Next Article in Journal
Envenomation with Snake Venoms as a Cause of Death: A Forensic Investigation of the Decomposition Stages and the Impact on Differential Succession Pattern of Carcass-Attracted Coleopteran Beetles
Previous Article in Journal
Toxicity of Flonicamid to Diaphorina citri (Hemiptera: Liviidae) and Its Identification and Expression of Kir Channel Genes
Previous Article in Special Issue
Stability and Assembly Mechanisms of Butterfly Communities across Environmental Gradients of a Subtropical Mountain
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Insects
Volume 15
Issue 11
10.3390/insects15110901
insects-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 thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Predicted Spatial Patterns of Suitable Habitats for Troides aeacus Under Different Climate Scenarios

by
Biyu Liu
†,
Xinqi Deng
†,
Zhiqian Liu
,
Xinju Wei
,
Honghua Zhang
,
Danping Xu
* and
Zhihang Zhuo
*
College of Life Science, China West Normal University, Nanchong 637002, China
*
Authors to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Insects 2024, 15(11), 901; https://doi.org/10.3390/insects15110901
Submission received: 26 October 2024 / Revised: 13 November 2024 / Accepted: 15 November 2024 / Published: 18 November 2024
(This article belongs to the Collection Butterfly Diversity and Conservation)
Browse Figures
Versions Notes

Simple Summary

This study was conducted to examine how the future distribution of Troides aeacus, a butterfly species, may be affected by climate change under different scenarios. It was found that temperature and precipitation are the main factors influencing its suitable habitat. By the 2050s and 2090s, both low and highly suitable areas are expected to expand, particularly under severe warming conditions (SSP5-8.5), while moderately suitable areas are likely to shrink. This shift could result in potential risks, such as habitat loss, increased competition, and local population declines. To ensure the species’ survival, conservation strategies should be focused on protecting optimal habitats and effectively managing marginal areas.

Abstract

Troides aeacus is the largest butterfly in China and is highly valued for its ornamental beauty. Due to T. aeacus being classified as a national second-class protected species in China, studying its spatial distribution is crucial for developing effective conservation measures. In this study, a total of 490 distribution points were obtained, and the potential distribution areas of the golden-sheathed T. aeacus were analyzed by using the maximum entropy model (MaxEnt) based on three different greenhouse gas emission scenarios, namely, SSP1-2.6, SSP2-4.5, and SSP5-8.5, in combination with nine important environmental variables. The results indicate that temperature and precipitation are the primary environmental factors influencing the suitable habitat of T. aeacus, with key variables including the minimum temperature of the coldest month (bio6), temperature annual range (bio7), mean temperature of the warmest quarter (bio10), annual precipitation (bio12), precipitation of the coldest quarter (bio19), and slope. The height distribution of T. aeacus in my country is in the area south of the Huaihe River in the Qinling Mountains, with a total area of 270.96 × 104 km2, accounting for 28.23% of the total area of China. According to future climate change conditions, as climate warming progresses, both low- and high-suitability areas show an expansion trend in most scenarios, particularly under the SSP5-8.5 scenario, where highly suitable areas increase significantly while moderately suitable areas gradually shrink. To address future climate change, conservation strategies should focus on protecting highly suitable areas and strengthening the management of marginal habitats to enhance the adaptability and survival chances of T. aeacus.

1. Introduction

Troides aeacus is the most northerly species in the Troides/Trogonoptera/Ornithoptera clade, known for its large size and golden–yellow hindwings. It is divided into five subspecies: the Troides aeacus aeacus, the Troides aeacus formosanus, the Troides aeacus insularis, the Troides aeacus malaiianus, and the Troides aeacus szechwanus [1]. T. aeacus is internationally recognized as one of the most ornamental butterfly species [2]. Its larvae feed on Aristolochiaceae plants, particularly those of the genus Aristolochia, such as Aristolochia acuminata and Aristolochia foveolata. Adults prefer to live in hot jungles, valleys, hills, and subtropical climates [3]. Due to differences in habitat and diet between larval and adult stages and sensitivity to environmental changes, T. aeacus’s survival is affected by numerous factors [4,5]. Consequently, it is listed as a protected species under the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) and as a second-class protected animal in the China National Key Protected Wildlife List [6]. Due to a decrease in habitat adaptability, the population of T. aeacus has been continuously declining [7]. Thus far, research on the potential suitable distribution range of T. aeacus has been relatively limited [8]. Therefore, conducting studies on the suitable distribution areas of T. aeacus in China is particularly important.
Climate change has the potential to affect species’ geographic distribution, abundance, distribution patterns, interspecies relationships, phenology, and photosynthesis, and it disrupts the balance of ecosystems [9]. Warmer temperatures can lead to increased metabolic rates, affecting the survival and reproduction of species. For some, this may mean population booms, while for others, it could signal a decline or even local extinction. The distribution patterns of species are further complicated by these changes, as habitats that were once suitable may no longer support the same diversity of life. For wildlife, climate change-induced habitat loss, food shortages, and invasive species pose significant survival challenges. Research indicates that the loss of various life forms has substantial impacts on the structure and function of ecosystems and affects ecosystem services [10]. This is particularly concerning for rare and endangered wildlife, as their lower abundance and often restricted and fragmented geographic ranges make them more vulnerable to environmental changes and extinction if not adequately protected [11,12].
The relationship between climate change and species distribution has long been a focal point of research. Predicting changes in suitable habitats for protected species under climate change is crucial for understanding species development patterns and establishing endangered species protection systems [13,14]. Therefore, studying the distribution dynamics of endangered species is particularly important for species assessment and conservation [15]. However, relying solely on field survey data often fails to capture the overall distribution trends of species comprehensively. By collecting known geographic distributions and corresponding environmental variables of species, suitable habitats can be predicted. Recently, Species Distribution Models (SDMs) have been widely used for predicting potential habitat distributions and have become an important tool for studying species suitability [16,17]. The most commonly used Species Distribution Models include Generalized Linear Models (GLM), Genetic Algorithm for Rule-set Prediction (GARP), Maximum Entropy (MaxEnt), HABITAT, and BIOCLIM. Each model has its advantages and limitations due to differences in their principles and algorithms [18,19]. When the relationship between species and environmental conditions is complex or the species’ distribution range is large, the Maximum Entropy model (MaxEnt) often provides better predictive accuracy compared to other models [20]. MaxEnt software (v3.4.4), which models species distribution based on recorded points, performs well even with small sample sizes and sparse distribution points and is widely recognized in the field for its short running time, simplicity of use, and minimal sample requirements [21,22,23].
Currently, research on the T. aeacus is still limited, with existing studies primarily focusing on its biological characteristics and habitat requirements. Research on the suitability assessment and protection of potential habitat distributions has not been given sufficient attention [24]. Therefore, this study uses the MaxEnt model to explore the relationship between the distribution of the T. aeacus and climatic conditions in China. It predicts the current and future potential distributions of the species and analyzes the centroid shift trend of its potential geographic distribution, providing important references and theoretical foundations for developing effective conservation measures.

