Climate risk prediction method and system for key growth period of crops

Abstract

The invention discloses a climate risk prediction method and system for a key growth period of crops, and relates to the technical field of climate risk prediction, and the method comprises the following steps: collecting meteorological data, remote sensing data and agricultural information data; performing space-time sequence analysis according to the meteorological data, the remote sensing data and the agricultural information data to obtain multi-source data time sequence characteristics considering the key growth period of crops; and performing prediction according to the time sequence characteristics of the multi-source data by adopting the prediction model to obtain a climate risk prediction result and a geostatistical analysis result. The weather data, the remote sensing data and the agricultural information data are collected, and the space-time sequence analysis and prediction model is combined, so that the relevance between the climate risk prediction and the key growth period of the crops is improved, and the reliability of the climate risk prediction is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of climate risk prediction, in particular to a climate risk prediction method and system for a key growth period of crops. BACKGROUND

[0002] The key growth period of crops is extremely sensitive to climate fluctuations. Drought, flooding, extreme temperature and other climate risks during this period directly affect crop yield and quality, and are related to food security and stable agricultural economy. Current climate risk prediction relies on single meteorological data, and lacks integration of crop growth period spatio-temporal characteristics and agricultural information, resulting in weak correlation between prediction and actual production needs, low reliability, and difficulty in effectively guiding disaster prevention and mitigation. Therefore, research is needed on climate risk prediction technology for the key growth period of crops.

[0003] In the prior art, Chinese patent CN113762768A discloses an agricultural drought dynamic risk assessment method based on a weather generator and a crop model. A weather generator is used to randomly simulate a large number of future daily meteorological data samples to describe the random uncertainty of future drought occurrence. A crop model is driven by measured and randomly simulated meteorological data to assess crop yield loss under different drought scenarios. Based on the risk definition, an expected yield loss rate index is constructed to assess agricultural drought dynamic risk.

[0004] However, the above prior art does not consider the correlation between climate risk prediction and the key growth period of crops, and the reliability of climate risk prediction is poor. SUMMARY

[0005] The present application provides a climate risk prediction method and system for the key growth period of crops to solve the problem that the prior art does not consider the correlation between climate risk prediction and the key growth period of crops, and the reliability of climate risk prediction is poor.

[0006] In one aspect, the present application provides a climate risk prediction method for the key growth period of crops, comprising the following steps: Step one, collecting meteorological data, remote sensing data and agricultural information data.

[0007] Step two, performing spatio-temporal sequence analysis based on the meteorological data, the remote sensing data and the agricultural information data to obtain multi-source data time sequence characteristics considering the key growth period of crops.

[0008] Step three, using a prediction model to predict based on the multi-source data time sequence characteristics to obtain climate risk prediction results and geostatistical analysis results.

[0009] In one possible implementation, in step one, the collected meteorological data, remote sensing data and agricultural information data are subjected to data cleaning and data fusion.

[0010] In a possible implementation, step two includes: dynamically identifying a crop growth period range according to the remote sensing data.

[0011] constructing a crop key growth period spatio-temporal sequence based on the crop growth period range.

[0012] extracting multi-source data time sequence features including meteorological features, remote sensing features, and agricultural features from the meteorological data, the remote sensing data, and the agricultural information data according to the crop key growth period spatio-temporal sequence.

[0013] In a possible implementation, in step two, on the basis of the multi-source data time sequence features, trend analysis is performed to obtain a data spatio-temporal feature change trend.

[0014] In a possible implementation, in step three, the prediction model includes a spatial interpolation sub-model, a random forest sub-model, a risk level prediction sub-model, and a geo-statistical analysis sub-model.

[0015] The spatial interpolation sub-model is configured to interpolate the multi-source data time sequence features of point data into raster data of a plane scale.

[0016] The random forest sub-model is configured to analyze, according to the raster data, an importance degree of the meteorological features in the multi-source data time sequence features on crop growth in the crop key growth period, to obtain a meteorological feature weight.

[0017] The risk level prediction sub-model is configured to perform risk prediction according to the meteorological feature weight, to obtain a climate risk prediction result.

[0018] The geo-statistical analysis sub-model is configured to statistically analyze a raster area of a given risk level attribute according to the climate risk prediction result, to obtain a geo-statistical analysis result.

[0019] In a possible implementation, the climate risk prediction method for the crop key growth period further includes step four: visualizing and applying the climate risk prediction result and the geo-statistical analysis result.

[0020] In a possible implementation, step four includes: dynamically displaying the geo-statistical analysis result.

[0021] dynamically displaying the climate risk prediction result.

[0022] dynamically visualizing response decision information corresponding to different spaces of the geo-statistical analysis result and the climate risk prediction result, and dynamically reminding real-time early warning information under different climate risk levels.

[0023] The real-time updated meteorological data, remote sensing data, agricultural information data, crop critical growth period spatio-temporal sequence, climate risk prediction results, and geostatistical analysis results are integrated into a crop growth period risk early warning system to provide information query and prediction early warning services.

[0024] In a possible implementation, the climate risk prediction method for the critical growth period of crops further includes: step five, managing and updating the data and models in steps one to four.

[0025] In a possible implementation, step five includes: managing the data of the meteorological data, the remote sensing data, the agricultural information data, the multi-source data time sequence features, the climate risk prediction results, and the geostatistical analysis results.

[0026] optimizing and updating the prediction model.

[0027] In another aspect, the present application provides a climate risk prediction system for the critical growth period of crops, which adopts the climate risk prediction method for the critical growth period of crops described above, and includes a data collection and preprocessing module, a spatio-temporal sequence analysis module, a prediction model module, a visual display and application module, and a system maintenance and update module.

[0028] The data collection and preprocessing module is configured to collect meteorological data, remote sensing data, and agricultural information data, and to perform data cleaning and data fusion on the collected meteorological data, remote sensing data, and agricultural information data.

[0029] The spatio-temporal sequence analysis module is configured to perform spatio-temporal sequence analysis based on the meteorological data, the remote sensing data, and the agricultural information data, to obtain multi-source data time sequence features considering the critical growth period of crops, and to perform trend analysis based on the multi-source data time sequence features to obtain a data spatio-temporal feature change trend.

[0030] The prediction model module is configured to use a prediction model to perform prediction based on the multi-source data time sequence features, to obtain climate risk prediction results and geostatistical analysis results.

[0031] The visual display and application module is configured to visually display and apply the climate risk prediction results and the geostatistical analysis results.

[0032] The system maintenance and update module is configured to manage the data of the meteorological data, the remote sensing data, the agricultural information data, the multi-source data time sequence features, the climate risk prediction results, and the geostatistical analysis results, and to optimize and update the prediction model.

[0033] The climate risk prediction method and system for key crop growth periods described in this application have the following advantages: By collecting meteorological data, remote sensing data, and agricultural information data, and combining them with spatiotemporal series analysis and prediction models, the correlation between climate risk prediction and the critical growth period of crops has been improved, thereby enhancing the reliability of climate risk prediction.

[0034] By dynamically identifying the range of crop growth periods based on remote sensing data, a spatiotemporal sequence of key crop growth periods is constructed. Meteorological, remote sensing, and agricultural characteristics are extracted from this sequence, improving the correlation between these characteristics and the key crop growth periods, thus providing a foundation for subsequent climate risk prediction.

[0035] By conducting trend analysis based on the temporal characteristics of multi-source data, the spatiotemporal characteristics of the data are obtained, providing a spatiotemporal basis for subsequent climate risk prediction.

[0036] By combining spatial interpolation sub-models, random forest sub-models, risk level prediction sub-models, and geostatistical analysis sub-models, the accuracy of climate risk prediction results and geostatistical analysis results has been improved.

[0037] By visualizing and applying climate risk prediction results and geostatistical analysis results, timely information query and forecasting and early warning services can be provided.

[0038] By optimizing and updating the prediction model, the adaptability of climate risk prediction has been improved. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 A flowchart illustrating the climate risk prediction method for critical growth stages of crops provided in this application embodiment; Figure 2 A schematic diagram illustrating the spatial variation trend of a certain meteorological factor of a certain crop within a certain spatial range during a specific growth period, provided as an embodiment of this application; Figure 3 This is a schematic diagram of the climate risk prediction results provided in the embodiments of this application. Detailed Implementation

[0041] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0042] like Figure 1 As shown in the embodiments of this application, a method for predicting climate risks during critical growth stages of crops is provided, including the following steps: Step 1: Collect meteorological data, remote sensing data, and agricultural information data.

[0043] Step two: Perform spatiotemporal sequence analysis on the meteorological data, the remote sensing data, and the agricultural information data to obtain the time series characteristics of multi-source data considering the key growth periods of crops.

[0044] Step 3: Using a prediction model, predictions are made based on the time-series characteristics of the multi-source data to obtain climate risk prediction results and geostatistical analysis results.

[0045] Specifically, in this embodiment, the collection of meteorological data includes the dynamic collection and updating of meteorological element data such as hourly rainfall, temperature, relative humidity, and sunshine duration from meteorological stations across the country; the collection of remote sensing data includes the collection and acquisition of land use image data, crop planting area data, and crop type and growth data; and the collection of agricultural information data includes the collection of information such as crop varieties and sowing periods.

[0046] For example, in step one, the collected meteorological data, remote sensing data, and agricultural information data are cleaned and fused.

[0047] Specifically, in this embodiment, data cleaning involves processing meteorological data, remote sensing data, and agricultural information data for missing and outlier values; data fusion involves fusing meteorological data, remote sensing data, and agricultural information data transmitted from different time scales, spatial scales, and sensor attributes to ensure data consistency.

[0048] For example, step two includes: The range of crop growth period is dynamically identified based on the remote sensing data.

[0049] Based on the range of crop growth periods, a spatiotemporal sequence of key growth periods of crops is constructed.

[0050] Based on the spatiotemporal sequence of the key growth period of crops, multi-source data time series features, including meteorological features, remote sensing features, and agricultural features, are extracted from the meteorological data, the remote sensing data, and the agricultural information data.

[0051] Specifically, in this embodiment, the calculation formula for dynamically identifying the range of crop growth periods is as follows: .

[0052] Wherein, NDVI(t) represents the NDVI value at time t, A1 and A2 represent the amplitude of the growth period and the amplitude of the maturity period, respectively, k1 and k2 represent the slope parameters of the growth period curve and the maturity period curve, respectively, and t1 and t2 represent the inflection point dates corresponding to the start and end times of the critical growth period.

[0053] For example, in step two, trend analysis is performed based on the time-series characteristics of the multi-source data to obtain the spatiotemporal characteristic change trend of the data.

[0054] Specifically, in this embodiment, the trend analysis employs the Mann-Kendall trend test, combined with the ArcGIS geographic information platform, to analyze the spatiotemporal characteristics of the data and determine their changing trends. The calculation process of the Mann-Kendall trend test is as follows: .

[0055] .

[0056] .

[0057] .

[0058] Where S represents the trend test statistic, x i x j Let x and y represent the i-th and j-th time series data points, respectively, n represent the time series length, and z represent the input value (x, y, z). j - x i Var(S) represents the variance of the statistic S, and Z represents the standardized test statistic. Trend judgment: |Z|>1.96 indicates a significant trend at the α=0.05 level. Plotting the obtained Z values, Z>0 indicates an upward trend, Z<0 indicates a downward trend, Z>1.96 indicates a significant increase at the α=0.05 level, and Z<-1.96 indicates a significant decrease at the α=0.05 level. The spatial variation trend of a certain meteorological factor for a certain crop within a certain spatial range during a specific growth period is shown below. Figure 2 As shown.

[0059] For example, in step three, the prediction model includes: a spatial interpolation sub-model, a random forest sub-model, a risk level prediction sub-model, and a geostatistical analysis sub-model.

[0060] The spatial interpolation sub-model is used to interpolate the multi-source temporal features of point data into raster data at the area scale.

[0061] The random forest sub-model is used to analyze the influence of meteorological features in the time series features of the multi-source data on crop growth during the critical growth period of crops based on the raster data, and to obtain the meteorological feature weights.

[0062] The risk level prediction sub-model is used to predict risks based on the meteorological feature weights to obtain climate risk prediction results.

[0063] The geostatistical analysis sub-model is used to calculate the grid area of ​​a given risk level attribute based on the climate risk prediction results, and obtain the geostatistical analysis results.

[0064] Specifically, in this embodiment, meteorological feature weights refer to the importance of meteorological features (rainfall, temperature, relative humidity, sunshine duration, etc.) on crop growth during critical growth stages (such as the wheat waxy ripening stage to full maturity stage), and these features are ranked and weighted accordingly. Climate risk prediction results refer to the spatial distribution of climate risk at different time scales, and based on real-time transmitted meteorological data, remote sensing data, and crop information, climate risk during critical growth stages of crops is predicted in real time. Geostatistical analysis results are based on the climate risk prediction results to determine the regional area under different risk levels, in order to better provide response decisions.

[0065] In this embodiment, spatial interpolation uses inverse distance weighting interpolation, which interpolates point data into raster data within a given area of ​​surface data, with a raster spatial scale of 30m × 30m. Before interpolating based on the point data and area data range, it is necessary to ensure that the coordinate systems of the point data and area data in SHP format are consistent. If the coordinate systems are not consistent, the coordinate systems need to be projected first. The formula for inverse distance weighting interpolation is: .

[0066] Where Z(x0) represents the predicted value of the point (raster) to be interpolated, Z(x i ) represents the observed value at the i-th meteorological station, d i represents the Euclidean distance from the point to be interpolated to the i-th station, p represents the distance attenuation coefficient (taken as 2), and n represents the number of neighboring stations participating in the interpolation.

[0067] In this embodiment, the random forest sub-model is implemented using Python code. It calculates the importance of different meteorological data, and based on the importance results, normalizes them to obtain their weights. The formula for calculating the importance is: .

[0068] Among them, X j represents the j-th meteorological feature (rainfall, temperature, relative humidity, sunshine hours), N_trees represents the number of decision trees in the random forest, T represents a single decision tree, t represents the splitting node in tree T, and ΔGini(t, X j ) represents the reduction in Gini impurity when splitting at node t using the feature X j .

[0069] The calculation formula for the meteorological feature weight is as follows: .

[0070] Among them, w j represents the normalized weight of the feature X j , and m represents the total number of features (rainfall, temperature, relative humidity, sunshine hours).

[0071] In this embodiment, the formula for the risk level prediction sub-model to calculate the climate risk value is as follows: .

[0072] Among them, R represents the climate risk value, w j represents the normalized weight of the feature X j , f(X j ) represents the standardization function of the feature X j , and m represents the number of features.

[0073] Based on the climate risk value, through the reclassification function in the ArcGIS geographic information platform, the following grade划分 is carried out: R ≤ 0.2: Grade 1 (low risk); 0.2 < R ≤ 0.4: Grade 2 (lower risk); 0.4 < R ≤ 0.6: Grade 3 (medium risk); 0.6 < R ≤ 0.8: Grade 4 (higher risk); R > 0.8: Grade 5 (high risk).

[0074] In this embodiment, the geostatistical analysis sub-model, according to the risk level assessment and prediction results, by statistically analyzing the area of the grid with the given risk level attribute, for example: if the risk level assessment result is 1 and 4, and the area with the given risk assessment result of 4 is the area at risk, then it is necessary to statistically analyze the grid area value with the grid attribute value of 2. The calculation formula for the grid area is as follows: .

[0075] Among them, A k represents the area of the region with the risk level of k, N k represents the number of grids with the risk level of k, and Res represents the grid resolution (30m × 30m = 900m²).

[0076] For example, the climate risk prediction method for key crop growth periods also includes: step four, visualizing and applying the climate risk prediction results and the geostatistical analysis results.

[0077] For example, step four includes: The statistical analysis results are displayed dynamically.

[0078] The climate risk prediction results are displayed dynamically.

[0079] The system dynamically visualizes and displays the response decision information corresponding to different spaces based on the geostatistical analysis results and the climate risk prediction results, and provides real-time dynamic reminders for early warning information under different climate risk levels.

[0080] The system integrates real-time updated meteorological data, remote sensing data, agricultural information data, spatiotemporal sequences of key crop growth periods, climate risk prediction results, and geostatistical analysis results into the crop growth period risk early warning system, providing information query and forecast early warning services.

[0081] Specifically, in this embodiment, the dynamic display of the geostatistical analysis results mainly involves plotting the area under different risk levels as a line graph using real-time dynamically updated data, better showcasing the dynamic changes in the area of ​​regions at different risk levels over time; the dynamic display of the climate risk prediction results mainly involves displaying the dynamic risk assessment prediction results in a rolling manner using the results obtained from the risk level prediction module; and the dynamic visualization of response decision-making information mainly refers to the visualization of response decision-making information in different spaces based on the results of displaying dynamic area changes and the spatial distribution of risk, and providing real-time dynamic reminders for early warning information under different climate risk levels. An example of the climate risk prediction results is as follows... Figure 3 As shown.

[0082] For example, the climate risk prediction method for key crop growth periods also includes: step five, managing and updating the data and models from steps one to four.

[0083] For example, step five includes: Data management is performed on the meteorological data, the remote sensing data, the agricultural information data, the time series characteristics of the multi-source data, the climate risk prediction results, and the geostatistical analysis results.

[0084] The prediction model is then optimized and updated.

[0085] On the other hand, this application provides a climate risk prediction system for critical growth stages of crops, which adopts the above-mentioned climate risk prediction method for critical growth stages of crops, including: a data acquisition and preprocessing module, a spatiotemporal series analysis module, a prediction model module, a visualization and application module, and a system maintenance and update module.

[0086] The data acquisition and preprocessing module is used to collect meteorological data, remote sensing data, and agricultural information data, and to perform data cleaning and data fusion on the collected meteorological data, remote sensing data, and agricultural information data.

[0087] The spatiotemporal sequence analysis module is used to perform spatiotemporal sequence analysis based on the meteorological data, the remote sensing data, and the agricultural information data to obtain the time series characteristics of multi-source data considering the key growth periods of crops. Based on the time series characteristics of the multi-source data, trend analysis is performed to obtain the change trend of the spatiotemporal characteristics of the data.

[0088] The prediction model module is used to make predictions based on the time-series characteristics of the multi-source data, and obtain climate risk prediction results and geostatistical analysis results.

[0089] The visualization and application module is used to visualize and apply the climate risk prediction results and the geostatistical analysis results.

[0090] The system maintenance and update module is used to manage the meteorological data, remote sensing data, agricultural information data, time series characteristics of multi-source data, climate risk prediction results, and geostatistical analysis results, and to optimize and update the prediction model.

[0091] This application's embodiments improve the correlation between climate risk prediction and critical growth periods of crops by collecting meteorological data, remote sensing data, and agricultural information data, combined with spatiotemporal series analysis and prediction models, thereby enhancing the reliability of climate risk prediction.

[0092] By dynamically identifying the range of crop growth periods based on remote sensing data, a spatiotemporal sequence of key crop growth periods is constructed. Meteorological, remote sensing, and agricultural characteristics are extracted from this sequence, improving the correlation between these characteristics and the key crop growth periods, thus providing a foundation for subsequent climate risk prediction.

[0093] By conducting trend analysis based on the temporal characteristics of multi-source data, the spatiotemporal characteristics of the data are obtained, providing a spatiotemporal basis for subsequent climate risk prediction.

[0094] By combining spatial interpolation sub-models, random forest sub-models, risk level prediction sub-models, and geostatistical analysis sub-models, the accuracy of climate risk prediction results and geostatistical analysis results has been improved.

[0095] By visualizing and applying climate risk prediction results and geostatistical analysis results, timely information query and forecasting and early warning services can be provided.

[0096] By optimizing and updating the prediction model, the adaptability of climate risk prediction has been improved.

[0097] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0098] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A climate risk prediction method for key crop growth stages, characterized in that, Includes the following steps: Step 1: Collect meteorological data, remote sensing data, and agricultural information data; Step 2: Perform spatiotemporal sequence analysis on the meteorological data, the remote sensing data, and the agricultural information data to obtain the time series characteristics of multi-source data considering the key growth periods of crops; Step 3: Using a prediction model, predictions are made based on the time-series characteristics of the multi-source data to obtain climate risk prediction results and geostatistical analysis results.

2. The climate risk prediction method for key crop growth stages according to claim 1, characterized in that, In step one, the collected meteorological data, remote sensing data, and agricultural information data are cleaned and fused.

3. The climate risk prediction method for key crop growth stages according to claim 1, characterized in that, Step two includes: The range of crop growth period is dynamically identified based on the remote sensing data. Construct a spatiotemporal sequence of key growth periods of crops based on the range of crop growth periods; Based on the spatiotemporal sequence of the key growth period of crops, multi-source data time series features, including meteorological features, remote sensing features, and agricultural features, are extracted from the meteorological data, the remote sensing data, and the agricultural information data.

4. The climate risk prediction method for key crop growth stages according to claim 1, characterized in that, In step two, based on the time-series characteristics of the multi-source data, trend analysis is performed to obtain the spatiotemporal characteristic change trend of the data.

5. The climate risk prediction method for key crop growth stages according to claim 1, characterized in that, In step three, the prediction model includes: a spatial interpolation sub-model, a random forest sub-model, a risk level prediction sub-model, and a geostatistical analysis sub-model; The spatial interpolation sub-model is used to interpolate the multi-source temporal features of point data into raster data at the area scale; The random forest sub-model is used to analyze the influence of meteorological features in the time series features of the multi-source data on crop growth during the critical growth period of crops based on the raster data, and to obtain the meteorological feature weights. The risk level prediction sub-model is used to predict risks based on the meteorological feature weights to obtain climate risk prediction results. The geostatistical analysis sub-model is used to calculate the grid area of ​​a given risk level attribute based on the climate risk prediction results, and obtain the geostatistical analysis results.

6. The climate risk prediction method for key crop growth stages according to claim 1, characterized in that, It also includes: Step four, visualizing and applying the climate risk prediction results and the geostatistical analysis results.

7. The climate risk prediction method for key crop growth stages according to claim 6, characterized in that, Step four includes: The statistical analysis results are displayed dynamically. The climate risk prediction results are displayed dynamically. The system dynamically visualizes the response decision information corresponding to the geostatistical analysis results and the climate risk prediction results in different spaces, and provides real-time dynamic reminders for early warning information under different climate risk levels. The system integrates real-time updated meteorological data, remote sensing data, agricultural information data, spatiotemporal sequences of key crop growth periods, climate risk prediction results, and geostatistical analysis results into the crop growth period risk early warning system, providing information query and forecast early warning services.

8. The climate risk prediction method for key crop growth stages according to claim 6, characterized in that, It also includes step five, which involves managing and updating the data and models from steps one through four.

9. The climate risk prediction method for key crop growth stages according to claim 8, characterized in that, Step five includes: Data management is performed on the meteorological data, the remote sensing data, the agricultural information data, the time series characteristics of the multi-source data, the climate risk prediction results, and the geostatistical analysis results; The prediction model is then optimized and updated.

10. A climate risk prediction system for key crop growth stages, characterized in that, The climate risk prediction method for key growth stages of crops as described in any one of claims 1 to 9 includes: a data acquisition and preprocessing module, a spatiotemporal series analysis module, a prediction model module, a visualization and application module, and a system maintenance and update module. The data acquisition and preprocessing module is used to acquire meteorological data, remote sensing data, and agricultural information data, and to perform data cleaning and data fusion on the acquired meteorological data, remote sensing data, and agricultural information data. The spatiotemporal sequence analysis module is used to perform spatiotemporal sequence analysis based on the meteorological data, the remote sensing data, and the agricultural information data to obtain the time series characteristics of multi-source data considering the key growth period of crops. Based on the time series characteristics of the multi-source data, trend analysis is performed to obtain the change trend of the spatiotemporal characteristics of the data. The prediction model module is used to make predictions based on the time-series characteristics of the multi-source data using a prediction model, and to obtain climate risk prediction results and geostatistical analysis results. The visualization and application module is used to visualize and apply the climate risk prediction results and the geostatistical analysis results. The system maintenance and update module is used to manage the meteorological data, remote sensing data, agricultural information data, time series characteristics of multi-source data, climate risk prediction results, and geostatistical analysis results, and to optimize and update the prediction model.

Images