Winter wheat drought damage quantitative dynamic risk estimation method, medium and program product

Abstract

The invention discloses a winter wheat drought damage quantitative dynamic risk estimation method, a medium and a program product, and belongs to the technical field of agricultural meteorological disaster risk assessment. The method comprises the following steps: firstly, establishing a vulnerability relation model of a historical cumulative meteorological drought comprehensive index and a winter wheat yield reduction rate; then, current and future drought disaster-inducing factor intensity is calculated by using real-time meteorological data and forecast data; then, winter wheat planting space distribution is extracted based on a satellite remote sensing technology; then, according to the disaster risk principle, the drought disaster-causing risk, the winter wheat exposure degree and the vulnerability model result are fused, the winter wheat yield reduction rates of different time scales in the future are calculated, and risk grades are divided; and finally, respectively generating a current drought influence assessment product and a risk estimation product in a future time period. According to the method, quantitative, dynamic and spatial assessment and estimation of winter wheat drought disaster risks can be realized, and the method can be widely applied to scenes such as agricultural drought disaster monitoring and assessment, disaster reduction scheduling and risk management.

Description

Technical Field

[0001] The present invention belongs to the technical field of agricultural meteorological disaster risk assessment, and relates to meteorological data processing, remote sensing monitoring and crop yield modeling technology. Specifically, it provides a quantitative dynamic risk estimation method for winter wheat drought damage, a computer-readable storage medium and a program product. Background Art

[0002] Winter wheat is a staple food crop in my country, and drought is one of the main meteorological disasters during its production. Against the backdrop of global climate change, the regional impact of drought disasters is becoming increasingly significant, posing an increasing threat to winter wheat production. If persistent drought occurs during the critical growth period of winter wheat, it will significantly affect key growth processes such as tillering, heading, and grain filling, leading to varying degrees of yield losses. To address this challenge, research on winter wheat drought risk pre-assessment technology has important theoretical value and practical significance: on the one hand, this research can provide a scientific basis and technical support for my country's agricultural disaster prevention and mitigation, helping to enhance agricultural disaster resilience; on the other hand, the research results will provide effective guarantees for the country's food security sustainable development strategy, and are of great significance for stabilizing food production and reducing disaster losses.

[0003] Previous winter wheat risk assessments were mostly static, based on historical meteorological and socioeconomic data. These assessments selected indicators to assess drought hazard, exposure, and vulnerability. By assigning different weights to these three factors, a risk assessment index was constructed and risk levels were determined. However, this method's weighting of the three factors is subjective, and the risk level can only indicate a relative level of risk. The specific meaning of risk is unclear, resulting in a low degree of quantification. Furthermore, these assessment systems typically produce static outputs that are difficult to dynamically update based on real-time meteorological data or future forecasts, hindering early warning responses and refined management.

[0004] In recent years, some studies have attempted to combine weather forecasts and climate predictions to conduct drought risk estimation. This method comprehensively considers previous and future precipitation and temperature conditions and constructs a meteorological drought index to estimate possible future droughts. However, in the application process, this type of method still mainly relies on expert experience to manually qualitatively or semi-quantitatively judge the possible impact of drought, making it difficult to objectively quantify the actual impact of drought on winter wheat yields. In addition, these methods usually lack dynamic monitoring of the exposure of disaster-prone objects and cannot reflect real-time changes in winter wheat planting area, resulting in limited accuracy and timeliness of risk estimation.

[0005] In addition, existing technologies have also proposed solutions for multi-source data fusion, but they still have significant limitations. For example, methods based on step-by-step correction of the leaf area index (such as Chinese patent CN118378419A) rely on the WOFOST mechanism model coupled with experimental data from specific regions. The local correction of model parameters is complex and does not address the problem of identifying dynamic exposure over a large area. Although the automated remote sensing classification system (CN110135677A) enables rapid drought assessment, the indicator design is still based primarily on traditional meteorological data and lacks a quantitative correlation between the cumulative effects of drought during the crop growth period and yield reduction rates. Although the comprehensive risk assessment model (CN113849763A) introduces multi-factor weight analysis, it does not construct a dynamic hazard-vulnerability relationship model and relies on historical statistical data, making it difficult to achieve a rolling risk forecast for the next 30-60 days.

[0006] In summary, existing winter wheat drought risk assessment and prediction methods suffer from strong subjectivity, low quantification, and poor real-time performance, making them inadequate for modern agricultural disaster prevention and mitigation. Therefore, improving the objectivity, quantification, and real-time performance of winter wheat drought risk prediction to provide more scientific and effective decision-making support for agricultural production is a pressing technical challenge. Summary of the Invention

[0007] (1) Purpose of the invention In response to the above-mentioned problems, the present invention aims to provide a technical method, medium and program product for quantitative dynamic risk estimation of winter wheat drought losses based on future climate forecasts, identification of winter wheat dynamic disaster-bearing body exposure using satellite remote sensing, and a winter wheat drought loss vulnerability relationship model. The present invention comprehensively utilizes the daily Meteorological Composite Index (MCI), remote sensing inversion planting distribution data and historical yield data to construct a quantitative mapping model between drought intensity and yield reduction rate, and integrates meteorological forecast information for the next 30 to 60 days with GIS spatial analysis technology to achieve dynamic quantitative assessment of winter wheat drought risk in the temporal and spatial dimensions, thereby improving the objectivity, quantification and real-time capabilities of dynamic risk estimation of winter wheat drought, and providing a scientific basis and technical support for agricultural disaster reduction decision-making and food security management.

[0008] (2) Technical solution In order to achieve the purpose of the invention and solve the technical problems, the present invention adopts the following technical solutions: The first object of the present invention is to provide a method for quantitative dynamic risk estimation of winter wheat drought damage, which is used to conduct a quantitative, dynamic, and spatial risk assessment of drought disaster risks in winter wheat planting areas over future time scales based on meteorological forecast data, remote sensing planting distribution information, and historical yield data. The method comprises at least the following steps: S100. Modeling the relationship between drought-causing factors and yield reduction vulnerability: For the target assessment area, a meteorological comprehensive drought index (MCI) was calculated based on historical daily meteorological data. The cumulative MCI index of moderate drought and above during the winter wheat growing period was then calculated, and the yield reduction rate was calculated based on historical winter wheat yield data. Based on the historical cumulative meteorological comprehensive drought index and yield reduction rate, a quantitative vulnerability relationship model between drought-causing factors and yield reduction rate was fitted. S200. Dynamic statistical analysis of winter wheat drought risk: Based on the daily meteorological drought comprehensive index of the past and future, the cumulative MCI index of moderate drought and above from the sowing of winter wheat to the present time is calculated to indicate the current intensity of drought-causing factors; and the cumulative meteorological drought comprehensive index of moderate drought and above from the sowing to a period of time in the future is calculated to indicate the estimated future intensity of drought-causing factors; S300. Obtaining the spatial grid distribution of winter wheat planting using satellite remote sensing: The spatial grid distribution of winter wheat plantings in the target assessment area is identified using medium- and high-resolution multi-temporal satellite remote sensing imagery. Remote sensing image processing and classification techniques are used to extract the rasterized location and area data of the winter wheat planting areas within geographic space. This data serves as a spatial quantitative representation of the exposure of the winter wheat hazard-bearing bodies in the current year in the risk assessment, providing an input basis for subsequent disaster-loss coupled analysis. S400. Quantitative spatial assessment of drought damage and risk classification: The winter wheat drought disaster risk, remote sensing planting area exposure, and vulnerability model results were integrated to calculate the potential winter wheat yield reduction rate due to drought in the current and future time periods at a grid scale. Risk levels were then classified according to the estimated winter wheat yield reduction rate and set thresholds. This generated a spatial distribution map of drought risk levels for winter wheat yield reduction in the current and future time periods, along with statistical results of the coverage area. S500. Winter wheat drought impact assessment and risk prediction product output: Based on the constructed drought risk level distribution map and estimation results, multi-time-scale winter wheat drought damage assessment and prediction products are generated and output in standard format layers, including current drought impact assessment maps, drought risk prediction maps for future periods, and corresponding regional statistical results, providing decision support for agricultural disaster warning, emergency response, and policy making.

[0009] The second object of the present invention is to provide a computer program product comprising computer instructions for executing the above-mentioned method for quantitative dynamic risk estimation of winter wheat drought losses.

[0010] The third invention object of the present invention is to provide a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned quantitative dynamic risk estimation method for winter wheat drought damage is implemented.

[0011] (3) Technical effects Compared with the existing technology, the technical method, medium and program product of the present invention for quantitative dynamic risk estimation of winter wheat drought damage have the following beneficial and significant technical effects: (1) The present invention adopts a drought disaster factor and yield reduction vulnerability relationship model constructed based on historical meteorological data and crop yield data, combined with the calculation of the daily meteorological drought comprehensive index (MCI) and the spatial distribution information of the planted area inverted from satellite remote sensing images, to achieve a fusion modeling of the three elements of drought disaster risk, exposure of the disaster-bearing body and crop vulnerability, and realize quantitative assessment of risks. It can more accurately predict the losses caused by drought, overcome the artificial weights set in traditional methods, reduce the influence of subjective factors on the risk assessment results, and improve the objectivity of the assessment.

[0012] (2) This paper makes full use of climate forecast data for future time periods, combines spatial rasterization modeling with standardized risk classification methods, and constructs a dynamic drought risk prediction mechanism for current and future time scales. This supports the formation of risk map products with timeliness, spatial continuity, and multi-format output capabilities. This method can not only assess the impact of current droughts but also estimate the risk of future droughts. It can be widely applied in scenarios such as agricultural disaster prevention and mitigation, food security monitoring, and agricultural information services, and has strong versatility and promotion value. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a flowchart of the technical method for quantitative dynamic risk estimation of winter wheat drought damage of the present invention; Figure 2 This is a diagram showing the drought impact assessment for winter wheat in a certain province on April 10, 2024; Figure 3 This is a diagram showing the drought risk estimate for winter wheat in a certain province over the next 30 days on April 10, 2024. Figure 4 This is a diagram showing the drought risk estimate for winter wheat in a certain province in the next 60 days on April 10, 2024. DETAILED DESCRIPTION

[0014] The present invention aims to provide a technical method, medium, and program product for quantitative dynamic risk estimation of winter wheat drought damage based on future climate forecasts, identification of winter wheat dynamic disaster-bearing body exposure using satellite remote sensing, and a winter wheat drought damage vulnerability relationship model. To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions in the embodiments of the present invention will be described in more detail below in conjunction with the accompanying drawings in the embodiments of the present invention. The described embodiments are only some of the embodiments of the present invention, not all of them, and the described embodiments are exemplary and intended to be used to explain the present invention, and should not be understood as limiting the present invention.

[0015] As a specific example, Figure 1 As shown, the quantitative dynamic risk estimation method for winter wheat drought damage provided by the embodiment of the present invention is used to perform quantitative, dynamic, and spatial risk assessment of drought disaster risks in winter wheat planting areas within a future time scale based on climate forecast data, remote sensing planting distribution information, and historical yield data. The implementation mainly includes the following steps: S100. Modeling the relationship between drought-causing factors and yield reduction vulnerability: For the target assessment area, the MCI of historical moderate drought and above levels during the winter wheat growing period was calculated based on historical daily meteorological data, and the yield reduction rate was calculated based on historical winter wheat yield data. Based on the historical cumulative MCI and yield reduction rate, a quantitative vulnerability relationship model between drought-causing factors and yield reduction rate was fitted.

[0016] More specifically, the modeling of the relationship between drought-causing factors and yield reduction vulnerability in this step at least includes: S110. Preparation of historical meteorological data and production data: For the selected target assessment area, collect historical daily ground-based meteorological observation data and winter wheat yield data for at least 30 consecutive years or more. Daily meteorological data shall at least include daily average temperature and daily precipitation. Winter wheat yield data shall be statistical actual yield data at the county level or smaller administrative units. The collected meteorological data and yield data shall be pre-processed for temporal and spatial consistency. At the same time, collect regional disaster type records and agricultural disaster information for the same period, and clarify the years in which non-drought disasters such as floods, frost damage, pests and diseases occurred, as well as the extent of their impact, for subsequent identification and elimination of samples affected by non-drought disasters. S120. Statistics of drought disaster factor intensity index: Based on the collected historical daily meteorological data, the daily meteorological drought comprehensive index (MCI) was calculated according to the national standard GB / T20481-2017. Drought is divided into four levels: light drought, moderate drought, severe drought, and extreme drought. The cumulative MCI values ​​of moderate drought and above in each historical year during the winter wheat growing period are calculated as the intensity index of winter wheat drought disaster factors. S130. Statistical calculation of winter wheat yield reduction rate: Based on the historical yield data of winter wheat, the decomposition method is usually used to calculate the yield reduction rate. Y Decomposition into trend yield Y t , climate yield Y c and random output Δ Y ,Right now Y = Y t + Y c +Δ Y . Due to the random output Δ Y Affected by random factors, it is usually difficult to quantify and can therefore be ignored. Climate yield can be expressed as Y c = Y - Y t , the relative climate yield can be expressed as a = Y c / Y t ×100%. Calculate the trend yield of each year based on the fitting function of winter wheat yield in previous years Y t , and derive climate yield based on relevant formulas Y c and relative climate yield a . For relative climate yield a For years less than 0, the relative climate yield a The absolute value of the reduction rate is defined as Y d , as a quantitative indicator of winter wheat disaster losses; further cross-check disaster information on years with reduced production, eliminate years with abnormal yields due to non-drought factors (such as floods, pests and diseases, frost damage, etc.), and retain only years with reduced production mainly due to drought for subsequent model construction; S140. Construction of a quantitative relationship model for vulnerability: The obtained historical drought disaster factor intensity and yield reduction rate data are paired, and the historical drought disaster factor intensity is used as the independent variable, and the historical yield reduction rate obtained after eliminating abnormal yield reduction due to non-drought factors is used as the dependent variable. Statistical regression fitting methods (such as linear regression fitting) are used to construct a linear or nonlinear quantitative vulnerability relationship model between the two. Y d = f The reliability and stability of the model were verified through significance tests and goodness of fit evaluations. The cumulative effects of drought disasters were considered to quantify the impact of drought on winter wheat yield during the growing period.

[0017] It should be pointed out that in order to improve the accuracy of vulnerability modeling, a multi-factor response mechanism can be further introduced. By constructing phased input characteristics that include drought intensity in different growth periods (such as the MCI cumulative values ​​of the sowing-greening period, jointing-filling period, etc.), a multivariate combination modeling system can be established to enhance the model's ability to depict the temporal nature of drought effects. At the same time, based on robust methods such as Bayesian regression and partial least squares regression (PLSR), model complexity control and high-dimensional regression adaptability can be taken into account to further improve the stability and generalizability of the model in the context of limited samples.

[0018] S200. Dynamic statistical analysis of winter wheat drought risk: At any time during the winter wheat growing season, the intensity of drought-causing factors is calculated based on the daily meteorological drought comprehensive index, both past and future. The cumulative meteorological drought comprehensive index for moderate drought and above from the time of winter wheat sowing to the current time is calculated, indicating the intensity of drought-causing factors in the previous period and used for dynamic drought impact assessment. The cumulative meteorological drought comprehensive index for moderate drought and above from the time of winter wheat sowing to the current time and the next 30 or 60 days is also calculated, indicating the intensity of drought-causing factors from the time of winter wheat sowing to the future period and used for dynamic drought risk assessment.

[0019] More specifically, the dynamic statistical analysis of winter wheat drought disaster risk in this step at least includes: S210. Real-time weather data acquisition: Obtain real-time daily meteorological data for the current winter wheat growing season, as well as daily climate forecast data for the next 30 to 60 days, including at least daily mean temperature and precipitation, to support historical and future statistical analysis of drought intensity. Real-time observation data are obtained from ground-based meteorological stations or regional automatic weather stations, while future climate forecast data are derived from the output products of global or regional climate prediction models. Necessary spatial downscaling, error correction, and intelligent aggregation processes are performed to ensure data temporal and spatial continuity and forecast reliability. S220. Calculation of meteorological drought comprehensive index: Based on the real-time meteorological data and future climate forecast data obtained in step S210, the daily MCI value from the winter wheat sowing date to the current time point and the daily predicted MCI value for the next 30 days and 60 days are calculated in accordance with the national standard GB / T20481-2017, and the drought level of each day is determined based on the MCI value, including light drought, moderate drought, severe drought, and extreme drought. S230. Statistics on the intensity of drought-causing factors in the early stage: Based on the real-time daily MCI values ​​calculated in step S220, the cumulative MCI values ​​of moderate drought and above from the time of winter wheat sowing to the current time are calculated as the current intensity index of the drought disaster factor, which is used to reflect the immediate impact of the existing drought intensity on crop growth. S240. Estimation of the intensity of future drought hazards: Based on the predicted daily MCI values ​​for the next 30 and 60 days calculated in step S220, the cumulative MCI values ​​of moderate drought or above from the time of winter wheat sowing to the next 30 days and the cumulative MCI values ​​of moderate drought or above from the time of winter wheat sowing to the next 60 days are respectively counted to quantify the intensity of future drought disaster factors that winter wheat may experience, providing quantitative indicators for drought risk estimation in future periods; S250. Generate spatial distribution of drought risk: Based on the statistical results of step S230 and step S240, combined with the geographical distribution information of meteorological stations, the spatial distribution of the intensity of drought disaster factors in the early stage and the spatial distribution of the intensity of drought disaster factors in the next 30 days and 60 days covering the target assessment area are generated through spatial interpolation technology, providing rasterized drought disaster risk spatial distribution data with geographic reference information, realizing the synchronous expression of drought intensity in time and space dimensions, and providing a spatial distribution basis for subsequent disaster loss model calls.

[0020] S300. Obtaining the spatial grid distribution of winter wheat planting using satellite remote sensing: Real-time dynamic inversion of medium- and high-resolution multi-temporal satellite remote sensing images is used to identify the spatial raster distribution information of the current year's winter wheat planting in the target assessment area. Remote sensing image processing and classification techniques are used to extract the raster position and area data of the winter wheat planting area in geographic space, which serves as the spatial quantitative representation of the exposure of the winter wheat disaster-prone body in the current year in risk assessment, providing an input basis for subsequent disaster-loss coupling analysis.

[0021] More specifically, in this step, obtaining the satellite remote sensing spatial grid distribution of winter wheat planting includes at least: S310. Remote Sensing Data Acquisition and Preprocessing: Obtain medium- and high-resolution multi-temporal satellite remote sensing images (preferably Sentinel-2, Landsat-8, or Gaofen series satellite data) of the target assessment area during the current winter wheat growing season. Perform radiometric calibration, atmospheric correction, geometric correction, and cloud removal on the acquired raw images to ensure spatial accuracy and spectral consistency of the remote sensing data, thereby improving the data quality of the remote sensing images. S320. Multi-temporal remote sensing feature extraction: Based on preprocessed multi-temporal remote sensing images, multidimensional feature parameters including vegetation index, texture characteristics, and temporal characteristics are extracted. The spectral characteristics and phenological characteristics of winter wheat at different growth stages are comprehensively utilized to construct a feature parameter set for winter wheat identification, thereby enhancing the separability of winter wheat from other land features. S330. Winter wheat distribution identification and classification: Based on the extracted multidimensional feature parameters, supervised classification, decision tree classification, and / or machine learning algorithms are used to identify and classify winter wheat planting areas within the target assessment area, generate binary classification results for the current year's winter wheat planting areas, and determine the spatial distribution range of the winter wheat planting areas; S340. Classification Accuracy Verification and Optimization: The winter wheat classification results were verified by field surveys of sample points, visual interpretation of high-resolution images, or existing crop distribution data. The overall accuracy, producer accuracy, user accuracy, and / or Kappa coefficient were calculated, and the classification results were optimized and adjusted based on the verification results. S350. Generation of rasterized spatial distribution data of winter wheat: The winter wheat classification results that have been verified and optimized for accuracy are converted into rasterized spatial distribution data that match the spatial distribution of drought disaster risk, including the spatial location, area and distribution density information of the winter wheat planting area, which serves as a quantitative representation of the exposure of winter wheat disaster-prone bodies in the assessment area, and provides basic data support for subsequent spatial overlay analysis with drought disaster risk and vulnerability models.

[0022] S400. Quantitative spatial assessment of drought damage and risk classification: The drought risk of winter wheat, remote sensing planting area exposure and vulnerability model results are integrated to calculate the winter wheat yield reduction rate due to drought in the current and future periods at the raster scale. The risk level is divided according to the set threshold based on the estimated winter wheat yield reduction rate, and a spatial drought risk level distribution map of winter wheat yield reduction in the current and future periods and its coverage area statistics are generated.

[0023] More specifically, quantitative spatial assessment of drought damage and risk classification should include at least: S410. Spatial matching and registration of risk factor data: The spatial distribution data of drought disaster risk obtained in step S200 is spatially registered with the spatial grid distribution data of winter wheat planting obtained in step S300 to ensure that the two types of data are consistent in spatial resolution, projection coordinate system, and spatial range, providing a consistent spatial basic unit for subsequent fusion calculations; S420. Rasterized calculation of drought loss rate based on disaster risk principle: In each spatial grid cell, based on the disaster risk assessment formula R=H×E×V, where R represents risk, H represents disaster hazard, E represents exposure of the hazard-bearing body, and V represents vulnerability, the potential yield reduction rate of winter wheat due to drought in the current, next 30 days, and next 60 days is calculated by integrating drought intensity H, crop area exposure E, and the yield reduction function value V estimated by the vulnerability model. S430. Drought risk classification: Based on the gridded yield reduction rate results and according to the preset risk level classification standards, the yield reduction rate value of each spatial grid is mapped to the corresponding risk level, and spatial distribution maps of winter wheat drought risk levels are generated for the current, next 30 days, and next 60 days time scales. Different colors or patterns are used to intuitively represent areas with different risk levels. S440. Statistics and analysis of risk coverage: Based on the generated spatial distribution map of risk levels, the coverage areas of different risk levels in the target assessment area and their proportion to the total winter wheat planting area are counted, and the spatial agglomeration characteristics and evolution trends of areas with different risk levels are analyzed to provide quantitative statistical analysis results for the subsequent generation of risk assessment products. At the same time, for administrative division units, the areas and proportions of different risk levels in each administrative area are calculated to provide support for regional drought prevention and control decisions.

[0024] As a preference, as shown in Table 1 below, the preset drought risk levels for winter wheat yield reduction rate are: Y d ≤5% is low risk, 5%< Y d ≤10% is considered low risk, 10%< Y d ≤15% is medium risk, Y d >15% is considered high risk. Based on this threshold standard, the calculated winter wheat yield reduction rate is divided into risk levels, so that the risk level has a clear quantitative meaning, which facilitates horizontal and vertical comparison of risk conditions in different regions and different time periods.

[0025] Table 1 Winter wheat yield reduction rate and drought risk level

[0026] S500. Winter wheat drought impact assessment and risk prediction product output: Based on the constructed drought risk level distribution map and estimation results, multi-time-scale winter wheat drought damage assessment and prediction products are generated and output in standard format layers, including current drought impact assessment maps, drought risk prediction maps for future periods, and corresponding regional statistical results, providing decision support for agricultural disaster warning, emergency response, and policy making.

[0027] As a preferred product, multi-time scale winter wheat drought damage assessment and prediction products include: winter wheat drought dynamic impact assessment product, which reflects the estimation of the yield reduction rate and impact assessment of the drought in the previous period of the current time on winter wheat (such as Figure 2 Winter wheat drought dynamic risk prediction product for the next 30 days (as shown in Figure 3 ), reflecting the estimated yield reduction rate and risk forecast of winter wheat caused by drought in the current period and the next 30 days; the dynamic risk forecast product of winter wheat drought in the next 60 days (such as Figure 4 ), reflecting the estimated yield reduction and risk profile of winter wheat due to drought conditions in the current period and the next 60 days. Furthermore, the risk assessment product is presented as a raster map, including information on the geographic spatial reference system, risk level grading, spatial distribution characteristics, and coverage statistics. Risk statistics can also be aggregated by administrative divisions at different levels as needed, providing decision support information for different levels.

[0028] Figure 2 This is a diagram showing the impact assessment of drought on winter wheat in a certain province on April 10, 2024. Figure 2 As shown in the figure, the drought risk assessment results for winter wheat-growing areas in the province exhibit distinct spatial distribution patterns. The assessment period, from October 1, 2023, to April 10, 2024, covers the critical growth period of winter wheat, from sowing to jointing and heading. As can be seen, the drought risk level for most of the province's winter wheat-growing areas is rated "low," highlighted in green. This indicates a low cumulative meteorological drought index during the assessment period, and the estimated winter wheat yield reduction is expected to be no more than 5%. The absence of red, orange, or yellow areas in the figure indicates that no high-risk (yield reduction > 15%), medium-risk (10% < 15% ≤ 15%), or low-risk (5% < 10% ≤ 10%) areas occurred in the province during the assessment period. The assessment results indicate that as of April 10, 2024, winter wheat production in the province was generally minimally impacted by drought, and agricultural production was generally good. The white areas within the province are non-winter wheat-growing areas and are not included in the drought risk assessment.

[0029] Figure 3This chart shows the estimated drought risk for winter wheat in the province over the next 30 days (October 1, 2023, to May 10, 2024), based on an assessment starting point of April 10, 2024. The estimate covers the entire province's winter wheat-growing area and spans the critical growth period from jointing and heading to flowering and grain filling. Spatially, all winter wheat-growing areas in the province are at low risk (green), indicating an estimated yield reduction of no more than 5%. High-risk (red), medium-risk (orange), and low-risk (yellow) areas are largely absent from the chart. This indicates that, based on an analysis of current cumulative drought conditions and 30-day meteorological forecast data, the risk of drought affecting winter wheat in the province over the next 30 days is low, and no significant drought losses are expected.

[0030] Figure 4 The results show the estimated results of the winter wheat drought risk in the province for the next 60 days (i.e., from October 1, 2023 to June 9, 2024) based on the assessment starting point of April 10, 2024. The estimated period covers the key growth stages of winter wheat from jointing and heading to maturity and harvest. Figure 2 and Figure 3 compared to, Figure 4 The spatial distribution of risk is more complex. Most winter wheat-growing areas in the province remain at low risk (green), with an estimated yield reduction of no more than 5%. However, a relatively low-risk area (yellow) has emerged in the southwestern part of the province, with an estimated yield reduction of 5-10%. This change suggests that as the forecast timescale increases, the cumulative effects of meteorological drought may lead to an increase in drought risk levels in some areas. The absence of high-risk (red) or medium-risk (orange) areas in the map indicates that over the longer 60-day forecast period, winter wheat in most parts of the province will remain safe from severe drought threats.

[0031] The above embodiments fully and effectively achieve the objectives of the present invention. Those skilled in the art will appreciate that the present invention includes, but is not limited to, the contents described in the accompanying drawings and the above specific embodiments. Although the present invention has been described with reference to the embodiments currently considered to be the most practical and preferred, it should be understood that the present invention is not limited to the disclosed embodiments, and any modifications that do not deviate from the functional and structural principles of the present invention are intended to be included within the scope of the claims.

Claims

1. A method for quantitative dynamic risk estimation of winter wheat drought damage, characterized in that: include: S100. For the target assessment area, calculate the Meteorological Comprehensive Drought Index (MCI) based on historical daily meteorological data. Then, calculate the cumulative MCI index for moderate drought and above during the winter wheat growing season. Combined with historical winter wheat yield data, calculate the yield reduction rate. Based on the historical cumulative MCI index and yield reduction rate, fit a quantitative vulnerability relationship model between drought-causing factors and yield reduction rate. S200. Based on the daily meteorological drought composite index of the past and future, calculate the cumulative MCI index of moderate drought or above from the time of winter wheat sowing to the present time, indicating the current intensity of drought-causing factors. Also calculate the cumulative meteorological drought composite index of moderate drought or above from the time of sowing to the future period, indicating the estimated future intensity of drought-causing factors. S300. Use medium- and high-resolution multi-temporal satellite remote sensing imagery to identify the spatial grid distribution of winter wheat plantings in the target assessment area. Use remote sensing image processing and classification techniques to extract the rasterized location and area data of the winter wheat planting areas within geographic space. This serves as a spatial quantitative representation of the exposure of the winter wheat hazard-prone areas in the current year for risk assessment, providing input for subsequent disaster-loss coupled analysis. S400. Integrate the winter wheat drought risk, remote sensing planted area exposure, and vulnerability model results to calculate the potential yield reduction rate due to drought at the grid scale for the current and future periods. Based on the estimated yield reduction rate, risk levels are classified according to set thresholds. This generates a spatial distribution map of drought risk levels for winter wheat yield reduction in the current and future periods, along with statistical results of the area covered. S500. Based on the constructed drought risk level distribution map and estimation results, generate and output multi-timescale winter wheat drought damage assessment and prediction products in standard format layers, including current drought impact assessment maps, future drought risk prediction maps, and corresponding regional statistical results.

2. The method for quantitative dynamic risk estimation of winter wheat drought damage according to claim 1, characterized in that: In step S100, the modeling of the relationship between drought disaster factors and yield reduction rate vulnerability includes at least: S110. Preparation of Historical Meteorological and Yield Data: For the selected target assessment area, collect at least 30 years of continuous daily surface meteorological observation data and winter wheat yield data. Also, collect records of regional disaster types and agricultural disaster data for the same period, clarifying the years when non-drought disasters occurred and the extent of their impact. S120. Drought Intensity Index Statistics: Based on collected daily historical meteorological data, the daily meteorological drought comprehensive index (MCI) is calculated. For the winter wheat growing season, the cumulative MCI values ​​for moderate drought and above levels are calculated for each historical year. S130. Statistical calculation of winter wheat yield reduction rate: Decomposing actual winter wheat yield Y into trend yield Y t and climate yield Y c Based on the winter wheat yield data of previous years, the trend yield of each year was calculated using the fitting function Y t According to the formula Y c = Y - Y t Calculate climate yield according to the formula a = Y c / Y t ×100% to calculate relative climate yield, when a When it is less than 0, its absolute value is defined as the production reduction rate Y d Further cross-check disaster information on years with reduced production, and eliminate years with abnormal production due to non-drought factors; S140. Construction of a quantitative vulnerability relationship model: Using the historical drought-causing factor intensity as the independent variable and the historical yield reduction rate after eliminating abnormal yield reductions due to non-drought factors as the dependent variable, a linear or nonlinear quantitative response relationship model is constructed using statistical regression fitting methods. Y d = f (MCI), which quantifies the impact of drought on winter wheat yield during the growing season.

3. The method for quantitative dynamic risk estimation of winter wheat drought damage according to claim 1, characterized in that: In step S200, the dynamic statistical analysis of winter wheat drought disaster risk includes at least: S210. Real-time weather data acquisition: Acquires real-time daily weather data for the current winter wheat growing season, as well as daily climate forecast data for the next 30 to 60 days. S220. Calculation of the Meteorological Drought Comprehensive Index: Based on real-time meteorological data and future climate forecasts, the MCI value is calculated daily from the winter wheat sowing date to the current time, as well as the daily forecast MCI values ​​for the next 30 and 60 days. The drought severity for each day is determined based on the MCI value. S230. Statistics of the intensity of drought-causing factors in the past: Based on the real-time daily MCI values ​​calculated in step S220 up to the current time, the cumulative MCI values ​​of moderate drought and above since the winter wheat sowing time up to the current time are calculated as the current intensity indicator of the drought-causing factors; S240. Estimation of the intensity of future drought hazards: Based on the predicted daily MCI values ​​for the next 30 and 60 days calculated in step S220, the cumulative MCI values ​​for moderate drought or above from the time of winter wheat sowing to the next 30 days, and the cumulative MCI values ​​for moderate drought or above from the time of winter wheat sowing to the next 60 days, are calculated to quantify the intensity of future drought hazards that the winter wheat may experience. S250. Generate spatial distribution of drought disaster risk: Based on the statistical results of steps S230 and S240, combined with the geographical distribution information of meteorological stations, generate the spatial distribution of the intensity of drought disaster factors in the early stage and the spatial distribution of the intensity of drought disaster factors in the next 30 days and 60 days covering the target assessment area through spatial interpolation technology, provide rasterized spatial distribution data of drought disaster risk with geographical reference information, and realize the synchronous expression of drought intensity in time and space dimensions.

4. The method for quantitative dynamic risk estimation of winter wheat drought damage according to claim 1, characterized in that: In step S300, obtaining the satellite remote sensing spatial grid distribution of winter wheat planting includes at least: S310. Remote Sensing Data Acquisition and Preprocessing: Obtain medium- and high-resolution multi-temporal satellite remote sensing images of the target assessment area during the current winter wheat growing season. Perform radiometric calibration, atmospheric correction, geometric correction, and cloud removal on the acquired raw images. S320. Multi-temporal Remote Sensing Feature Extraction: Extract multidimensional feature parameters from pre-processed multi-temporal remote sensing images. Utilizing the spectral and phenological characteristics of winter wheat at different growth stages, a feature parameter set is constructed to identify winter wheat, enhancing its separability from other land features. S330. Winter Wheat Distribution Identification and Classification: Based on the extracted multidimensional feature parameters, classification techniques are used to identify and classify winter wheat planting areas within the target assessment area. Binary classification results are generated for the current year's winter wheat planting areas, and the spatial distribution range of the winter wheat planting areas is determined. S340. Classification Accuracy Verification and Optimization: Verify the accuracy of winter wheat classification results through field surveys of sample points, visual interpretation of high-resolution imagery, or existing crop distribution data, and optimize and adjust the classification results based on the verification results. S350. Generation of rasterized spatial distribution data of winter wheat: Convert the accuracy-verified and optimized winter wheat classification results into rasterized spatial distribution data that matches the spatial distribution of drought disaster risk, including the spatial location, area, and distribution density information of winter wheat planting areas, as a quantitative representation of the exposure of winter wheat disaster-prone bodies in the assessment area.

5. The method for quantitative dynamic risk estimation of winter wheat drought damage according to claim 1, characterized in that: In step S400, the quantitative spatial assessment of drought damage and risk level classification at least includes: S410. Spatial matching and registration of risk factor data: Spatial registration is performed on the drought risk spatial distribution data obtained in step S200 and the winter wheat planting spatial grid distribution data obtained in step S300 to ensure consistency between the two data types in spatial resolution, projection coordinate system, and spatial extent. S420. Grid-based calculation of drought loss rates based on disaster risk principles: Within each spatial grid cell, drought intensity, crop area exposure, and yield reduction rates estimated by vulnerability models are integrated to calculate the potential winter wheat yield reduction rate due to drought for the current, next 30 days, and next 60 days. S430. Drought Risk Classification: Based on the gridded yield reduction rate results and according to the preset risk classification criteria, the yield reduction rate value of each spatial grid is mapped to the corresponding risk level. Spatial distribution maps of winter wheat drought risk levels are generated for the current, next 30 days, and next 60 days time scales. Different colors or patterns are used to visually represent areas of different risk levels. S440. Statistics and analysis of risk coverage area: Based on the spatial distribution map of risk levels, statistics are compiled on the coverage areas of different risk levels within the target assessment area and their proportion to the total winter wheat planting area.

6. The method for quantitative dynamic risk estimation of winter wheat drought damage according to claim 1 or 5, characterized in that: In step S400, the preset winter wheat yield reduction rate drought risk level is: winter wheat yield reduction rate Y d ≤5% is low risk, 5%< Y d ≤10% is considered low risk, 10%< Y d ≤15% is medium risk, Y d >15% is considered high risk. Based on this threshold standard, the calculated winter wheat yield reduction rate is divided into risk levels, giving the risk level a clear quantitative meaning.

7. The method for quantitative dynamic risk estimation of winter wheat drought damage according to claim 1, characterized in that: In step S500, the multi-time-scale winter wheat drought damage assessment and prediction products include: a winter wheat drought dynamic impact assessment product, which reflects the estimated yield reduction rate and impact assessment of the drought in the current time period on winter wheat; a winter wheat drought dynamic risk prediction product for the next 30 days, which reflects the estimated yield reduction rate and risk assessment of the drought in the current time period and the next 30 days on winter wheat; and a winter wheat drought dynamic risk prediction product for the next 60 days, which reflects the estimated yield reduction rate and risk assessment of the drought in the current time period and the next 60 days on winter wheat.

8. The method for quantitative dynamic risk estimation of winter wheat drought damage according to claim 1 or 7, characterized in that: In step S500, the risk assessment product is presented in the form of a raster map, which includes a geographic spatial reference system, risk level classification representation, spatial distribution characteristics and coverage statistics, and risk statistics are summarized by different administrative division units as needed.

9. A computer program product comprising computer instructions, characterized in that Used to implement the quantitative dynamic risk estimation method for winter wheat drought damage according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for quantitative dynamic risk estimation of winter wheat drought damage according to any one of claims 1 to 8 is implemented.

Images