2. Materials and Methods

2.1. Species Distribution Data

All location data for this study are from China. The MaxEnt model simulation requires species distribution data and environmental data. Distribution data for the T. aeacus were obtained from the following: (1) specimens in research institution collections; (2) relevant literature records; and (3) the Global Biodiversity Information Facility (GBIF, http://www.gbif.org/) (accessed on 10 October 2024). Using Google Maps (http://ditu.google.cn) (accessed on 10 October 2024), these data were converted into geographic coordinates, resulting in a total of 490 sample points. To prevent model overfitting, if the distance between two distribution points was less than the environmental factor grid resolution (1 km), one of the points was removed [25]. Ultimately, 234 distribution records of the T. aeacus were selected. These distribution points were compiled in an Excel sheet with species name, longitude, and latitude in decimal format and saved as a “CSV” file for MaxEnt model calculations (Figure 1).

2.2. Environmental Variables and Data Processing

In this study, 22 environmental variables were selected, including 19 bioclimatic variables (bio1–bio19) and three terrain factors (Table 1). Environmental data were downloaded from WorldClim (https://www.worldclim.org/) (accessed on 10 October 2024), with “current period” data from 1970 to 2000 and future climate data for the 2050s (2041–2060) and 2090s (2081–2100) [26]. For future climates, this study chose SSP1-2.6, SSP2-4.5, and SSP5-8.5, representing low, medium, and high greenhouse gas emission scenarios, respectively, provided by the Beijing Climate Center’s new generation climate model BCC-CSM2MR [27]. Due to strong autocorrelation among environmental variables and the fact that not all variables are necessary for species distribution prediction [28,29], bioclimatic variables with high similarity to others or lower predictive capability were excluded to avoid overfitting and improve model accuracy [30]. Pearson correlation coefficients (R) were used to identify multicollinearity, and variables contributing significantly to model development were selected to minimize the impact of covariance on modeling and result interpretation [31,32]. When the absolute correlation coefficient between two ecological factors was ≥0.8, one representative variable was retained [33,34].

2.3. MaxEnt Model Construction and Result Evaluation

In this study, we used MaxEnt version 3.4.1 to model the current and future (2050s and 2090s) potential habitats of the T. aeacus. We imported the filtered distribution points into ArcGIS 10.8 and used the “Toolbox/Spatial Analyst Tools/Extraction/Sample” tool to sample and interpolate the distribution points across 22 environmental variables. The same method was applied to the initial MaxEnt prediction model for further sampling and interpolation of the distribution points. In SPSS, we conducted multicollinearity analysis using the Variance Inflation Factor (VIF) and Spearman correlation analysis on the interpolated data, initially selecting environmental factors with a VIF of less than 100 and a correlation of less than 0.8. VIF, also known as the reciprocal of tolerance, indicates multicollinearity: when VIF < 10, there is no multicollinearity among factors; when 10 < VIF < 100, moderate multicollinearity exists; and when VIF > 100, severe multicollinearity is present [35,36]. Environmental variables are crucial parameters in constructing niche models, as redundant variables can lead to overfitting, reducing model accuracy [37]. In this study, we used ENMTools.pl to analyze the correlation among the 22 environmental variables, defining variables with a Pearson correlation coefficient > |0.8| as highly correlated [38]. By comparing the contribution rates of variables in the initial model, we retained those with higher contributions and easier interpretability [39,40].
Based on statistical significance (partial receiver operating characteristic, ROC, with 500 iterations), predictive capability (omission rate, OR), and model complexity (AICc), we optimized the MaxEnt model parameters using the kuenm_ceval function [41]. According to the “OR_AICc” standard, a significant candidate model is defined as one with an omission rate below the threshold (e.g., ≤0.05, where applicable) and the lowest ΔAICc value (≤2), which is considered the final model with optimal parameters [42]. MaxEnt model parameters include feature combinations (FC), regularization multipliers (RM), and the maximum number of background points (BC), among others [43]. MaxEnt supports five feature types: linear (L), quadratic (Q), hinge (H), product (P), and threshold (T) [44]. By default, the RM value is set to 1, and the selection of feature combinations depends on the number of species occurrence points. Generally, the linear feature is always enabled; quadratic features are used when occurrence points exceed 10, hinge features are used when points exceed 15, and threshold and product features are applied when points exceed 80 [45]. However, studies have shown that the default MaxEnt settings are not always suitable for predicting all species distributions, potentially resulting in overfitting and difficulties in interpreting the predictions [46].
In this study, we used the ENMeval package in R 3.6.3 to optimize the MaxEnt model. Initially, we used a block partitioning method, dividing the distribution points of T. aeacus into four parts, with three parts used for training and one part for testing [47,48]. Subsequently, we utilized the kuenm package in R 3.6.3 to compare various combinations of the two key parameters (feature class and regularization multiplier) to identify the optimal combination [49]. The five MaxEnt features yielded 31 feature combinations, while the regularization multiplier was tested across four values ranging from 0.1 to 4 in 0.1 intervals. A total of 1240 candidate models were evaluated, including all combinations of 40 regularization multiplier settings, 31 feature classes, and a set of 11 environmental variables.
The receiver operating characteristic (ROC) curve is used to assess the accuracy of model predictions, with the Area Under the Curve (AUC) under the ROC curve serving as a performance indicator for MaxEnt predictions [50]. The AUC evaluation criteria are: 0.5 ≤ AUC < 0.6 indicates poor prediction; 0.6 ≤ AUC < 0.7 indicates fair prediction; 0.7 ≤ AUC < 0.8 indicates good prediction; 0.8 ≤ AUC < 0.9 indicates very good prediction; AUC > 0.9 indicates excellent prediction [51]. The closer the AUC value is to 1.0, the higher the accuracy of the model, allowing for the identification of the best predictive model.

