental changes have been shown to affect species distribution. For example, the southern latitudinal boundary of South African mangroves is predicted to shift southward due to multiple environmental factors [17]. In Mexico, the suitable habitat area for Tagetes lucida Cav. (Asteraceae) is expected to migrate northward and decrease in size due to significant climate change [18]. As such, research predicting species’ responses to future environmental scenarios is necessary [19], and species distribution models (SDMs) have emerged as a tool for this purpose. SDMs are mathematical models that estimate species’ niche requirements based on distribution data and environmental variables, thus predicting potential species distributions and assessing the contribution of environmental variables. These models have been widely used in habitat suitability assessments, biological invasion risk evaluations, and species distribution change studies [20]. Various SDMs are available, including the Maximum Entropy Model (MaxEnt) [21], CLIMEX [22], Ecological Niche Factor Analysis (ENFA) [23], and others. The Biomod2 platform, based on R language [24], integrates 10 commonly used SDMs, including the Generalized Linear Model (GLM) [25], Generalized Boosted Regression Model (GBM) [26], Generalized Additive Model (GAM) [25], Classification Tree Analysis (CTA) [27], Artificial Neural Network (ANN) [28], Surface Range Envelope (SRE) [29], Flexible Discriminant Analysis (FDA) [30], Multivariate Adaptive Regression Splines (MARS) [31], Random Forest (RF) [32], and MaxEnt [21]. This platform allows for comprehensive species evaluation and integrates the best-performing models to predict species distributions, thereby maximizing prediction accuracy and enhancing the reliability of future distribution forecasts. It is currently one of the most mature multi-model platforms [27,33]. Ensemble models, which combine results from multiple models, have been shown to offer certain advantages [27,33,34]. Although research using ensemble models remains relatively limited [35], it is anticipated that future studies will increasingly adopt this approach.
Currently, there have been no reported studies on the distribution prediction of S. suberectus, nor have the factors affecting its distribution in China been extensively explored. Additionally, there is a lack of research predicting the future distribution of S. suberectus in China under climate change scenarios. Species distribution predictions can reflect changes in species’ adaptability to specific regions, the risk of overexploitation, and evolutionary trends [36]. The absence of such studies on S. suberectus could hinder efforts to monitor its exploitation rate and limit insights into its evolutionary patterns. Therefore, this study utilizes the Biomod2 ensemble model to predict the suitable habitat areas of S. suberectus in China over different time periods, aiming to analyze the changes in suitable areas under various climatic conditions, assess the migration trends of S. suberectus under climate change, and identify the key drivers influencing its distribution. This study aims to fill the gap in the distribution prediction of S. suberectus.

2. Materials and Methods

2.1. Acquisition and Screening of Species Distribution Data

The species distribution point coordinates of S. suberectus in China used in this study were primarily sourced from the following: (1) the China Digital Herbarium (https://www.cvh.ac.cn/, accessed on 7 August 2024); (2) the Global Biodiversity Information Facility (GBIF, https://doi.org/10.15468/dl.hy9gcv, accessed on 7 August 2024); and (3) literature retrieval (https://www.x-mol.com/, accessed on 7 August 2024). The obtained distribution data were imported into ArcGIS, and unreasonable distribution points were excluded based on relevant literature. A total of 118 distribution points of S. suberectus in China were initially compiled. To avoid model overfitting, the ENMTool was used to remove redundant data and verify and remove duplicate occurrence points based on spatial distance criteria, retaining only one distribution point within a 5 km radius; 118 distribution points were analyzed one by one [37]. Every time a point whose distance meets the redundancy standard was encountered, it was eliminated. After such a screening process, a large number of repeated or highly similar points in the spatial position were removed, and finally 53 distribution data which can more effectively represent the real distribution of S. suberectus in China without redundancy were obtained (Figure 1).

2.2. Acquisition and Screening of Environmental Data

This study utilized two sets of environmental data, comprising 19 climate variables, 11 soil variables, and 3 topographic factors, totaling 33 environmental variables. The climate data were obtained from the WorldClim database (https://www.worldclim.org/, accessed on 7 August 2024), soil quality data were sourced from the Global Soil Database (https://gaez.fao.org/pages/hwsd, accessed on 7 August 2024), and the Human Influence Index was derived from the Earth Observing System Data and Information System (EOSDIS) managed by NASA, in collaboration with the Socioeconomic Data and Applications Center (SEDAC) (https://sedac.ciesin.columbia.edu/, accessed on 7 August 2024). Combined with relevant data on factors influencing the distribution of S. suberectus in China, 11 soil variables were selected and extracted in ArcGIS.
The Biomod2 software was used to predict the potential geographical distribution of Spatholobus suberectus in China under current and future (2050s, 2070s, and 2090s) climate scenarios (SSP1-2.6, SSP2-4.5, and SSP5-8.5). Among them, the SSP1-2.6 scenario represents a combination of low emissions, low vulnerability, low mitigation pressure, and low radiative forcing; the SSP2-4.5 scenario represents an intermediate development pathway, a combination of moderate social vulnerability and moderate radiative forcing; and the SSP5-8.5 scenario represents a high-emission and high social development scenario. The selection of environmental variables is a crucial step in building a robust Biomod2 model. Therefore, a scientific screening process was applied: the extracted data were resampled in ArcGIS, and the resolution was standardized to 2.5’ (5 km × 5 km). The 19 bioclimatic factors and 11 soil variables were subjected to principal component analysis (PCA) and correlation heatmap analysis (Figure A1). Variables with high contributions in PCA1 and PCA2 and a correlation lower than 0.8 were selected. Subsequently, variables with correlations greater than 0.8 were removed, along with those based on soil variable PCA and correlation analysis heatmaps (Table 1).

2.3. Model Construction and Evaluation of Combined Model Accuracy

In this study, the Biomod2 package was used to model the distribution prediction of S. suberectus in China. Eleven different species prediction models, GLM, GBM, GAM, CTA, ANN, SRE, FDA, MARS, RF, MaxEnt, and XGBOOST, were applied from the Biomod2 package. To improve the accuracy of the model, the DISK method was used to create pseudo-absences. First, the locations of presence points of Spatholobus suberectus in China were determined based on the screened distribution points, and these data are represented by green triangles in the figures. Then, true absences were created according to the positions of the presence points and based on the geographical scope of China. Finally, when the number of true absence points was insufficient, three sets of pseudo-absence datasets (PA1, PA2, and PA3) were created at the boundary of the distribution range of Spatholobus suberectus, with 1000 points in each set. A random 25% of the distribution data of Spatholobus suberectus in China was selected for model testing, and 75% was used for model training. Each model was repeated 10 times, and a total of 300 model simulation results were finally established [38].
The contribution rates of each environmental factor to the model were comprehensively evaluated by the Biomod2 software package. The accuracy of the model was evaluated by both the True Skill Statistic (TSS) and Key Activity Performance Analysis (KAPPA) (Table 2). TSS is a test of sensitivity and specificity, and it is not affected by the size of the validation data, which can make up for the shortcomings of KAPPA in the evaluation process [39]. Both TSS and KAPPA are in the range of [0, 1], and the closer the results are to 1, the better the model construction effect. Models with TSS > 0.9 and KAPPA > 0.9 were selected to build the ensemble model and re-model, and the accuracy of the ensemble model was evaluated by TSS and KAPPA values. In this study, the results generated by the Biomod2 model were converted into raster data and visualized by using the Reclassify function of ArcGIS. The ensemble model obtained the results of suitable habitat probability values in the range of [0, 1]. Based on the cutoff value as the boundary, combined with the TSS threshold of the ensemble model, the habitat suitability index (HSI) and the physiological characteristics of the species, the potential suitable habitat areas were divided into four grades: unsuitable habitat area (0 ≤ p ≤ 0.263), low-suitability habitat area (0.263 < p ≤ 0.525), moderately suitable habitat area (0.525 < p ≤ 0.763), and highly suitable habitat area (0.763 < p ≤ 1) [40].

3. Results

3.1. Model Evaluation

The Kappa value of the combined model was 0.908, and the TSS value was 0.916. Regarding the Biomod2 package, the RF model and XGBOOST model had the highest mean Kappa and TSS values. Although the mean values of KAPPA and TSS for the RF model and XGBOOST model are both higher than those of the combined model, the RF model or XGBOOST model, as a single model, has limitations in certain aspects. These results indicate that the combined model provided more accurate predictions of the geographic distribution of S. suberectus. Additionally, by integrating the predictions from multiple individual models, the combined model effectively reduced the potential fluctuations caused by the inherent limitations of any single model, thereby yielding more stable prediction results (Figure 2).

3.2. Selection of Environmental Factors and Contribution Rate Analysis

Based on the correlation analysis between environmental factors and the contribution rate of each factor to the model, the percentage contribution of each environmental factor was obtained. This study focused on the five environmental factors with the highest contribution rates for further discussion (Figure 3). The results indicated that the top five environmental factors in terms of contribution rate were as follows: the average temperature of the coldest quarter (bio11), with a contribution rate of 20%; the minimum temperature of the coldest month (bio06), with a contribution rate of 11.81%; the annual average temperature (bio01), with a contribution rate of 9.83%; temperature seasonality (bio04), with a contribution rate of 9.27%; and the precipitation of the driest season (bio17), with a contribution rate of 8.05%.
While determining the environmental variables, the combined model also analyzed the response curves of S. suberectus to environmental factors (Figure A2). The adaptation range of S. suberectus to each environmental variable was roughly identified: the average temperature of the coldest quarter was around 8–14 °C, with an optimal value at approximately 12 °C; the minimum temperature of the coldest month ranged from 0 to 10 °C, with an optimal value around 7 °C; the annual average temperature ranged from 14 to 22 °C, with an optimal value around 16 °C; temperature seasonality ranged from 0 to 650 °C, with an optimal value around 400 °C; and the precipitation of the driest season ranged from 0 to 140 mm, with an optimal value around 50 mm.

3.3. Potential Distribution of Suitable Habitat for S. suberectus in the Current Period

Based on the results of the combined Biomod2 model prediction (Figure 4), it was found that under the current scenario, the potential distribution of S. suberectus was primarily concentrated in the southern and southeastern coastal regions of China, mostly within tropical monsoon and subtropical monsoon climate zones. These areas are typically characterized by high temperatures, abundant precipitation, ample sunlight, and high humidity. The high-suitability habitat was mainly distributed in Yunnan Province, western Guangxi, southern Sichuan, and with scattered occurrences in Taiwan, southern Guizhou, eastern Guangdong, Central Hainan, southern Fujian, southeast Tibet, and western Chongqing, covering an area of 225.52 × 103 km2. The medium-suitability habitat was mainly distributed in southeast Tibet, Taiwan, Hainan Province, southern Fujian, Yunnan Province, western Guangxi, the border areas between Guangxi and Guangdong, and eastern Guangdong, with scattered occurrences in Jiangxi Province, Hunan Province, southern Guizhou, and western Chongqing, covering an area of 285.93 × 103 km2. The low-suitability habitat was found in southern China and the southeastern coastal areas, covering an area of 493.97 × 103 km2.

3.4. Potential Distribution of Suitable Habitat for S. suberectus in Future Periods

Under the SSP1-2.6, SSP2-4.5 and SSP5-8.5 scenarios, based on the combined model predictions for the 2050s, 2070s, and 2090s (Figure 5, Figure 6 and Figure 7), it was observed that the suitable habitat for S. suberectus would change in the future. In the future, the high-suitability habitat for S. suberectus would still be concentrated in Yunnan Province, where the temperature variation is small and precipitation is abundant, making it suitable for its growth. The following paragraphs describe the scenarios compared to the current period.
Under the SSP1-2.6 scenario, the increase in the low-suitability habitat is mainly concentrated in Sichuan Province, Hunan Province, Jiangxi Province, and Zhejiang Province, which is the higher in the 2090s by 533.11 km2; however, there is a reduction in the low-suitability areas in some regions, such as Yunnan and Hainan. The main distribution of the medium-suitability habitat is in Fujian, Guangxi, and Guangdong Provinces, which would increase by 77.82 × 103 km2 in the 2050s; the high-suitability habitat would remain similar to the current distribution.
Under the SSP2-4.5 scenario, the distribution of the low-suitability habitat increased significantly, with Shanxi, Henan, and Jiangsu Provinces added. The medium-suitability habitat in Sichuan and Guangxi would increase; the high-suitability habitat showed no significant fluctuation.
Under the SSP5-8.5 scenario, the low-suitability area as a whole expands northward to Henan Province, Anhui Province, and Jiangsu Province. The distribution of moderately suitable areas in Taiwan, Guizhou Province, and Chongqing has increased, while the highly suitable areas in Guangxi have decreased slightly, but the overall distribution has increased (Table A1).

3.5. Area Changes of Suitable Habitat for S. suberectus in Future Periods

The distribution patterns of suitable habitat for S. suberectus in the current and future periods were compared and analyzed to evaluate the changes in suitable habitat area and the expansion and contraction of suitable habitat. Under the three future periods and three scenarios, the main expansion areas included Hunan Province, Jiangxi Province, Zhejiang Province, Shaanxi Province, Henan Province, and Jiangsu Province, with the expansion area peaking in the 2090s under the SSP5-8.5 scenario at 1290.31 × 103 km2. The area of the suitable area of chicken blood vine shrank in some areas, and the main shrinkage area was Guangxi Province under the SSP5-8.5 scenario. Conversely, the suitable habitat area contracted in certain regions, particularly in Guangxi with the SSP5-8.5 scenario.

3.6. Changes in the Suitable Habitat Centroid for S. suberectus in the Future

Under the SSP1-2.6, SSP2-4.5, and SSP5-8.5 scenarios, the changes in the centroid of the high-suitability habitat for S. suberectus were analyzed by combining the centroid shift maps (Figure 8). Over time, by the 2090s, the centroid of the high-suitability habitat generally shifted northwestward, followed by a northeastward shift. The centroid remained in Yunnan Province throughout all periods. In the current period, the high-suitability centroid was located in Yuxi Pingdian Township, Yunnan Province, with coordinates of 24.04° N, 101.87° E. In the SSP2-4.5 scenario, the centroids of the 2050s, 2070s, and 2090s were moved to Laochang Township, Yuxi, Yunnan, with coordinates of 24.26° N, 101.77° E, 24.28° N, 101.66° E and 24.32° N and 101.64° E, respectively. In the SSP5-8.5 scenario, the center of mass in the 2050s moved to Laochang Township, Yuxi, Yunnan Province, with coordinates 24.29° N, 101.66° E, the center of mass in the 2070s moved to Ainishan Township, Shuangbai County, Yunnan Province, with coordinates 24.41° N, 101.49° E, and the center of mass in the 2090s moved to Ejia Township, Shuangbai County, Yunnan Province, with coordinates of 24.47° N, 101.37° E.

4. Discussion

4.1. Evaluation of the Combined Model Performance

Currently, many studies tend to use single models to predict species distributions, but this approach carries the risk of overfitting or underfitting. Research has shown that combining the predictions of single models can reduce the prediction uncertainty of individual models, which will significantly enhance the robustness of prediction outcomes [41]. Furthermore, numerous studies have demonstrated that ensemble models can provide more reliable and stable predictions of species habitat suitability [42,43]. Biomod2 is particularly advantageous, as it can fit and compare different models, facilitating model selection [44]. Therefore, based on model screening, this study selected 11 single models to participate in the construction and accuracy evaluation of the ensemble model. Ultimately, six models with excellent TSS evaluation metrics (>0.9) (RF, XGBOOST, GAM, GBM, GLM, and MARS) were chosen to construct the ensemble model for predicting the geographical distribution of S. suberectus in China under global climate change scenarios. Although the evaluation results of single models indicated that the average Kappa and TSS values of the RF and XGBOOST models were slightly higher than those of the final ensemble model, this does not imply that single models are superior to the ensemble model in all aspects. The core value of ensemble modeling lies in its integration of predictions from multiple models with different algorithmic principles and inherent biases. This effectively reduces predictive fluctuations and uncertainties arising from reliance on a single model structure or specific dataset. Consequently, when faced with complex environmental variables and future climate change scenarios, ensemble models generally provide more robust and reliable predictions of distribution patterns. Therefore, although individual single models may perform exceptionally well on the training dataset, the ensemble model exhibits significant advantages in generalization ability and predictive stability, which is particularly critical for predicting the potential distribution of species under future climate change.
Overall, Biomod2, by integrating multiple models, can provide more robust prediction results even with relatively limited input distribution data [37,45,46]. The results of this study can provide a theoretical foundation for the conservation and utilization of S. suberectus in the future. Furthermore, this study mainly focuses on the influence of climatic factors (such as temperature and precipitation) on the distribution of S. suberectu. However, in the future, species distribution may also be significantly affected by non-climatic factors, such as land-use changes, human activity interference, soil properties, etc. [47,48]. Urbanization and deforestation may lead to habitat fragmentation or loss, resulting in actual distribution ranges being smaller than those predicted by climate models. For example, certain areas, although climatically suitable, may become unavailable for Spatholobus suberectus due to land development. Additionally, harvesting for traditional medicine may exert direct pressure on local populations, and even in climatically suitable regions, overharvesting could lead to population decline. Spatholobus suberectus also has specific requirements for soil conditions (e.g., pH, nutrient content, and drainage), which may restrict root development in karst regions or nutrient-poor lateritic soils, rendering climatically suitable areas uninhabitable in practice. Current models, based solely on climatic factors, may produce overly optimistic predictions. Therefore, future studies should integrate high-resolution land-use data to quantify habitat fragmentation effects or employ hybrid modeling approaches (e.g., mechanistic models coupling climate and soil moisture) to correct biases. Further integration of these non-climatic factors will enhance the comprehensiveness and accuracy of model predictions.

4.2. Influence of Environmental Factors on the Geographical Distribution of S. suberectus

This study investigates the environmental factors influencing the distribution of S. suberectus, with the contribution rate results indicating that the environmental factors mainly affecting the potential suitable area distribution of S. suberectus during three periods were temperature factors (such as the mean temperature of the coldest quarter, the minimum temperature of the coldest month, the annual mean temperature, and temperature seasonality) and precipitation factors (such as the precipitation during the driest season). This is mainly because S. suberectus grows in tropical and subtropical regions, where the warm and humid climate provides optimal conditions for temperature and precipitation for its growth. Temperature is one of the most critical environmental factors in plant growth and development, exerting multifaceted effects on plants. Research has shown that moderate temperatures can more effectively induce the accumulation of secondary metabolites (including phenolic and flavonoid compounds) in plants [49]. Drought and heat stress have adverse effects on plant physiological processes, leading to oxidative stress, cellular damage, reduced photosynthesis, and decreased productivity, which, in severe cases, can result in cell death [50]. Cold temperatures as low as −10 °C can effectively break dormancy in temperate woody plants [51]. At the same time, the results of this study are consistent with previous findings: precipitation plays a crucial role in shaping plant species distributions in mountainous ecosystems [52]. The optimal precipitation for S. suberectus occurs during the driest season at approximately 50 mm; both excessively high and low precipitation levels are detrimental to its growth and distribution. For S. suberectus, a suitable temperature is particularly critical for its growth, as it is highly sensitive to temperature. In this study, the contribution rate of temperature factors among environmental variables is 62.62%. In model predictions, the mean temperature of the coldest quarter contributes the most to the overall model. The response curve shows that when the mean temperature of the coldest quarter approaches 10 °C, its impact on the suitable growth regions of S. suberectus is maximal, promoting growth. When the minimum temperature of the coldest month is between 0 and 10 °C, growth is still promoted. This is because when plants are exposed to cold stress, they can gain tolerance to subsequent freezing conditions by temporarily being exposed to cold but non-freezing temperatures [53]. Additionally, key transcription factors within the plant help balance growth and stress tolerance [54]. Moreover, S. suberectus contains selenium, which aids in maintaining photosynthesis under temperature stress, enhancing the plant’s antioxidant capacity and mitigating oxidative damage under low temperatures [55]. When the mean temperature of the coldest quarter exceeds 0 °C but remains below 10 °C, or when the minimum temperature of the coldest month is below 0 °C, the suitability of S. suberectus for growth declines, and growth is inhibited. This indicates that S. suberectus has relatively low cold tolerance. Tolerance is one of the types of resistance responses plants exhibit to environmental challenges [56], and the response depends on the type and intensity of the environmental stress. One of the most prominent challenges is temperature variation, which significantly impacts the physiological and biochemical processes of plants [57,58]. In this study, S. suberectus is widely distributed in southern China, typically growing in tropical monsoon and subtropical monsoon climates. In warm climates, its growth rate is rapid. The results show that an annual mean temperature of around 16 °C is the optimal temperature for S. suberectus survival. From the results, it is clear that temperature factors are the dominant influence on the growth process of S. suberectus, and under the context of global warming, temperature changes will have profound effects on the growth and distribution of S. suberectus.

4.3. Future Changes in the Potential Suitable Distribution Areas of S. suberectus

This study employs the Biomod2 ensemble modeling framework to project dynamic changes in the potential suitable habitats of Spatholobus suberectus across China under three future climate scenarios. As a thermophilic and hygrophilous species, S. suberectus primarily inhabits sparse forests, shrublands, slopes, and valleys at elevations of 800–1700 m [6]. Under current climatic conditions, its distribution is concentrated in southern and southeastern coastal China, predominantly within tropical and subtropical monsoon climate zones. Empirical evidence indicates that without substantial greenhouse gas mitigation, global surface temperatures will rise by ≥2.1 °C by 2100, while the global mean sea level is projected to increase by 0.28–0.55 m by 2100 even under the lowest-emission scenario (SSP1-2.6) [59]. Elevation exerts a significant influence on the species’ distributional response under climate change. Rising temperatures are projected to drive upward shifts in its optimal elevational range, consistent with documented patterns of montane plants migrating to higher elevations to track thermally suitable conditions [60]. Although temperature typically declines with increasing altitude, high-elevation zones under global warming exhibit reduced frost frequency, enhancing the thermal suitability for S. suberectus [61]. For instance, in southwestern China, populations historically distributed at 800–1700 m may expand to 1700–2500 m, while habitat suitability at lower elevations (<800 m) may decline due to thermal stress.
Future temperature increases and sea-level rise will substantially alter distribution patterns. Under the SSP1-2.6, SSP2-4.5, and SSP5-8.5 scenarios, the centroids of highly suitable habitats remain anchored in Yunnan Province but exhibit divergent trajectories: a minor southward shift toward Yuxi under SSP1-2.6 versus northwestward shifts toward Shuangbai County under SSP2-4.5 and SSP5-8.5. This directional pattern corresponds to established biogeographic principles of poleward migration under warming [62,63], validating the Biomod2 model’s reliability. The pronounced northwestward shift under SSP5-8.5 likely reflects adaptive responses to arid thermal stress through seeking habitats with optimal hydrothermal equilibrium [64]. Scenario-specific analyses reveal distinct dynamics: Under SSP1-2.6 (low radiative forcing), suitable habitats exhibit moderate expansion with low-suitability areas extending northward to Sichuan, Hunan, and Jiangxi, peaking in the 2090s—consistent with extended growing seasons and positive responses of southwestern high-elevation vegetation [65]. For SSP2-4.5 (intermediate forcing), medium-suitability areas expand in Sichuan and Guangxi while highly suitable habitats remain stable, aligning with studies showing suppressed habitat suitability where summer temperatures exceed thermal tolerance thresholds in eastern China [64]. Under SSP5-8.5 (high radiative forcing), low-suitability areas shift northward to Henan, Anhui, and Jiangsu with maximal expansion in the 2090s, while highly suitable habitats in the Guangxi contract significantly—corresponding to projected heatwaves (>25 °C) exceeding the species’ thermal stress threshold (optimal growth ≤22 °C) [66]. Climate change disproportionately enhances habitat suitability in southwestern China (Yunnan and Guizhou), where future warming shifts temperatures toward the species’ optimum. The region’s warm, humid climate and complex topography provide substantial habitat heterogeneity, while temperature increases may extend growing seasons at higher elevations and enhance plateau productivity [67]. Conversely, South China (Guangxi and Guangdong) exhibits marked degradation of highly suitable habitats under SSP5-8.5 due to synergistic stressors: extreme summer temperatures (>25 °C) and precipitation variability. This high-temperature–drought co-stress induced contraction mirror vulnerability patterns in global tropical–subtropical transition zones [18]. Collectively, multi-scenario projections indicate the net expansion of suitable habitats, with Yunnan persisting as the core high-suitability area. These findings provide critical insights for conservation planning and the sustainable utilization of S. suberectus.

4.4. Optimization of Cultivation Zoning for S. suberectus in China Under Climate Change Scenarios

S. suberectus possesses significant pharmacological value. Due to its slow growth in the wild and overharvesting, its natural resources are becoming increasingly scarce. To meet the growing medicinal demand, it is essential to implement scientific cultivation planning, optimize production regions, and adapt to climate change for rational resource development. The results of this study indicate that in the three scenarios and the future three periods, the contraction areas are mainly concentrated in South China. In these regions, future conservation strategies should prioritize reducing reliance on wild resources by transitioning to cultivated plantations, protecting existing populations, and preventing habitat fragmentation [68]. The expansion areas are mainly concentrated in southwest, central, and East China. Notably, in the three scenarios and the future three periods, the center of gravity of the high-suitability areas will always remain in Yunnan Province. To mitigate the negative impact of climate change on the kudzu vine in the future, Yunnan should be designated as a core protection zone, with measures implemented to curb overharvesting and promote sustainable resource utilization. Previous studies have demonstrated that climate warming significantly influences species distribution, with the rate of warming positively correlated to the speed of range shifts [69]. In response to the rapid climatic adaptation of species, a dynamic monitoring network should be established to mitigate cultivation risks [70]. Geographic Information System (GIS) technology can be employed to update suitable habitat boundaries in real time, optimize harvesting zones, and align management strategies with centroid migration trajectories [71]. In conclusion, given the sustained impacts of climate change on the distribution of medicinal plant resources, it is imperative to implement rational planning for future production zones of S. suberectus.

5. Conclusions

This study predicts and analyzes the future expansion and contraction of the potential suitable distribution areas of Spatholobus suberectus in China under the influence of climate and environmental data changes and puts forward some suggestions on origin planning using the Biomod2 ensemble model. Under the SSP1-2.6 scenario, the distribution of Spatholobus suberectus in the future three periods shows a trend of first expansion and then contraction. In the 2070s, the suitable habitat area has significantly shrunk, with the contracted areas mainly concentrated in the southwest and central China regions. Under the SSP2-4.5 and SSP5-8.5 scenarios, the distribution of Spatholobus suberectus in the future three periods shows an expansion trend. The expanded areas are mainly concentrated in the southwest, central, and eastern China regions, with a small part located in the western Xinjiang and western Tibet regions. Compared with the current suitable habitat distribution, the suitable habitat range of Spatholobus suberectus in different future periods under the three scenarios is still mainly concentrated in the tropical monsoon climate and subtropical monsoon climate regions. The primary factors influencing the distribution of S. suberectus in China are climate factors, though topography and soil factors also play a synergistic role. Overall, climate change has a significant impact on the potential distribution areas of Spatholobus suberectus in the border areas of the southwest provinces and central China. Future development and utilization of S. suberectus in China should closely monitor the impacts of climate change while also preventing the future contraction of its suitable areas to ensure ecological balance and sustainable development.

Author Contributions

Conceptualization, Y.G., Y.P. and Y.M.; Methodology, S.L. and X.H.; Software, X.H. and C.W.; Validation, J.L.; Formal analysis, Y.L. and Q.L.; Investigation, C.W. and J.L.; Resources, D.X.; Data curation, D.X.; Writing—original draft, Y.L., Q.L. and S.L.; Writing—review & editing, D.X.; Visualization, Y.C. and X.Y.; Supervision, D.X.; Funding acquisition, D.X. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the China West Normal University Support Program (20A007, 20E051, 21E040, and 22kA011).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data that support the findings of this study will be available in figshare at https://doi.org/10.6084/m9.figshare.28559795 (accessed on 7 August 2024) (GBIF, https://doi.org/10.15468/dl.hy9gcv, accessed on 7 August 2024).

Conflicts of Interest

The authors declare no conflicts of interest.

Appendix A

Table A1. Suitable habitat area of S. suberectus in different periods.
Table A1. Suitable habitat area of S. suberectus in different periods.
EpochLow-Suitability Habitat (×103 km2)Moderate-Suitability Habitat (×103 km2)High-Suitability Habitat (×103 km2)
Current493.97285.9322.552
SSP1-2.6 2050S948.17363.75258.05
SSP1-2.6 2070S911.89320.15247.58
SSP1-2.6 2090S1027.08334.02245.95
SSP2-4.5 2050S1043.92360.41258.66
SSP2-4.5 2070S1207.60386.84249.91
SSP2-4.5 2090S1331.57423.99258.78
SSP5-8.5 2050S1194.47353.42256.30
SSP5-8.5 2070S1456.96403.35262.96
SSP5-8.5 2090S1598.92434.87261.94
Biology 14 01071 g0a1
Figure A1. Principal component analysis (PCA) of 19 climates: (A) PCA of climate variables (PCA2); (B) heatmap of the correlation analysis of climate variables. (C) PCA of soil variables (PCA2), (D) Heatmap of the correlation analysis of soil variables.
Figure A1. Principal component analysis (PCA) of 19 climates: (A) PCA of climate variables (PCA2); (B) heatmap of the correlation analysis of climate variables. (C) PCA of soil variables (PCA2), (D) Heatmap of the correlation analysis of soil variables.
Biology 14 01071 g0a1
Biology 14 01071 g0a2
Figure A2. Influence curves of environmental factors.
Figure A2. Influence curves of environmental factors.
Biology 14 01071 g0a2

References

  1. Hieu, N.N.; Van Tuan, V.; Tung, P.H.T.; Le, D.Q.; Tung, N.H. Ethnopharmacology, Phytochemistry, and Pharmacological Activities of Spatholobus suberectus Vine Stem. Nat. Prod. Commun. 2022, 17, 1934578X221142724. [Google Scholar] [CrossRef]
  2. Editorial Committee of Flora of China. Flora of China; Science Press: Beijing, China, 1987; Volume 42, pp. 1–192. [Google Scholar]
  3. Xia, C.; Zhao, Y.; Liu, C.; Gao, Y. Diversity of fungal community and quality evaluation of Spatholobus Suberectus Dunn during the process of mildew. AMB Express 2024, 14, 12. [Google Scholar] [CrossRef] [PubMed]
  4. Zhang, F.; Ganesan, K.; Liu, Q.; Chen, J. A review of the pharmacological potential of Spatholobus suberectus Dunn on cancer. Cells 2022, 11, 2885. [Google Scholar] [CrossRef] [PubMed]
  5. Bo, F.; Guijae, Y.; Jun, L.P.; Yinda, Q.; Hoon, L.S.; Min, Y.H.; Namki, C. Network Pharmacology-Based Strategyto Investigate the Anti-Breast CancerMechanisms of Spatholobus suberectus Dunn. Nat. Prod. Commun. 2022, 17, 1934578X2210778. [Google Scholar]
  6. Pang, J.; Guo, J.; Jin, M.; Chen, Z.; Wang, X.; Li, J. Antiviral effects of aqueous extract from Spatholobus suberectus Dunn. against coxsackievirus B3 in mice. Chin. J. Integr. Med. 2011, 17, 764–769. [Google Scholar] [CrossRef] [PubMed]
  7. Chen, S.; Wang, A.; Lin, L.; Qiu, H.; Wang, Y.; Wang, Y. In vitro study on anti-hepatitis C virus activity of Spatholobus suberectus Dunn. Molecules 2016, 21, 1367. [Google Scholar] [CrossRef] [PubMed]
  8. Liu, X.; Zhang, Y.; Yang, X.; Yang, Y.; Xu, W.; Zhao, W.; Peng, K.; Gong, Y.; Liu, N.; Zhang, P. Anti-inflammatory activity of some characteristic constituents from the vine stems of Spatholobus suberectus. Molecules 2019, 24, 3750. [Google Scholar] [CrossRef]
  9. Huang, X.; Fei, Q.; Yu, S.; Liu, S.; Zhang, L.; Chen, X.; Cao, L.; Wang, Z.; Shan, M. A comprehensive review: Botany, phytochemistry, traditional uses, pharmacology, and toxicology of Spatholobus suberectus vine stems. J. Ethnopharmacol. 2023, 312, 116500. [Google Scholar] [CrossRef] [PubMed]
  10. Liu, Y.; Dai, S.; Xu, Y.; Xiang, Y.; Zhang, Y.; Xu, Z.; Sun, L.; Zhang, G.C.X.; Shu, Q. Integration of Network Pharmacology and Experimental Validation to Explore Jixueteng—Yinyanghuo Herb Pair Alleviate Cisplatin-Induced Myelosuppression. Integr. Cancer Ther. 2024, 23, 1573553105. [Google Scholar] [CrossRef]
  11. Shuyu, Z.; Ting, Y.; Xiaomei, L.; Hong, L.; Yuying, Z. Interactions between thrombin and natural products of Millettia speciosa Champ. using capillary zone electrophoresis. Electrophoresis 2008, 29, 3391–3397. [Google Scholar] [CrossRef] [PubMed]
  12. Chen, J.; Huilian, H.; Yuqin, H.E.; Ling, Z. Chemical constituents from the stem of Callerya nitida Benth. var. hirsutissima Z. Wei. J. China Pharm. Univ. 2021, 52, 177–185. [Google Scholar]
  13. Wruck, C.J.; Claussen, M.; Fuhrmann, G.; Romer, L.; Schulz, A.; Pufe, T.; Waetzig, V.; Peipp, M.; Herdegen, T.; Gotz, M.E. Luteolin Protects Rat PC 12 and C6 Cells Against MPP+ Induced Toxicity via an ERK Dependent Keapl-Nrf2-ARE Pathway; Springer: Berlin/Heidelberg, Germany, 2007. [Google Scholar]
  14. Lee, H.; Kim, M.; Han, J. Alleviating effects of lupeol on postprandial hyperglycemia in diabetic mice. Toxicol. Res. 2021, 10, 495–500. [Google Scholar] [CrossRef]
  15. Ma, X.; Dong, S.; Zhang, S.; Jia, N.; Ou, S. Taraxerone triterpene inhibits cancer cell growth by inducing apoptosis in non-small cell lung cancer cells. Bangladesh J. Pharmacol. 2016, 11, 342–347. [Google Scholar] [CrossRef]
  16. Senner, N.R.; Stager, M.; Cheviron, Z.A. Spatial and temporal heterogeneity in climate change limits species’ dispersal capabilities and adaptive potential. Ecography 2018, 41, 1428–1440. [Google Scholar] [CrossRef]
  17. Quisthoudt, K.; Adams, J.; Rajkaran, A.; Dahdouh-Guebas, F.; Koedam, N.; Randin, C.F. Disentangling the effects of global climate and regional land-use change on the current and future distribution of mangroves in South Africa. Biodivers. Conserv. 2013, 22, 1369–1390. [Google Scholar] [CrossRef]
  18. Kurpis, J.; Serrato-Cruz, M.A.; Arroyo, T.P.F. Modeling the effects of climate change on the distribution of Tagetes lucida Cav. (Asteraceae). Glob. Ecol. Conserv. 2019, 20, e747. [Google Scholar] [CrossRef]
  19. Zhao, G.; Cui, X.; Sun, J.; Li, T.; Wang, Q.I.; Ye, X.; Fan, B. Analysis of the distribution pattern of Chinese Ziziphus jujuba under climate change based on optimized biomod2 and MaxEnt models. Ecol. Indic. 2021, 132, 108256. [Google Scholar] [CrossRef]
  20. Xie, P.; Gu, Y.; Sui, W.; Tao, G.; Sun, S. Big data challenges for species distribution models in predicting potential distribution of marine species. Mar. Inf. 2019, 34, 51–61. [Google Scholar]
  21. Phillips, S.J.; Anderson, R.P.; Schapire, R.E. Maximum entropy modeling of species geographic distributions. Ecol. Model. 2006, 190, 231–259. [Google Scholar] [CrossRef]
  22. Song, H.M.; Zhang, Q.F.; Han, X.M.; Xu, P.; Xu, R.M. CLIMEX: Professional biological software for predicting potential distribution of species. Entomol. Knowl. 2004, 41, 379–386. [Google Scholar]
  23. Hirzel, A.H.; Hausser, J.; Chessel, D.; Perrin, N. Ecological-Niche Factor Analysis: How to Compute Habitat-Suitability Maps without Absence Data? Ecology 2002, 83, 2027–2036. [Google Scholar] [CrossRef]
  24. Gu, R.; Wei, S.; Li, J.; Zhwng, S.; Li, Z.; Liu, G.; Fan, S. Predicting the impacts of climate change on the geographic distribution of moso bamboo in China based on biomod2 model. Eur. J. For. Res. 2024, 143, 1499–1512. [Google Scholar] [CrossRef]
  25. Guisan, A.; Edwards, T.C., Jr.; Hastie, T. Generalized linear and generalized additive models in studies of species distributions: Setting the scene. Ecol. Model. 2002, 157, 89–100. [Google Scholar] [CrossRef]
  26. Elith, J.; Leathwick, J.R.; Hastie, T. A working guide to boosted regression trees. J. Anim. Ecol. 2008, 77, 802–813. [Google Scholar] [CrossRef]
  27. Thuiller, W.; Araújo, M.B.; Lavorel, S. Generalized models vs. classification tree analysis: Predicting spatial distributions of plant species at different scales. J. Veg. Sci. 2003, 14, 669–680. [Google Scholar] [CrossRef]
  28. Jinming, Z.; Yi, H.; Sung-Sau, S. Overview of artificial neural networks. Methods Mol. Biol. 2008, 458, 15–23. [Google Scholar]
  29. Booth, T.H.; Nix, H.A.; Hutchinson, M.F.; Jovanic, T. Niche analysis and tree species introduction. For. Ecol. Manag. 1988, 23, 47–59. [Google Scholar] [CrossRef]
  30. Hastie, T.; Tibshirani, R.; Buja, A. Flexible Discriminant Analysis by Optimal Scoring. J. Am. Stat. Assoc. 2012, 89, 1255–1270. [Google Scholar] [CrossRef]
  31. Sekulic, S.; Kowalski, B.R. MARS—A TUTORIAL. J. Chemom. 1992, 6, 199–216. [Google Scholar] [CrossRef]
  32. Breiman, L. Random Forests. Mach. Learn. 2001, 45, 5–32. [Google Scholar] [CrossRef]
  33. Thuiller, W.; Lafourcade, B.; Engler, R.; Araújo, M.B. BIOMOD–a platform for ensemble forecasting of species distributions. Ecography 2009, 32, 369–373. [Google Scholar] [CrossRef]
  34. Hao, T.; Elith, J.; Guillera Arroita, G.; Lahoz-Monfort, J.J. A review of evidence about use and performance of species distribution modelling ensembles like BIOMOD. Divers. Distrib. 2019, 25, 839–852. [Google Scholar] [CrossRef]
  35. Chang, B.; Huang, C.; Chen, B.; Wang, Z.; He, X.; Chen, W.; Huang, Y.; Zhang, Y.; Yu, S. Predicting the potential distribution of Taxus cuspidata in northeastern China based on the ensemble model. Ecosphere 2024, 15, e4965. [Google Scholar] [CrossRef]
  36. McLaughlin, S.P. Floristic plant geography: The classification of floristic areas and floristic elements. Prog. Phys. Geogr. 1994, 18, 185–208. [Google Scholar] [CrossRef]
  37. Velazco, S.J.E.; Rose, M.B.; de Andrade, A.F.A.; Minoli, I.; Franklin, J. Flexsdm: An R package for supporting a comprehensive and flexible species distribution modelling workflow. Methods Ecol. Evol. 2022, 13, 1661–1669. [Google Scholar] [CrossRef]
  38. Barbet-Massin, M.; Jiguet, F.; Albert, C.H.; Thuiller, W. Selecting pseudo-absences for species distribution models: How, where and how many? Methods Ecol. Evol. 2012, 3, 327–338. [Google Scholar] [CrossRef]
  39. Bebber, D.P.; Field, E.; Gui, H.; Mortimer, P.; Holmes, T.; Gurr, S. Many unreported crop pests and pathogens are probably already present. Glob. Change Biol. 2019, 25, 2703–2713. [Google Scholar] [CrossRef] [PubMed]
  40. Pan, Y.; Guan, Y.; Lv, S.; Huang, X.; Lin, Y.; Wei, C.; Xu, D. Assessing the Potential Distribution of Lonicera japonica in China Under Climate Change: A Biomod2 Ensemble Model-Based Study. Agriculture 2025, 15, 393. [Google Scholar] [CrossRef]
  41. Araújo, M.B.; New, M. Ensemble forecasting of species distributions. Trends Ecol. Evol. 2007, 22, 42–47. [Google Scholar] [CrossRef]
  42. Hao, T.; Elith, J.; Lahoz-Monfort, J.J.; Guillera-Arroita, G. Testing whether ensemble modelling is advantageous for maximising predictive performance of species distribution models. Ecography 2020, 43, 549–558. [Google Scholar] [CrossRef]
  43. Valavi, R.; Guillera-Arroita, G.; Lahoz-Monfort, J.J.; Elith, J. Predictive performance of presence-only species distribution models: A benchmark study with reproducible code. Ecol. Monogr. 2022, 92, e1486. [Google Scholar] [CrossRef]
  44. Li, R.; Hu, X.; Li, Q.; Liu, L.; He, Y.; Chen, C. Gap analysis of Firmiana danxiaensis, a rare tree species endemic to southern China. Ecol. Indic. 2024, 158, 111606. [Google Scholar] [CrossRef]
  45. Wisz, M.S.; Guisan, A. Do pseudo-absence selection strategies influence species distribution models and their predictions? An information-theoretic approach based on simulated data. BMC Ecol. 2009, 9, 8. [Google Scholar] [CrossRef]
  46. Chefaoui, R.M.; Lobo, J.M. Assessing the effects of pseudo-absences on predictive distribution model performance. Ecol. Model. 2008, 210, 478–486. [Google Scholar] [CrossRef]
  47. Guo, Y.; Wang, S.; Xu, Y.; Liu, Y.; Yan, X.; Wang, Y. Maternal effects of human-mediated disturbances on the growth and reproductive traits of different ecotypes of an annual invasive plant Erigeron annuus. Environ. Exp. Bot. 2024, 218, 105588. [Google Scholar] [CrossRef]
  48. Lopez, B.E.; Urban, D.; White, P.S. Testing the effects of four urbanization filters on forest plant taxonomic, functional, and phylogenetic diversity. Ecol. Appl. 2018, 28, 2197–2205. [Google Scholar] [CrossRef]
  49. Johnson, M.A.; Kumar, M.; Thakur, S. Effect of Variation in Temperature and Light Duration on Morpho-physiology and Phytochemical Content in Sprouts and Microgreens of Common Buckwheat (Fagopyrum esculentum Moench). Plant Foods Hum. Nutr. 2024, 79, 875–885. [Google Scholar] [CrossRef] [PubMed]
  50. Guo, H.; Sun, Y.; Peng, X.; Wang, Q.; Marvin, H. Up-regulation of abscisic acid signaling pathway facilitates aphid xylem absorption and osmoregulation under drought stress. J. Exp. Bot. 2016, 67, 681–693. [Google Scholar] [CrossRef]
  51. Wang, H.; Bai, W.; Hu, Z.; Lin, S.; Ge, Q. Effectiveness of freezing temperatures on dormancy release of temperate woody species. Ann. Bot. 2024, 134, 615–630. [Google Scholar]
  52. Moustafa, A.E.A.; Zayed, A. Effect of environmental factors on the flora of alluvial fans in southern Sinai. J. Arid. Environ. 1996, 32, 431–443. [Google Scholar] [CrossRef]
  53. Kim, J.S.; Kidokoro, S.; Yamaguchi-Shinozaki, K.; Shinozaki, K. Regulatory networks in plant responses to drought and cold stress. Plant Physiol. 2024, 195, 170–189. [Google Scholar] [CrossRef] [PubMed]
  54. Wang, X.; Song, Q.; Liu, Y.; Marian, B.; Yang, X. The network centered on ICEs play roles in plant cold tolerance, growth and development. Planta 2022, 255, 81. [Google Scholar] [CrossRef]
  55. Ramezan, D.; Zargar, M.; Nakhaev, M.R.; Akhmadovich, K.A.; Bayat, M.; Ghaderi, A. Selenium alleviates growth characteristics, plant pigments, photosynthetic and antioxidant capacity of basil (Ocimum basilicum L.) under low temperature. Biocatal. Agric. Biotechnol. 2024, 58, 103198. [Google Scholar] [CrossRef]
  56. Nievola, C.C.; Carvalho, C.P.; Carvalho, V.; Rodrigues, E. Rapid responses of plants to temperature changes. Temperature 2017, 4, 371–405. [Google Scholar] [CrossRef] [PubMed]
  57. Hasanuzzaman, M.; Nahar, K.; Alam, M.M.; Roychowdhury, R.; Fujita, M. Physiological, biochemical, and molecular mechanisms of heat stress tolerance in plants. Int. J. Mol. Sci. 2013, 14, 9643–9684. [Google Scholar] [CrossRef]
  58. Szymańska, R.; Ślesak, I.; Orzechowska, A.; Kruk, J. Physiological and biochemical responses to high light and temperature stress in plants. Environ. Exp. Bot. 2017, 139, 165–177. [Google Scholar] [CrossRef]
  59. Ma, Z.; Ren, J.; Chen, H.; Jiang, H.; Gao, Q.; Liu, S.; Yan, W.; Li, Z. Analysis and recommendations of IPCC working group. Res. Environ. Sci. 2022, 35, 2550–2558. [Google Scholar]
  60. Lenoir, J.; Gégout, J.C.; Marquet, P.A.; de Ruffray, P.; Brisse, H. A Significant Upward Shift in Plant Species Optimum Elevation During the 20th Century. Science 2008, 320, 1768–1771. [Google Scholar] [CrossRef]
  61. Schai-Braun, S.C.; Jenny, H.; Ruf, T.; Hackländer, K. Temperature increase and frost decrease driving upslope elevational range shifts in Alpine grouse and hares. Glob. Change Biol. 2021, 27, 6602–6614. [Google Scholar] [CrossRef]
  62. Zhao, G.; Cui, X.; Wang, Z.; Jing, H.; Fan, B. Prediction of Potential Distribution of Ziziphus jujuba var. spinosa in China under Context of Climate Change. Sci. Silvae Sin. 2021, 57, 158–168. [Google Scholar]
  63. Thuiller, W.; Lavorel, S.; Araujo, M.B.; Sykes, M.T.; Prentice, I.C. Climate change threats to plant diversity in Europe. Proc. Natl. Acad. Sci. USA 2005, 102, 8245–8250. [Google Scholar] [CrossRef]
  64. Zhou, Z.; Nguyen-Xuan, T.; Liao, H.; Qiu, L.; Im, E.-S. Characterization of temperature and humidity effects on extreme heat stress under global warming and urban growth in the Pearl and Yangtze River Deltas of China. Weather. Clim. Extrem. 2024, 44, 100659. [Google Scholar] [CrossRef]
  65. Wang, H.; Liu, H.; Huang, N.; Bi, J.; Ma, X.; Ma, Z.; Shangguan, Z.; Zhao, H.; Feng, Q.; Liang, T. Satellite-derived NDVI underestimates the advancement of alpine vegetation growth over the past three decades. Ecology 2021, 102, e3518. [Google Scholar] [CrossRef]
  66. Liao, Z.; Yuan, Y.F.; Chen, Y.; Zhai, P.M. Extraordinary hot extreme in summer 2022 over the Yangtze River basin modulated by the La Nina condition under global warming. Adv. Clim. Change Res. 2024, 15, 21–30. [Google Scholar] [CrossRef]
  67. Sun, L.; Cai, Y.; Zhou, Y.; Shi, S.; Zhao, Y.; Gunnarson, B.E.; Jaramillo, F. Radial Growth Responses to Climate of Pinus yunnanensis at Low Elevations of the Hengduan Mountains, China. Forests 2020, 11, 1066. [Google Scholar] [CrossRef]
  68. Ticktin, T. The Ecological Implications of Harvesting Non-Timber Forest Products. J. Appl. Ecol. 2004, 41, 11–21. [Google Scholar] [CrossRef]
  69. Chen, I.; Hill, J.K.; Ohlemüller, R.; Roy, D.B.; Thomas, C.D. Rapid Range Shifts of Species Associated with High Levels of Climate Warming. Science 2011, 333, 1024–1026. [Google Scholar] [CrossRef] [PubMed]
  70. Gaston, K.J.; Jackson, S.F.; Cantú-Salazar, L.; Cruz-Piñón, G. The Ecological Performance of Protected Areas. Annu. Rev. Ecol. Evol. Syst. 2008, 39, 93–113. [Google Scholar] [CrossRef]
  71. Turner, W.; Spector, S.; Gardiner, N.; Fladeland, M.; Sterling, E.; Steininger, M. Remote sensing for biodiversity science and conservation. Trends Ecol. Evol. 2003, 18, 306–314. [Google Scholar] [CrossRef]
Biology 14 01071 g001
Figure 1. Distribution of Spatholobus suberectus in China (the green triangular points in the figure represent the coordinates of the current distribution points of Spatholobus suberectus).
Figure 1. Distribution of Spatholobus suberectus in China (the green triangular points in the figure represent the coordinates of the current distribution points of Spatholobus suberectus).
Biology 14 01071 g001
Biology 14 01071 g002
Figure 2. KAPPA–ROC curve and ROC–TSS curve of the ensemble model and individual models. The average Kappa value and TSS value of the RF model and the XGBOOST model are the highest, and the values are the same. Therefore, the two models in the figure are on the same horizontal line, which causes one of the graph lines to be blocked.
Figure 2. KAPPA–ROC curve and ROC–TSS curve of the ensemble model and individual models. The average Kappa value and TSS value of the RF model and the XGBOOST model are the highest, and the values are the same. Therefore, the two models in the figure are on the same horizontal line, which causes one of the graph lines to be blocked.
Biology 14 01071 g002
Biology 14 01071 g003
Figure 3. Contribution rates of environmental factors. The higher the bar chart, the greater the contribution rate.
Figure 3. Contribution rates of environmental factors. The higher the bar chart, the greater the contribution rate.
Biology 14 01071 g003
Biology 14 01071 g004
Figure 4. Predicted results of the ensemble model for S. suberectus in the current period. Yellow dots are data distribution points.
Figure 4. Predicted results of the ensemble model for S. suberectus in the current period. Yellow dots are data distribution points.
Biology 14 01071 g004
Biology 14 01071 g005
Figure 5. Predicted results of the ensemble model for S. suberectus in the 2050s, 2070s, and 2090s under the SSP1-2.6 scenario.
Figure 5. Predicted results of the ensemble model for S. suberectus in the 2050s, 2070s, and 2090s under the SSP1-2.6 scenario.
Biology 14 01071 g005
Biology 14 01071 g006
Figure 6. Predicted results of the ensemble model for S. suberectus in the 2050s, 2070s, and 2090s under the SSP2-4.5 scenario.
Figure 6. Predicted results of the ensemble model for S. suberectus in the 2050s, 2070s, and 2090s under the SSP2-4.5 scenario.
Biology 14 01071 g006
Biology 14 01071 g007
Figure 7. Predicted results of the ensemble model for S. suberectus in the 2050s, 2070s, and 2090s under the SSP5-8.5 scenario.
Figure 7. Predicted results of the ensemble model for S. suberectus in the 2050s, 2070s, and 2090s under the SSP5-8.5 scenario.
Biology 14 01071 g007
Biology 14 01071 g008
Figure 8. Changes in the centroid of the highly suitable habitat.
Figure 8. Changes in the centroid of the highly suitable habitat.
Biology 14 01071 g008
Table 1. Environmental variables.
Table 1. Environmental variables.
Environment VariablesVariableUnit
gm_lc_v3Soil cover type_
gm_ve_v2Vegetation cover percentage_
hf_v2geo1Human footprint and activity impact index_
annual_mean_uv-bAnnual average ultraviolet radiationkWh/m2
elevAltitudem
slopeSlope°
aspectAspect_
bio01Annual average temperature°C
bio03Isothermality_
bio04Temperature seasonality_
bio05Maximum temperature of the warmest month°C
bio06Minimum temperature of the coldest month°C
bio09Average temperature of the driest season°C
bio11Average temperature of the coldest season°C
bio17Precipitation of the driest seasonmm
d1_clayClay contentcps
d1_cn_ratioCarbon-to-nitrogen ratio_
d1_elec_condElectrical conductivity_
d1_ph_waterpH levelpH
d1_sandSand contentkg/m3
d1_swrSoil moisture status_
d1_total_nTotal nitrogen contentmg/L
d1_usdaSoil texture classification_
Table 2. TSS and KAPPA evaluation criteria.
Table 2. TSS and KAPPA evaluation criteria.
TSSKAPPAEvaluation Criteria
0 ≤ TSS ≤ 0.40 ≤ KAPPA ≤ 0.4Lose
0.4 < TSS ≤ 0.550.4 < KAPPA ≤ 0.55Normal
0.55 < TSS ≤ 0.70.55 < KAPPA ≤ 0.7Good
0.7 < TSS ≤ 0.850.7 < KAPPA ≤ 0.85Fine
0.85 < TSS 0.85 < KAPPASuperb
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Lin, Y.; Liu, Q.; Lv, S.; Huang, X.; Wei, C.; Li, J.; Guan, Y.; Pan, Y.; Mi, Y.; Cheng, Y.; et al. Assessing the Potential Distribution of the Traditional Chinese Medicinal Plant Spatholobus suberectus in China Under Climate Change: A Biomod2 Ensemble Model-Based Study. Biology 2025, 14, 1071. https://doi.org/10.3390/biology14081071

AMA Style

Lin Y, Liu Q, Lv S, Huang X, Wei C, Li J, Guan Y, Pan Y, Mi Y, Cheng Y, et al. Assessing the Potential Distribution of the Traditional Chinese Medicinal Plant Spatholobus suberectus in China Under Climate Change: A Biomod2 Ensemble Model-Based Study. Biology. 2025; 14(8):1071. https://doi.org/10.3390/biology14081071

Chicago/Turabian Style

Lin, Yijun, Quanwei Liu, Shan Lv, Xiaoyu Huang, Chaoyang Wei, Jun Li, Yijie Guan, Yaxuan Pan, Yijia Mi, Yanshu Cheng, and et al. 2025. "Assessing the Potential Distribution of the Traditional Chinese Medicinal Plant Spatholobus suberectus in China Under Climate Change: A Biomod2 Ensemble Model-Based Study" Biology 14, no. 8: 1071. https://doi.org/10.3390/biology14081071

APA Style

Lin, Y., Liu, Q., Lv, S., Huang, X., Wei, C., Li, J., Guan, Y., Pan, Y., Mi, Y., Cheng, Y., Yang, X., & Xu, D. (2025). Assessing the Potential Distribution of the Traditional Chinese Medicinal Plant Spatholobus suberectus in China Under Climate Change: A Biomod2 Ensemble Model-Based Study. Biology, 14(8), 1071. https://doi.org/10.3390/biology14081071

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop