Jan 21, 2025

Abstract Ormosia microphylla is a nationally prioritized wild plant in China but effects of likely future climate change have been poorly studied. Here distribution data of O. microphylla and environmental data with an optimized MaxEnt maximum entropy model were used to predict potentially suitable areas under current and future climate scenarios. The results showed that with future climate warming, the total area suitable for O. microphylla might gradually increase. In the three future periods (2030s (2021–2040), 2050s (2041–2060) and 2090s (2081–2100)), the medium and high suitable areas under different climate scenarios generally showed an expanding trend, while the low suitable areas mostly showed a decreasing trend. At the same time, the potential suitable areas for O. microphylla in China have shown a certain degree of migration trend towards higher latitudes in the north and northwest, as well as a trend towards higher altitudes. The research results will provide data support for the protection of germplasm resources and the development of artificial cultivation techniques for O. microphylla, and provide a theoretical basis for the protection of other rare and endangered plants. Introduction Ormosia microphylla, a nationally protected wild plant in China at Grade I. It is mainly found in evergreen broadleaf forests and deciduous broadleaf mixed forests with abundant precipitation and humid climate. Due to long term over-harvesting, the O. microphylla has become an endangered species. Currently, research on O. microphylla mainly focuses on resource distribution and germplasm resource protection, with relatively little research on suitable distribution areas and climate change response. The only studies have considered climate related factors 1 , 2 , but the growth of a species is closely related to factors such as soil, terrain, and human activity 3 in addition to climate factors. Additionally, species diversity is an important indicator of the variety of species within a community or region, and informs environmental conservation actions 4 , 5 . Climate change has had a profound impact on biodiversity 6 , mainly through changes in temperature and precipitation reducing or even extinguishing suitable habitats for plants. In China, significant shifts in climate patterns have been observed over recent decades. The average temperature in China has risen at a rate higher than the global average, leading to alterations in species distributions and phenological events. For instance, northern regions are experiencing more rapid warming, causing some plant species to shift northward. Precipitation patterns have also changed markedly, with increased frequency of extreme weather events such as heavy rainfall and droughts. Some regions, like the Yangtze River basin, have seen increased precipitation, while northern and western areas face reduced rainfall and prolonged drought conditions. These changes affect plant growth, reproduction, and the suitability of habitats. Some scholars have conducted relevant research. For example, Asase and Peterson studied the possible impacts of future climate change processes on the suitable geographic regions of the important medicinal plant resource, Alstonia boonei De Wild. They found that the availability of habitat for this species was expected to be lower in the northern and southern extreme regions where its geographic distribution was known, while it was higher in the eastern half 7 . Predicting the impact of climate change on the suitable distribution areas of rare and endangered species and their future trends is of great reference value for formulating in-situ and ex-situ conservation strategies, and it aids in the more effective protection of rare and endangered plants 8 , 9 . There are various species distribution models (SDMs) used to predict species distributions, such as BIOCLIM, GARP, and MaxEnt. Among them, the Maximum Entropy Model (MaxEnt) is a representative ecological niche model 10 , 11 , 12 . Its core idea is to use existing sample data to predict unknown distribution status 13 , 14 . The model is able to synthesize information on the latitude and longitude of the species’ current presence as well as a variety of factors such as climatic, geographic, and biological characteristics to make predictions of species distribution 15 . The MaxEnt model is robust even with small sample sizes due to its ability to handle presence-only data and its high predictive accuracy. Therefore, numerous researchers have chosen to use the Maxent model to simulate the potential fitness zones of species 16 , 17 . In particular, for endangered and rare plants with fewer records and complex growing environments, Maxent model has been applied due to its excellent applicability 18 . Therefore, the main objectives of this article are: (1) to comprehensively consider relevant environmental factors, use optimized parameters to construct a Maxent model, and predict the suitable distribution of O. microphylla in China under current and future climate scenarios. (2) to analyze the area changes in the suitable regions and the trend of centroid migration of O. microphylla under different climate scenarios in different periods; (3) to