2.4. Suitable Grade Zoning

Based on the actual situation of the T. aeacus, we used MaxEnt modeling and ArcGIS software to generate the probability distribution map of T. aeacus in China. To classify the levels of distribution values and their corresponding distribution ranges, the “Reclassify” function in ArcGIS was used according to the probability classification method recommended by the Intergovernmental Panel on Climate Change (IPCC) report [52]. Suitable habitats were categorized into four levels, with specific classification criteria as follows: 0–0.05 (unsuitable area); 0.05–0.33 (low-suitability area); 0.33–0.66 (moderate-suitability area); and 0.66–1 (high-suitability area) [53]. In the map, white, blue, orange, and red represent unsuitable areas, low-suitability areas, moderate-suitability areas, and high-suitability areas, respectively. Additionally, the SDMtoolbox in ArcGIS 10.8 (http://www.sdmtoolbox.org/) (accessed on 10 October 2024) was used to analyze the centroid positions and migration trends in T. aeacus’s potential distribution under different future climate scenarios [54].

3. Results

3.1. Model Performance and Key Environmental Variables

By running the kuenm package in R 3.6.3, we identified only one model that met both the OR and AICc criteria. Consequently, for the MaxEnt model settings of T. aeacus, we selected the model M_0.1_F_pq_Set_1 (regularization multiplier = 0.1, feature combination = P and Q). Using 234 current distribution records and 9 environmental variables, we simulated the potential geographic distribution of T. aeacus in China with the MaxEnt software. The AUC value for the training data was 0.984 (Figure 2), representing an “excellent” level of performance. This indicates that the MaxEnt model’s predictions are accurate and reliable, demonstrating high predictive capability.

3.2. The Main Environmental Factors Influencing the Distribution of T. aeacus

The MaxEnt model results identified the key environmental factors influencing the distribution range of T. aeacus and their respective contributions. The contribution rates of various environmental variables reflect their importance in determining species distribution. Among all the variables, the annual temperature range (bio7) plays a dominant role with a contribution rate of 26.50%, highlighting the significant influence of temperature variation on habitat selection. This is followed by annual precipitation (bio12), with a contribution rate of 20.10%. The minimum temperature of the coldest month (bio6) and slope contribute 15.00% and 10.00%, respectively, indicating that temperature extremes and terrain features also significantly impact habitat suitability. Additionally, the precipitation of the coldest quarter (bio19) contributes 10.00%, while the mean temperature of the warmest quarter (bio10) and precipitation of the driest month (bio14) contribute 7.90% and 6.70%, respectively, underscoring the role of seasonal climate in species adaptation. In contrast, altitude (elev) and aspect have lower contribution rates of 3.10% and 0.80%, respectively, suggesting a relatively minor influence on species distribution. Overall, temperature and precipitation emerge as the key determinants of species distribution, while terrain and other environmental variables have a comparatively secondary role (Table 2). The Jackknife test results indicated that when using single environmental variables, the five most influential variables were as follows: temperature annual range (bio5–bio6) (bio7), annual precipitation (bio12), min. temperature of the coldest month (bio6), slope (slope), and precipitation of the coldest quarter (bio19) (Figure 3).

3.3. Environmental Variables Influencing the Geographic Distribution of the T. aeacus

Based on the response curves of the probability distribution of the T. aeacus depicted by the MaxEnt model (Figure S1), the value ranges for variables affecting the future distribution of T. aeacus were determined. The results show that the range for temperature annual range (bio5–bio6) is 0.16 °C to 22.84 °C. Annual precipitation ranges from 2080.5 mm to 6324.6 mm. The range for the minimum temperature of the coldest month is 6.89 °C to 25.03 °C. Slope varies between 5.37° and 30.12°. The precipitation of the coldest quarter ranges from 81.83 mm to 2410.38 mm. Variations in these influencing factors have a significant impact on the presence probability of T. aeacus. If values exceed these ranges, the probability of the species’ distribution will gradually decrease.

3.4. Potential Distribution of the T. aeacus in the Current Period

Based on the MaxEnt model simulations using nine key environmental variables and distribution data of the T. aeacus, the potential distribution areas for the current period were classified into four suitability levels: high suitability, medium suitability, low suitability, and unsuitable. The results, which are illustrated in Figure 4, show a clear spatial gradient in the geographic distribution of suitable habitat for T. aeacus. The high-suitability areas are mainly concentrated in southern China and the southeastern coastal regions, including Guangdong, Guangxi, Hainan, Yunnan, Guizhou, and parts of the mid-to-lower Yangtze River basin, where environmental conditions are most favorable. Medium-suitability areas are widely distributed across the Yangtze River basin and its surrounding central and eastern regions, such as Hunan, Jiangxi, northern Zhejiang, and Jiangsu, where conditions are slightly less optimal but still supportive of the species’ growth. Low-suitability areas are scattered across the southern margins of North China, transition zones north of the Yangtze River, and the high-altitude edges of the southwestern region, where harsher climatic conditions limit the species’ wider distribution. Overall, the suitable habitat for T. aeacus displays a “southern concentration, northern dispersion” pattern, with suitability gradually decreasing as latitude increases.
The total area of suitable habitat for the T. aeacus in the contemporary period is 270.96 × 104 km2 (Table 3). Among these, high-suitability areas, medium-suitability areas, and low-suitability areas cover 74.2 × 104 km2, 113.03 × 104 km2, and 83.734 km2, respectively, representing 27.38%, 41.71%, and 30.90% of the total suitable habitat area.

3.5. Potential Future Distribution of the T. aeacus

Figure 5 illustrates the predicted distribution of suitable habitats for the T. aeacus under three climate change scenarios—SSP1-2.6, SSP2-4.5, and SSP5-8.5—for the 2050s (2041–2060) and 2090s (2081–2100). In addition, the areas of different suitability zone classes were also calculated (Table 3). In the 2050s, under different climate scenarios, the suitable habitat of T. aeacus shows a significant trend of change. The low-suitability area expands across all scenarios, with the most notable increase in the SSP5-8.5 scenario, where the area grows by 171,600 km2, representing a 20.49% increase compared to the current distribution. The SSP1-2.6 and SSP2-4.5 scenarios exhibit smaller expansions, with increases of 1400 km2 (0.17%) and 300 km2 (0.04%), respectively. The medium-suitability area shows minimal change in the 2050s, with decreases of 300 km2 (−0.03%) and 200 km2 (−0.02%) under SSP1-2.6 and SSP2-4.5 scenarios, respectively. However, it declines by 69,800 km2 (−6.18%) under the SSP5-8.5 scenario. In contrast, the high-suitability area expands overall, with increases of 2000 km2 (0.27%) and 487,900 km2 (65.77%) under SSP2-4.5 and SSP5-8.5, while remaining stable under SSP1-2.6.
In the 2090s, changes in the suitable area intensify across all scenarios. The low-suitability area expands significantly, with the largest increase under the SSP5-8.5 scenario, where it grows by 263,300 km2, representing a 31.43% increase from the current extent. SSP1-2.6 and SSP2-4.5 also show considerable expansions of 154,100 km2 (18.41%) and 196,700 km2 (23.49%), respectively. Meanwhile, the medium-suitability area decreases more sharply, with reductions of 59,200 km2 (−5.24%) and 76,200 km2 (−6.74%) under SSP1-2.6 and SSP2-4.5, respectively, and a substantial decline of 203,900 km2 (−18.04%) under SSP5-8.5. The high-suitability area experiences significant expansion in the 2090s, particularly under SSP5-8.5, increasing by 555,000 km2 (74.78%). It also grows under SSP2-4.5 and SSP1-2.6, with increases of 451,600 km2 (60.86%) and 142,800 km2 (19.25%), respectively.

