A method for predicting a suitable area of plant diseases and insect pests based on GIS and MaxEnt

Abstract

The application discloses a kind of based on GIS and MaxEnt prediction plant disease and insect pest suitable area method, comprising the following steps: S1: the natural geographical distribution point data of the target disease and insect pest is obtained;S2: the environmental climate dataset under current and future climate conditions is obtained;S3: the biological climate variable screened in environmental climate dataset is obtained;S4: the main biological climate variable that influences the target disease and insect pest distribution is calculated and analyzed using MaxEnt model;S5: using ArcGIS software obtains the suitable area prediction graph of the target disease and insect pest under current and future climate scenarios;The beneficial effects of the application are to provide a kind of comprehensive prediction method combining modern GIS technology and advanced MaxEnt model;Second, it can better understand and predict the distribution change of disease and insect pest under different climate scenarios;Three is, the application of the application helps to improve agricultural production efficiency, reduce the loss caused by disease and insect pest.

Description

Technical Field

[0001] This invention relates to the field of predicting suitable areas for plant diseases and pests, and specifically to a method for predicting suitable areas for plant diseases and pests based on GIS and MaxEnt. Background Technology

[0002] Plant diseases and pests have a serious impact on agricultural production, and predicting their suitable habitats is crucial for developing effective control strategies. However, a significant problem in current plant disease and pest forecasting and management practices is the lack of a scientifically accurate method for predicting potential suitable habitats for diseases and pests.

[0003] Current forecasting methods typically rely on empirical judgments or simple climate models, which often fail to accurately reflect the complex relationship between pests and diseases and the environment, leading to inaccurate and unreliable predictions. This inadequacy is particularly evident in the context of global climate change. Furthermore, existing technologies lack the capacity to comprehensively analyze and predict pests and diseases using advanced statistical models and geographic information system tools. They often focus only on the distribution of suitable habitats, lacking assessments of suitability levels and probabilities, thus failing to provide comprehensive forecast results. This results in insufficient information for developing pest and disease management and control strategies, thereby impacting the effectiveness of pest and disease control.

[0004] Therefore, there is an urgent need to develop a method that integrates GIS and advanced statistical models such as MaxEnt to predict suitable areas for plant diseases and pests. This method can not only improve the accuracy and reliability of disease and pest prediction, but also help to better understand the relationship between disease and pest distribution and environmental factors, providing a scientific basis for effective disease and pest management and control. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides a method for predicting suitable habitats for plant diseases and pests based on GIS and MaxEnt. This method can quickly and effectively predict the current and future distribution of suitable habitats for plant diseases and pests.

[0006] The method for predicting suitable areas for plant diseases and pests based on GIS and MaxEnt includes the following steps: S1: Identify the target pests and diseases, and obtain data on the natural geographical distribution points of the target pests and diseases; S2: Obtain environmental climate datasets under current and future climate conditions; S3: Obtain the bioclimatic variables after centralized filtering of the environmental climate dataset; S4: Based on the selected bioclimatic variables, use the MaxEnt model to calculate and analyze the main bioclimatic variables that affect the distribution of the target pests and diseases, and output the MaxEnt model prediction results. S5: Based on the prediction results of the MaxEnt model, use ArcGIS software to obtain the predicted maps of the suitable habitats of the target pests and diseases under the current and future climate scenarios, and evaluate the changes in the distribution of the suitable habitats of the target pests and diseases. Preferably, in step S1, the methods for obtaining the natural geographical distribution data of the target pests and diseases include field surveys, literature searches, and electronic database searches. Further, in step S1, the natural geographical distribution point data of the target pests and diseases are subjected to geographical filtering and sparsification processing to obtain the effective distribution points of the target pests and diseases, and to obtain the geographical distribution dataset of the target pests and diseases. Preferably, in step S2, an environmental climate dataset under current and future climate conditions is downloaded from the WorldClim global climate database; Preferably, the specific operation of step S3 is as follows: import the environmental climate dataset and the geographic distribution dataset of the target pest into ArcGIS, extract the bioclimatic variable data of the distribution points, perform Pearson correlation analysis, remove bioclimatic variables with a contribution rate of 0, remove bioclimatic variables with an absolute value of correlation coefficient greater than 0.9, small contribution rate and no clear biological significance, and the remaining bioclimatic variables in the environmental climate dataset are the variables used for the MaxEnt model of the target pest. Preferably, the specific operation of step S4 is as follows: Step S41: MaxEnt software parameter settings are as follows: Set the training set to 75%, the test set to 25%, the regularization multiplier to be determined using the ENMeval package in R software, the maximum number of iterations to 5000, the number of repeated training iterations to 10, the repeated iteration method to "Subsample", enable the function of plotting response curves and knife cutting method, the output file format is Logistic, and the output file type is selected asc. Step S42, Model Prediction Performance Evaluation, specifically includes: Step S421: Plot the ROC curve; Step S422: Calculate the AUC value, where the AUC value is the area enclosed by the ROC curve and the coordinate axis; Step S423: The evaluation criteria represented by different AUC values ​​are as follows: below 0.6, the model simulation results fail and have no application value; 0.6-0.7, the model simulation results are poor; 0.7-0.8, the model prediction performance is average; 0.8-0.9, the model prediction performance is good; 0.9-1.0, the model prediction performance is excellent. Step S43: Obtain the main bioclimatic variables that affect the distribution of the target pests and diseases. The main bioclimatic variables that affect the distribution of the target pests and diseases are bioclimatic variables with a high contribution rate and clear biological significance. Step S44: Output the MaxEnt model prediction results. The MaxEnt model prediction results are output in an ASC layer format. The value of each point in the layer represents the suitability probability of the target pest or disease, and the range is [0,1]. Preferably, the specific operation of step S5 is as follows: Step S51: Based on the MaxEnt model, predict the suitability probability of the target pests and diseases, and divide the suitable areas into four levels: highly suitable areas, moderately suitable areas, lowly suitable areas, and unsuitable areas. Step S52: Import the prediction results of the MaxEnt model into ArcGIS software to form a prediction map of the suitable habitat for the pests and diseases; Step S53: Based on the predicted suitable habitat maps of the target pests and diseases under current and future climate scenarios, analyze and evaluate the changes in the distribution of suitable habitats of the target pests and diseases.

[0007] This invention provides a method for predicting suitable habitats for plant diseases and pests based on GIS and MaxEnt, overcoming the shortcomings of existing technologies in terms of prediction accuracy and scientific rigor. This method leverages the combined advantages of advanced statistical models and geographic information system tools to significantly improve the accuracy and reliability of suitable habitat prediction. Furthermore, this invention can analyze and evaluate in detail the suitability probability and suitability level of diseases and pests, providing a comprehensive scientific basis for formulating effective disease and pest control strategies.

[0008] The advantages and application prospects of this invention are mainly reflected in the following aspects: First, this method provides a comprehensive prediction method combining modern GIS technology and advanced MaxEnt models, which is a breakthrough in existing technologies. Second, this method considers the impact of various environmental factors, such as climate change, on the distribution of pests and diseases, enabling a better understanding and prediction of the distribution changes of pests and diseases under different climatic scenarios. Third, the application of this invention helps improve agricultural production efficiency, reduce losses caused by pests and diseases, and has broad application prospects for agricultural ecosystem management, pest and disease control, and climate change adaptation research. Therefore, this invention not only solves the key problems of existing technologies but also provides effective technical support for the scientific management of plant pests and diseases and the sustainable development of agriculture. Attached Figure Description

[0009] Figure 1 A flowchart illustrating the method for predicting suitable areas for plant diseases and pests based on GIS and MaxEnt provided in an embodiment of the present invention; Figure 2 Contribution rate and correlation analysis of 19 environmental factors for the natural geographical distribution of anthrax in Camellia oleifera; Figure 3 ROC curve for the MaxEnt model of anthrax in Camellia oleifera; Figure 4 Knife-cut plot analysis of bioclimatic variables for the MaxEnt model of anthrax in Camellia oleifera; Figure 5 The response curves of the MaxEnt model for anthrax in Camellia oleifera to major bioclimatic variables; Detailed Implementation

[0010] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and examples. Obviously, the described embodiments are merely some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the protection scope of this invention. Example 1

[0011] The target pest and disease is anthracnose in camellia oleifera, such as Figure 1 As shown, the method for predicting suitable areas for plant diseases and pests based on GIS and MaxEnt includes the following specific steps: Step S1: Identify the target pests and diseases and obtain data on the natural geographical distribution points of the target pests and diseases.

[0012] The natural geographical distribution data of Camellia oleifera anthrax were obtained from field surveys, relevant literature, and databases such as the National Specimen Resource Platform (NSII, www.nsii.org.cn / ) and the Teaching Specimen Resource Sharing Platform (http: / / mnh.scu.edu.cn / ). Geographic filtering and sparsification were performed on the natural geographical distribution data of Camellia oleifera anthrax to obtain 146 valid distribution points.

[0013] Step S2: Obtain environmental climate datasets under current and future climate conditions.

[0014] Preferably, in step S2, an environmental climate dataset under current and future climate conditions is downloaded from the WorldClim global climate database. The bioclimate variables in the environmental climate dataset are shown in Table 1.

[0015] Table 1 Bioclimatic Variables in WorldClim Data ; Step S3: Obtain the bioclimate variables after centralized filtering of the environmental climate dataset.

[0016] Preferably, step S3 specifically involves: importing the environmental climate dataset and the geographic distribution dataset of the target pest into ArcGIS, extracting the bioclimatic variable data of the distribution points, performing Pearson correlation analysis, removing bioclimatic variables with a contribution rate of 0, and removing bioclimatic variables with an absolute value of correlation coefficient greater than 0.9, a small contribution rate, and no clear biological significance. The remaining bioclimatic variables in the environmental climate dataset are the variables used in the MaxEnt model of the target pest.

[0017] according to Figure 2 The contribution rate and correlation analysis of 19 environmental factors at the natural geographical distribution points of Camellia oleifera anthracnose led to the selection of 9 bioclimatic variables for the MaxEnt model of Camellia oleifera anthracnose: monthly average diurnal temperature range (Bio2), isotherm (Bio3), maximum temperature of the hottest month (Bio5), minimum temperature of the coldest month (Bio6), average temperature of the wettest season (Bio8), annual precipitation (Bio12), precipitation of the wettest season (Bio13), coefficient of variation of precipitation (Bio15), and precipitation of the hottest season (Bio17).

[0018] S4: Based on the selected bioclimatic variables, use the MaxEnt model to calculate and analyze the main bioclimatic variables affecting the distribution of the target pests and diseases, and output the MaxEnt model prediction results. The specific operation is as follows: Step S41: MaxEnt software parameter settings are as follows: Set the training set to 75%, the test set to 25%, the regularization multiplier to be determined using the ENMeval package in R software, the maximum number of iterations to 5000, the number of repeated training iterations to 10, the repeated iteration method to "Subsample", enable the function of plotting response curves and knife cutting method, the output file format is Logistic, and the output file type is selected asc.

[0019] Step S42, Model Prediction Performance Evaluation, specifically includes: Step S421: Plot the ROC curve.

[0020] Step S422: Calculate the AUC value, where the AUC value is the area enclosed by the ROC curve and the coordinate axis.

[0021] Step S423: The evaluation criteria represented by different AUC values ​​are as follows: below 0.6, the model simulation results fail and have no application value; 0.6-0.7, the model simulation results are poor; 0.7-0.8, the model prediction performance is average; 0.8-0.9, the model prediction performance is good; 0.9-1.0, the model prediction performance is excellent.

[0022] like Figure 3As shown, the AUC value of the MaxEnt model for anthrax in Camellia oleifera is 0.943, and the model's predictive performance reaches the "excellent" standard.

[0023] Step S43: Obtain the main bioclimatic variables that affect the distribution of the target pests and diseases. The main bioclimatic variables that affect the distribution of the target pests and diseases are bioclimatic variables with a high contribution rate and clear biological significance.

[0024] according to Figure 4 The results of the bioclimatic variable knife plot analysis of the MaxEnt model for anthrax in Camellia oleifera show that the distribution of anthrax in Camellia oleifera is mainly affected by the precipitation in the hottest season (Bio17), annual precipitation (Bio12), maximum temperature in the hottest month (Bio5), and monthly average of diurnal temperature range (Bio2).

[0025] The response curves of the main bioclimatic variables in the MaxEnt model of anthrax in Camellia oleifera are as follows: Figure 5 As shown.

[0026] Step S44: Output the MaxEnt model prediction results. The MaxEnt model prediction results are output in an ASC layer format. The value of each point in the layer represents the suitability probability of the target pest or disease, ranging from [0,1].

[0027] S5: Based on the prediction results of the MaxEnt model, use ArcGIS software to obtain the predicted suitable habitat maps of the target pests and diseases under the current and future climate scenarios, and evaluate the changes in the distribution of the suitable habitats of the target pests and diseases.

[0028] Step S51: Based on the MaxEnt model, predict the suitability probability of the target pests and diseases, and divide the suitable areas into four levels: highly suitable areas, moderately suitable areas, lowly suitable areas, and unsuitable areas.

[0029] Camellia oleifera anthrax is classified as follows: suitable growth probability ( P A value less than 0.1 indicates an unsuitable growing area, and 0.10 ≤ P <0.3 indicates a low-yield suitable habitat; 0.3≤ P <0.5 indicates a moderately suitable growing area. P A value of ≥0.5 indicates a highly suitable habitat.

[0030] Step S52: Import the prediction results of the MaxEnt model into ArcGIS software to form a prediction map of the suitable habitat for the pests and diseases.

[0031] Step S53: Based on the predicted suitable habitat maps of the target pests and diseases under current and future climate scenarios, analyze and evaluate the changes in the distribution of suitable habitats of the target pests and diseases.

[0032] Compared to the current climate scenario, under the future climate scenario, the suitable area for anthrax in Camellia oleifera will increase to varying degrees, and the spatial location will shift to varying degrees, as shown in Table 2.

[0033] ; This invention provides a scientific and accurate method for predicting suitable habitats for plant diseases and pests by comprehensively utilizing GIS and MaxEnt models. By analyzing and integrating climate data, environmental variables, and disease and pest distribution data, this invention achieves high-precision prediction of suitable habitats for plant diseases and pests, providing valuable methodological references and data support for related fields such as agricultural ecology, plant protection, and environmental science.

[0034] Those skilled in the art should understand that the embodiments and descriptions of this invention are merely preferred embodiments of the invention, used to describe the basic principles and main features of the invention. The embodiments described herein are intended to help readers understand the implementation methods of the invention and should be understood as not limiting the scope of protection of the invention to such specific statements and embodiments. Those skilled in the art can modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features; however, these modifications or substitutions do not cause the essence of the corresponding technical solutions to depart from the spirit and scope of the technical solutions of the embodiments of this invention.

Claims

1. A method for predicting suitable areas for plant diseases and pests based on GIS and MaxEnt, characterized in that: Includes the following steps: S1: Identify the target pests and diseases, and obtain data on the natural geographical distribution points of the target pests and diseases; S2: Obtain environmental climate datasets under current and future climate conditions; S3: Obtain the bioclimatic variables after centralized filtering of the environmental climate dataset; S4: Based on the selected bioclimatic variables, use the MaxEnt model to calculate and analyze the main bioclimatic variables that affect the distribution of the target pests and diseases, and output the MaxEnt model prediction results; S5: Based on the prediction results of the MaxEnt model, use ArcGIS software to obtain the predicted suitable habitat maps of the target pests and diseases under the current and future climate scenarios, and evaluate the changes in the distribution of the suitable habitats of the target pests and diseases.

2. The method for predicting suitable areas for plant diseases and pests based on GIS and MaxEnt according to claim 1, characterized in that: In step S1, the methods for obtaining the natural geographical distribution data of the target pests and diseases include field surveys, literature searches, and electronic database searches.

3. The method for predicting suitable areas for plant diseases and pests based on GIS and MaxEnt according to claim 1, characterized in that: In step S1, the natural geographical distribution point data of the target pests and diseases are subjected to geographical filtering and sparsification processing to obtain the effective distribution points of the target pests and diseases, and thus obtain the geographical distribution dataset of the target pests and diseases.

4. The method for predicting suitable areas for plant diseases and pests based on GIS and MaxEnt according to claim 1, characterized in that: In step S2, environmental climate datasets under current and future climate conditions are downloaded from the WorldClim global climate database.

5. The method for predicting suitable areas for plant diseases and pests based on GIS and MaxEnt according to claim 1, characterized in that: The specific operation of step S3 is as follows: import the environmental climate dataset and the geographic distribution dataset of the target pest into ArcGIS, extract the bioclimatic variable data of the distribution points, perform Pearson correlation analysis, remove bioclimatic variables with a contribution rate of 0, remove bioclimatic variables with an absolute value of correlation coefficient greater than -.9, with a small contribution rate and no clear biological significance, and the remaining bioclimatic variables in the environmental climate dataset are the variables used in the MaxEnt model of the target pest.

6. The method for predicting suitable areas for plant diseases and pests based on GIS and MaxEnt according to claim 1, characterized in that: The specific operation of step S4 is as follows: Step S41: MaxEnt software parameter settings are as follows: Set the training set to 75%, the test set to 25%, the regularization multiplier to be determined using the ENMeval package in R software, the maximum number of iterations to 5000, the number of repeated training iterations to 10, the repeated iteration method to "Subsample", enable the function of plotting response curves and knife cutting method, the output file format is Logistic, and the output file type is selected asc. Step S42, Model Prediction Performance Evaluation, specifically includes: Step S421: Plot the ROC curve; Step S422: Calculate the AUC value, where the AUC value is the area enclosed by the ROC curve and the coordinate axis; Step S423: The evaluation criteria for different AUC values ​​are as follows: below 0.6, the model simulation results fail and have no application value; 0.6-0.7, the model simulation results are poor; 0.7-0.8, the model prediction performance is average; 0.8-0.9, the model prediction performance is good. The model's predictive performance is excellent, ranging from 0.9 to 1.

0. Step S43: Obtain the main bioclimatic variables that affect the distribution of the target pests and diseases. The main bioclimatic variables that affect the distribution of the target pests and diseases are bioclimatic variables with a high contribution rate and clear biological significance. Step S44: Output the MaxEnt model prediction results. The MaxEnt model prediction results are output in an ASC layer format. The value of each point in the layer represents the suitability probability of the target pest or disease, and the range is [0, 1].

7. The method for predicting suitable areas for plant diseases and pests based on GIS and MaxEnt according to claim 1, characterized in that: The specific operation of step S5 is as follows: Step S51: Based on the MaxEnt model, predict the suitability probability of the target pests and diseases, and divide the suitable areas into four levels: highly suitable areas, moderately suitable areas, lowly suitable areas, and unsuitable areas. Step S52: Import the prediction results of the MaxEnt model into ArcGIS software to form a prediction map of the suitable habitat for the pests and diseases; Step S53: Based on the predicted suitable habitat maps of the target pests and diseases under current and future climate scenarios, analyze and evaluate the changes in the distribution of suitable habitats of the target pests and diseases.

Images