Method for evaluating habitat suitability and diffusion trend of codling moths and application

Abstract

The invention relates to the technical field of pest control, in particular to a method for evaluating habitat suitability and diffusion trend of codling moths and application. The method comprises the following steps: S1, collecting occurrence distribution data of codling moths; s2, acquiring a plurality of pieces of bioclimate variable data, human activity influence indexes and apple yield data of codling moth occurrence distribution, constructing a multi-dimensional environmental factor data set according to the plurality of pieces of bioclimate variable data, the human activity influence indexes and the apple yield data, and performing cross validation on the data set; and S3, based on a Biomod2 set model, establishing a plurality of single ecological models according to the occurrence distribution data and the multi-dimensional environmental factor data set, selecting high-accuracy models according to TSS and AUC values of the single models, integrating the models to generate a set model, and obtaining a habitat suitability result of codling moths and a contribution rate of main environmental factors. Scientific reference is provided for early warning, monitoring and prevention of codling moths.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of pest control, more particularly, to a method for evaluating habitat suitability and diffusion trend of Cydia pomonella and application thereof. BACKGROUND

[0002] Species distribution models have wide application prospects in predicting potential geographic distribution of species and biodiversity conservation. However, although the Biomod2 ensemble model has made significant achievements in the field of plant distribution and biodiversity conservation, its application to the evaluation of habitat suitability and diffusion trend of the major invasive pest Cydia pomonella is still in a blank state.

[0003] Understanding the potential distribution of the invasive species Cydia pomonella and the dominant factors affecting its occurrence is an important basis and basis for monitoring, early warning and control of the pest. Traditionally, the monitoring and control of Cydia pomonella rely on field observation, which requires a large amount of manpower, material resources and time, and it is difficult to meet the needs of large-scale and rapid response. Species distribution models have become an important tool for predicting the impact of future climate change on species and developing effective protection strategies. However, the existing technology has significant deficiencies in the application to the study of Cydia pomonella, mainly in the following four aspects: 1. Single model selection: Most of the predictions of suitable areas for Cydia pomonella in China only use a single model, such as GARP, MaxEnt and ClIMEX. The prediction performance of a single model is highly sensitive to data quality, quantity and distribution. If the training data is biased, incomplete or has a lot of noise, the model may not accurately learn the true rules in the data, affecting the prediction effect. In addition, a single model is difficult to fully capture the complex niche characteristics of Cydia pomonella, limiting the accuracy and reliability of the prediction.

[0004] 2. Limited application of models: Ensemble models (such as Biomod2) are less used in the potential distribution of insects, especially the ensemble model for Cydia pomonella has not been reported. Ensemble models can effectively reduce the uncertainty of single models and improve the stability and accuracy of predictions by integrating the results of multiple single niche models. However, this advantage has not been fully utilized in the study of Cydia pomonella, resulting in a high degree of uncertainty in the prediction of the potential distribution of Cydia pomonella in existing studies.

[0005] 3. Single selection of environmental variables: Existing studies rely mainly on 19 bioclimatic variables (such as temperature, precipitation), but Cydia pomonella as an invasive pest, its spread is not only related to climate conditions, but also highly related to the distribution of host plants (such as apples) and human activities (such as land use, transportation network, etc.). However, current research often ignores these key factors, resulting in inaccurate prediction of suitable areas for Cydia pomonella by the model.

[0006] 4. The timescale of invasion and spread events is not precise: Most predictions for the suitable habitat of the codling moth in my country only divide the time into two stages: past and future, ignoring the complex dynamic trends of the codling moth invasion and spread events. This rough time division cannot accurately capture the spread characteristics and potential risks of the codling moth in different time periods, thus limiting a comprehensive understanding and accurate prediction of the codling moth invasion process. Summary of the Invention

[0007] In view of the shortcomings of the existing technology, this invention proposes an innovative method to overcome the limitations of the existing technology and provide a more scientific, comprehensive and accurate tool for assessing the habitat suitability and spread trend of the codling moth and making control decisions.

[0008] To achieve the above objectives, the present invention is implemented through the following technical solution: A method for assessing the habitat suitability and dispersal trend of the codling moth includes the following steps: S1. Collect data on the occurrence and distribution of codling moths; S2. Obtain data on several bioclimatic variables, human activity impact index, and apple yield for the occurrence and distribution of codling moths. Construct a multidimensional environmental factor dataset based on the aforementioned bioclimatic variables, human activity impact index, and apple yield data, and then cross-validate the dataset using Biomod2.

[0009] S3. Based on the Biomod2 ensemble model, multiple individual ecological models are established according to the occurrence distribution data and multidimensional environmental factor dataset. Based on the TSS and AUC values ​​of individual models, models with high accuracy are selected and integrated to generate an ensemble model. Finally, the habitat suitability results of the codling moth and the contribution rate of the main environmental factors are obtained.

[0010] In the preferred embodiment, in step S1, the occurrence and distribution data of codling moth are collected, and the collected occurrence and distribution data are cleaned using ENMTools software, retaining one occurrence point in each grid cell; the occurrence and distribution data includes at least the geographical location information of codling moth.

[0011] Under the preferred scheme, the occurrence and distribution data of the codling moth are obtained through one or more methods: (1) Access to the online database of the Global Biodiversity Information Facility; (2) Extract the location information of published articles and maps on the occurrence of codling moths; (3) Government documents, reports and related supporting materials.

[0012] In the preferred embodiment, correlation analysis is performed on each two of the 19 bioclimatic variables, and the bioclimatic variable data highly correlated with at least one bioclimatic variable is retained to obtain the several bioclimatic variable data in step S2.

[0013] In the preferred embodiment, the 19 bioclimatic variables of the current era corresponding to all the codling moth distribution points are extracted, the Pearson correlation coefficient between each two bioclimatic variables is calculated r , and the bioclimatic variable data with a correlation analysis result of r > 0.8 is retained.

[0014] In the preferred embodiment, the 19 bioclimatic variables include annual mean temperature, monthly mean diurnal temperature range, isothermality, temperature seasonality, maximum temperature of the warmest month, minimum temperature of the coldest month, annual range of temperature, mean temperature of the wettest quarter, mean temperature of the driest quarter, mean temperature of the warmest quarter, mean temperature of the coldest quarter, annual precipitation, precipitation of the wettest month, precipitation of the driest month, precipitation seasonality, precipitation of the wettest quarter, precipitation of the driest quarter, precipitation of the warmest quarter, and precipitation of the coldest quarter.

[0015] In the preferred embodiment, in step S2, the Biomod2 cross-validation dataset is used, and myBiomodCV<- BIOMOD_CrossValidation(bm.format = myBiomodData) is input into Biomod2 for verification, comprehensive evaluation of model performance, effective avoidance of overfitting, and improvement of model generalization ability.

[0016] In the preferred embodiment, in step S3, GLM, GBM, GAM, MARS, CTA, ANN, SRE, FDA, RF, and Maxent are established according to the occurrence distribution data and the multi-dimensional environmental factor system, model accuracy evaluation is performed on the established models, the true skill statistic (TSS) and the area under the receiver operating characteristic curve (AUC) of each model are calculated, and the single ecological model with a true skill statistic and an area under the receiver operating characteristic curve value greater than 0.9 is integrated.

[0017] In the preferred embodiment, in the process of establishing each single ecological model, the occurrence distribution data and the multi-dimensional environmental factor system are divided into a training set and a test set, multiple training is performed, and 1000 pseudo-absent points are randomly selected for multiple training.

[0018] Under the preferred scheme, the habitat suitability of the codling moth is judged according to the set model result P, the value range of P is 0-1, when 0.64 <= P <= 1.0, it indicates a highly suitable area; when 0.40 <= P < 0.64, it indicates a moderate suitable area; when 0.25 <= P < 0.40, it indicates a low suitable area; and when 0.00 <= P < 0.25, it indicates a non-suitable area.

[0019] In a second aspect of the present application, the application of the method for evaluating the habitat suitability and diffusion trend of the codling moth according to any one of the preceding schemes is provided.

[0020] The present application has the following beneficial effects: The Biomod2 model based on the R language platform has significant advantages in evaluating the habitat suitability and diffusion trend of the invasive pest codling moth, and can provide strong technical support for the monitoring and control of invasive pests, including the following specific aspects: 1. Innovative environmental variable system significantly improves prediction accuracy: By innovatively integrating the human impact index (HII) and apple yield data, a multi-dimensional environmental factor system is constructed, which significantly improves the prediction accuracy of the model for the habitat suitability and diffusion trend of the codling moth. This multi-dimensional environmental factor system not only considers climatic factors, but also combines the influence of host plant distribution and human activities, and can more comprehensively reflect the habitat requirements of the codling moth.

[0021] 2. Application of ensemble model enhances the stability and reliability of the model: The Biomod2 ensemble model effectively reduces the uncertainty of single models (such as GLM, GAM, RF, MARS, etc.) by integrating multiple single models. This integrated method can combine the advantages of different models and reduce the prediction error caused by data bias or unreasonable model assumptions.

[0022] 3. Provides long-term early warning function: Combined with future climate scenario data, this method can realize long-term early warning of the suitable area of the codling moth. This provides a scientific basis for the monitoring and control of invasive pests, helping relevant departments to plan and deploy control measures in advance. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 is a Pearson correlation analysis heat map of 19 bioclimatic variables.

[0024] Figure 2 is the accuracy evaluation of the model algorithm of the codling moth at different invasion periods (a: 1957-1987, b: 1957-2006, c: 1957-2012, d: 1957-2022).

[0025] Figure 3is the precision evaluation of the model algorithm of the codling moth under the future climate scenarios of 2021-2040 and 2041-2060 (a: 2021-2040-SSP126, b: 2021-2040-SSP245, c: 2021-2040-SSP585, d: 2041-2060-SSP126, e: 2041-2060-SSP245, f: 2041-2060-SSP585).

[0026] Figure 4 is the suitable area of the codling moth in China from 1957 to 2022.

[0027] Figure 5 is the suitable area of the codling moth in China under three climate scenarios of 2021-2040 and 2041-2060 (a: 2021-2040, b: 2041-2060). DETAILED DESCRIPTION

[0028] The technical solutions in the embodiments of the present application will be described below in a clear and complete manner. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should fall within the scope of protection of the present application.

[0029] The main purposes of the present application are as follows: 1. Optimizing the prediction ability of the model: by integrating the results of multiple single-niche models, the advantages of the Biomod2 ensemble model are utilized to significantly improve the accuracy and reliability of the prediction of the suitable area of the codling moth.

[0030] 2. Building a multi-dimensional environmental factor system: the human impact index (HII) and apple yield data are introduced, combined with traditional climate variables, to build a more comprehensive multi-dimensional environmental factor system, fully considering the complex mechanism of the spread of the codling moth.

[0031] 3. Accurately analyzing the spread and harm of the codling moth in the main apple-producing areas in China, the present application divides the past four different invasion periods (1957-1987, 1957-2006, 1957-2012 and 1957-2022) and the potential suitable area distribution prediction of the codling moth in China under three scenarios (SSP126, SSP370 and SSP585) in the future two time periods (2021-2040 and 2041-2060).

[0032] 4. Provide scientific prevention and control basis: Through accurate assessment of the habitat suitability and diffusion trend of the apple codling moth, provide strong support for formulating scientific and effective prevention and control strategies, meet the needs of large-scale and rapid response, reduce monitoring and control costs, and improve prevention and control efficiency.

[0033] Therefore, the Biomod2 ensemble model is applied to the habitat assessment of the apple codling moth in the present application, the results of multiple single ecological niche models are integrated, and the complex diffusion mechanism of the apple codling moth is optimized; the human influence index (HII) and apple yield data are introduced, combined with traditional climate variables, to construct a multi-dimensional environmental factor system, which significantly improves the accuracy of the model in predicting the suitable habitat of the apple codling moth, and provides a more scientific basis for prevention and control strategies. The present application aims to fill the gaps in the prior art and provide an innovative, practical and scientific solution for the ecological risk assessment and prevention and control decision of the invasive pest apple codling moth.

[0034] Embodiment 1

[0035] A method for assessing the habitat suitability and diffusion trend of the apple codling moth, comprising the following steps: S1, data collection and cleaning: As shown in Table 1, the occurrence distribution records of the apple codling moth are collected from multiple sources to ensure the completeness and reliability of the data, and the data is cleaned to remove redundant and repeated occurrence distribution records.

[0036] In step S1, the occurrence records of the apple codling moth mainly refer to geographical location information, which comes from the following ways: (1) Existing open source data: obtained through the Global Biodiversity Information Facility online database.

[0037] (2) Use published articles and maps of apple codling moth occurrence to extract occurrence location information.

[0038] (3) Government documents, reports and related auxiliary materials.

[0039] In step S1, the data can be screened using ENMTools software to ensure that there is at most one occurrence point in each grid, and the cleaned occurrence distribution data is saved as a csv file for future use.

[0040] S2, environmental factor data acquisition and screening: The environmental factor data that may affect the occurrence and distribution of the codling moth are acquired, including bioclimatic variables (such as temperature, precipitation), host plant distribution (such as apple yield), and human activity impact index (such as land use, transportation network, etc.). Through correlation analysis and stepwise screening, redundant variables are removed, and environmental factors that significantly affect the distribution of the codling moth are retained, a multi-dimensional environmental factor data set is constructed, and the data set is cross-validated by Biomod2. In step S2, the bioclimatic variables in the environmental factor data refer to the current and future 19 bioclimatic variables, which are from the WorldClim database (https: / / worldclim.org / ). The future bioclimatic variables are selected from the bottom maps of the sustainable development path (SSP126), the intermediate path (SSP245) and the fossil fuel-based development path (SSP585) of the CMIP6 model adopted in the Sixth Assessment Report of IPCC for the period from 2021 to 2040 and from 2041 to 2060, and the spatial resolution is 2.5 min.

[0041] In addition to the 19 bioclimatic variables provided by the WorldClim database, the present application adds the following two types of key variables: human impact index (HII): from the Socioeconomic Data and Applications Center (https: / / sedac.ciesin.columbia.edu / ), quantifying the promoting effect of human activities (such as land use, transportation network) on the spread of the codling moth; apple yield data (app): from EarthStat (http: / / www.earthstat.org / ), reflecting the driving effect of the host on the colonization of the codling moth.

[0042] In step S2, ArcGIS 10.8 is used to extract the values of the 19 bioclimatic variables of the current era corresponding to all the distribution points of the codling moth, and SPSS (26.0) is used to calculate the Pearson correlation coefficient between each two bioclimatic variables to evaluate the multicollinearity between the bioclimatic variables. For two bioclimatic variables with a correlation coefficient | r |>0.8, only the one with higher biological correlation is retained for subsequent analysis.

[0043] In step S2, the data set is cross-validated by Biomod2, and myBiomodCV<-BIOMOD_CrossValidation(bm.format = myBiomodData) can be input into Biomod2 for validation.

[0044] S3, ensemble model construction and precision evaluation: The Biomod2 ensemble model was used to integrate the results of multiple single-niche models (such as GARP, MaxEnt, RF, etc.), reducing the uncertainty of single models and improving the stability and accuracy of the prediction. The habitat suitability results of the codling moth and the contribution rate of the main environmental factors were obtained to comprehensively analyze the influence of each environmental factor on the distribution of the codling moth.

[0045] In step S3, the ensemble model was constructed and the accuracy was evaluated. First, based on the occurrence data of the codling moth in China and the environmental variables, 10 different models were used in Biomod2: GLM (generalized linear model), GBM (propulsion regression tree model), GAM (generalized additive model), MARS (multivariate adaptive regression spline), CTA (classification tree analysis model), ANN (artificial neural network), SRE (surface distribution differentiation chamber model), FDA (flexible discriminant analysis model), RF (random forest model), and Maxent (maximum entropy model). For each modeling process, the parameter settings were made in Biomod2, and 75% of the distribution records were randomly selected as the training data set, and the remaining 25% as the test data set. The training and test data were repeated 5 times. In addition, 1000 pseudo-absent points were randomly selected and repeated 3 times. The code get_evaluations(myBiomodEM) was input in Biomod2 to obtain the true skill statistic (TSS) and the area under the receiver operating characteristic curve (AUC) values, and the accuracy of the model was tested based on the true skill statistic (TSS) and the area under the receiver operating characteristic curve (AUC). TSS takes into account the average error of missing data and is not affected by the size of the validation data set. Generally, the higher the values of these two indicators, the higher the accuracy of the model results. The AUC and TSS values of the 10 single models were compared and analyzed, and the single models with AUC and TSS values both >0.9 were selected to construct the ensemble model.

[0046] In step S3, the habitat suitability of the codling moth was determined based on the results of the ensemble model. The ensemble model result picture was imported into ArcGIS 10.7, and the P value could be seen. The value of P ranged from 0 to 1. According to the description of the IPCC-CMIP5 (IPCC, 2014) on the "probability" of species, the suitable area was divided into four types: high suitability area (0.64 ≤ P ≤ 1.0), moderate suitability area (0.40 ≤ P <0.64), low suitability area (0.25 ≤ P <0.40), and non-suitability area (0.00 ≤ P <0.25).

[0047] In step S3, the habitat suitability and diffusion trend results also included the area calculation of different suitable areas. The attribute table of the codling moth suitable area distribution map was opened using ArcGIS 10.8, and the area of the suitable area was calculated by multiplying the number of pixel units by the area of each pixel unit.

[0048] In step S3, the main environmental factor contribution rate is calculated by inputting the code get_variables_importance(myBiomodEM) in Biomod2, and then the data is arranged in excel.

[0049] The present application first applies the Biomod2 ensemble model to the habitat suitability and diffusion trend evaluation of the apple codling moth, fills the gap in the research of invasive pests in this field. By innovatively integrating the human influence index (HII) and apple yield data, a multi-dimensional environmental factor system is constructed, which significantly improves the prediction accuracy. In addition, the present application also realizes the long-term early warning of the apple codling moth suitable area under future multiple climate scenarios, which provides a strong scientific basis for the monitoring and control of invasive pests.

[0050] Example 2

[0051] Based on the method provided in Example 1, the model of the present application uses 225 pieces of apple codling moth distribution data obtained from 9 provinces (regions, cities) and 204 counties (cities) in China. The distribution points are screened using ENMTools, so that there is only one apple codling moth distribution point in a 2.5 km x 2.5 km grid. As shown in Table 1, 203 pieces of apple codling moth geographic distribution data are finally obtained. As shown in Table 1, 203 pieces of apple codling moth geographic distribution data are finally obtained. Figure 1 As shown in Table 2, the Pearson correlation analysis heat map of 19 bioclimatic variables is shown. The correlation analysis of 19 bioclimatic variables is performed to eliminate multicollinearity. For any two highly correlated environmental variables (| r |>0.8), the variable with more biological significance is retained, and finally 9 environmental factor data including apple yield data and human influence index are used. Among them, the retained bioclimatic variables are shown in Table 2.

[0052] Table 1: Final retained apple codling moth occurrence distribution records

[0053]

[0054]

[0055]

[0056]

[0057]

[0058]

[0059] Table 2: 19 bioclimatic variables and retained bioclimatic variables

[0060] The evaluation accuracy of ANN, CTA, FDA, GAM, GBM, GLM, MARS, MAXENT, RF and SRE was calculated, among which the generalized linear model (GBM), generalized boosted model (GLM), random forest (RF) and multivariate adaptive regression spline (MARS) were the optimal models for predicting the habitat distribution of the codling moth, and the surface distribution area division model (SRE) was the worst model and failed to pass the model accuracy test. The four models with AUC and TSS values exceeding 0.9 were selected to construct an ensemble model. The average AUC and TSS of the ensemble model were 0.995 and 0.948, respectively, which improved the fitting accuracy of the ensemble model and reduced the uncertainty of the ensemble model fitting, indicating that the prediction result was reliable.

[0061] (1) Habitat suitability and expansion trend of the codling moth in China from 1957 to 2022 As shown in Table 3, from 1957 to 1987, the codling moth spread to the entire fruit production area in Xinjiang, with a total suitable area of 120.21×10 4 km 2 , including a highly suitable area of 39.24×10 4 km 2 , a moderately suitable area of 41.99×10 4 km 2 , and a lowly suitable area of 38.98×10 4 km 2 . From 1957 to 2006, the codling moth invaded Gansu Province and spread southeastward, reaching Inner Mongolia bordering Gansu. The total suitable area reached 173.75×10 4 km 2 , including a highly suitable area of 62.56×10 4 km 2 , a moderately suitable area of 54.02×10 4 km 2 , and a lowly suitable area of 57.17×10 4 km 2 . From 1957 to 2012, the codling moth continued to spread in the western part of China. In the eastern part of China, in 2006, the suitable area of the codling moth was distributed in Heilongjiang Province bordering Russia, and it spread westward and southward to Jilin Province. In this stage, the total suitable area reached 314.12×10 4 km 2 , an increase of 140.37×10 4 km 2; the area of high suitable habitat was 105.99×10 4 km 2 ; the area of moderate suitable habitat was 111.74×10 4 km 2 ; the area of low suitable habitat was 96.39×10 4 km 2 . From 1957 to 2022, the distribution of the apple codling moth in the eastern region showed a jump-like expansion, and the suitable habitat was distributed in Liaoning, Jilin, Tianjin, Hebei, etc. In this stage, the total suitable habitat area reached 390.65×10 4 km 2 , an increase of 76.53×10 4 km 2 compared with 2012, the area of high suitable habitat was 149.63×10 4 km 2 , an increase of 43.64×10 4 km 2 compared with 2012; the area of moderate suitable habitat was 125.16×10 4 km 2 , and the rest was low suitable habitat.

[0062] Under the future climate scenario, the apple codling moth expansion prediction in China is shown in Table 4. From 2021 to 2040, the total suitable habitat area under the three climate scenarios decreased compared with the current situation. Among them, the area of moderate and low suitable habitat decreased, but the area of the most suitable habitat increased, with the most suitable habitat area under the SSP126, SSP245, and SSP585 climate scenarios increasing by 8.53×10 4 km 2 , 24.66×10 4 km 2 , and 10.61×10 4 km 2 , respectively. From 2041 to 2060, the total suitable habitat area decreased significantly, especially in the northeast and north China. Under the SSP126 climate scenario, the area of low suitable habitat was 99.90×10 4 km 2 , the area of moderate suitable habitat was 94.74×10 4 km 2 , and the area of the most suitable habitat was 88.75×10 4 km 2 . Under the SSP245 climate scenario, the area of low suitable habitat was 93.68×10 4 km 2 , the area of moderate suitable habitat was 139.10×10 4 km 2, the low suitable area was 89.60 x 10 4 km 2 ; under SSP585 climate scenario, the low suitable area was 107.38 x 10 4 km 2 , the moderate suitable area was 108.13 x 10 4 km 2 , and the most suitable area was 79.80 x 10 4 km 2 (Table 3).

[0063] Table 3: Current and future different suitable area of C. pomonella

[0064] Table 4: Current and future different suitable area distribution of C. pomonella

[0065]

[0066]

[0067]

[0068]

[0069] Note: “—” represents no distribution data.

[0070] Through the contribution rate analysis of environmental variables of C. pomonella geographic distribution, we found that the contribution rate of each variable in different time periods was different (Table 5). Overall, HII and precipitation-related variables (such as Bio16, Bio17) had higher contribution rates in most periods, indicating that human activities and precipitation conditions were the key factors affecting the geographic distribution of C. pomonella, while the contribution rates of temperature-related variables (such as Bio07, Bio10) and apple yield were relatively low and fluctuated greatly. Under different climate scenarios in the future, the contribution rates of various factors to the distribution of C. pomonella suitable area were different. Overall, the influence of climate factors and human activities on the distribution of C. pomonella suitable area was different, among which HII, Bio11 and Bio16 were outstanding in most scenarios.

[0071] Table 5: Contribution rate of environmental variables to the distribution of C. pomonella suitable area

[0072] The application creatively proposes an apple codling moth habitat suitability assessment method based on Biomod2 model combined with ArcGIS on the R language platform, which is the first time in this field. The method comprehensively considers multiple key factors affecting the spread of apple codling moth, not only considers the application of 19 bioclimatic variables, but also specially applies to the distribution of host plants (apple) and the influence index of human activities (land use, traffic network and other factors), and comprehensively analyzes the role mechanism of environmental factors in the spread of apple codling moth. Through model accuracy evaluation, a set model is constructed by selecting a single model with high TSS and AUC values, and the habitat suitability and spread trend of apple codling moth can be accurately evaluated, thereby providing strong support for formulating scientific and effective prevention and control strategies. This innovative method not only fills the research gap in the field of apple codling moth habitat assessment, but also provides a new paradigm for the ecological risk assessment of other invasive pests, which has important theoretical and practical value.

[0073] The above is only the preferred specific embodiment of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can make equivalent replacement or change according to the technical solution and the inventive concept of the present application within the technical range disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A method of assessing the habitat suitability and spread tendency of the codling moth, characterized in that, The method comprises the following steps: S1, collecting occurrence distribution data of the codling moth; S2, obtaining a plurality of biological climate variable data, a human activity influence index and apple yield data of the occurrence distribution of the codling moth, constructing a multi-dimensional environmental factor data set according to the plurality of biological climate variable data, the human activity influence index and the apple yield data, and verifying the data set by Biomod2; S3, based on the Biomod2 ensemble model, establishing a plurality of single ecological models according to the occurrence distribution data and the multi-dimensional environmental factor data set, selecting a model with high accuracy according to the TSS and AUC values of the single model, and integrating the selected model to generate an ensemble model, thereby obtaining a habitat suitability result of the codling moth and a contribution rate of a main environmental factor.

2. The method of assessing the habitat suitability and spread tendency of the codling moth according to claim 1, characterized in that, In step S1, the occurrence distribution data of the codling moth is collected, and the collected occurrence distribution data is cleaned by using ENMTools software, and one occurrence point is retained in each grid. The occurrence distribution data at least includes geographical position information of the codling moth.

3. A method of assessing the habitat suitability and spread tendency of the codling moth according to claim 1 or 2, characterized in that, The occurrence distribution data of the codling moth is obtained through one or more of the following ways: (1) Global Biodiversity Information Facility online database; (2) Published articles and maps on occurrence of the codling moth, and extraction of occurrence location information; (3) Government documents, reports and related auxiliary materials.

4. The method of claim 1, wherein the habitat suitability and spread tendency of the codling moth is evaluated, and Correlation analysis is performed on each two of the 19 biological climate variable data, and the biological climate variable data highly correlated with at least one biological climate variable is retained, thereby obtaining the plurality of biological climate variable data in step S2.

5. The method of claim 4, wherein the habitat suitability and spread tendency of the codling moth is evaluated by using the following formula: ###0001### wherein, S = habitat suitability of the codling moth, D = spread tendency of the codling moth, and P = the number of the codling moth eggs laid per female. The Pearson correlation coefficient r between each two biological climate variables is calculated by extracting the 19 biological climate variable values of the current era corresponding to all distribution points of the codling moth, and retaining the biological climate variable data with a correlation analysis result of |r|>0.8 with at least one biological climate variable; The 19 biological climate variable data includes annual average temperature, monthly average diurnal temperature difference, isothermality, temperature seasonal change, maximum temperature in the warmest month, minimum temperature in the coldest month, annual temperature variation range, average temperature in the wettest quarter, average temperature in the driest quarter, average temperature in the warmest quarter, average temperature in the coldest quarter, annual rainfall, rainfall in the wettest month, rainfall in the driest month, rainfall seasonal change, rainfall in the wettest quarter, rainfall in the driest quarter, rainfall in the warmest quarter and rainfall in the coldest quarter.

6. The method of claim 1, wherein the habitat suitability and spread trend of the codling moth is evaluated, and In step S2, the data set is verified by Biomod2, the model performance is comprehensively evaluated, overfitting is effectively avoided, and the generalization ability of the model is improved.

7. The method for assessing the habitat suitability and dispersal trend of the codling moth according to claim 1, characterized in that, In step S3, GLM, GBM, GAM, MARS, CTA, ANN, SRE, FDA, RF and Maxent are established according to the occurrence distribution data and the multi-dimensional environmental factor system, the accuracy of the established models is evaluated, the true skill statistic (TSS) and the area under the receiver operating characteristic curve (AUC) of each model are calculated, and the single ecological model with a true skill statistic and an area under the receiver operating characteristic curve value greater than 0.9 is integrated.

8. The method of claim 7, wherein the habitat suitability and spread tendency of the codling moth is evaluated, and In each single ecological model, the occurrence distribution data and multi-dimensional environmental factor system are divided into training set and test set, and multiple training is performed, and then 1000 pseudo-absent points are randomly selected for multiple training.

9. The method of claim 1, wherein the habitat suitability and spread tendency of the codling moth is evaluated, and According to the set model result P, the habitat suitability of the codling moth is judged, the value range of P is 0-1, when 0.64≤P≤1.0, it indicates a highly suitable area; when 0.40≤P<0.64, it indicates a moderately suitable area; when 0.25≤P<0.40, it indicates a low suitable area; and when 0.00≤P<0.25, it indicates a non-suitable area. ​ 10. An application of the method for evaluating the habitat suitability and diffusion trend of the codling moth according to any one of claims 1-9.

Images