3.6. Migration of the Centroid of Potential Distribution for the T. aeacus

Under different climate scenarios, the centroid of the suitable habitat for T. aeacus shows significant spatial changes. Overall, as climate warming intensifies, the centroid tends to shift from northwest to southeast or northeast, but the specific trends vary across scenarios (Figure 6). In the SSP1-2.6 scenario, the centroid first shifts slightly northwest by 28.01 km from the present to the 2050s, with a direction of 297.3 degrees. However, from the 2050s to the 2090s, it shifts southeast by 94.99 km, with a direction of 108.92 degrees, indicating a transition from northwest to southeast due to future warming. In the SSP2-4.5 scenario, the centroid consistently moves northeast over all three periods: 115.89 km from the present to the 2050s, 116.58 km from the 2050s to the 2090s, and a total of 232.47 km from the present to the 2090s, with directions of 53.62, 54.27, and 53.74 degrees, respectively, suggesting a stable northeast shift under moderate warming conditions. In the SSP5-8.5 scenario, the centroid moves rapidly northeast by 160.04 km before the 2050s, with a direction of 27.52 degrees, representing the most pronounced early shift among all scenarios. However, from the 2050s to the 2090s, it shifts southeast by 47.5 km, with a direction of 151.58 degrees, indicating a temporal change in the expansion direction under extreme warming. Over the entire period from the present to the 2090s, the centroid moves 139.58 km northeast, with a direction of 43.98 degrees, demonstrating a long-term trend in expansion toward the northeast (Table 4).
The direction of centroid movement in the suitable habitat varies under different climate scenarios. In the SSP1-2.6 scenario, the centroid shifts northwest in the short term but transitions southeast in the long term. In the SSP2-4.5 scenario, the centroid exhibits a consistent shift toward the northeast. In the SSP5-8.5 scenario, the centroid rapidly expands to the northeast before the 2050s but gradually shifts southeast by the 2090s.

4. Discussion

This study, based on the MaxEnt model and ArcGIS geographic information technology, analyzes the current suitable distribution locations of the protected wildlife species in China, the T. aeacus, and predicts the future potential suitable areas while exploring the ecological characteristics of these regions. The model evaluation shows an AUC value of 0.962, indicating high accuracy. The results classify the T. aeacus into four different habitat suitability zones: high-suitability, moderate-suitability, low-suitability, and unsuitable zones. The projected climate change scenarios indicate that T. aeacus-suitable habitats will experience significant spatial shifts by 2050 and 2090. Climate change is expected to substantially expand the species’ highly suitable habitat areas, particularly under the SSP5-8.5 scenario, where the high-suitability range is anticipated to increase by 555,000 km2 by 2090—a 74.78% growth. This expansion of highly suitable areas suggests that T. aeacus may benefit from more optimal habitats under warming conditions, supporting population stability and growth. Furthermore, the dynamic nature of climate change and the increase in extreme weather events introduce uncertainties, potentially causing spatial and temporal fluctuations in habitat suitability. While the significant expansion of highly suitable areas may offer short-term advantages for population growth, the ecological stability of these expanded regions remains challenged by the impacts of ongoing climate change. Future research should incorporate these complex factors to better understand how shifts in habitat suitability may influence the long-term survival of T. aeacus. Although this study did not incorporate anthropogenic factors, human activities such as habitat loss, land-use changes, pollution, and resource competition may significantly impact the habitat suitability of T. aeacus, leading to habitat fragmentation, reduced ecological adaptability, and increased survival pressure. Therefore, future research should consider these human-induced disturbances to more comprehensively assess their potential effects on the distribution and population dynamics of T. aeacus.
Generally, the ecological niche of a species remains relatively stable over short historical periods, with minimal evolutionary changes [55]. Environmental variables affecting species distribution at varying spatial scales often show that, at larger scales, species interactions weaken and climate variables play a predominant role [56]. Research indicates that the choice of environmental variables can impact the predictions of ecological niche models [57]. Using the 22 bioclimatic factors from the WorldClim database, this study addressed the issue of unavoidable autocorrelation and multicollinearity among variables by performing correlation analysis and selection. Jackknife tests combined with Pearson correlation coefficients identified nine key environmental variables that limit the distribution of the T. aeacus: bio7, bio12, bio6, slope, bio19, bio10, bio14, elev, and aspect. The MaxEnt model was reconstructed to reduce redundant information and improve accuracy. Results showed that temperature and precipitation are significant factors influencing the distribution of T. aeacus [58]. Previous studies have shown that the T. aeacus prefers to live in hot jungles and valleys, being most commonly found during the rainy season. These areas have an annual temperature range of 16 °C–35 °C and an annual rainfall of 1500 mm to 2000 mm. The slightly increased annual precipitation observed in this study compared to past studies may be linked to global warming, which allows the atmosphere to hold more moisture, often resulting in increased precipitation.
As climate change and human activities impact the environment, biodiversity crises are occurring globally, with habitat fragmentation and loss being major drivers of biodiversity decline and species extinction [59]. Understanding the potential geographic distribution of threatened species and their current habitat suitability is crucial for effective conservation efforts. However, many species, especially endangered ones like the T. aeacus, have limited geographic distribution data. Based on the MaxEnt model predictions, the northward shift of the centroid suggests that the species’ suitable habitat may gradually expand to cooler northern regions in response to climate change or other environmental pressures, potentially as an adaptation to new ecological conditions or challenges brought about by changing climates.