identify the main environmental factors that affect the distribution of suitable growth areas for O. microphylla. This study will contribute to the development of more effective protection strategies for O. microphylla. Materials and methods Distribution data of Ormosia microphylla The distribution data of O. microphylla are mainly sourced from the Chinese Virtual Herbarium ( http://www.cvh.ac.cn/ ), Global Biodiversity Information Facility ( https://www.gbif.org/ ), field investigations, and the published literature. For some distribution data with specific locations but without latitude and longitude, the corresponding coordinates were obtained through the Baidu Coordinate Picker System ( https://api.map.baidu.com/lbsapi/getpoint/ ). Specifically, we input the place names or detailed addresses into the Baidu Coordinate Picker, which displays the location on the map. By clicking on the exact location on the map, the system provides the precise latitude and longitude coordinates. This method ensures accuracy in geocoding the distribution points that lack coordinate information. To reduce the errors caused by clustering effects during modeling, a series of duplicate and incorrect samples were removed using the ENMeval package, resulting in a dataset of 38 distribution sites of O. microphylla for modeling (e.g. Fig. 1 ). Fig. 1 Record of distribution points of O. microphylla. The map was prepared by Pan Huang, Zhiwei Gong and Yingfang Zhu in ArcGIS 10.6 ( https://www.esri.com/en-us/arcgis/geospatial-platform/overview ). Environmental factors This paper mainly adopted four environmental factors, including climate factors, soil factors, terrain factors and human activity intensity factors. Climate variables and terrain data, especially altitude information, were sourced from WorldClim ( http://www.worldclim.org ), with a unified data resolution of 2.5 arc-minutes. To ensure the accuracy of the model, we analyzed the distances between distribution points and found that the minimum distance between any two distribution points exceeds the environmental data resolution. This prevents different distribution points from being assigned to the same habitat, which could cause inappropriate results. The slope and direction data in terrain factors were obtained through fine reclassification and surface analysis of altitude data by using ArcGIS software. For the prediction of future climate data, four greenhouse gas emission scenarios under the BCC-CSM2-MR model are adopted, which are low level (SSP1-2.6), medium level (SSP2-4.5), medium-high level (SSP3-7.0) and high level (SSP5-8.5) 19 . At the same time, this paper considered the climate variables of three future periods, namely 2030s (2021–2040), 2050s (2041–2060) and 2090s (2081–2100), to more accurately predict the potential suitable areas of O. microphylla in the future under different climate scenarios. Soil data were sourced from the World Soil Database, with a coordinate system of WGS84 20 . The grid size of the database is approximately 1 square kilometer, covering 15 soil-related factors. Human activity intensity data were sourced from the dataset of human footprint published by the Center of International Earth Science Information Network (CIESIN). The dataset covers multi-dimensional information, including population density data, comprehensively revealing the intensity of human activities and their spatial distribution 21 . The collected environmental variable data were converted into grid format and imported into ArcGIS 10.6 software along with the final data distribution points. After converting the distribution point data into grid format, we extracted values for 38 environmental variables at each point. To evaluate the correlation among environmental factors and detect multicollinearity, we conducted Variance Inflation Factor (VIF) analysis and Pearson correlation tests. If two variables had a Pearson correlation coefficient greater than 0.8, we compared their ecological significance to the distribution of O. microphylla. The variable with higher ecological relevance was retained, while the other was discarded. Variables with VIF values greater than 10 were considered to exhibit multicollinearity and were removed from the analysis. After strict screening, we retained the environmental variables that were most relevant to the species’ distribution for further analysis 22 . Finally, 14 variables were selected as the main factors for the study, as shown in Table  1 . Table 1 Selected environmental factors for predicting the geographical distribution of O. microphylla. Construction and optimization of the MaxEnt Model The MaxEnt model fully respects known distribution information and remains neutral to unknown distribution information without making any subjective assumptions 23 . For the target region X, it is considered to be composed of a certain number of grids. Let \(\:{\uppi\:}\left(\text{x}\right)\) be the probability of \(\:{\uppi\:}\) distribution on each network, then\(\:\:\sum\:{\uppi\:}\left(\text{x}\right)=1.