Italian locust pest occurrence probability prediction method and device, medium and equipment

Abstract

The invention provides an Italian locust pest occurrence probability prediction method and device, a medium and equipment, and relates to the technical field of agricultural insect pests, and the method comprises the steps: obtaining historical occurrence point data and environmental data of Italian locust pests, carrying out the relative registration and preprocessing, and obtaining environmental factors; through a MaxEnt model and a Pearson correlation coefficient, environment element variables related to Italian locust damage are extracted, and key environment factors are obtained; decision analysis is selected to carry out reclassification and weighting processing on the key environment factors to obtain environment factor indexes, and an Italian locust pest occurrence probability prediction model is constructed based on a multi-criterion analysis method; and the weather forecast data and the vegetation data of the to-be-evaluated region are sent to the Italian locust pest occurrence probability prediction model, a corresponding Italian locust pest occurrence prediction result is obtained, and a grade prediction result is obtained in combination with a set grade division rule. And a new thought and reference are provided for remote sensing dynamic monitoring and prediction of grassland locusts.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural pest control, and particularly relates to a method, device, medium and equipment for predicting the occurrence probability of Calliptamus italicus pests. Background Art

[0002] Desert grassland is the zonal grassland type in a certain area in the west. Its locust area has the largest distribution area, which is 20,581 km 2 , accounting for 52% of the total locust area distribution in the west of this area, ranking first. Such grasslands are often inhabited by dominant species such as Calliptamus italicus, Pseudochorthippus parallelus, and Calliptamus barbarus.

[0003] Calliptamus italicus (Linnaeus, 1758) is a common pest in grasslands, deserts, and semi-deserts in Central Asia and surrounding areas. Due to its long-distance migration characteristics, it has caused serious damage to the local grassland ecosystem and agricultural production. The outbreak of locusts is often international, cross-border, and mobile, which will cause devastating damage to agriculture and animal husbandry and destroy the balance of the ecosystem. As one of the regions most sensitive to global change, the intensity and frequency of extreme climate events in a certain area in the west both show an increasing trend, which makes the habitat of locusts change violently and increases the possibility of locust plagues. The hatching period of Calliptamus italicus is in May. The nymph development period is from early May to mid-July. It emerges in early and mid-June and lays eggs in early and mid-July, and sometimes it can be delayed until August. It mainly feeds on Trigonella foenum-graecum, Medicago falcata, Cirsium setosum, corn, rice, cotton, etc. The hatching of Calliptamus italicus eggs has a certain relationship with weather conditions, as well as soil temperature and humidity. During the hatching period, if the weather turns cloudy and the temperature drops, the hatching rate will decrease significantly. In extreme weather such as rainy or snowy days, the locust eggs will not hatch, but after the rain turns clear and the temperature rises, the hatching rate will increase significantly.

[0004] The latest research finds that climate change will lead to a general increase in locust outbreaks. New hotspots have emerged in Central Asia, bringing additional challenges to global locust control coordination. In recent years, the temperature in most parts of a certain area in the west has been relatively high, and the locust occurrence period is 7-10 days earlier than that of the previous year. The development period is advanced, resulting in a relatively large occurrence area in some local areas, bringing many difficulties to the prevention and control work. In addition, the cost of spraying chemical agents is high, and the labor intensity is large. The most terrible thing is that it may also cause adverse changes to the environment. Therefore, the monitoring and early warning of grassland locusts are an important part of ensuring national food and ecological security. Especially in the context of global warming, biological invasions caused by climate change cannot be ignored.

[0005] The work of locust monitoring and early warning must be carried out before the pests spread and migrate. Accurately extracting the suitable habitats of typical locusts, dynamically predicting their spread, delimiting their migration routes, and then determining the control areas and control trends are the keys to effective locust control. At present, locust monitoring faces various difficulties, including a wide distribution range, which requires a large amount of manpower for investigation; changes in ecological conditions make the habitats and migration patterns of locusts unpredictable, increasing the complexity of monitoring; many areas lack effective early warning mechanisms, resulting in the inability to respond in a timely manner when agriculture and social economy are damaged; the research on the biological characteristics of locusts such as their life cycle and breeding habits is still not deep enough, affecting the formulation of monitoring and control methods, etc. Remote sensing technology can monitor through satellite images and drones, covering large geographical areas, and quickly track the migration and spread of locusts, especially in cases where locusts are widely distributed and traditional monitoring methods cannot fully cover. Summary of the Invention

[0006] In order to overcome the above-mentioned disadvantages that traditional monitoring methods cannot fully cover when locusts are widely distributed, the main purpose of the present invention is to provide a method, device, medium and equipment for predicting the occurrence probability of Calliptamus italicus pests.

[0007] To achieve the above object, the present invention adopts the following technical solutions. A method for predicting the occurrence probability of Calliptamus italicus pests includes the following steps:

[0008] Obtain the historical occurrence point data and environmental data of Calliptamus italicus pests, perform relative registration and preprocessing on the historical occurrence point data and the environmental data to obtain environmental factors;

[0009] Extract the environmental factors related to Calliptamus italicus pests through the MaxEnt model and Pearson correlation coefficient to obtain key environmental factors;

[0010] Reclassify and assign weights to the key environmental factors to obtain environmental factor indicators. Based on the multi-criteria analysis method, combine the environmental factor indicators and the corresponding weights to construct a prediction model for the occurrence probability of Calliptamus italicus pests;

[0011] Obtain the corresponding environmental factor indicators of the area to be evaluated, input them into the prediction model for the occurrence probability of Calliptamus italicus pests, and obtain the corresponding prediction results for the occurrence of Calliptamus italicus pests.

[0012] It also includes: after obtaining the corresponding prediction results for the occurrence of Calliptamus italicus pests, combine the set grade division rules to obtain the grade prediction results for the occurrence of Calliptamus italicus pests.

[0013] The environmental data includes: meteorological data, vegetation data, soil data and terrain data.

[0014] The obtaining of the key environmental factors includes:

[0015] The jackknife method in the MaxEnt model was used to evaluate the relative contributions of environmental factors, and the main environmental factors affecting the geographical distribution of Calliptamus italicus were obtained;

[0016] Based on the Pearson correlation coefficient, the correlation between the main environmental factors was obtained. According to the set prefabricated range, key environmental factors were selected until all key environmental factors were obtained.

[0017] The above-mentioned selection and analysis decision reclassifies and assigns weight to the key environmental factors to obtain environmental factor indicators; it includes the following steps:

[0018] Select AHP, based on the indicators affecting the occurrence and survival of Calliptamus italicus; reclassify in combination with the attributes of the key environmental factors to obtain impact indicators;

[0019] Based on the relative importance of each impact indicator, a hierarchical structure arrangement is made to construct a judgment matrix;

[0020] According to the judgment matrix, the weights of each impact indicator are obtained;

[0021] Based on the impact indicators and the corresponding weights, environmental factor indicators are obtained.

[0022] The set grading rules are as follows:

[0023] When the prediction result of the occurrence of Calliptamus italicus pests is between 0.8 - 1.0, the area to be evaluated is highly suitable;

[0024] When the prediction result of the occurrence of Calliptamus italicus pests is between 0.6 - 0.8, the area to be evaluated is moderately suitable. When the prediction result of the occurrence of Calliptamus italicus pests is between 0.4 - 0.6, the area to be evaluated is lowly suitable;

[0025] When the prediction result of the occurrence of Calliptamus italicus pests is between 0 - 0.4, the area to be evaluated is unsuitable.

[0026] A device for predicting the occurrence probability of Calliptamus italicus pests, comprising:

[0027] An impact factor acquisition module; used to obtain the historical occurrence point data and environmental data of Calliptamus italicus pests, perform relative registration and preprocessing to obtain environmental factors; through the MaxEnt model and the Pearson correlation coefficient, extract environmental element variables related to Calliptamus italicus pests to obtain key environmental factors;

[0028] A prediction model construction module, used to reclassify and assign weights to the key environmental factors to obtain environmental factor indicators, and based on the multi-criteria analysis method, combine the environmental factor indicators and the corresponding weights to construct a prediction model for the occurrence probability of Calliptamus italicus pests;

[0029] A probability prediction module is used to obtain the corresponding environmental factor indicators of the area to be evaluated, input them into the Italian locust pest occurrence probability prediction model, obtain the corresponding prediction results of the Italian locust pest occurrence, and combine the set grade division rules to obtain the grade prediction results of the Italian locust pest occurrence.

[0030] Compared with the prior art, the beneficial effects of the present invention are as follows: In the areas where Italian locusts are widely distributed, based on fully considering the environmental factors for the occurrence of Italian locusts, the environmental element variables related to Italian locust pests are obtained as key environmental factors. Then, the analytic hierarchy process is selected to assign weights to the key environmental factors, and a multi-criteria analysis method is used to construct an Italian locust pest occurrence probability prediction model. According to the actual investigation occurrence sample points of Italian locust pests, the Italian locust pest occurrence prediction model is determined. Real-time meteorological numerical forecasts are obtained, combined with the constructed Italian locust pest occurrence probability prediction model, and finally, the short-term prediction of Italian locust pest occurrence is realized, providing new ideas and references for the remote sensing dynamic monitoring and prediction of grassland locusts. Brief Description of the Drawings

[0031] The drawings described herein are used to provide a further understanding of the present application and form a part of the present application. The schematic embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation to the present application.

[0032] Figure 1 It is a flowchart of a method for predicting the occurrence probability of Italian locust pests according to the present invention. Detailed Embodiments

[0033] In order to make the objectives, technical solutions and advantages of the present invention clearer, the following will Figure 1 , in conjunction with the appended

[0034] drawings, the detailed embodiments of the present invention will be described in detail. It should be understood that the specific embodiments described herein are only used to more clearly illustrate the technical solutions of the present invention and do not limit the protection scope of the present invention. That is, the specific embodiments described are only a part of the embodiments of the present invention, rather than all of the specific embodiments. Usually, the components of the specific embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations, and the present invention can also have other embodiments.

[0035] To further understand the content, features and effects of the present invention, a certain autonomous region in China where Calliptamus italicus is widely distributed was taken as the study area, and the method of the present invention was used to verify the method in the study area. The steps are as follows:

[0036] In the first step, data on historical occurrence points of Calliptamus italicus, meteorological data, vegetation data, soil data and terrain data were collected, and relative registration and preprocessing of the data in the study area were carried out to obtain the preprocessed data;

[0037] 1) Define the specific study area and collect the accurate vector range of the study area;

[0038] 2) According to the scope and size of the specific study area, data on historical occurrence points of Calliptamus italicus pests were collected. The data were sourced from the Global Biodiversity Information Facility, field sampling data and extracts from public literature. The data on historical occurrence points of Calliptamus italicus were collected in accordance with the principle of "more rather than less". A total of 502, 504 and 676 historical occurrence point data of Calliptamus italicus were collected from 2015 to 2017.

[0039] 3) Data cleaning was carried out on the collected data on historical occurrence points of Calliptamus italicus pests by deleting duplicate occurrence records and deleting missing georeference records; the data on historical occurrence points of Calliptamus italicus pests after data cleaning were spatially filtered within a 5 km × 5 km area using the "Spatially Sparse Occurrence Data" plug-in in the SDMtoolbox tool to reduce spatial correlation and eliminate spatial clustering positions during model calibration and evaluation;

[0040] 4) According to the scope and size of the specific study area, meteorological data, vegetation data, soil data and terrain data related to the occurrence of Calliptamus italicus pests were collected, and the data were downloaded from multiple relevant official platforms respectively. The monthly average temperature, monthly relative humidity, monthly average precipitation, monthly soil temperature (0 - 30 cm), monthly soil humidity (0 - 30 cm), normalized difference vegetation index NDVI, soil pH value, soil sand content, soil silt content, soil clay content and DEM elevation data covering the study area were obtained. For details, see Table 1.

[0041] Table 1 Introduction to environmental factor data

[0042]

[0043] 5) Use the multi-dimensional tool in ArcGIS10.7 software - create a NetCDF raster layer to convert the NetCDF data of meteorological data into Geotiff format.

[0044] 6) Obtain the latest MOD09A1 data on the Google Earth Engine (GEE) platform, synthesize the data within the study area, and calculate the NDVI. The definition of NDVI is as follows:

[0045]

[0046] In the formula, Nir and Red are the reflectance of the 5th and 4th bands respectively, and their value ranges are between [-1, 1]. Usually, positive values indicate areas covered by vegetation, and the larger the value, the higher the vegetation coverage; 0 indicates areas with basically no vegetation coverage; negative values generally represent non-vegetation-covered areas such as water bodies, bare soil, and rocks.

[0047] 7) Use the Resampling tool in the Data Management Tools - Raster - Raster Processing in ArcGIS 10.7 software. Based on the spatial resolution of the study area, perform simple upscaling and downscaling on the raster datasets from multiple information sources to unify the study scale (WGS_1984_UTM_Zone_45N, 1000m);

[0048] 8) Use the Slope - Aspect tool in the 3D Analyst tool - Raster Surface in ArcGIS 10.7 software based on the DEM elevation data to extract the slope and aspect elements of the study area and unify them to the same coordinate system;

[0049] 9) Use the Clip tool in the Data Management Tools - Raster - Raster Processing in ArcGIS 10.7 software to clip the environmental factor datasets of relevant raster types based on the accurate vector range of the study area to reduce data redundancy.

[0050] In the second step, combine the MaxEnt model and the Pearson correlation coefficient to analyze the substantial impact of environmental factor variables on the locust pest in Italy, extract the environmental variables with the greatest influence on the locust pest in Italy, and construct an environmental dataset for the occurrence of the locust pest in Italy, specifically including:

[0051] 1) Use the MaxEnt model to analyze the preprocessed environmental factors, evaluate the relative contributions of environmental variables through the Jackknife method, and explore the main environmental factors affecting the geographical distribution of the locust in Italy, specifically including:

[0052] ① Use the Raster to ASCll (Folder) tool in the Basic Tools of SDMToolbox v2.6 to convert the preprocessed environmental data into the ASCII format required by the MaxEnt model;

[0053] Import the initial environmental factors and the historical occurrence location data of Calliptamus italicus into the MaxEnt model. Open the MaxEnt 3.4.4 software. In the file selection box of "Samples" in the upper left corner of the interface, select the species distribution data of Calliptamus italicus and check the target species. In the file selection box of "Environment allayers" in the upper right corner of the interface, select the folder storing the environmental variables of Calliptamus italicus, and the variables can be automatically read. Check all the environmental factors, select "Create response curves" to generate response curves, "Make picture of predictions" to make prediction maps, and "Do jackknife to measure variable importance" to measure the importance of each factor using the jackknife method. For the "Output format", select Logistic, and for the "Output file type", select asc, and set the output path.

[0054] ②Open the "Setting" option, select the "Basic" interface, and check the random seed "Random seed", give visual warnings "Give visual warnings", show tooltips "Show tooltips", ask before overwriting "Ask before overwriting", remove duplicate presence records "Remove duplicate presence records", write clamp grid when projecting "Write clamp grid when projecting", do MESS analysis when projecting "Do MESS analysis when projecting". Randomly select 75% of the distribution data of Calliptamus italicus for model training, and the remaining 25% of the data is used for testing. Set 10,000 background points, execute 10 times repeatedly, and run repeatedly for cross-validation to ensure the accuracy of the model. Select the "Advanced" interface, and check add samples to background "Add samples to background", write plot data "Write plot data", extrapolate "Extrapolate", do clamping "Do clamping", write output grids "Write output grids", write plots "Write plots", cache ascii files "Cache ascii files". Set the maximum number of iterations "Maximum iterations" to 5,000 times, and select the maximum training sensitivity plus specificity "Maximum training sensitivity plus specificity" for the apply threshold rule "Apply threshold rule". After closing the "Setting" option, click Run "Run".

[0055] ③ Use the area under the receiver operating characteristic curve (AUC) value to evaluate the quality of the model simulation results. The size of the AUC value reflects the prediction ability of the model. The larger the AUC value, the better the prediction ability of the model. Refer to the criteria of Swets (1988): 0.5 < AUC < 0.7 indicates general prediction ability; 0.7 < AUC < 0.9 indicates good prediction ability; 0.9 < AUC < 1 indicates excellent prediction ability. AUC is used to measure the overall performance of the model. Its value is generally between 0.5 and 1. When it is equal to 0.5, it is equal to the AUC value of the random prediction model. When it is equal to 1, it has the best judgment ability. However, the value of AUC can also be lower than the range of 0.5 - 1, which means that the model used for prediction is worse than the random prediction model. In the present invention, the AUC values of the MaxEnt model running from 2015 to 2017 are 0.945, 0.959, and 0.946 respectively, indicating that the model running results are good.

[0056] 2) To eliminate redundancy between variables, obtain the correlation between environmental factors based on Pearson correlation. To reduce the autocorrelation between environmental factors, when the Pearson correlation coefficient is less than 0.8, the environmental variable is retained; otherwise, the variable is screened. When the Pearson coefficients of two variables are both greater than 0.8, the environmental variable with relatively smaller contribution is excluded according to the contribution degree. The calculation expression of the Pearson correlation coefficient r is:

[0057]

[0058] where m i and z i respectively represent the values of the two variables at the i-th observation value, and respectively represent the means of the two variables;

[0059] 3) Through repeated screening of the relative contributions of environmental variables and correlation coefficient analysis until the correlation coefficients of all environmental factors are < 0.8. Finally, relative humidity (RH), elevation (DEM), slope, normalized difference vegetation index (NDVI), average temperature (Tem), soil silt content (T_SILT), soil pH (T_PH_H20), soil clay content (T_CLAY), aspect, and soil sand content (T_SAND) are determined as the environmental factors for constructing the model.

[0060] In the third step, reclassify and assign weights to the key environmental factors through the analytic hierarchy process (AHP) and the average method, construct a prediction model for the occurrence probability of Calliptamus italicus pests based on the multi-criteria analysis (MCA), extract the prediction results of the occurrence of Calliptamus italicus pests, and divide the prediction results into four grades: high, medium, low, and unsuitable habitat areas according to the actual occurrence situation of Calliptamus italicus.

[0061] 1) Due to the different units and numerical thresholds of various factors, it is necessary to unify the standards of various factors before constructing the model. The present invention defines five levels of 1, 4, 6, 8, and 10 to represent very unsuitable, unsuitable, suitable, relatively suitable, and very suitable respectively, so as to reflect the survival status of Calliptamus italicus under the corresponding environmental conditions. According to this standard, the attributes of each factor are reclassified and corresponding weights are assigned to obtain the final standardized factor indicators for participating in the model construction as shown in Table 2.

[0062] Table 2 Classification Standards of Each Factor

[0063]

[0064] 2) In the stage of assigning weights to each factor, the present invention adopts two methods. One is the AHP method. A judgment matrix is constructed according to the relative importance of each factor to the occurrence of locusts to determine the weight score assigned to each factor. The other method is the average method, assuming that each influencing factor has the same important influence on the occurrence of locusts, that is, the weight scores of climate, vegetation, soil, and terrain factors are equal.

[0065] ① In this embodiment, the AHP method is used to arrange each factor in a hierarchical structure according to its relative importance. Its weighting process is based on constructing a square matrix, comparing all pairwise paired factors, and then comprehensively calculating the score of each factor. A high score indicates a high contribution rate of the factor to the research target. On the contrary, a low score indicates a low contribution rate to the research target. Usually, the smaller the CR value, the better the consistency of the judgment matrix. Generally, when the CR value is less than 0.1, the judgment matrix meets the consistency test; if the CR value is greater than 0.1, it means that there is no consistency and the judgment matrix should be appropriately adjusted and then analyzed again. For the 10-order judgment matrix calculated in this study, the CI value is 0.084, and the RI value is found from the table to be 1.490. Therefore, the calculated CR value is 0.084 < 0.1, which means that the judgment matrix of this study meets the consistency test and the calculated weights are consistent. In this process, the definition reference of the judgment matrix scale between two evaluation indicators is shown in Table 3.

[0066] Table 3 Definition of Judgment Matrix Scale

[0067]

[0068] ② In this embodiment, the average method is used to assume that each influencing factor has the same important influence on the occurrence of locusts, that is, the weight scores of climate, vegetation, soil, and terrain factors are equal. Among them, the climate factor includes the average temperature and relative humidity; the vegetation factor is the normalized difference vegetation index; the soil factor includes the soil silt content, soil pH, soil clay content, and soil sand content; the terrain factor includes elevation, slope, and aspect.

[0069] ③Finally, the weight scores assigned by the AHP method and the average method are shown in Table 4.

[0070] Table 4 Weight scores assigned by AHP and the average method

[0071]

[0072] 3) Based on the above indicators of environmental factors and the corresponding weight scores, a model for the occurrence of Calliptamus italicus pests can be constructed.

[0073] ①AHP method:

[0074] Locust = 0.2756*X1 + 0.2176*X2 + 0.1886*X3 + 0.0564*X4 + 0.0296*X5 + 0.0479*X6 + 0.0479*X7 + 0.0945*X8 + 0.0219*X9 + 0.0201*X10

[0075] ②Average method:

[0076] Locust = 0.125*X1 + 0.125*X2 + 0.25*X3 + 0.0625*X4 + 0.0625*X5 + 0.0625*X6 +

[0077] 0.0625*X7 + 0.0833*X8 + 0.0833*X9 + 0.0833*X10

[0078] Among them: Locust represents the occurrence of locusts in the current year; X1 is the average temperature from May to July in the current year; X2 is the average relative humidity from May to July in the current year; X3 is the vegetation coverage from June to August in the current year; X4 is the silt content of the upper soil layer; X5 is the acidity and alkalinity of the upper soil layer; X6 is the clay content of the upper soil layer; X7 is the sand content of the upper soil layer; X8 is the elevation of the corresponding geographical location; X9 is the slope, reflecting the steepness of the corresponding geographical location; X10 is the slope aspect, reflecting the sunny orientation of the corresponding geographical location.

[0079] 4) S3.4. Classify the prediction results into four categories according to the actual occurrence of Calliptamus italicus; areas with values less than the threshold are considered unsuitable (Value < 0.4), values higher than the threshold indicate low suitability (0.4 < Value < 0.6), medium suitability (0.6 < Value < 0.8), and high suitability (0.8 < Value < 1). Use ArcGIS 10.7 software to conduct a visual analysis of the results of the probability of Calliptamus italicus pest occurrence.

[0080] Fourth step, based on the actual occurrence sample points of Calliptamus italicus pests, use the prediction results of Calliptamus italicus pests, and respectively evaluate the prediction accuracy rates of the analytic hierarchy process and the average method through probability statistics to determine the optimal model for predicting the occurrence of Calliptamus italicus pests;

[0081] 1) Determine the actual occurrence point data of Calliptamus italicus pest in 2015 - 2017 as the inevitable occurrence points of Calliptamus italicus pest, and use them as the actual reference for verifying the prediction model of the multi-criterion analysis method. Corresponding to the highly suitable and moderately suitable habitat grades constructed by the model, among which the sample points in 2015 - 2017 are 307, 266, and 395 respectively.

[0082] 2) Use the spatial analysis tool in ArcGIS 10.7 software - Extract by Mask - Extract Values to Points. Take the actual occurrence points in 2015 - 2017 as the input point features, and the suitable habitat grade of Calliptamus italicus pest as the input raster. Set the output point features, and check the interpolation at the point position to extract the suitable habitat grades corresponding to the actual occurrence points of Calliptamus italicus in 2015 - 2017 for verifying the prediction model method of Calliptamus italicus.

[0083] 3) Import the extracted verification results into Excel software, delete the outliers and null values, and analyze the overall accuracy of the analytic hierarchy process (AHP) and the average method respectively through probability statistics to determine the optimal model for predicting the occurrence of Calliptamus italicus pest.

[0084] By deleting the outliers and null values after extraction in 2015 - 2017, there are 296, 263, and 390 verification points left respectively. The accuracy verification results show that the overall accuracy of both methods is above 70%, indicating that the prediction methods of Calliptamus italicus pest constructed based on the analytic hierarchy process and the average method can be used for extracting the suitable habitats of Calliptamus italicus. Generally speaking, the average accuracies of the AHP method and the average method are 78.6% and 77.47% respectively, and the average accuracy of the AHP method is higher than that of the average method, especially the prediction effect in 2017.

[0085] Table 5 Model Accuracy Verification

[0086]

[0087] In the fifth step, by obtaining the real-time meteorological numerical forecast products of the occurrence of Calliptamus italicus pest, use the optimal prediction model of the occurrence probability of Calliptamus italicus pest in step S4 to conduct simulation experiments for short-term prediction of the occurrence probability of Calliptamus italicus pest; specifically including:

[0088] 1) Download the latest 15-day meteorological numerical forecasts including surface temperature and relative humidity from the ECMWF, a medium-term weather forecast center in a certain continent; download the latest NDVI data (MOD09A1) from the EARTHDATA website of NASA.

[0089] 2) Perform operations of data synthesis, format conversion, relative data registration, and preprocessing on meteorological and NDVI data in step S15 to obtain preprocessed environmental data. The number of rows and columns, pixel size, pixel type, pixel depth, coordinate system, and unit of all data must be consistent.

[0090] 3) Input meteorological data and vegetation data into the optimal model for predicting the occurrence probability of Calliptamus italicus pests, extract the short-term prediction results of Calliptamus italicus pests, and at the same time divide the prediction results into four levels: high, medium, low, and unsuitable habitats.

[0091] Generally speaking, compared with the prior art, a remote sensing prediction method for the occurrence probability of Calliptamus italicus pests provided by the present invention has the following beneficial effects:

[0092] The present invention can take the areas where Calliptamus italicus is widely distributed as the research area. On the basis of fully considering the environmental factors for the occurrence of Calliptamus italicus, assign weights to key environmental factors through the analytic hierarchy process and the averaging method, construct a prediction model for the occurrence probability of Calliptamus italicus pests based on the multi-criteria analysis method, evaluate the prediction accuracy of the analytic hierarchy process and the averaging method respectively through probability statistics according to the actual occurrence sample points of Calliptamus italicus pests, and determine the optimal model for predicting the occurrence of Calliptamus italicus pests; obtain meteorological numerical forecast products in real time, and finally realize the short-term prediction of the occurrence of Calliptamus italicus pests, providing new ideas and references for the remote sensing dynamic monitoring and prediction of grassland locusts.

[0093] It should be noted that in the present invention, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

[0094] The above embodiments are only illustrative examples of the present invention and do not constitute a limitation on the protection scope of the present invention. Any design identical or similar to the present invention falls within the protection scope of the present invention.

Claims

1. A method for predicting the probability of locust pest occurrence in Italy, characterized in that: The following steps are involved: Obtaining historical occurrence point data and environmental data of locust pests in Italy, relatively aligning and preprocessing the historical occurrence point data and the environmental data to obtain environmental factors; The environmental factors related to locust infestation in Italy were extracted through the MaxEnt model and Pearson correlation coefficient to obtain key environmental factors; Reclassify and weight the key environmental factors to obtain environmental factor indicators. Based on the multi-criteria analysis method, combine the environmental factor indicators and the corresponding weights to build a probability prediction model for locust pests in Italy. Obtain the corresponding environmental factor indicators of the area to be evaluated, input them into the Italian locust pest occurrence probability prediction model, and obtain the corresponding Italian locust pest occurrence prediction results.

2. The Italian locust pest occurrence probability prediction method as claimed in claim 1, characterized in that: Also includes: After obtaining the corresponding prediction results of the locust infestation in Italy, the level prediction results of the locust infestation in Italy are obtained by combining the set level classification rules.

3. The Italian locust pest occurrence probability prediction method as claimed in claim 1, characterized in that: The environmental data includes: meteorological data, vegetation data, soil data and terrain data.

4. The method for predicting the probability of locust pest occurrence in Italy as claimed in claim 1, characterized in that: The key environmental factors are obtained, including: The relative contribution of environmental factors was evaluated using the jackknife method in the MaxEnt model to obtain the main environmental factors affecting the geographical distribution of locusts in Italy; The correlation between the main environmental factors is obtained based on the Pearson correlation coefficient, and the key environmental factors are selected according to the preset range until all the key environmental factors are obtained.

5. The method for predicting the probability of locust pest occurrence in Italy as claimed in claim 1, characterized in that: The reclassification and weighting of key environmental factors to obtain environmental factor indicators includes the following steps: AHP was selected based on the indicators that affect the occurrence and survival of locusts in Italy; the attributes of key environmental factors were combined to reclassify them and obtain the impact indicators; Construct a judgment matrix based on the relative importance of each influencing indicator by arranging the hierarchical structure; According to the judgment matrix, multiple influencing index weights are obtained; Based on the impact indicators and the corresponding impact indicator weights, the environmental factor indicators are obtained.

6. The method for predicting the probability of locust pest occurrence in Italy as claimed in claim 1, characterized in that: The setting level division rules are: When the prediction result of locust infestation in Italy is between 0.8 and 1.0, the area to be assessed is highly suitable; When the prediction result of locust infestation in Italy is between 0.6 and 0.8, the area to be assessed is of medium suitability. When the prediction result of locust infestation in Italy is between 0.4 and 0.6, the area to be assessed is of low suitability; When the prediction result of locust infestation in Italy is between 0-0.4, the area to be assessed is unsuitable.

7. A device for predicting the probability of locust pests in Italy, characterized in that: include: Impact factor acquisition module; It is used to obtain the historical occurrence point data and environmental data of locust pests in Italy, perform relative registration and preprocessing, and obtain environmental factors; through the MaxEnt model and Pearson correlation coefficient, extract environmental factor variables related to locust pests in Italy and obtain key environmental factors; The prediction model building module is used to reclassify and weight key environmental factors, obtain environmental factor indicators, and build a probability prediction model for locust pests in Italy based on multi-criteria analysis, combining environmental factor indicators and corresponding weights; The probability prediction module is used to obtain the corresponding environmental factor indicators of the area to be evaluated, input them into the probability prediction model of the locust pest in Italy, obtain the corresponding prediction results of the locust pest in Italy, and combine them with the set grade classification rules to obtain the grade prediction results of the locust pest in Italy.

8. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

9. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method described in any one of claims 1 to 6 is implemented.

Images