5. Conclusions

Based on the MaxEnt model and species data distribution, the current and future suitable habitat distribution areas of the T. aeacus in China were determined. The results show that the min. temperature of the coldest month (bio6), temperature annual range (bio7), mean temperature of the warmest quarter (bio10), annual precipitation (bio12), precipitation of the coldest quarter (bio19), and slope affect its distribution. The main environmental factors of the pattern. Climate change will significantly affect its distribution pattern. Temperature and precipitation are the main factors affecting its distribution pattern. Habitat expansion and contraction are closely related to these indicators. Under the current climate conditions, the T. aeacus is mainly distributed in the area south of the Huaihe River in the Qinling Mountains. The high-suitability areas are mainly distributed in Guangxi, Guangdong, Hong Kong, and Taiwan, with the highly suitable area reaching 74.2 × 104 square kilometers. Under future climate conditions, the low- and high-suitability areas are expected to expand significantly, especially under the SSP5-8.5 scenario in the 2090s; they increased by 26.33% and 55.5%, respectively. Therefore, relevant managers can use these habitat suitability maps to identify high-risk areas, thereby prioritizing conservation actions in these areas and facilitating the smooth implementation of habitat assessment and protection work.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/insects15110901/s1. Figure S1: Response Curves of the presence probability of the T. aeacus to key environmental variables.

Author Contributions

Conceptualization, Z.Z.; methodology, B.L. and X.D.; software, B.L.; formal analysis, Z.L.; investigation, X.W.; data curation, X.W. and H.Z.; writing—original draft preparation, B.L.; writing—review and editing, B.L. and D.X.; supervision, Z.Z. All authors have read and agreed to the published version of the manuscript.

Funding

This work was funded by the National Key R&D Program of China (2022YFE0115200), the Sichuan Province Science and Technology Support Program (2022NSFSCO986), and the China West Normal University Support Program (20A007, 20E051, 21E040, and 22kA011).

Data Availability Statement

The data supporting the results are available in a public repository at: GBIF.org (12 April 2024), GBIF Occurrence Download https://doi.org/10.15468/dl.4gyzgp; T. aeacus occurrence data: 10.6084/m9.figshare.27309102. Inquiries regarding code availability and further inquiries can be directed to the corresponding author.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Wu, I.; Yang, P.; Liu, C.; Yeh, W. Genetic differentiation of Troides aeacus formosanus (Lepidoptera: Papilionidae), based on cytochrome oxidase I sequences and amplified fragment length polymorphism. Ann. Entomol. Soc. Am. 2010, 103, 1018–1024. [Google Scholar] [CrossRef]
  2. Chen, P.; Arikawa, K.; Yang, E. Diversity of the photoreceptors and spectral opponency in the compound eye of the Golden Birdwing, Troides aeacus formosanus. PLoS ONE 2013, 8, e62240. [Google Scholar] [CrossRef] [PubMed]
  3. Hsieh, K.; Kuo, Y.; Perng, J.; Lee, T.; Lee, H. Oviposition preference and larval survival of Troides aeacus formasanus (Lepidoptera: Papilionidae) on Aristolochia zollingeriana in different environments in the Kenting area. Taiwan J. For. Sci. 2010, 25, 353–368. [Google Scholar]
  4. Koh, L.P. Impacts of land use change on South-east Asian forest butterflies: A review. J. Appl. Ecol. 2007, 44, 703–713. [Google Scholar] [CrossRef]
  5. Cao, Y.; Li, R.; Zhou, S.; Song, L.; Quan, R.; Hu, H. Ethnobotanical study on wild edible plants used by three trans-boundary ethnic groups in Jiangcheng County, Pu’er, Southwest China. J. Ethnobiol. Ethnomed. 2020, 16, 1–23. [Google Scholar] [CrossRef]
  6. Li, X.; Luo, Y.; Zhang, Y.; Schweiger, O.; Settele, J.; Yang, Q. On the conservation biology of a Chinese population of the birdwing Troides aeacus (Lepidoptera: Papilionidae). J. Insect Conserv. 2010, 14, 257–268. [Google Scholar] [CrossRef]
  7. Huang, Y.; Chen, T.; Chang, Z.; Wang, T.; Lee, S.J.; Kim, J.C.; Kim, J.S.; Chiu, K.; Nai, Y. Genomic sequencing of Troides aeacus nucleopolyhedrovirus (TraeNPV) from golden birdwing larvae (Troides aeacus formosanus) to reveal defective Autographa californica NPV genomic features. BMC Genom. 2019, 20, 419. [Google Scholar] [CrossRef]
  8. Fang, L.; Zhang, Y.; Gao, K.; Ding, C.; Zhang, Y. Butterfly communities along the Heihe River Basin in Shaanxi Province, a biodiversity conservation priority area in China. J. Insect Conserv. 2019, 23, 873–883. [Google Scholar] [CrossRef]
  9. Lenoir, J.; Gégout, J.; 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]
  10. Parmesan, C.; Yohe, G. A globally coherent fingerprint of climate change impacts across natural systems. Nature 2003, 421, 37–42. [Google Scholar] [CrossRef]
  11. Pizzi, F.; Caroli, A.M.; Landini, M.; Galluccio, N.; Mezzelani, A.; Milanesi, L. Conservation of endangered animals: From biotechnologies to digital preservation. Nat. Sci. 2013, 5, 903–913. [Google Scholar] [CrossRef]
  12. Pilling, D.; Hoffmann, I. Climate change and animal genetic resources for food and agriculture: State of knowledge, risks and opportunities. FAO CGRFA Backgr. Study Pap. 2011, 53, 28. [Google Scholar]
  13. Rumpf, S.B.; Hülber, K.; Klonner, G.; Moser, D.; Schütz, M.; Wessely, J.; Willner, W.; Zimmermann, N.E.; Dullinger, S. Range dynamics of mountain plants decrease with elevation. Proc. Natl. Acad. Sci. USA 2018, 115, 1848–1853. [Google Scholar] [CrossRef]
  14. Thomas, C.D.; Cameron, A.; Green, R.E.; Bakkenes, M.; Beaumont, L.J.; Collingham, Y.C.; Erasmus, B.F.; de Siqueira, M.F.; Grainger, A.; Hannah, L. Extinction risk from climate change. Nature 2004, 427, 145–148. [Google Scholar] [CrossRef] [PubMed]
  15. Thorn, J.S.; Nijman, V.; Smith, D.; Nekaris, K. Ecological niche modelling as a technique for assessing threats and setting conservation priorities for Asian slow lorises (Primates: Nycticebus). Divers. Distrib. 2009, 15, 289–298. [Google Scholar] [CrossRef]
  16. Yu, D.; Chen, M.; Zhou, Z.; Eric, R.; Tang, Q.; Liu, H. Global climate change will severely decrease potential distribution of the East Asian coldwater fish Rhynchocypris oxycephalus (Actinopterygii, Cyprinidae). Hydrobiologia 2013, 700, 23–32. [Google Scholar] [CrossRef]
  17. Coro, G.; Vilas, L.G.; Magliozzi, C.; Ellenbroek, A.; Scarponi, P.; Pagano, P. Forecasting the ongoing invasion of Lagocephalus sceleratus in the Mediterranean Sea. Ecol. Model. 2018, 371, 37–49. [Google Scholar] [CrossRef]
  18. 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]
  19. Yoon, S.; Lee, W. Methodological analysis of bioclimatic variable selection in species distribution modeling with application to agricultural pests (Metcalfa pruinosa and Spodoptera litura). Comput. Electron. Agric. 2021, 190, 106430. [Google Scholar] [CrossRef]
  20. Merow, C.; Smith, M.J.; Silander Jr, J.A. A practical guide to MaxEnt for modeling species’ distributions: What it does, and why inputs and settings matter. Ecography 2013, 36, 1058–1069. [Google Scholar] [CrossRef]
  21. Bosso, L.; Di Febbraro, M.; Cristinzio, G.; Zoina, A.; Russo, D. Shedding light on the effects of climate change on the potential distribution of Xylella fastidiosa in the Mediterranean basin. Biol. Invasions 2016, 18, 1759–1768. [Google Scholar] [CrossRef]
  22. Han, Y.; Wang, Y.; Xiang, Y.; Ye, J. Prediction of potential distribution of Bursaphelenchus xylophilus in China based on Maxent ecological niche model. J. Nanjing For. Univ. 2015, 58, 6. [Google Scholar]
  23. Yang, J.; Huang, Y.; Jiang, X.; Chen, H.; Liu, M.; Wang, R. Potential geographical distribution of the edangred plant Isoetes under human activities using MaxEnt and GARP. Glob. Ecol. Conserv. 2022, 38, e2186. [Google Scholar] [CrossRef]
  24. Beirão, M.V.; Campos-Neto, F.C.; Pimenta, I.A.; Freitas, A.V. Population biology and natural history of Parides burchellanus (Papilionidae: Papilioninae: Troidini), an endangered Brazilian butterfly. Ann. Entomol. Soc. Am. 2012, 105, 36–43. [Google Scholar] [CrossRef]
  25. Zhang, D.H.; Hu, Y.M.; Liu, M. Potential distribution of Spartinal alterniflora in China coastal areas based on Maxent niche model. Ying Yong Sheng Tai Xue Bao = J. Appl. Ecol. 2019, 30, 2329–2337. [Google Scholar]
  26. Wei, X.; Xu, D.; Zhuo, Z. Predicting the impact of climate change on the geographical distribution of leafhopper, Cicadella viridis in China through the MaxEnt model. Insects 2023, 14, 586. [Google Scholar] [CrossRef]
  27. Wang, R.; Li, Q.; He, S.; Liu, Y.; Wang, M.; Jiang, G. Modeling and mapping the current and future distribution of Pseudomonas syringae pv. actinidiae under climate change in China. PLoS ONE 2018, 13, e192153. [Google Scholar]
  28. Zhu, G.; Petersen, M.; Bu, W. Selecting Biological Meaningful Environmental Dimensions of Low Discrepancy among Ranges to Predict Potential Distribution of Bean Plataspid Invasion. PLoS ONE 2012, 7, e46247. [Google Scholar] [CrossRef]
  29. Zhu, G.; Guoqing, L.; Bu, W.; Yubao, G. Ecological niche modeling and its applications in biodiversity conservation. Biodivers. Sci. 2013, 21, 90–98. [Google Scholar]
  30. Rose, C. Pearson A, Field J, Jordan Z. Evidence-based clinical practice in nursing and health care: Assimilating research, experience and expertise. Oxford: Blackwell, 2007. Evid. Based Med. 2007, 12, 156. [Google Scholar] [CrossRef]
  31. Jones, M.C.; Dye, S.R.; Fernandes, J.A.; Frölicher, T.L.; Pinnegar, J.K.; Warren, R.; Cheung, W.W. Predicting the impact of climate change on threatened species in UK waters. PLoS ONE 2013, 8, e54216. [Google Scholar] [CrossRef] [PubMed]
  32. Ab Lah, N.Z.; Yusop, Z.; Hashim, M.; Mohd Salim, J.; Numata, S. Predicting the habitat suitability of Melaleuca cajuputi based on the MaxEnt species distribution model. Forests 2021, 12, 1449. [Google Scholar] [CrossRef]
  33. Yang, X.; Kushwaha, S.; Saran, S.; Xu, J.; Roy, P.S. Maxent modeling for predicting the potential distribution of medicinal plant, Justicia adhatoda L. in Lesser Himalayan foothills. Ecol. Eng. 2013, 51, 83–87. [Google Scholar] [CrossRef]
  34. Abolmaali, S.M.; Tarkesh, M.; Bashari, H. MaxEnt modeling for predicting suitable habitats and identifying the effects of climate change on a threatened species, Daphne mucronata, in central Iran. Ecol. Inform. 2018, 43, 116–123. [Google Scholar] [CrossRef]
  35. Liu, Z.; Peng, Y.; Xu, D.; Zhuo, Z. Meta-Analysis and MaxEnt Model Prediction of the Distribution of Phenacoccus solenopsis Tinsley in China under the Context of Climate Change. Insects 2024, 15, 675. [Google Scholar] [CrossRef]
  36. Gao, X.; Lin, F.; Li, M.; Mei, Y.; Li, Y.; Bai, Y.; He, X.; Zheng, Y. Prediction of the potential distribution of a raspberry (Rubus idaeus) in China based on MaxEnt model. Sci. Rep. 2024, 14, 24438. [Google Scholar] [CrossRef]
  37. Peterson, A.T. Predicting species’ geographic distributions based on ecological niche modeling. The condor 2001, 103, 599–605. [Google Scholar] [CrossRef]
  38. Jing, W.; Qi, G.; Jun, M.A.; Ren, Y.; Rui, W.; McKirdy, S. Predicting the potential geographic distribution of Bactrocera bryoniae and Bactrocera neohumeralis (Diptera: Tephritidae) in China using MaxEnt ecological niche modeling. J. Integr. Agric. 2020, 19, 2072–2082. [Google Scholar]
  39. Elith, J.; Phillips, S.J.; Hastie, T.; Dudík, M.; Chee, Y.E.; Yates, C.J. A statistical explanation of MaxEnt for ecologists. Divers. Distrib. 2011, 17, 43–57. [Google Scholar] [CrossRef]
  40. Warren, D.L.; Glor, R.E.; Turelli, M. ENMTools: A toolbox for comparative studies of environmental niche models. Ecography 2010, 33, 607–611. [Google Scholar] [CrossRef]
  41. Syfert, M.M.; Smith, M.J.; Coomes, D.A. The effects of sampling bias and model complexity on the predictive performance of MaxEnt species distribution models. PLoS ONE 2013, 8, e55158. [Google Scholar] [CrossRef]
  42. Cobos, M.E.; Townsend Peterson, A.; Barve, N.; Osorio-Olvera, L. kuenm: An R package for detailed development of ecological niche models using Maxent. PeerJ 2019, 7, 6281. [Google Scholar] [CrossRef] [PubMed]
  43. Zhu, G.P.; Yuan, X.J.; Fan, J.Y.; Wang, M.L. Effects of model parameters in MaxEnt modeling of ecological niche and geographic distribution: Case study of the brown marmorated stink bug, Halyomorpha haly. J. Biosaf. 2018, 27, 46–51. [Google Scholar]
  44. Zhu, G.; Qiao, H. Effect of the Maxent model’s complexity on the prediction of species potential distributions. Biodivers. Sci. 2016, 24, 1189–1196. [Google Scholar] [CrossRef]
  45. Phillips, S.J.; Anderson, R.P.; Dudík, M.; Schapire, R.E.; Blair, M.E. Opening the black box: An open-source release of Maxent. Ecography 2017, 40, 887–893. [Google Scholar] [CrossRef]
  46. Radosavljevic, A.; Anderson, R.P. Making better Maxent models of species distributions: Complexity, overfitting and evaluation. J. Biogeogr. 2014, 41, 629–643. [Google Scholar] [CrossRef]
  47. Muscarella, R.; Galante, P.J.; Soley Guardia, M.; Boria, R.A.; Kass, J.M.; Uriarte, M.; Anderson, R.P. ENM eval: An R package for conducting spatially independent evaluations and estimating optimal model complexity for Maxent ecological niche models. Methods Ecol. Evol. 2014, 5, 1198–1205. [Google Scholar] [CrossRef]
  48. Sharma, R.; Khan, S.; Kaul, V. Predicting the potential habitat suitability and distribution of “Weed-Onion” (Asphodelus tenuifolius Cavan.) in India under predicted climate change scenarios. J. Agric. Food Res. 2023, 14, 100697. [Google Scholar] [CrossRef]
  49. Cobos, M.E.; Peterson, A.T.; Osorio-Olvera, L.; Jiménez-García, D. An exhaustive analysis of heuristic methods for variable selection in ecological niche modeling and species distribution modeling. Ecol. Inform. 2019, 53, 100983. [Google Scholar] [CrossRef]
  50. Yan, Y.; Zhao, C.; Xie, Y.; Jiang, X. Nature reserves and reforestation expend the potential habitats for endangered plants: A model study in Cangshan, China. J. Nat. Conserv. 2024, 77, 126533. [Google Scholar] [CrossRef]
  51. Araujo, M.B.; Pearson, R.G.; Thuiller, W.; Erhard, M. Validation of species–climate impact models under climate change. Glob. Change Biol. 2005, 11, 1504–1513. [Google Scholar] [CrossRef]
  52. Yilmaz, M.; Erkan Buğday, S. Planiranje šumskih rekreacijskih usluga na razini provincije s višekriterijskim pristupom: Slučaj Turske. Šumarski List 2024, 148, 29–37. [Google Scholar] [CrossRef]
  53. Heng, S.; Li, N.; Yang, Q.; Liang, J.; Liu, X.; Wang, Y. Effects of environment and human activities on rice planting suitability based on MaxEnt model. Int. J. Biometeorol. 2024, 68, 2413–2429. [Google Scholar] [CrossRef] [PubMed]
  54. Brown, J.L. SDM toolbox: A python-based GIS toolkit for landscape genetic, biogeographic and species distribution model analyses. Methods Ecol. Evol. 2014, 5, 694–700. [Google Scholar] [CrossRef]
  55. Holt, R.D. Bringing the Hutchinsonian niche into the 21st century: Ecological and evolutionary perspectives. Proc. Natl. Acad. Sci. USA 2009, 106 (Suppl. S2), 19659–19665. [Google Scholar] [CrossRef]
  56. Guisan, A.; Thuiller, W. Predicting species distribution: Offering more than simple habitat models. Ecol. Lett. 2005, 8, 993–1009. [Google Scholar] [CrossRef]
  57. Guisan, A.; Zimmermann, N.E. Predictive habitat distribution models in ecology. Ecol. Model. 2000, 135, 147–186. [Google Scholar] [CrossRef]
  58. Harsch, M.A.; HilleRisLambers, J. Climate warming and seasonal precipitation change interact to limit species distribution shifts across Western North America. PLoS ONE 2016, 11, e0159184. [Google Scholar] [CrossRef]
  59. Pardini, R.; Nichols, E.; Püttker, T. Biodiversity response to habitat loss and fragmentation. Encycl. Anthr. 2017, 3, 229–239. [Google Scholar]
Insects 15 00901 g001
Figure 1. Distribution record of T. aeacus in China. (The red dots indicate the distribution points of T. aeacus in the current period).
Figure 1. Distribution record of T. aeacus in China. (The red dots indicate the distribution points of T. aeacus in the current period).
Insects 15 00901 g001
Insects 15 00901 g002
Figure 2. ROC curve of the potential distribution of T. aeacus.
Figure 2. ROC curve of the potential distribution of T. aeacus.
Insects 15 00901 g002
Insects 15 00901 g003
Figure 3. Jackknife test of the importance of environmental variables for the T. aeacus.
Figure 3. Jackknife test of the importance of environmental variables for the T. aeacus.
Insects 15 00901 g003
Insects 15 00901 g004
Figure 4. Distribution of suitable areas for the T. aeacus in China under current climatic conditions. (dark blue: high-suitability area; purple: medium-suitability area; blue: low-suitability area; and light blue: unsuitable area. The yellow solid dots represent the distribution points of T. aeacus).
Figure 4. Distribution of suitable areas for the T. aeacus in China under current climatic conditions. (dark blue: high-suitability area; purple: medium-suitability area; blue: low-suitability area; and light blue: unsuitable area. The yellow solid dots represent the distribution points of T. aeacus).
Insects 15 00901 g004
Insects 15 00901 g005
Figure 5. Comparison of potential distribution areas for T. aeacus between 2050 and 2090. (a) SSP1-2.6 (2050s); (b) SSP1-2.6 (2090s); (c) SSP2-4.5 (2050s); (d) SSP2-4.5 (2090s); (e) SSP5-8.5 (2090s); and (f) SSP5-8.5 (2090s).
Figure 5. Comparison of potential distribution areas for T. aeacus between 2050 and 2090. (a) SSP1-2.6 (2050s); (b) SSP1-2.6 (2090s); (c) SSP2-4.5 (2050s); (d) SSP2-4.5 (2090s); (e) SSP5-8.5 (2090s); and (f) SSP5-8.5 (2090s).
Insects 15 00901 g005
Insects 15 00901 g006
Figure 6. The change in the centroid of the potential distribution area of T. aeacus in China.( The black dots represent the center of mass of the current period, the green dots represent the movement of the center of mass of SSP1-2.6, the blue dots represent the movement of the center of mass of SSP2-4.5, and the red dots represent the movement of the center of mass of SSP5-8.5).
Figure 6. The change in the centroid of the potential distribution area of T. aeacus in China.( The black dots represent the center of mass of the current period, the green dots represent the movement of the center of mass of SSP1-2.6, the blue dots represent the movement of the center of mass of SSP2-4.5, and the red dots represent the movement of the center of mass of SSP5-8.5).
Insects 15 00901 g006
Table 1. Environmental variables of potential geographical distribution of T. aeacus.
Table 1. Environmental variables of potential geographical distribution of T. aeacus.
SymbolEnvironmental VariableUnit
Annual mean temperaturebio1°C
Mean diurnal range (monthly mean) (max. temp–min. temp)bio2°C
Isothermality (bio2/bio7) (×100)bio3°C
Temperature seasonality (standard deviation × 100)bio4°C
Max. temperature of the Warmest Monthbio5°C
Min. temperature of the Coldest Monthbio6°C
Temperature annual range (bio5–bio6)bio7°C
Mean temperature of the Wettest Quarterbio8°C
Mean temperature of the Driest Quarterbio9°C
Mean temperature of the Warmest Quarterbio10°C
Mean temperature of the Coldest Quarterbio11°C
Annual precipitationbio12mm
Precipitation of the Wettest Monthbio13mm
Precipitation of the Driest Monthbio14mm
Precipitation seasonality (coefficient of variation)bio15%
Precipitation of the Wettest Quarterbio16mm
Precipitation of the Driest Quarterbio17mm
Precipitation of the Warmest Quarterbio18mm
Precipitation of the Coldest Quarterbio19mm
Altitudeelevm
Slopeslopedegree
Aspectaspectdegree
Table 2. Permutation importance of model variables.
Table 2. Permutation importance of model variables.
CodeEnvironmental VariablesContribution Rate
bio7Temperature annual range (bio5–bio6)26.50%
bio12Annual precipitation20.10%
bio6Min. temperature of the Coldest Month15.00%
slopeSlope10.00%
bio19Precipitation of the Coldest Quarter10.00%
bio10Mean temperature of the Warmest Quarter7.90%
bio14Precipitation of the Driest Month6.70%
elevAltitude3.10%
aspectAspect0.80%
Table 3. Comparison of suitable habitat area for T. aeacus under current and future climate conditions.
Table 3. Comparison of suitable habitat area for T. aeacus under current and future climate conditions.
Predicted Area (104 km2)Comparison with Current Distribution (%)
PeriodScenariosLow-SuitabilityMedium-SuitabilityHigh-SuitabilityLow-SuitabilityMedium-SuitabilityHigh-Suitability
Current-83.73113.0374.2---
2050sSSP1-2.695.53109.4674.220.14−0.030
SSP2-4.586.53110.4389.410.03−0.020.2
SSP5-8.598.1105.14110.417.16−6.9848.79
2090sSSP1-2.696.63107.4184.8215.41−1.8714.28
SSP2-4.5100.2104.42107.7119.67−7.6245.16
SSP5-8.5105.7889.98115.3826.33−20.3955.5
Table 4. Trajectory of centroid shift in the suitable habitat of T. aeacus under climate change scenarios.
Table 4. Trajectory of centroid shift in the suitable habitat of T. aeacus under climate change scenarios.
ScenePeriodAngle (°)DirectionDisplacement (km)
SSP1-2.6Current to 2050s297.3northwest28.01
2050s to 2090s108.92southeast94.99
Current to 2090s105.6southeast67.39
SSP2-4.5Current to 2050s53.62northeast115.89
2050s to 2090s54.27northeast116.58
Current to 2090s53.74northeast232.47
SSP5-8.5Current to 2050s27.52northeast160.04
2050s to 2090s151.58southeast47.5
Current to 2090s43.98northeast139.58
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