\) Therefore, the entropy of the estimated distribution\(\:{\:{\uppi\:}}^{{\prime\:}}\:\)is\(\:\:\text{H}\left({{\uppi\:}}^{{\prime\:}}\right)=-\sum\:{\uppi\:}{\prime\:}\left(\text{x}\right)\text{l}\text{n}{\uppi\:}{\prime\:}\left(\text{x}\right)(x\in\:X)\). Thus, the known species distribution point information actually constitutes constraint conditions. Among the distributions that can satisfy these specific constraints, there exist multiple possibilities 24 . The MaxEnt model selects the distribution with the maximum entropy as the most ideal distribution state, also known as the optimal distribution 25 , 26 . To simplify the model structure and improve prediction accuracy, this study optimizes the MaxEnt model using the ENMeval package in the Rstudio environment. The optimization mainly involves adjusting two key parameters: the regularization multiplier (RM) and the feature combination (FC). The RM adjustment range is set between 0.5 and 4, with a step size of 0.5, forming 8 different RM values. Additionally, 8 different feature combination methods are adopted, including L, LQ, LQP, QHP, LQH, LQHP, QHPT, and LQHPT. A comprehensive evaluation of these 64 parameter combinations is conducted based on the ENMeval package. During the parameter selection process, the Akaike Information Criterion (AICc) is used as the evaluation standard, and the AICc value is utilized to assess the complexity and fitness of the model. A lower AICc value indicates higher prediction accuracy of the model 27 , especially when the AICc value is 0, which signifies that the model performance has reached an optimal state 28 . After determining the best parameter combination, the screened species distribution point data and environmental variable data are input into the MaxEnt model, replacing the model’s default settings with these optimized parameters. Through this method, a more accurate species distribution model for O. microphylla is successfully constructed to predict its current and future potential suitable areas under the combined influence of climate and environmental factors. Model accuracy and suitability area classification The Receiver Operating Characteristic curve (ROC) is a widely recognized diagnostic test evaluation metric and has been extensively used in assessing potential species suitable areas 29 . In this study, the area under the ROC curve (AUC) is calculated to analyze the model’s accuracy. The AUC value ranges from 0 to 1, with values closer to 1 indicating higher predictive accuracy of the model. According to AUC evaluation standards, the predictive performance can be divided into the following levels: 0.50–0.60 represents invalid prediction, 0.60–0.70 indicates poor prediction, 0.70–0.80 indicates moderate prediction, 0.80–0.90 indicates good prediction, and 0.90-1.00 indicates excellent prediction. When the AUC value exceeds 0.9, the model’s prediction results can be considered highly accurate 30 . In ArcGIS software, the prediction results from the MaxEnt model are loaded, and the natural breaks classification method is employed to categorize the suitability index into four levels: highly suitable, moderately suitable, low suitable, and unsuitable areas. To prevent inconsistent classification standards across different scenarios and time periods, a uniform classification process is applied to all prediction results from various scenarios and periods. Specifically, all result data are merged, and natural breaks classification is performed based on the combined dataset, ensuring that the threshold values for each suitability level remain consistent across different scenarios and time periods. Subsequently, the area statistics function is utilized to calculate the actual area of each suitability level for O. microphylla, and a distribution map of its potential suitable areas is created. Spatial distribution pattern changes of suitable areas The reclassification tool in ArcGIS 10.6 software is used to binarize potential geographical distribution data for O. microphylla under current and future climate change scenarios, establishing a presence/absence matrix for the potential geographical distribution of O. microphylla. Suitable areas are represented by the value ‘1’ indicating species presence, while unsuitable areas are represented by ‘0’ indicating species absence. Based on this, the changes in the spatial pattern of suitable areas for O. microphylla under current and future climate scenarios are further analyzed, classifying the distribution into newly suitable areas, lost suitable areas, and stable suitable areas. The changes in the spatial pattern of potential suitable areas under current and future climate scenarios are defined as follows: matrix value 0→1 represents new suitable areas, 1→0 represents retracted suitable areas, and 1→1 represents stable suitable areas 28 , 31 . The areas of lost, newly gained, and stable suitable regions for O. microphylla under different scenarios are calculated by the number of pixels. Using the SDMTool toolbox, the centroid positions of the suitable areas for O. microphylla under current and future climate scenarios are calculated to explore the spatial distribution changes and migration direction of the suitable areas. Additionally, based on the latitude and longitude data of the centroid points in each period, the migration distance of the centroids of O. microphylla under different climate scenarios is