Method for predicting future suitable habitat change trend of medicinal plant diversity

Abstract

The invention discloses a method for predicting a future suitable habitat change trend of medicinal plant diversity, relates to the technical field of species distribution prediction, and aims to solve the problem that a habitat of medicinal plants is seriously damaged and under the background that part of the medicinal plants are eradicated or endangered. A future distribution prediction framework is constructed based on a MaxEnt model on the premise that the medicinal plant diversity is urgently required to be protected, and the suitable habitat change trend of the medicinal plant diversity in the future is predicted by using the MaxEnt model, so that species eradication can be effectively avoided, sustainable development of medicinal plant resources is guaranteed, and meanwhile ecological balance is maintained. The MaxEnt model has the characteristics of high accuracy and simplicity and convenience in operation, can accurately predict the suitable habitat of the medicinal plant diversity even when sample distribution points are few, and can provide reference for local governments to cope with international sustainable development and diversity protection policies.

Description

Technical Field

[0001] The present invention relates to the technical field of species distribution prediction, and in particular to a method for predicting future suitable habitat change trends of medicinal plant diversity. Background Art

[0002] Biodiversity is fundamental to human survival and plays a key role in maintaining ecosystem stability. Furthermore, biodiversity represents a complex ecosystem, formed by the interactions between organisms and their environment, encompassing a wide range of ecological processes. The component of biodiversity that studies the relationship between plants and their surroundings is called plant diversity. Protecting plant diversity is a fundamental aspect and a key area of ​​biodiversity conservation, with medicinal plant diversity being of particular importance. Medicinal plant diversity is a crucial component of plant diversity, providing treatments and preventive measures for human disease. Medicinal plants are plants that contain specific medicinal components that have preventative, therapeutic, and health-promoting properties, and are also a crucial component of Traditional Chinese Medicine. In recent years, efforts have been made to correct missing data, update existing information, and discover new plant species. Unfortunately, due to overexploitation, the spread of invasive alien species, and the impacts of climate change, the habitats of medicinal plants have been severely damaged, and some species have become extinct or are threatened with extinction. Predicting future trends in medicinal plant diversity is crucial. This is of great significance for the conservation, development, utilization, sustainable development, and ecosystem stability of medicinal plant resources.

[0003] Currently, the most widely used species distribution model for species distribution prediction is the MaxEnt model. Built on the principle of maximum entropy, the MaxEnt model offers high accuracy and ease of use compared to other single models, achieving accurate predictions even with a small number of sample points. However, most research on predicting the future distribution of medicinal plants focuses on single species, while research on predicting the future distribution of medicinal plant diversity is lacking. Therefore, it is necessary to combine the MaxEnt model with medicinal plant distribution points to explore the distribution trends of medicinal plant diversity under climate change. Summary of the Invention

[0004] The main purpose of the present invention is to provide a method for predicting the future trend of changes in suitable habitats for medicinal plant diversity, which can effectively solve the problems in the background technology.

[0005] To achieve the above object, the technical solution adopted by the present invention is:

[0006] A method for predicting future trends in suitable habitat changes for medicinal plant diversity, comprising the following steps:

[0007] S1. Obtaining medicinal plant distribution data;

[0008] S2. Selection of environmental variables;

[0009] S3.MaxEnt model parameter settings;

[0010] S4. Prediction of currently suitable potential habitats for medicinal plants;

[0011] S5. Overlay and normalize the prediction result raster layers;

[0012] S6. Delineation of currently suitable potential habitats for medicinal plant diversity;

[0013] S7. Prediction and delineation of potential suitable habitats for medicinal plant diversity under future climate conditions;

[0014] S8. Comparing and analyzing the changing trends of the current potential suitable habitats in step S6 and the future suitable habitats in step S7;

[0015] The S3 specifically includes: This step focuses on the overall goal of understanding the spatial distribution pattern of medicinal plant diversity. Therefore, MaxEnt is set as follows: The maximum number of iterations of the model is 10 5 times, with a convergence threshold of 0.0005, and using "Crossvalidate" as the model rerun type, a total of 10 runs; to ensure the accuracy of model predictions, the area under the curve (AUC) was used as the evaluation metric; generally, an AUC value exceeding 0.75 indicates that the model predictions are robust, which means that the results are highly reliable; therefore, the prediction results for each species were carefully checked to confirm that the AUC value exceeded the threshold of 0.75; the output generated by Logistic regression ranges from 0 to 1, which is highly consistent with the probability of species distribution, which is in line with the research objectives; therefore, Logistic regression was selected as the output method, and the default settings of the remaining parameters were retained.

[0016] Preferably, the S1 specifically includes: the investigation follows the "four principles": there are investigation records, voucher specimens, actual photos, and medicinal evidence; sampling data are collected according to the investigation route and sampling survey, and combined with modern technologies such as GPS and GIS, each sampling point represents the occurrence record of the species and performs deduplication and other operations.

[0017] Preferably, the S2 specifically includes: a comprehensive set of data, including 19 bioclimatic variables and elevation data; current and future climate data are downloaded from the Worldclim database with a resolution of 2.5 arcmins; future climate data are selected from four climate scenarios (SSP126, SSP245, SSP370 and SSP585) in the BCC-CSM2-MR model for 2030s (2021-2040) and 2070s (2061-2080), representing future carbon emissions from low to high respectively; elevation data is derived from SRTM elevation data; it is worth noting that since there is no information about future predicted elevation, it is assumed that the current or future elevation does not change; due to the large number of species in this study, the most widely accepted data were used, that is, all the above data were used to predict the distribution of medicinal plant diversity under current and future climate conditions.

[0018] Preferably, the step S4 specifically includes: using step S3 in combination with the distribution points of the medicinal plants and current environmental factors to predict the current potential suitable habitat of the medicinal plants.

[0019] Preferably, S5 specifically includes: outputting the prediction results of the MaxEnt model in ASC format; converting the results into tiff format using ArcGIS 10.8; overlaying the separate suitable habitat maps using the raster calculator in ArcGIS 10.8, and normalizing the overlaid layers.

[0020] Preferably, S6 specifically includes: using ArcGIS 10.8 to classify the medicinal plant diversity results into four categories: unsuitable (0-0.25], low suitability (0.25-0.5], moderate suitability (0.5-0.75], and high suitability (0.75-1). Using 0.25 as a threshold to convert the current potential suitable habitats of medicinal plant diversity into binary (presence / absence).

[0021] Preferably, the S7 specifically includes: using step S3 to combine the medicinal plant distribution points and environmental factors under the future climate to predict the potential suitable habitats of medicinal plants under the future climate; at the same time, using steps S5 and S6 to convert the potential suitable habitats of medicinal plant diversity under the future climate into binary (presence / absence).

[0022] Preferably, S8 specifically includes: using Distribution Changes Between Binary SDMs in SDM toolbox to compare and analyze the changing trend of suitable habitats in the future. -1 = range expansion; 0 = no occupancy (absence in both); 1 = no change (presence in both); 2 = range contraction.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] The present invention provides a method for predicting future trends in suitable habitat changes for medicinal plant diversity. This method uses more accurate species distribution point data combined with environmental variables. By using the MaxEnt model to predict future trends in suitable habitat changes for medicinal plant diversity, it can effectively avoid species extinction, ensure the sustainable development of medicinal plant resources, and maintain ecological balance.

[0025] This study uses the MaxEnt model, which offers greater accuracy and ease of use than other single models. It can achieve accurate predictions even when sampling from a limited number of locations. This model can accurately predict suitable habitats for medicinal plant diversity, providing a reference for local governments in addressing international sustainable development and biodiversity conservation policies. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a flow chart of an embodiment of the present invention;

[0027] Figure 2 A distribution map of medicinal plant sampling points in the study area of ​​an embodiment of the present invention;

[0028] Figure 3 A map of currently suitable potential habitats for medicinal plant diversity according to an embodiment of the present invention;

[0029] Figure 4 A graph showing the changing trend of potential suitable habitats for medicinal plant diversity under future climate conditions according to an embodiment of the present invention;

[0030] In the picture: Figure 3 (A) widely distributed medicinal plants; (B) rare medicinal plants; (C) endangered medicinal plants;

[0031] Figure 4 The purple box represents widespread distribution; the blue box represents rare; and the orange box represents endangered. DETAILED DESCRIPTION

[0032] In order to make the technical means, creative features, objectives and effects achieved by the present invention easier to understand, the present invention is further described below in conjunction with specific implementation methods.

[0033] In the description of the present invention, it should be noted that the terms "under climate" and "under future climate" and other terms indicating periods or eras are based on the periods or eras shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They do not indicate or imply that the application time referred to must be within a specific period or era, and therefore should not be construed as limiting the present invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0034] Example

[0035] A method for predicting future trends in suitable habitat changes for medicinal plant diversity, comprising the following steps:

[0036] S1. Obtaining medicinal plant distribution data;

[0037] S2. Selection of environmental variables;

[0038] S3.MaxEnt model parameter settings;

[0039] S4. Prediction of currently suitable potential habitats for medicinal plants;

[0040] S5. Overlay and normalize the prediction result raster layers;

[0041] S6. Delineation of currently suitable potential habitats for medicinal plant diversity;

[0042] S7. Prediction and delineation of potential suitable habitats for medicinal plant diversity under future climate conditions;

[0043] S8. Comparing and analyzing the changing trends of the current potential suitable habitats in step S6 and the future suitable habitats in step S7;

[0044] The S3 specifically includes: This step focuses on the overall goal of understanding the spatial distribution pattern of medicinal plant diversity. Therefore, MaxEnt is set as follows: The maximum number of iterations of the model is 10 5times, with a convergence threshold of 0.0005, and using "Crossvalidate" as the model rerun type, a total of 10 runs; to ensure the accuracy of model predictions, the area under the curve (AUC) was used as the evaluation metric; generally, an AUC value exceeding 0.75 indicates that the model predictions are robust, which means that the results are highly reliable; therefore, the prediction results for each species were carefully checked to confirm that the AUC value exceeded the threshold of 0.75; the output generated by Logistic regression ranges from 0 to 1, which is highly consistent with the probability of species distribution, which is in line with the research objectives; therefore, Logistic regression was selected as the output method, and the default settings of the remaining parameters were retained.

[0045] Example 1:

[0046] The following is a prediction and division of the specific implementation process and the changing trend of suitable habitats for medicinal plant diversity in Yinshan Mountain (China) based on this method;

[0047] like Figure 1 A method for predicting future suitable habitat changes for medicinal plant diversity according to an embodiment of the present invention includes the following steps:

[0048] S1. Obtaining medicinal plant distribution data:

[0049] The data on the distribution points of medicinal plants in the Yinshan area are derived from field surveys from 2012 to 2022. The survey followed the "four principles", namely, survey records, voucher specimens, actual photos, and medicinal evidence to ensure the scientificity and accuracy of the results. Sampling data were collected according to the survey route and sampling survey, and combined with modern technologies such as GPS and GIS, each sampling point represents the occurrence record of the species. At the same time, operations such as deduplication were performed. In the end, 10,301 data points remained. Combining the spatial analysis function and visualization drawing function of ArcGIS, a sampling point distribution map was drawn ( Figure 2 ).

[0050] S2. Selection of environment variables:

[0051] The study aimed not only to predict the distribution of medicinal plant diversity under current climate conditions but also to gain insight into how various socio-sharing economy pathways might influence the future spatial distribution of medicinal plant diversity. To achieve this goal, a comprehensive dataset was incorporated, including 19 bioclimatic variables and elevation data. This multifaceted approach enabled the exploration of potential future changes in medicinal plant diversity patterns and their associations with different socio-sharing economy pathways. Current and future climate data were downloaded from the Worldclim database at a resolution of 2.5 arcmins. Future climate data were selected from four climate scenarios (SSP126, SSP245, SSP370, and SSP585) from the BCC-CSM2-MR model for the 2030s (2021-2040) and 2070s (2061-2080), representing future carbon emissions, from low to high. Elevation data were derived from SRTM elevation data. Of note, since no information on future projected elevations was available, no change in current or future elevations was assumed. Due to the large number of species in this study, the most widely accepted data were used, i.e., all the above data were used to predict the distribution of medicinal plant diversity under current and future climate conditions, respectively.

[0052] S3.MaxEnt model parameter settings:

[0053] Because this study collected a large dataset and the results could be affected by complex data biases, a rigorous screening process was performed. This meticulous approach aimed to identify the most representative distributions of medicinal plant species, focusing on three distinct categories. First, the "widespread medicinal plants of the Yinshan Mountains" category was considered. This included species found at multiple sampling sites, particularly those with more than 60 sampling sites. Second, the "rare medicinal plants of the Yinshan Mountains" category encompassed species with more restricted distributions, typically occurring at only 10 to 30 sampling sites each. Finally, the "endangered medicinal plants of the Yinshan Mountains" category was examined, which are listed as critically endangered plants. The primary goal of this study was to reveal the spatial distribution characteristics of medicinal plant diversity. Therefore, the "Jackknife" approach, commonly used in general MaxEnt modeling studies, was not employed. This approach is typically used to assess the impact of various environmental factors on the distribution of individual species or to create response curves to quantify the relationship between environmental factors and the distribution probability of individual species. Instead, our efforts focused on the overall goal of understanding the spatial distribution patterns of medicinal plant diversity. MaxEnt settings were as follows: the maximum number of model iterations was 10 5times, with a convergence threshold of 0.0005, and using "Crossvalidate" as the model rerun type, for a total of 10 runs. To ensure the accuracy of the model predictions, the area under the curve (AUC) was used as the evaluation metric. Typically, an AUC value exceeding 0.75 indicates that the model predictions are robust, meaning the results are highly reliable. Therefore, the prediction results for each species were carefully checked to confirm that the AUC value exceeded the threshold of 0.75. The output produced by Logistic regression ranges from 0 to 1, which is highly consistent with the probability of species distribution, which is consistent with the research objectives. Therefore, Logistic regression was selected as the output method, and the default settings of the remaining parameters were retained.

[0054] S4. Prediction of currently suitable habitats for medicinal plants:

[0055] Step S3 is used to combine the distribution points of medicinal plants and current environmental factors to predict the current potential suitable habitats of medicinal plants.

[0056] S5. Overlay and normalization of prediction result raster layers:

[0057] The MaxEnt model outputs predictions in ASC format. Subsequently, the results were converted to TIFF format using ArcGIS 10.8. The individual suitable habitat maps were overlaid using the Raster Calculator in ArcGIS 10.8. The overlays were normalized for consistency.

[0058] S6. Delineation of currently suitable habitats for medicinal plant diversity:

[0059] The results of medicinal plant diversity were divided into four categories using ArcGIS 10.8: unsuitable (0-0.25], low suitability (0.25-0.5], moderate suitability (0.5-0.75], and high suitability (0.75-1). The current potential suitable habitats for medicinal plant diversity were converted into binary (presence / absence) using 0.25 as the threshold ( Figure 3 ).

[0060] S7. Prediction and delineation of potential suitable habitats for medicinal plant diversity under future climate conditions:

[0061] Step S3 combines the distribution points of medicinal plants with environmental factors under the future climate to predict the potential suitable habitats for medicinal plants under the future climate. Simultaneously, steps S5 and S6 convert the potential suitable habitats for medicinal plant diversity under the future climate into a binary value (presence / absence).

[0062] S8. Compare and analyze the changing trends of the current potential suitable habitats in step S6 and the future suitable habitats in step S7:

[0063] Use the Distribution Changes Between Binary SDMs in SDMtoolbox to compare and analyze the trend of future suitable habitat changes ( Figure 4 ). -1=range expansion; 0=no occupancy (absence in both); 1=no change (presence in both); 2=range contraction.

[0064] The above embodiments are intended only to illustrate the design concepts and features of the present invention. Their purpose is to enable those skilled in the art to understand the contents of the present invention and implement them accordingly. The scope of protection of the present invention is not limited to the above embodiments. Therefore, any equivalent changes or modifications made based on the principles and design concepts disclosed in the present invention are within the scope of protection of the present invention.

Claims

1. A method for predicting the future trend of changes in suitable habitats for medicinal plant diversity, characterized in that: The following steps are involved: S1. Obtaining medicinal plant distribution data; S2. Selection of environmental variables; S3.MaxEnt model parameter settings; S4. Prediction of currently suitable potential habitats for medicinal plants; S5. Overlay and normalize the prediction result raster layers; S6. Delineation of currently suitable potential habitats for medicinal plant diversity; S7. Prediction and delineation of potential suitable habitats for medicinal plant diversity under future climate conditions; S8. Comparing and analyzing the changing trends of the current potential suitable habitats in step S6 and the future suitable habitats in step S7; The S3 specifically includes: This step focuses on the overall goal of understanding the spatial distribution pattern of medicinal plant diversity. Therefore, MaxEnt is set as follows: The maximum number of iterations of the model is 10 5 times, with a convergence threshold of 0.0005, and using "Crossvalidate" as the model rerun type, for a total of 10 runs; to ensure the accuracy of the model predictions, the area under the curve (AUC) was used as the evaluation metric; generally, an AUC value exceeding 0.75 indicates that the model predictions are robust, meaning that the results are highly reliable; therefore, the prediction results for each species were carefully checked to confirm that the AUC value exceeded the threshold of 0.75; the output produced by Logistic regression ranges from 0 to 1, which is highly consistent with the probability of species distribution, which is consistent with the research objectives; therefore, Logistic regression was selected as the output method, and the default settings of the remaining parameters were retained.

2. The method for predicting future trends in suitable habitats for medicinal plant diversity according to claim 1, characterized in that: The S1 specifically includes: the investigation follows the "four principles": investigation records, voucher specimens, actual photos, and medicinal evidence; sampling data are collected according to the investigation route and sampling survey, and combined with modern technologies such as GPS and GIS, each sampling point represents the occurrence record of the species and performs deduplication and other operations.

3. The method for predicting future trends in suitable habitats for medicinal plant diversity according to claim 1, characterized in that: The S2 specifically includes: a comprehensive set of data, including 19 bioclimatic variables and elevation data; current and future climate data were downloaded from the Worldclim database with a resolution of 2.5 arcmins; future climate data were selected from four climate scenarios (SSP126, SSP245, SSP370 and SSP585) in the BCC-CSM2-MR model for 2030s (2021-2040) and 2070s (2061-2080), representing future carbon emissions from low to high; elevation data came from SRTM elevation data; it is worth noting that since there is no information on future predicted elevation, it is assumed that there is no change in current or future elevation; due to the large number of species in this study, the most widely accepted data were used, that is, all the above data were used to predict the distribution of medicinal plant diversity under current and future climate conditions.

4. The method for predicting future suitable habitat changes for medicinal plant diversity according to claim 1, characterized in that: Said S4 specifically includes: using step S3 in combination with the medicinal plant distribution points and current environmental factors to predict the current potential suitable habitat of the medicinal plant.

5. The method for predicting future suitable habitat changes for medicinal plant diversity according to claim 1, characterized in that: The S5 specifically includes: outputting the prediction results of the MaxEnt model in ASC format; converting the results into tiff format using ArcGIS 10.8; overlaying the separate suitable habitat maps using the raster calculator in ArcGIS 10.8, and normalizing the overlaid layers.

6. The method for predicting future suitable habitat changes for medicinal plant diversity according to claim 1, characterized in that: The S6 specifically includes: using ArcGIS 10.8 to classify the medicinal plant diversity results into four categories: unsuitable (0-0.25], low suitability (0.25-0.5], moderate suitability (0.5-0.75], and high suitability (0.75-1). Using 0.25 as the threshold, the current potential suitable habitats of medicinal plant diversity were converted into binary (presence / absence).

7. The method for predicting future suitable habitat changes for medicinal plant diversity according to claim 1, characterized in that: Said S7 specifically includes: using step S3 to combine the medicinal plant distribution points and environmental factors under the future climate to predict the potential suitable habitats of medicinal plants under the future climate; at the same time, using steps S5 and S6 to convert the potential suitable habitats of medicinal plant diversity under the future climate into binary (presence / absence).

8. The method for predicting future suitable habitat changes for medicinal plant diversity according to claim 1, characterized in that: S8 specifically includes: using Distribution Changes Between Binary SDMs in SDM toolbox to compare and analyze the changing trend of suitable habitats in the future. -1 = range expansion; 0 = no occupancy (absence in both); 1 = no change (presence in both); 2 = range contraction.

Images