Liu, B.; Deng, X.; Liu, Z.; Wei, X.; Zhang, H.; Xu, D.; Zhuo, Z. Predicted Spatial Patterns of Suitable Habitats for Troides aeacus Under Different Climate Scenarios. Insects 2024, 15, 901. https://doi.org/10.3390/insects15110901

AMA Style

Liu B, Deng X, Liu Z, Wei X, Zhang H, Xu D, Zhuo Z. Predicted Spatial Patterns of Suitable Habitats for Troides aeacus Under Different Climate Scenarios. Insects. 2024; 15(11):901. https://doi.org/10.3390/insects15110901

Chicago/Turabian Style

Liu, Biyu, Xinqi Deng, Zhiqian Liu, Xinju Wei, Honghua Zhang, Danping Xu, and Zhihang Zhuo. 2024. "Predicted Spatial Patterns of Suitable Habitats for Troides aeacus Under Different Climate Scenarios" Insects 15, no. 11: 901. https://doi.org/10.3390/insects15110901

APA Style

Liu, B., Deng, X., Liu, Z., Wei, X., Zhang, H., Xu, D., & Zhuo, Z. (2024). Predicted Spatial Patterns of Suitable Habitats for Troides aeacus Under Different Climate Scenarios. Insects, 15(11), 901. https://doi.org/10.3390/insects15110901

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