calculated to better understand the dynamic changes in its distribution. Results Optimization of environmental factors and model accuracy verification The optimized parameters significantly reduce the complexity and fitting of the model, enhancing the accuracy and reliability of the prediction results. When the regularization multiplier (RM) was set to 2.5 and the feature classes (FC) were set to LQH, the delta.AICc value approached 0 (Fig. 2 ), indicating that the model is optimal at this setting. Thus, RM = 2.5 and FC = LQH are adopted as the parameters for model construction in this study. Using these parameters, ten simulation training runs produced an average AUC of 0.963 (Fig. 3 ), indicating a high degree of prediction accuracy. Fig. 2 Discussion In this study, we investigated the potential distribution of O. microphylla under current and future climate scenarios using the MaxEnt model. Our findings highlight the significant influence of specific environmental factors on the species’ distribution and provide insights into its potential migration patterns in response to climate change. Influence of environmental factors on the geographical distribution of Ormosia microphylla The results of Table  3 show that the lowest temperature of the coldest month (bio6) is the most critical factor affecting the distribution of O. microphylla, with a contribution rate of 54.3%. This emphasizes the species’ sensitivity to low-temperature extremes, which can severely impact seed germination, plant survival, and flowering processes. Table 3 Contribution rate and importance value of important environmental factors. Soil factors such as clay content and base saturation also play significant roles, contributing 11.7% and 8.5% to the model, respectively. High clay content enhances soil water retention, providing a stable moisture supply essential for seed germination and early seedling development, especially during dry periods 35 . Adequate soil moisture facilitates the activation of metabolic pathways required for radicle emergence and seedling establishment 36 . Base saturation reflects soil fertility, indicating the availability of essential nutrients like calcium, magnesium, and potassium, which are crucial for plant growth, flowering, and fruiting 37 . Optimal nutrient availability supports robust vegetative growth and enhances reproductive success. The average diurnal temperature range (bio2) influences the daily thermal fluctuations experienced by the plant. A suitable bio2 range of 5.07 °C to 8.09 °C suggests that O. microphylla prefers environments with moderate temperature variations in Fig. 8 . Excessive fluctuations can cause thermal stress, affecting stomatal regulation, transpiration rates, and photosynthesis efficiency, potentially leading to flower bud abortion or reduced pollen viability 38 . Precipitation in the driest month (bio14), contributing 5.2% to the model, is vital for sustaining O. microphylla during periods of water scarcity. Adequate precipitation during critical growth phases supports metabolic activities, prevents desiccation, and ensures successful flowering and seed development 39 . Drought stress can lead to stomatal closure, reducing carbon assimilation and affecting energy availability for growth and reproduction. Explanation of migration patterns under climate change Climate change projections suggest a northward and upward shift in the suitable habitat of O. microphylla by the 2050s and 2090s. This migration pattern is primarily driven by increasing global temperatures, altering the climatic suitability of regions currently outside the species’ range. As temperatures rise, southern regions may exceed the upper thermal limits of O. microphylla, making them less suitable for its survival. The species is expected to migrate towards higher latitudes and elevations where the temperatures fall within its optimal bio6 range. This shift allows O. microphylla to maintain its physiological processes without the stress of excessive heat. However, the migration is challenged by the species’ limited seed dispersal mechanisms, which may not keep pace with the rapid rate of climate change. Physical barriers and habitat fragmentation due to human activities further impede natural migration routes. Additionally, newly colonized areas may present different soil properties and biotic communities, affecting establishment success. Soil differences in clay content and base saturation could influence water retention and nutrient availability, while competition with established species and exposure to unfamiliar pests or diseases may pose additional challenges. Altered precipitation patterns, particularly changes in bio14, may also impact migration. Increased frequency of droughts or shifts in rainfall timing can exacerbate water stress during seed germination and flowering, reducing reproductive success and seedling survival. Conservation strategies and recommendations Given the predicted shifts in suitable habitats, conservation strategies for O. microphylla should be adaptive and proactive, addressing both current and future challenges. In-situ conservation with a focus on climate adaptation is crucial. Establishing climate-informed protected areas by protecting regions projected to remain or become suitable habitats can ensure the long-term conservation of O. microphylla populations. Utilizing climate models helps prioritize these areas effectively. Enhancing habitat connectivity through the development of ecological corridors can connect fragmented habitats, facilitating gene flow and species migration. This approach enhances genetic diversity and increases the species’ adaptability to changing conditions. Mitigating anthropogenic barriers is also essential; implementing land-use policies that reduce habitat fragmentation caused by urban development and agriculture, and encouraging sustainable land management practices, can minimize the impact on migration routes. Ex-situ conservation and assisted migration are vital components of a comprehensive conservation strategy. Collecting and storing seeds from various populations for seed banking and germplasm preservation conserves genetic diversity, serving as a resource for future restoration and research efforts. Developing propagation techniques for O. microphylla supports reintroduction into suitable habitats, especially in areas where natural regeneration is limited. Assisted migration, which involves carefully translocating individuals to new suitable areas, should be considered following thorough risk assessments to prevent ecological imbalances. Community engagement and sustainable management play critical roles in conservation efforts. Promoting education and awareness about O. microphylla’s ecological importance and vulnerability to climate change among local communities can foster support for conservation initiatives. Enforcing regulations that control the exploitation of O. microphylla encourages sustainable use and reduces pressure on wild populations. Monitoring and research are essential for informed conservation management. Establishing long-term monitoring programs to track changes in population dynamics, distribution, and health status enables timely management interventions. Investigating the species’ physiological and genetic responses to environmental stressors can identify traits that confer resilience, informing conservation priorities and aiding in the development of adaptive strategies. Integration with broader biodiversity conservation efforts Conserving O. microphylla contributes to the overall health of ecosystems and supports biodiversity. Protecting its habitat safeguards other species with similar ecological requirements, enhancing ecosystem resilience. Collaborative conservation approaches are necessary for effective implementation. Fostering multi-stakeholder partnerships among government agencies, non-governmental organizations, researchers, and local communities can lead to the development of comprehensive conservation strategies. Aligning O. microphylla conservation efforts with regional and national biodiversity action plans ensures cohesive and effective policy integration. Engaging in transboundary cooperation is crucial when the species’ range extends across borders, facilitating shared research and management efforts. Implementing ecosystem-based adaptation strategies can provide co-benefits that extend beyond O. microphylla. Such measures enhance carbon sequestration, mitigate climate impacts, and improve habitat connectivity, benefiting both O. microphylla and other species within the ecosystem. By integrating O. microphylla conservation into broader biodiversity efforts, we can promote ecosystem resilience and contribute to global conservation goals. Conclusion Ormosia microphylla is a species that not only depends on sufficient water for its growth, but also has strict requirements for suitable temperatures. Studies have shown that the minimum temperature of the coldest month is a key environmental factor for growth, suggesting that O. microphylla is extremely sensitive to low temperatures. In addition, the clay content and alkali saturation of the soil also have important effects on its growth. With global warming, conditions suitable for the growth of O. microphylla are gradually emerging in northern China, expanding the potential habitat of this species to higher latitudes. However, the soil conditions in the northern part of the country are significantly different from those in the original habitat, and the species may face new competitive pressures, resulting in a more complex community structure in the potential habitat in the future. Based on the prediction of potential habitat areas and migration pathways of O. microphylla, we can develop targeted conservation strategies using an ecosystem-based approach, which not only helps to sequester carbon and mitigate the impacts of climate change, but also is essential for the conservation of the species and its ecosystem, and makes an important contribution to the global conservation of the species. Data availability The data source has been explained in the article. Further inquiries are available on request from Pan Huang. References Wei, L. et al. Predicting suitable habitat for the endangered tree Ormosia microphylla in China. Sci. Rep. 14, 10330. https://doi.org/10.1038/s41598-024-61200-5 (2024). Rights and permissions Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ .