Next Article in Journal
Research into the Mechanical Tillage Regulation Mechanisms of the Soil Structure in Black Soil Paddy Fields
Previous Article in Journal
Effect of Caragana korshinskii Plantation Succession on Community Stability in Alpine Sandy Regions
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Agriculture
Volume 15
Issue 11
10.3390/agriculture15111144
agriculture-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

Optimized MaxEnt Model Predicts Future Suitable Habitats for Chinese Caterpillar Fungus Under Climate Change

by
Yaqin Peng
,
Zhihang Zhuo
,
Qianqian Qian
and
Danping Xu
*
College of Life Science, China West Normal University, Nanchong 637002, China
*
Author to whom correspondence should be addressed.
Agriculture 2025, 15(11), 1144; https://doi.org/10.3390/agriculture15111144
Submission received: 14 April 2025 / Revised: 16 May 2025 / Accepted: 16 May 2025 / Published: 26 May 2025
(This article belongs to the Topic Sustainable Utilization of Insect Biodiversity in Agriculture Forestry and Food Systems Resource Utilization, Food Security and Policy Integration)
Browse Figures
Versions Notes

Abstract

:
The Chinese Caterpillar Fungus (CCF) is a precious and rare traditional Chinese medicinal material that is extremely sensitive to environmental changes, making wild resources scarce. Therefore, studying the impact of climate change on the potential distribution and changes of the CCF is of great significance. Employing an enhanced MaxEnt approach (optimized with ENMeval), this study determined the primary ecological constraints on CCF and mapped its potential present and future ranges. The results indicated that elevation, bio05, bio04, bio12, bio11, slope, d1_ph_water, and hf were the driving environmental factors influencing the survival of the CCF. The ideal habitat zones for the CCF were mainly distributed in the plateau and alpine climate zones of northwestern and southwestern China, covering an area of 7.42 × 104 km2. Compared with the current climate scenario, the area of suitable habitats for the CCF was expected to increase in the future. In the 2090s, under the SSP1–2.6 scenario, the highly suitable areas for the CCF will have increased the most, by 67.54%, while the low–suitability areas will have decreased by 6.87%. Overall, the highly suitable areas for the CCF will shift towards higher latitudes. The outcomes of this study can inform subsequent conservation strategies for CCF resources and facilitate research on other ecological variables affecting CCF distribution patterns.

1. Introduction

With global warming, it is widely believed that climate change poses the greatest threat to humanity [1]. In fact, the new report by the Intergovernmental Panel on Climate Change (IPCC) indicates that 3.3 billion people worldwide are highly vulnerable to the impacts of climate change, and the suitability of habitats for some species is also severely challenged [2]. According to IPCC projections, global temperatures will rise by 1.8–4 °C by the end of the 21st century [2]. The World Meteorological Organization (WMO) report on the state of the global climate in 2020 indicates that temperatures in 2020 were 1.2 ± 0.1 °C higher compared with the period from 1850 to 1900 [3]. Climate change will alter the suitable distribution areas of certain species, leading to habitat fragmentation and a reduction in global biodiversity [4].
The Chinese Caterpillar Fungus (CCF) is a precious traditional Chinese medicinal resource, known in Chinese as “Dong Chong Xia Cao”, which means “winter worm and summer grass” [5]. The origin of this name is related to its formation process. Ophiocordyceps sinensis (O. sinensis) (Clavicipitaceae) parasitizes the larvae of Hepialidae, consuming the host’s nutrients and destroying its internal structure. By winter, the host transforms into a mummified larva filled with mycelium. In the summer, a club–shaped fruiting body grows from the larva’s head, resembling grass [5,6]. The CCF is mainly distributed in Bhutan, India, Nepal, and China, with China being the main producing region. In China, the CCF is primarily found in the Tibet Autonomous Region, Qinghai Province, Gansu Province, Sichuan Province, and Yunnan Province [7].
The CCF contains a variety of bioactive components, including nucleosides, D–mannitol, sterols, flavonoids, fatty acids, amino acids, vitamins, peptides, amides, proximate, and mineral composition [5]. Among them, peptides, nucleosides, and polysaccharides are the main active components of the CCF, possessing functions such as immune regulation, antioxidation, anti–allergy, anti–tumor, anti–inflammatory, antibacterial, anti–malarial, and antifungal activities [8]. Therefore, the active components of the CCF are an important source for the production of drugs used to treat various diseases. As the pharmacological effects of the CCF become more widely known, the demand for the CCF is increasing. According to surveys, CCF resources greatly drive the economic development of CCF–producing areas. Income from selling the CCF accounts for 60–78% of the annual income of collectors’ families, while the income of non–collectors is 15–55% lower than that of collectors [9]. In recent years, an increasing number of reports indicate a sharp decline in CCF resources. This is not only attributed to human overexploitation but also to the complex formation mechanism of the CCF and its high environmental requirements and narrow geographical distribution. Since 1999, it has been listed as a nationally protected endangered species [8]. According to the latest assessment by China’s Ministry of Ecology and Environment, this species has been designated as vulnerable in the “China Biodiversity Red List—Fungi Volume” [8,10].
Scientific surveys demonstrate that the CCF is principally found in high–elevation areas (>3000 m) of the southeastern Tibetan Plateau and Himalayan ranges, colonizing both alpine meadows and shrub–dominated grasslands [7,8,11,12]. Many of these studies rely on data from several years ago, and, considering the unique habitat of the CCF, there may be some bias in the research results. Therefore, timely assessment and evaluation of the current status of the CCF are of significant importance for its conservation. The present study utilizes species distribution models (SDM) [13] to predict the current and future potential suitable distribution areas for the CCF. SDM is a model based on the quantitative relationship between environmental variables and species occurrence points, assisting in understanding the ecological niche requirements of species [14,15]. SDM is typically used to assess the habitat suitability of a species, determine environmental tolerances, and predict species responses to climate change and other disturbances [16,17]. Species distribution models are also used to predict potential new suitable habitats for finding rare or endangered species. Currently, numerous SDM methods are used to simulate species distributions, including CLIMEX, GARP, MaxEnt, and the mechanical niche model [18,19]. Among them, MaxEnt, a general machine learning method developed by Phillips et al. (2006), is one of the most widely used methods [20]. After using the MaxEnt model to predict the species’ suitable habitats, the results can be visualized using ArcGIS to obtain clear species distribution maps. The MaxEnt model demonstrates reliable predictive capability for species distribution modeling even when sample sizes are limited [21,22], and many studies have already demonstrated its reliability in predicting species distributions. The MaxEnt model has made valuable progress in identifying suitable areas for the conservation of endangered species, as demonstrated in studies on orchid species [23], Dipteronia sinensis [24], Zelkova schneideriana [25], and others.
In recent years, many studies have used the MaxEnt model to predict species distributions [21], but the choice of model parameters (such as feature combinations and regularization multipliers) significantly affects the results. The emergence of the ENMeval package has provided a standardized process for model optimization [20]. Although some research has focused on the distribution of the CCF [8,9,11], its distribution patterns and responses to climate change have not been fully studied. Utilizing a MaxEnt model refined through the ENMeval package with contemporary environmental variables and species distribution data, this study focuses on three core research aims: (1) introducing climatic, soil, and topographic factors to analyze the main environmental drivers affecting the distribution of the CCF; (2) predicting the potential suitable habitat distribution and changing trends of the CCF in China from the present to 2100; and (3) comparing the shifts in the centroid of highly suitable areas for the CCF from the present to 2100. Through this analysis, we aim to better understand the suitable range of the CCF and its responses to climate change, thereby providing a conceptual framework for the conservation, balanced development, and rational utilization of CCF resources.

2. Materials and Methods

2.1. Occurrence Records for Chinese Caterpillar Fungus

The occurrence data of the CCF primarily originated from searches of the literature, with data from the Global Biodiversity Information Facility (GBIF.org; https://doi.org/10.15468/dl.zekau6, accessed on 1 March 2024) serving as a reference [21]. When sampling points lacked accurate latitude and longitude information, Google Maps (http://ditu.google.cn/, accessed on 4 March 2024) was utilized to accurately locate the distribution points of the CCF [26]. A total of 424 distribution points for the CCF were collected. The presence of spatial autocorrelation in species distribution data violates MaxEnt’s independence assumption, potentially biasing the model’s predictive accuracy [27]. Therefore, duplicate data points were retained only once, and ENMTools 1.4 was used for distribution point filtering. The dataset was spatially rarefied by retaining only one occurrence point per 2.5 arc–minute (~4.5 km) grid cell to address potential autocorrelation issues while preserving ecological representativeness [28]. In the end, 350 occurrence data points for the CCF were obtained for analysis (Figure 1). The map data used in ArcMap 10.8 software was sourced from the National Geomatics Center of China (https://ngcc.cn/ngcc/, accessed on 4 March 2024), which includes vector maps of China’s administrative divisions and China’s geographical features.

2.2. Environment Data

The MaxEnt modeling framework incorporated 26 environmental predictors (Table 1), comprising: (a) 19 bioclimatic parameters, (b) 3 topographic features, and (c) 3 soil characteristics, along with (d) the Human Footprint Index and (e) the Human Influence Index as anthropogenic pressure indicators. The 19 bioclimatic variables were sourced from the WorldClim–Global Climate Database (WorldClim 2.1, http://worldclim.org/, accessed on 4 March 2024), including current (1970–2000) and future (2021–2100) climate data at a resolution of 2.5 arc–minutes (Table 1). The BCC–CSM2–MR model demonstrates strong capabilities in modeling climate change in China; hence, it was chosen for future climate projections [29]. The BCC–CSM2–MR model, as published in the Sixth Assessment Report of the Coupled Model Intercomparison Project (CMIP6), includes four Shared Socioeconomic Pathways (SSPs) scenarios: SSP1–2.6 (low greenhouse gas emissions), SSP2–4.5 (medium greenhouse gas emissions), SSP3–7.0 (high greenhouse gas emissions), and SSP5–8.5 (very high greenhouse gas emissions) [30]. In this study, climate data for the 2030s (2021–2040), 2050s (2041–2060), 2070s (2061–2080), and 2090s (2081–2100) under the current and SSP2–4.5 scenarios were selected to build the MaxEnt model. Three topographic datasets were obtained from the National Centers for Environmental Information (NOAANCEI, https://www.ngdc.noaa.gov/, accessed on 4 March 2024), including global elevation data (DEM). The spatial resolution of these datasets is 2.5 arc–minutes [31]. The soil variables were sourced from the World Soil Database (https://gaez.fao.org/pages/hwsd, accessed on 4 March 2024). The human influence index (HF) was obtained from the Center for International Earth Science Information Network (CIESIN) (http://www.ciesin.org/, accessed on 4 March 2024) [32]. The environmental factors were imported into ENMTools v1.4 (https://github.com/danlwarren/ENMTools, version 1.4, accessed on 10 March 2024) for correlation analysis. Environmental variables exhibiting pairwise Pearson’s correlation coefficients exceeding |0.8| were sequentially eliminated to mitigate multicollinearity effects in the modeling framework [33]. In the MaxEnt model, the distribution points of the CCF and the remaining environmental factors were iteratively processed. Environmental factors with a contribution rate greater than 1% were selected. After correlation analysis and variable selection, 13 out of the original 26 environmental variables were retained for final habitat suitability modeling of the CCF (Table 2 and Table 3).

2.3. Optimized Model

The MaxEnt model has sophisticated modeling techniques and is sensitive to sampling biases. Through ENMeval optimization, we quantified model complexity as a function of: (1) FC—defining the permitted relationships (linear, quadratic, product, threshold) between variables, and (2) RM—scaling the penalty term applied to the maximum entropy solution [34]. These two parameters significantly affect the accuracy of the MaxEnt model. Some researchers believe that complex machine learning algorithms can lead to overfitting, resulting in a lack of generality [35]. The default configurations in MaxEnt (feature classes = LQHP, RM = 1) represent generalized settings that may generate suboptimal predictions for specific taxa like the CCF, necessitating parameter customization to avoid systematic biases. Muscarella et al. [36] developed the R package ENMeval for optimizing MaxEnt model parameters. This tool primarily evaluates different Feature Combination (FC) and Regularization Multiplier (RM) settings to identify the optimal model configuration. In our study, we first partitioned all occurrence records into training (75%) and testing (25%) datasets [27]. We then employed the kuenm R package to optimize RM values ranging from 0.1 to 4 (incremented by 0.1), while testing 31 distinct feature combinations, ultimately generating 1240 candidate models [28]. Model selection was based on three rigorous criteria: (1) statistical significance (assessed via ROC analysis with 500 iterations); (2) predictive performance (evaluated by omission rate (OR), with OR < 5% as target); (3) model complexity (using the Akaike information criterion (AIC), retaining only models with a ΔAIC ≤ 2) [4]. The final optimal parameter combination (FC: “pth”, RM: (4)) was implemented in MaxEnt version 3.4.1 for modeling [37].

2.4. MaxEnt Model Development and Performance Assessment

Based on the species occurrence data and environmental data, MaxEnt V 3.4.1 was used to predict potential suitable habitats for the CCF. Of all the occurrence data for the CCF, 75% was randomly selected as training data, while the remaining 25% was selected as testing data [34]. In MaxEnt, prediction results were generated in Logistic format, with the run type set to Subsample, and repeated ten times. The Jackknife method was utilized to evaluate the impact of environmental factors on species occurrence. ROC (Receiver Operating Characteristic) curve analysis was conducted to assess the accuracy of the prediction results [38]. The ROC curve analysis method is highly diagnostic and widely used to evaluate the predictive ability of potential species distribution models. The Area Under the Receiver Operating Characteristic Curve (AUC) quantifies model discrimination capacity, with values spanning 0 (anti–prediction) to 1 (perfect prediction). Following established ecological modeling conventions, we interpreted the AUC scores as: 0.9–1.0 = excellent, 0.8–0.9 = good, 0.7–0.8 = acceptable, 0.6–0.7 = marginal, and <0.6 = inadequate discrimination capacity [39]. In the MaxEnt model, running the model ten times and selecting the average of the AUC values obtained from these ten runs served as the final evaluation result.

2.5. Potentially Suitable Area Partitions

The continuous habitat suitability surface from MaxEnt was post–processed in ArcGIS 10.8, where we applied species–specific thresholds to identify core habitats and then calculated areal statistics for each suitability class. The habitat suitability of the species was quantitatively assessed, ranging from 0 to 1. Combining natural breaks, records from the literature, and the habitat suitability of the CCF, the suitable habitat areas for the CCF were categorized into high habitat suitability, medium habitat suitability, low habitat suitability, and unsuitable habitat [11]. The habitat suitability thresholds for the high–suitability zone range from 0.66 to 1, for the medium–suitability zone range from 0.33 to 0.66, for the low–suitability zone range from 0.05 to 0.33, and for the unsuitable zone range from 0 to 0.05 [25]. Using ArcGIS tools, the area occupied by different suitability zones was calculated, and distribution maps were generated. The “Distribution Changes Between Binary SDMs” tool in SDM_Toolbox_v10.4_9 was employed to compare future suitable habitat distribution with current distribution, producing a change map. The distribution changes were categorized as: gain (areas newly becoming suitable in the future); unsuitable habitat (areas unsuitable in both current and future scenarios); stable (areas remaining suitable in both current and future scenarios); and loss (currently suitable areas that become unsuitable in the future).

2.6. Climate–Driven Range Core Displacement

This study utilized the SDM Toolbox’s advanced geospatial processing capabilities to enhance ecological pattern detection [40]. The SDM Toolbox was employed to compute geometric centroids for both current and projected suitable habitats, followed by vector–based trajectory analysis to quantify spatiotemporal shifts in distribution centers under climate change scenarios [41]. In this study, through the SDM Toolbox, the centroids of present and future highly suitable areas for the CCF were computed. The trajectory of centroid movement for high–suitability areas of the CCF was plotted over time (Figure 2).

3. Result

3.1. Crucial Environmental Factors Influencing the Distribution of the CCF

Figure 3 shows the ROC curve used to estimate the model prediction accuracy, with an average AUC value of 0.946 over repeated runs. The results indicated that the model performed well, demonstrating high prediction accuracy. The Jackknife method examination revealed the relative contribution of different environmental variables to the suitability of species distribution [42]. The Jackknife method was employed to calculate the impact of environmental variables on species distribution suitability under the scenarios “Without variable”, “With only variable”, and “With all variables.” When the blue bar representing “With only variable” was longer, it indicated a greater influence of that environmental variable on species distribution. Figure 4 illustrates the influence of different environmental variables on CCF distribution as determined by the Jackknife method examination. The relative importance of variables based on regularization training gain was determined to be, in decreasing order: elev, bio05, bio04, bio12, bio11, slope, d1_ph_water, and hf.
When the habitat adaptability threshold of the species is greater than 0.5, the environmental factors are suitable for the survival of the CCF. Elevation (elev) is the most important environmental factor influencing the distribution of the CCF. According to the response curve (Figure 5), the adaptability of the CCF initially increases with elevation, reaching its peak at 3000–4292 m, and then decreases as elevation continues to rise. The suitable environmental conditions for the CCF are as follows: max temperature of the warmest month (bio05): 14–20 °C; temperature seasonality (bio04): 580–800; annual precipitation (bio12): 490 mm–990 mm; mean temperature of the coldest quarter (bio11): −12.5 °C to 4 °C; pH (d1_ph_water): 6.8–7.5; and the human footprint and anthropogenic impact index (hf): 5–45.

3.2. Potential Suitable Habitat for the CCF Under Current Climate

A total of 350 occurrences of the CCF were recorded. As shown in Figure 6, the current potential distribution range predicted by the MaxEnt model for the CCF essentially covers its occurrence data. The total area of suitable distribution was 172.48 × 104 km2. The highly suitable habitats for the CCF are predominantly distributed in fragmented patches across the plateau and alpine climatic zones of northwestern and southwestern China, specifically encompassing southeastern Gansu Province, southeastern Qinghai Province, the eastern Tibet Autonomous Region, and limited western areas of Sichuan Province. These optimal habitats are geographically situated between latitude 25° N–39° N and longitude 87° E–105° E, covering a total area of 7.42 × 104 km2. The distribution pattern reflects the strong preference of the CCF for high–altitude ecosystems with specific bioclimatic conditions in these regions. The medium–suitability zone was mainly distributed in a belt–like pattern in the plateau and mountainous climate areas of China, including the eastern to southern edges of Qinghai Province, the central and southwestern parts of the Tibet Autonomous Region, and the northwestern part of Yunnan Province, covering an area of 75.87 × 104 km2. The low–suitability zone was mainly concentrated in the southwestern part of Gansu Province adjacent to Qinghai Province, the central part of Qinghai Province, the central part of the Tibet Autonomous Region, and the northwestern part of Xinjiang, with an area of 89.19 × 104 km2 (Table 4).

3.3. Distribution and Change of Future Potential Suitable Habitat

Using the same aforementioned criteria, the optimized MaxEnt model was employed to predict the potential distribution of the CCF for the 2030s and 2090s under four emission scenarios (SSP1–2.6, SSP2–4.5, SSP3–7.0, and SSP5–8.5). The future suitable habitats were spatially mapped using ArcGIS 10.8. In this study, the areas of four suitability levels for the CCF under the four scenarios in the 2030s and 2090s were calculated (Table 4). The total suitable area for the CCF was defined as the sum of low–, medium–, and high–suitability areas.
Projections indicate an overall increase in the total CCF–suitable area across future scenarios. Under the SSP2–4.5 scenario in the 2090s, the largest expansion (5.96%) in the total suitable area was observed. Meanwhile, under SSP1–2.6 in the 2090s, the high–suitability area increased the most (67.54%), while the low–suitability area decreased by 6.87%. This suggests a future trend of low–suitability areas transitioning into high–suitability zones. In terms of spatial distribution, the future range of the CCF shows minimal deviation from current distributions (Figure 7 and Figure 8). High–suitability areas remain fragmented, primarily concentrated in the southeastern Tibet Autonomous Region, western Sichuan Province, southeastern Qinghai Province (bordering Tibet and Sichuan), southern and southeastern Gansu Province (adjacent to Qinghai and Sichuan) and northwestern Yunnan Province.
A comparison of current and future suitable habitats (Figure 8) reveals that stable suitable areas (persisting in both periods) are mainly located in western Sichuan, eastern Tibet, southeastern Qinghai, and southern Gansu. Newly emerging suitable areas are predominantly distributed in the northwestern regions, while lost habitats are primarily found in northeastern Qinghai and northwestern Yunnan.

3.4. Centroid Displacement of Optimal Habitats Across Multiple Climate Projections

Under current climatic conditions, the distribution center of highly suitable habitats for the CCF is located in Dege County (99.13° E, 31.83° N) within the Ganzi Tibetan Autonomous Prefecture of Sichuan Province (Figure 9). In future scenarios, while the degree of shift varies, the distribution centers of highly suitable habitats consistently migrate northwestward. Under the SSP1–2.6 scenario, the shift is minimal, with the center ultimately relocating to Jiangda County (98.42° E, 31.95° N) in Qamdo City, Tibet Autonomous Region. Across the other three emission scenarios, the centroids of highly suitable habitats generally remain within Jiangda County, Qamdo City. Notably, the most pronounced shift occurs under the SSP5–8.5 scenario, where the distribution center moves approximately 130.9 km to a new location (97.74° E, 32.02° N).

4. Discussion

4.1. Model Accuracy

SDMs are widely used to mathematically model the ecological niche of species by integrating species occurrence data with environmental factors [43]. The MaxEnt model can accurately predict the potential distribution of species with a smaller sample size, handling both continuous and categorical environmental variables within a single model. Accordingly, recent years have witnessed extensive utilization of the MaxEnt model across three key research domains: predicting current and potential future species distributions, evaluating invasion risks of non–native species, and assessing climate change effects on species’ geographical ranges. Species–specific parameter tuning is often necessary as MaxEnt’s preset configurations may lack universal applicability. This study employed parameter optimization of the MaxEnt model through systematic tuning of regularization multipliers and feature class combinations to identify the optimal configuration for CCF distribution modeling. The AIC can be used to measure the information–theoretic fit of the model [44]. A ΔAIC ≤ 2 between the optimized and default models suggests substantial empirical support for the optimized configuration as a better approximating model. Finally, with the regularization multiplier set to 4 and the feature class parameter combination set to PTH, a ΔAIC ≤ 2, the optimal parameters for the MaxEnt model were obtained. The AUC value of this model is 0.946, indicating high prediction accuracy.

4.2. Model Limitations

In the model construction process, we selected the BCC–CSM2–MR climate model, developed by China’s National Climate Center, which has demonstrated strong performance in simulating East Asian climate variability [29]. However, this climate model has certain limitations. First, although the BCC–CSM2–MR performs well in regional climate simulations, the use of a single climate model may introduce biases and irreducible climate projection uncertainties. Subsequent studies will prioritize the use of CMIP6 multi–model ensembles to quantify variations in species distribution outcomes under different emission scenarios. Second, despite implementing bias correction, residual uncertainties in microclimate representation may persist due to the 1 km resolution limitation, leading to inaccuracies in small–scale predictions. Third, non–climatic factors such as interspecific competition and dispersal barriers were not explicitly modeled, which could result in overestimations of future suitable habitats.

4.3. Environmental Variables

The MaxEnt model was built using 26 climatic variables and 3 topographic factors. Pearson’s correlation coefficient was used to filter highly correlated environmental factors, and finally, low–contributing environmental factors were removed, totaling 13 environmental factors. According to the Jackknife test, elev, bio05, bio04, bio12, bio11, slope, d1_ph_water, and hf are the most significant environmental factors influencing CCF distribution. Therefore, the distribution of the CCF is mainly associated with altitude, temperature, and humidity. The CCF is capable of living in habitats with elevations ranging from 3000 m to 4292 m, a maximum temperature of the warmest month (bio05) ranging from 14 °C to 20 °C, and a mean temperature of the coldest quarter (bio11) ranging from −12.5 °C to 4 °C. This is consistent with the climatic characteristics of the current main distribution of the CCF in the plateau mountainous climate zone. Additionally, according to Uttam Babu Shrestha et al.’s [11] prediction, the CCF is capable of surviving in places with an average temperature ranging from −10 °C to 0 °C during the coldest quarter, which is similar to the results obtained in this study. The results of this study indicate that the CCF is suited to environments with a maximum temperature of the warmest month (bio05) ranging from 14 °C to 20 °C and annual precipitation (bio12) ranging from 490 mm to 990 mm. This finding is similar to the prediction by Yujing Yan et al. [7], who suggested that the CCF is adapted to habitats with mean temperatures ranging from 5 °C to 17 °C during the wettest quarter and average precipitation ranging from 200 mm to 600 mm during the warmest quarter. In summary, the CCF is well suited to environments characterized by high altitude, high precipitation, and low temperatures [8].

4.4. Changes in Suitable Habitats

In this study, for the first time, the potential suitable distribution areas for the CCF from 2021 to 2100 under the current and four Shared Socioeconomic Pathways scenarios were predicted using the MaxEnt model. Under the current climate conditions, the CCF is predominantly distributed in the Gansu, Qinghai, Sichuan, and Yunnan provinces, as well as the Tibet Autonomous Region. These areas are primarily located in the plateau and mountainous climate zones, characterized by high altitude, low temperatures, and abundant precipitation, which are conducive to the survival of the CCF. The distribution points of the CCF largely overlap with the predicted potential suitable distribution areas, indicating the accuracy of the model predictions. Our results are in agreement with Yanqiang Wei et al. [8] regarding the current suitable distribution range for the CCF. Our findings show that suitable habitats for the CCF are progressively expanding over time, matching the projected distribution shifts reported by Uttam Babu Shrestha et al. [11] in the Himalayas. The existing literature presents divergent projections regarding future suitable habitats for the CCF, with some studies indicating potential range contractions [8]. These discrepancies likely arise from methodological differences in environmental variable processing and model parameterization. Specifically, while certain studies utilized Principal Component Analysis (PCA) for dimensionality reduction of environmental variables, our approach employed Pearson’s correlation analysis for variable screening, potentially yielding more ecologically relevant predictors. Furthermore, our enhanced predictive accuracy may be attributed to comprehensive optimization of the MaxEnt model parameters, including feature class selection and regularization settings tailored to the species’ ecological characteristics. These methodological refinements likely explain why our projections differ from studies forecasting habitat reduction, as our modeling framework better captures the species’ environmental tolerances under climate change scenarios. The observed variation underscores the importance of model optimization and transparent methodology reporting in species distribution modeling.
Under future climate conditions, the high–suitability range of O. sinensis progressively migrates toward higher latitudes by the 2090s, likely due to global warming. This pattern is similar to the results of MaxEnt models for other species, such as Cynaeus angustus [45], pine wilt disease [46], and Chinese fir [47]. Overall, the suitability zone of O. sinensis shows minimal changes from the present to the future. The formation of the CCF is closely linked with O. sinensis and its host. Studies indicate that the optimal temperature for indoor cultivation of O. sinensis is 15–18 °C, while in the wild, the suitable temperature range for its growth and development is 7–12 °C [48]. Therefore, O. sinensis can survive at higher temperatures when the surrounding habitat is suitable. The activity, growth, and population size of the host are related to its feeding behavior. Studies suggest that the suitability zones of forests and shrubs distributed in highlands may increase in response to future climate warming, while the suitability zones of alpine meadows and grasslands may undergo less change [49].

4.5. Challenges

Based on the current CCF resources, there has been a sharp decline in CCF resources in recent years [50]. The microbial community in the environment and human exploitation are also important factors affecting the abundance of the CCF. Many studies have indicated that various other microorganisms exist in the natural environment where the CCF is formed, and these microorganisms may have either promoting or inhibitory effects on the formation of the CCF [51]. The interaction between the O. sinensis fungus and other microorganisms parasitizing within the larvae of the bat moth may not only affect the growth and physiological activities of the host larvae but could also be one of the key mechanisms underlying the occurrence of O. sinensis [52]. Human exploitation directly leads to a reduction in CCF resources. The fragile and sensitive ecosystems in high–altitude areas are susceptible to disturbance from human entry and exploitation activities, which can disrupt the original ecological environment. Therefore, the predicted suitable habitats may not perfectly encompass all possible occurrences in reality. To protect CCF resources, measures such as limiting exploitation and establishing natural reserves can be implemented to minimize the risk of loss.

5. Conclusions

This study presents the first application of an optimized MaxEnt model integrating climate, soil, and topographic factors to predict the current and future (2030s and 2090s under four scenarios) distribution patterns of the CCF in China. The results identify elevation, temperature, and humidity as the key environmental determinants, with the SSP1–2.6 scenario in the 2090s showing the greatest expansion of high–suitability areas (67.54% increase) alongside a 6.87% reduction in low–suitability zones. The high–suitability areas exhibit a gradual poleward migration while remaining predominantly fragmented across plateau and alpine regions in northwestern and southwestern China, suggesting these areas should be prioritized for establishing nature reserves or protected zones to conserve CCF germplasm resources. The findings not only validate MaxEnt’s effectiveness in predicting distributions of endangered species in China but also highlight the need for future research on microbial communities, anthropogenic factors, soil environments, and vegetation cover influencing CCF development.

Author Contributions

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

Funding

Funding for this research was provided by China West Normal University’s Fundamental Research Funds (20A007, 20E051, 21E040 and 22kA011).

Data Availability Statement

The data supporting the results are available in a public repository at: GBIF.org (1 March 2024) GBIF Occurrence Download https://doi.org/10.15468/dl.rsuphw, accessed on 1 March 2024.

Acknowledgments

The authors wish to thank China West Normal University for their financial support for this study, Zhihang Zhuo and Danping Xu for their guidance and supervision, and Qianqian Qian for his help.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Tiedje, J.M.; Bruns, M.A.; Casadevall, A.; Criddle, C.S.; Eloe-Fadrosh, E.; Karl, D.M.; Nguyen, N.K.; Zhou, J. Microbes and Climate Change: A Research Prospectus for the Future. mBio 2022, 13, e00800-22. [Google Scholar] [CrossRef]
  2. IPCC. Climate Change 2022: Impacts, Adaptation and Vulnerability; Cambridge University Press: Cambridge, UK, 2022. [Google Scholar]
  3. WMO. WMO Report on the State of the Global Climate 2020; WTO: Geneva, Switzerland, 2021; p. 38. [Google Scholar]
  4. Shi, X.D.; Wang, J.W.; Zhang, L.; Chen, S.X.; Zhao, A.L.; Ning, X.D.; Fan, G.R.; Wu, N.S.; Zhang, L.; Wang, Z.D. Prediction of the potentially suitable areas of Litsea cubeba in China based on future climate change using the optimized MaxEnt model. Ecol. Indic. 2023, 148, 110093. [Google Scholar] [CrossRef]
  5. Xu, C.H.; Wu, F.F.; Zou, Z.C.; Mao, L.Y.; Lin, S. Discovery of the chemical constituents, structural characteristics, and pharmacological functions of Chinese caterpillar fungus. Open Chem. 2023, 21, 20220337. [Google Scholar] [CrossRef]
  6. Wang, Z.; Li, M.; Ju, W.; Ju, W.B.; Ye, W.Q.; Xue, L.H.; Boufford, D.E.; Gao, X.F.; Yue, B.S.; Liu, Y.; et al. Corrigendum to “The entomophagous caterpillar fungus Ophiocordyceps sinensis is consumed by its lepidopteran host as a plant endophyte”. Fungal Ecol. 2021, 50, 101033. [Google Scholar] [CrossRef]
  7. Yan, Y.J.; Li, Y.; Wang, W.J.; He, J.S.; Yang, R.H.; Wu, H.J.; Wang, X.L.; Jiao, L.; Tang, Z.Y.; Yao, Y.J. Range shifts in response to climate, change of Ophiocordyceps sinensis, a fungus endemic to the Tibetan Plateau. Biol. Conserv. 2017, 206, 143–150. [Google Scholar] [CrossRef]
  8. Wei, Y.Q.; Zhang, L.; Wang, J.N.; Wang, W.W.; Niyati, N.; Guo, Y.L.; Wang, X.F. Chinese caterpillar fungus (Ophiocordyceps sinensis) in China: Current distribution, trading, and futures under climate change and overexploitation. Sci. Total Environ. 2020, 755, 142548. [Google Scholar] [CrossRef]
  9. Laha, A.; Badola, R.; Hussain, S.A. Earning a Livelihood from Himalayan Caterpillar Fungus in Kumaon Himalaya: Opportunities, Uncertainties, and Implications. Mt. Res. Dev. 2018, 38, 276–4741. [Google Scholar] [CrossRef]
  10. MEE. Redlist of China’s Biodiversity—Macrofungi; Ministry of Ecology and Environment, PRC: Beijing, China; Chinese Academy of Sciences: Beijing, China, 2018; p. 85. [Google Scholar]
  11. Shrestha, U.B.; Bawa, K.S. Impact of climate change on potential distribution of Chinese caterpillar fungus (Ophiocordyceps sinensis) in Nepal Himalaya. PLoS ONE 2014, 9, e106405. [Google Scholar] [CrossRef]
  12. Zhou, K.S.; Zhang, J.C.; Huang, X.Q.; Lei, Y.; Luo, Z.; Shi, L.; Wei, K. Suitability and regionalization of Ophiocordyceps sinensis in the Tibetan Plateau. Acta Ecol. Sin. 2018, 38, 2768–2779. [Google Scholar]
  13. Peterson, A.T. Uses and Requirements of Ecological Niche Models and Related Distributional Models. Biodivers. Inform. 2006, 3, 59–72. [Google Scholar] [CrossRef]
  14. Fernandes, R.F.; Scherrer, D.; Guisan, A. Effects of simulated observation errors on the performance of species distribution models. Divers. Distrib. 2019, 25, 400–413. [Google Scholar] [CrossRef]
  15. Hirzel, A.H.; Le Lay, G.; Helfer, V.; Randin, C.; Guisan, A. Evaluating the ability of habitat suitability models to predict species presences. Ecol. Model. 2006, 199, 142–152. [Google Scholar] [CrossRef]
  16. Peterson, A.T. Predicting the Geography of Species’ Invasions via Ecological Niche Modeling. Q. Rev. Biol. 2003, 78, 419–433. [Google Scholar] [CrossRef]
  17. Elith, J.; Leathwick, J.R. Species Distribution Models: Ecological Explanation and Prediction Across Space and Time. Annu. Rev. Ecol. Evol. Syst. 2009, 40, 677–697. [Google Scholar] [CrossRef]
  18. 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]
  19. Hou, Z.C.; Sun, Z.; Du, G.L.; Shao, D.K.; Zhong, Q.; Yang, S.P. Assessment of suitable cultivation region for Pepino (Solanum muricatum) under different climatic conditions using the MaxEnt model and adaptability in the Qinghai–Tibet plateau. Heliyon 2023, 9, e18974. [Google Scholar] [CrossRef]
  20. 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]
  21. Liu, C.Q.; Yang, M.; Li, M.; Jin, Z.N.; Yang, N.W.; Yu, H.; Liu, W.X.; Dai, S.P.; Zhao, Z.Z. Climate Change Facilitates the Potentially Suitable Habitats of the Invasive Crop Insect Ectomyelois ceratoniae (Zeller). Atmosphere 2024, 15, 119. [Google Scholar] [CrossRef]
  22. Ballardin, S.; Belladona, R.; Tiago, D.V.; Volnei, D.B.; Anderson, B.T.; Roehe, R.P.A. MaxEnt machine learning model predicts high groundwater potential areas in a fractured volcanic aquifer system. J. S. Am. Earth Sci. 2024, 135, 104794. [Google Scholar] [CrossRef]
  23. Wan, J.Z.; Wang, C.J.; Han, S.J. Planning the priority protected areas of endangered orchid species in northeastern China. Biodivers. Conserv. 2014, 23, 1395–1409. [Google Scholar] [CrossRef]
  24. Huang, Y.; Zeng, Y.; Jiang, P.; Chen, H.; Yang, J.T. Prediction of Potential Geographic Distribution of Endangered Relict Tree Species Dipteronia sinensis in China Based on MaxEnt and GIS. Pol. J. Environ. Stud. 2022, 31, 3597–3609. [Google Scholar] [CrossRef]
  25. Zhou, Y.R.; Lu, X.; Zhang, G.F. Potentially differential impacts on niche overlap between Chinese endangered Zelkova schneideriana and its associated tree species under climate change. Front. Ecol. Evol. 2023, 11, 1218149. [Google Scholar] [CrossRef]
  26. Zhang, X.M.; Zhou, Y.T.; Xu, T.; Zong, S.X. Impact of Climate Change on the Habitat Suitability of Monochamus saltuarius Gebler (Coleoptera; Cerambycidae) and Its Natural Enemies in China. Forests 2024, 15, 33. [Google Scholar] [CrossRef]
  27. Wang, S.Y.; Yao, W.T.; Ma, Y.; Shang, E.; Zhang, S.Y.; Chen, F.; Zeng, Y. Optimizing natural boundary definition and functional zoning in protected areas: An integrated framework encompassing species, landscapes and ecosystems. Glob. Ecol. Conserv. 2024, 49, e02781. [Google Scholar] [CrossRef]
  28. Li, D.; Li, B.; Hou, X.Y.; Wang, X.L.; Li, X.W.; Zhang, Y.X. Habitat suitability assessment for Saunders’s Gull (Saundersilarus saundersi) in the Yellow River Delta, China. Ecol. Inform. 2024, 79, 102393. [Google Scholar] [CrossRef]
  29. Gao, R.H.; Liu, L.; Zhao, L.J.; Cui, S.P. Potentially Suitable Geographical Area for Monochamus alternatus under Current and Future Climatic Scenarios Based on Optimized MaxEnt Model. Insects 2023, 14, 182. [Google Scholar] [CrossRef] [PubMed]
  30. Xia, Y.L.; Muhammad, K.; Nabeel, N.M.; Yang, Y.X.; Li, Q.; Li, T.; Xu, S.L.; Wang, Y.H.; Fan, X.C.; Zhao, J.P.; et al. Suitability changes of Citrus medica L. var. sarcodactylis Swingle, a medicine–food plants affected by climate warming using the optimized MaxEnt model. PLoS ONE 2023, 18, e282659. [Google Scholar] [CrossRef]
  31. Li, Y.X.; Shao, W.H.; Huang, S.Q.; Zhang, Y.Z.; Fang, H.F.; Jiang, J.M. Prediction of Suitable Habitats for Sapindus delavayi Based on the MaxEnt Model. Forests 2022, 13, 1611. [Google Scholar] [CrossRef]
  32. Li, J.J.; Fan, G.; He, Y. Predicting the current and future distribution of three Coptis herbs in China under climate change conditions, using the MaxEnt model and chemical analysis. Sci. Total Environ. 2020, 698, 134141. [Google Scholar] [CrossRef]
  33. West, A.M.; Kumar, S.; Brown, C.S.; Stohlgren, T.J.; Bromberg, J. Field validation of an invasive species Maxent model. Ecol. Inform. 2016, 36, 126–134. [Google Scholar] [CrossRef]
  34. Yan, H.Y.; He, J.; Xu, X.C.; Yao, X.Y.; Wang, G.Y.; Tang, L.G.; Feng, L.; Zou, L.M.; Gu, X.L.; Qu, Y.F.; et al. Prediction of Potentially Suitable Distributions of Codonopsis pilosula in China Based on an Optimized MaxEnt Model. Front. Ecol. Evol. 2021, 9, 773396. [Google Scholar] [CrossRef]
  35. Heikkinen, R.K.; Marmion, M.; Luoto, M. Does the interpolation accuracy of species distribution models come at the expense of transferability? Ecography 2012, 35, 276–288. [Google Scholar] [CrossRef]
  36. Muscarella, R.; Galante, P.J.; Mariano, S.G.; Bpria, R.A.; Jamie, K.M.; Maria, U.; Anderson, R.P. ENMeval: 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]
  37. Fang, B.; Zhao, Q.; Qin, Q.L.; Yu, J. Prediction of Potentially Suitable Distribution Areas for Prunus tomentosa in China Based on an Optimized MaxEnt Model. Forests 2022, 13, 381. [Google Scholar] [CrossRef]
  38. Gao, C.; Guo, S.L.; Ma, C.L.; Yang, J.X.; Kang, X.L.; Li, R. Impact of Climate Change on the Potential Geographical Distribution Patterns of Luculia pinceana Hook. f. since the Last Glacial Maximum. Forests 2024, 15, 253. [Google Scholar] [CrossRef]
  39. Yan, H.Y.; Feng, L.; Zhao, Y.F.; Li, F.; Wu, D.; Zhu, C.P. Prediction of the spatial distribution of Alternanthera philoxeroides in China based on ArcGIS and MaxEnt. Glob. Ecol. Conserv. 2020, 21, e856. [Google Scholar] [CrossRef]
  40. Brown, J.L. SDM toolbox: A python–basedGIS toolkit for landscape genetic, biogeographic and species distribution model analyses. Methods Ecol. Evol. 2014, 5, 694–700. [Google Scholar] [CrossRef]
  41. Brown, J.L.; Bennett, J.R.; French, C.M. SDMtoolbox 2.0: The next generation Python–based GIS toolkit for landscape genetic, biogeographic and species distribution model analyses. Peerj 2017, 5, e4095. [Google Scholar] [CrossRef] [PubMed]
  42. Chen, C.H.; Longzhu, D.J.; Lu, X.W.; Song, Z.C.; Miao, Q.; Sun, F.H.; Suo, N.J. Habitat suitability of Corydalis based on the optimized MaxEnt model in China. Acta Ecol. Sin. 2023, 43, 10345–10362. [Google Scholar]
  43. Guo, Y.L.; Zhao, Z.F.; Qiao, H.J.; Wang, R.; Wei, H.Y.; Wang, L.K.; Gu, W.; Li, X. Challenges and Development Trend of Species Distribution Model. Adv. Earth Sci. 2020, 35, 1292–1305. [Google Scholar] [CrossRef]
  44. Zhu, Y.H.; Chang, G.B.; Yang, M.S. Application of Akaike information criterion in selecting random error model for inertial measurement unit. J. Chin. Inert. Technol. 2024, 32, 180–186. [Google Scholar]
  45. Zhao, C.; Bai, C.Q.; Wang, D.X. Predicting current and future potential distribution of Cynaeus angustus (Coleoptera: Tenebrionidae) in global scale using the MaxEnt model. J. Stored Prod. Res. 2023, 101, 102089. [Google Scholar] [CrossRef]
  46. Wang, W.; Zhu, Q.; He, G.; Liu, X.Y.; Peng, W.T.; Cai, Y.L. Impacts of climate change on pine wilt disease outbreaks and associated carbon stock losses. Agr. Forest Meteorol. 2023, 334, 109426. [Google Scholar] [CrossRef]
  47. Zhao, Y.; Deng, X.W.; Xiang, W.H.; Chen, L.; Ouyang, S. Predicting potential suitable habitats of Chinese fir under current and future climatic scenarios based on Maxent model. Ecol. Inform. 2021, 64, 101393. [Google Scholar] [CrossRef]
  48. Lei, Z.; Shui, X.R. Ecological Response of Ophiocordyceps sinensis in Its Occurrence and Development. Hubei Agric. Sci. 2015, 54, 100–104. [Google Scholar] [CrossRef]
  49. Gao, Q.Z.; Guo, Y.Q.; Xu, H.M.; Ganjurjav, H.; Li, Y.; Wang, Y.F.; Qin, X.B.; Ma, X.; Liu, S. Climate change and its impacts on vegetation distribution and net primary productivity of the alpine ecosystem in the Qinghai–Tibetan Plateau. Sci. Total Environ. 2016, 554–555, 34–41. [Google Scholar] [CrossRef]
  50. Hopping, K.A.; Chignell, S.M.; Lambin, E.F. The demise of caterpillar fungus in the Himalayan region due to climate change and overharvesting. Proc. Natl. Acad. Sci. USA 2018, 115, 11489–11494. [Google Scholar] [CrossRef] [PubMed]
  51. Li, Y.L.; Yao, Y.S.; Zhang, Z.H.; Liu, X.; Xu, H.F.; Ma, S.L.; Wu, Z.W.; Zhu, J.S. Synergy of Fungal Complexes Isolated from the Intestines of Hepialus lagii Larvae in Increasing Infection Potency. J. Fungal Res. 2016, 14, 96–112. [Google Scholar] [CrossRef]
  52. Xu, M.; Xu, M.; Li, R.Q. Progress of several crucial aspects in the biological and ecological research on the Chinese caterpillar fungus, Ophiocordyceps sinensis. Acta Ecol. Sin. 2019, 39, 1853–1862. [Google Scholar] [CrossRef]
Agriculture 15 01144 g001
Figure 1. Geographical distribution points of Chinese Caterpillar Fungus in China.
Figure 1. Geographical distribution points of Chinese Caterpillar Fungus in China.
Agriculture 15 01144 g001
Agriculture 15 01144 g002
Figure 2. Flowchart of the CCF MaxEnt Model Construction Process.
Figure 2. Flowchart of the CCF MaxEnt Model Construction Process.
Agriculture 15 01144 g002
Agriculture 15 01144 g003
Figure 3. ROC curve of the CCF and the AUC values for the optimized MaxEnt model.
Figure 3. ROC curve of the CCF and the AUC values for the optimized MaxEnt model.
Agriculture 15 01144 g003
Agriculture 15 01144 g004
Figure 4. Jackknife test of the CCF for the environmental variables.
Figure 4. Jackknife test of the CCF for the environmental variables.
Agriculture 15 01144 g004
Agriculture 15 01144 g005
Figure 5. Single response curves of dominant environmental factors of the CCF (Blue line: variance; Red line: mean value).
Figure 5. Single response curves of dominant environmental factors of the CCF (Blue line: variance; Red line: mean value).
Agriculture 15 01144 g005
Agriculture 15 01144 g006
Figure 6. Distribution of suitable habitats for the CCF under the current climate.
Figure 6. Distribution of suitable habitats for the CCF under the current climate.
Agriculture 15 01144 g006
Agriculture 15 01144 g007
Figure 7. Potential distribution of O. sinensis based on different future climate scenarios. (a) 2030 s–SSP1–2.6; (b) 2030 s–SSP2–4.5; (c) 2030 s–SSP3–7.0; (d) 2030 s–SSP5–8.5; (e) 2090 s–SSP1–2.6; (f) 2090 s–SSP2–4.5; (g) 2090 s–SSP3–7.0; and (h) 2090 s–SSP5–8.5.
Figure 7. Potential distribution of O. sinensis based on different future climate scenarios. (a) 2030 s–SSP1–2.6; (b) 2030 s–SSP2–4.5; (c) 2030 s–SSP3–7.0; (d) 2030 s–SSP5–8.5; (e) 2090 s–SSP1–2.6; (f) 2090 s–SSP2–4.5; (g) 2090 s–SSP3–7.0; and (h) 2090 s–SSP5–8.5.
Agriculture 15 01144 g007
Agriculture 15 01144 g008
Figure 8. Potential distribution map for the CCF under the future climate scenario. (a) Comparison of 2030s SSP1–2.6 and current high–suitability area size. (b) Comparison of 2030s SSP2–4.5 and current high–suitability area size. (c) Comparison of 2030s SSP3–7.0 and current high–suitability area size. (d) Comparison of 2030s SSP5–8.5 and current high–suitability area size. (e) Comparison of 2090s SSP1–2.6 and current high–suitability area size. (f) Comparison of 2090s SSP2–4.5 and current high–suitability area size. (g) Comparison of 2090s SSP3–7.0 and current high–suitability area size. (h) Comparison of 2090s SSP5–8.5 and current high–suitability area size.
Figure 8. Potential distribution map for the CCF under the future climate scenario. (a) Comparison of 2030s SSP1–2.6 and current high–suitability area size. (b) Comparison of 2030s SSP2–4.5 and current high–suitability area size. (c) Comparison of 2030s SSP3–7.0 and current high–suitability area size. (d) Comparison of 2030s SSP5–8.5 and current high–suitability area size. (e) Comparison of 2090s SSP1–2.6 and current high–suitability area size. (f) Comparison of 2090s SSP2–4.5 and current high–suitability area size. (g) Comparison of 2090s SSP3–7.0 and current high–suitability area size. (h) Comparison of 2090s SSP5–8.5 and current high–suitability area size.
Agriculture 15 01144 g008
Agriculture 15 01144 g009
Figure 9. Shift in the centroids of highly suitable habitats for the CCF under future climate scenarios. (a) Distribution of centroids on the complete map of China; (b) Trajectory of centroid movement.
Figure 9. Shift in the centroids of highly suitable habitats for the CCF under future climate scenarios. (a) Distribution of centroids on the complete map of China; (b) Trajectory of centroid movement.
Agriculture 15 01144 g009
Table 1. Environmental variables related to the distributions.
Table 1. Environmental variables related to the distributions.
AbbreviationClimate VariablesUnit
Bio01Annual mean temperature°C
Bio02Mean diurnal range°C
Bio03Isothermality (bio02/bio07) (×100)
Bio04Temperature seasonality (standard deviation × 100)
Bio05Max temperature of warmest month°C
Bio06Min temperature of coldest month°C
Bio07Temperature annual range (bio05–bio06)°C
Bio08Mean temperature of wettest quarter°C
Bio09Mean temperature of driest quarter°C
Bio10Mean temperature of warmest quarter°C
Bio11Mean temperature of coldest quarter°C
Bio12Annual precipitationmm
Bio13Precipitation of wettest monthmm
Bio14Precipitation of driest monthmm
Bio15Precipitation seasonality (Coefficient of variation)
Bio16Precipitation of wettest quartermm
Bio17Precipitation of driest quartermm
Bio18Precipitation of warmest quartermm
Bio19Precipitation of coldest quartermm
ElevAltitude (elevation above sea level) (m)m
SlopeSlope°
AspectAspectrad
d1_ph_waterpH (chemistry)mol/L
d1_swrSoil moisture statusθg
d1_usdaClassification of soil texture–
hfHuman footprint and anthropogenic impact index–
Table 2. Environmental variables used in this study.
Table 2. Environmental variables used in this study.
AbbreviationClimate VariablesUnit
AspectAspectrad
Bio04Temperature seasonality (standard deviation × 100)
Bio05Max temperature of warmest month°C
Bio11Mean temperature of coldest quarter°C
Bio12Annual precipitationmm
Bio14Precipitation of driest monthmm
Bio15Precipitation seasonality (Coefficient of variation)
d1_ph_waterpH (chemistry)mol/L
d1_swrSoil moisture statusθg
d1_usdaClassification of soil texture–
hfHuman footprint and anthropogenic impact index–
ElevAltitude (elevation above sea level) (m)m
SlopeSlope°
Table 3. Percentage contribution of 6 environment variables of the Chinese Caterpillar Fungus (CCF).
Table 3. Percentage contribution of 6 environment variables of the Chinese Caterpillar Fungus (CCF).
VariableDescriptionPercent
Contribution (%)		Permutation
Importance
AspectAspect00
Bio04Temperature seasonality (standard deviation × 100)13.817.5
Bio05Max temperature of warmest month40.95.4
Bio11Mean temperature of coldest quarter5.55.1
Bio12Annual precipitation13.81.7
Bio14Precipitation of driest month0.51.1
Bio15Precipitation seasonality (Coefficient of variation)0.20.2
d1_ph_waterpH (chemistry)0.10
d1_swrSoil moisture status00
d1_usdaClassification of soil texture21.3
hfHuman footprint and anthropogenic impact index1.83
ElevAltitude (elevation above sea level) (m)18.961.2
SlopeSlope2.63.4
Table 4. Areas of suitable habitats for the CCF under future climate scenarios.
Table 4. Areas of suitable habitats for the CCF under future climate scenarios.
Decade ScenariosPredicted Area (×104 km2)Comparison with Current Distribution (%)
Low Habitat SuitabilityMedium Habitat SuitabilityHigh Habitat SuitabilityTotal AreaLow Habitat SuitabilityMedium Habitat SuitabilityHigh Habitat SuitabilityTotal Area
Current89.1975.877.42172.48
2030s–SSP1–2.682.7781.3011.03175.10−7.207.1548.771.52
2030s–SSP2–4.582.1679.8511.26173.26−7.895.2451.760.45
2030s–SSP3–7.083.6282.8011.02177.44−6.249.1348.602.88
2030s–SSP5–8.584.4584.4411.02179.91−5.3111.2948.534.31
2090s–SSP1–2.683.0683.5212.43179.00−6.8710.0767.543.78
2090s–SSP2–4.583.0687.8111.89182.76−6.8715.7360.365.96
2090s–SSP3–7.083.9486.779.29180.00−5.8914.3625.294.36
2090s–SSP5–8.587.3886.008.13181.51−2.0313.359.595.23
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

Peng, Y.; Zhuo, Z.; Qian, Q.; Xu, D. Optimized MaxEnt Model Predicts Future Suitable Habitats for Chinese Caterpillar Fungus Under Climate Change. Agriculture 2025, 15, 1144. https://doi.org/10.3390/agriculture15111144

AMA Style

Peng Y, Zhuo Z, Qian Q, Xu D. Optimized MaxEnt Model Predicts Future Suitable Habitats for Chinese Caterpillar Fungus Under Climate Change. Agriculture. 2025; 15(11):1144. https://doi.org/10.3390/agriculture15111144

Chicago/Turabian Style

Peng, Yaqin, Zhihang Zhuo, Qianqian Qian, and Danping Xu. 2025. "Optimized MaxEnt Model Predicts Future Suitable Habitats for Chinese Caterpillar Fungus Under Climate Change" Agriculture 15, no. 11: 1144. https://doi.org/10.3390/agriculture15111144

APA Style

Peng, Y., Zhuo, Z., Qian, Q., & Xu, D. (2025). Optimized MaxEnt Model Predicts Future Suitable Habitats for Chinese Caterpillar Fungus Under Climate Change. Agriculture, 15(11), 1144. https://doi.org/10.3390/agriculture15111144

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