Artificial precipitation enhancement operation demand index calculation method based on multi-source data fusion

Abstract

The invention discloses an artificial precipitation enhancement operation demand index calculation method based on multi-source data fusion. The method comprises the steps of obtaining multi-source heterogeneous data of a target area; carrying out normalization processing on the multi-source heterogeneous data for artificial precipitation enhancement operation requirements to obtain four types of normalized feature values; dividing the four types of normalized eigenvalues to district and county-level administrative districts through spatial interpolation or attribute aggregation to form four-dimensional eigenvectors corresponding to each district and county contained in the target region; and carrying out weighted fusion on the four-dimensional feature vector of each district and county, and introducing the drought level of the district and county to carry out constraint adjustment to obtain an artificial precipitation enhancement operation demand index of each district and county. According to the method, multi-source heterogeneous data of meteorology, environment, water conservancy, forestry and the like are comprehensively utilized, the artificial precipitation enhancement operation demand index refined to the district and county-level administrative district is calculated and output through intelligent weighted fusion, and a more scientific, accurate and timely quantitative decision basis is provided for artificial precipitation enhancement operation.

Description

Technical Field

[0001] This invention relates to the field of weather forecasting and decision support technology, and more specifically to a method for calculating the demand index of artificial rain enhancement operations based on multi-source data fusion. Background Technology

[0002] Artificial rain enhancement, as an important means of meteorological disaster prevention and mitigation and water resource regulation, has been widely applied in fields such as agricultural drought relief, reservoir water storage, forest fire prevention, and air pollution control. Scientifically and accurately identifying and quantifying the demand for artificial rain enhancement operations is a key prerequisite for improving operational efficiency, optimizing resource allocation, and avoiding ineffective or excessive operations. However, traditional methods of assessing artificial rain enhancement demand often rely on single meteorological factors (such as precipitation anomaly percentage, soil moisture, or meteorological drought index), lacking a systematic integration and quantitative assessment of actual needs across multiple fields, and thus failing to comprehensively reflect the urgency of regional integrated rain enhancement.

[0003] In recent years, with the continuous improvement of informatization levels in departments such as meteorology, environmental protection, water resources, and forestry, multi-source heterogeneous data has become increasingly abundant. For example, the meteorological drought composite index (MCI) released by the meteorological department can objectively characterize the degree of regional drought; the air quality index (AQI) released in real time by the ecological environment department reflects the state of air pollution, and there is often an emergency need to improve air quality through rain enhancement during periods of heavy pollution; the information on reservoir water levels, storage capacity, and water potential changes monitored by the water resources department is directly related to water resource allocation and water supply security; and the forest fire weather risk level released by the forestry department constitutes a significant need for rain enhancement and fire prevention during periods of high fire risk. These data reveal the potential application scenarios of artificial rain enhancement from different dimensions, but a unified and standardized fusion computing framework has not yet been formed, resulting in the "islanding" of information on the needs of various departments, making it difficult to support collaborative decision-making.

[0004] In existing technologies, some studies attempt to construct rain enhancement potential indices or operational condition indices, but these mostly focus on cloud physical conditions (such as cloud top temperature, liquid water content, and wind field structure), emphasizing the feasibility of "whether operations are possible," while paying insufficient attention to the comprehensive social-ecological-economic demand assessment of "whether operations are necessary." A few other methods introduce multi-factor weighting, but the weighting is highly subjective, lacks operational basis, and fails to consider threshold constraints for different demand factors under extreme events (e.g., even if other factors are high during mild drought, overall demand should still be limited). Furthermore, existing methods also have shortcomings in data scale processing, failing to effectively address the spatial matching and scale consistency issues between station observation data, areal administrative division data, and point facility (such as reservoir) data, thus affecting the accuracy of refined demand assessments at the district and county levels.

[0005] Therefore, providing a technical solution that can integrate multi-source operational data from meteorology, environmental protection, water conservancy, forestry, etc., and based on scientific normalization and dynamic weighting mechanisms to achieve quantitative assessment of artificial rain enhancement operation demand at the county and district level is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0006] In view of the above problems, this invention is proposed to provide a method for calculating the demand index of artificial rain enhancement operations based on multi-source data fusion to overcome or at least partially solve the above problems. This method not only solves the limitation of traditional methods that rely solely on a single meteorological indicator, but also constructs a standardized, scalable, and business-friendly artificial rain enhancement demand assessment system.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: This invention provides a method for calculating the demand index of artificial rain enhancement operations based on multi-source data fusion, including: S1. Obtain multi-source heterogeneous data of the target area; the multi-source heterogeneous data includes meteorological drought comprehensive index, air quality index, reservoir water status information and forest fire weather level; the target area includes multiple district and county-level administrative regions; S2. Normalize the multi-source heterogeneous data according to the needs of artificial rain enhancement operations to obtain four types of normalized feature values. S3. The four types of normalized feature values ​​are assigned to district / county-level administrative regions through spatial interpolation or attribute aggregation to form a four-dimensional feature vector corresponding to each district / county included in the target area. S4. The four-dimensional feature vectors of each district and county are weighted and fused, and the drought level of the district and county is introduced for constraint adjustment to obtain the artificial rain enhancement operation demand index of each district and county.

[0008] Further, step S2 specifically includes: ( a Based on the extreme values ​​of the historical meteorological drought composite index, the meteorological drought composite index is back-normalized to obtain the drought demand characteristic value. a 1; ( b Air Quality Index AQI Normalization is performed using a piecewise function to calculate the characteristic value of air demand. a 2; when 0≤ AQI <100, a 2= AQI / 100; when AQI When ≥100, a 2 = 1; ( cThe reservoir water potential information is mapped to numerical values, and the maximum value of the mapped values ​​of all reservoirs within each district / county-level administrative region is taken as the water demand characteristic value of the current district / county. a 3; ( d The forest fire weather risk level is mapped to a binary variable; when the fire risk level is Level 1, the fire risk demand characteristic value is... a 4=0; When the fire hazard level is level two or above, the characteristic value of fire hazard demand is... a 4 = 1.

[0009] Furthermore, based on the extreme values ​​of the historical meteorological drought composite index, the meteorological drought composite index is back-normalized to obtain the drought demand characteristic value. a 1; expressed by the formula:

[0010] in, MCI This represents the comprehensive meteorological drought index. Max hist This represents the historical maximum drought value. Min hist This represents the lowest historical drought value.

[0011] Furthermore, the numerical mapping rule for the reservoir water potential state information is as follows: When the reservoir water level status information is "falling", it is mapped to 1, indicating a high demand for artificial rain enhancement operations; When the reservoir water level status information is "flat", the mapping is 0, indicating a general demand for artificial rain enhancement operations; When the reservoir water level is "rising", it is mapped to -1, indicating a low demand for artificial rain enhancement operations.

[0012] Furthermore, step S3 specifically includes: Meteorological drought composite index MCI and air quality index AQI, As for site-type data, the corresponding feature values ​​are interpolated into continuous raster surfaces using the Kriging interpolation method, and then the average value of the corresponding feature values ​​in each district and county is obtained through regional statistics. The reservoir water level status information, as point data, is first located in the administrative region to which it belongs, and then the water conservancy demand characteristic values ​​are aggregated by district and county, and the maximum value is taken. Forest fire weather risk levels, as areal data, are directly correlated with fire risk demand characteristic values ​​according to administrative divisions.

[0013] Furthermore, in step S4, the four-dimensional feature vectors of each district / county are weighted and fused, as expressed by the formula:

[0014]

[0015] in, I j,init This represents the initial value of the demand index for artificial rain enhancement operations. a 1j Indicates the first j Drought demand characteristics of each district and county, a 2j Indicates the first j Air demand characteristics of each district and county, a 3j Indicates the first j Water demand characteristics of each district and county a 4j Indicates the first j Fire risk demand characteristics of each district and county, w 1. w 2. w 3 and w 4 represents the weights of drought demand, air demand, water demand, and fire risk demand, respectively.

[0016] Furthermore, in step S4, drought level constraint adjustments are introduced, specifically including: A drought level is introduced as a mandatory constraint to adjust the lower limit of the initial value of the artificial rain enhancement operation demand index: When the drought level of the corresponding district / county is mild drought, the final demand index for artificial rain enhancement operations is: I j =max( I j,init ,T1); When the drought level of the corresponding district / county is moderate drought, the final demand index for artificial rain enhancement operations is: I j =max( I j,init ,T2); When the corresponding county or district is classified as severely or extremely dry, the final demand index for artificial rain enhancement operations is: I j =max( I j,init ,T3); Where T1, T2 and T3 represent the predefined threshold values ​​for the demand index of artificial rain enhancement operations at different drought levels.

[0017] Furthermore, the demand index for artificial rain enhancement operations, The artificial rain enhancement operation needs are categorized into corresponding levels based on preset thresholds; these levels include none, low, moderate, and high needs. It is also used in conjunction with geographic information systems to output results including district and county codes, normalized components, artificial rain enhancement operation demand index and levels, and to generate visual maps.

[0018] As can be seen from the above technical solution, compared with the prior art, the present invention discloses a method for calculating the demand index of artificial rain enhancement operations based on multi-source data fusion, which has the following beneficial effects: 1. Achieve unified quantification and integration of the needs of multiple departments. For the first time, the operational needs of multiple fields, including meteorology (drought), environmental protection (air quality), water conservancy (reservoir water level), and forestry (forest fire risk), have been incorporated into the same assessment framework, breaking down data silos and enabling artificial rain enhancement decisions to shift from "single meteorological drive" to "multi-dimensional social-ecological demand synergistic drive," significantly improving the scientific and comprehensive nature of operational deployment.

[0019] 2. Improve the accuracy and targeting of artificial rain enhancement operations. By calculating the demand index at the district and county-level administrative unit scale, refined spatial positioning can be achieved, which can accurately identify high-demand areas, avoid "sprinkling pepper" operations, optimize the spatiotemporal allocation of operational resources (such as rockets, aircraft, and catalysts), and improve operational efficiency and input-output ratio.

[0020] 3. Introduce normalization and threshold constraint mechanisms to enhance logical rationality. To address the differences in dimensions and physical meanings of data from different sources, a customized normalization method tailored to rainfall enhancement needs is designed, for example... MCI Reverse mapping AQI Segmented processing and numericalization of reservoir status are implemented; at the same time, lower limit constraints are set for the index in conjunction with drought levels to ensure the basic business principle of "no rain enhancement without drought" or "low demand due to mild drought" and prevent misjudgment.

[0021] 4. Possesses good explainability and business operability. The index calculation process is transparent, and the contribution of each component (drought, pollution, water storage, fire risk) is clear and traceable, which facilitates consultation and judgment among multiple departments such as meteorology, environmental protection, water conservancy, and emergency response. The final output includes index values, level classifications, and a visual map, which can be directly embedded into existing weather modification business platforms to support rapid decision-making.

[0022] 5. Supports dynamic adjustment and scene adaptation, with strong scalability. The weighting coefficients can be flexibly configured according to regional characteristics (such as severe drought areas in the north vs. fire prevention areas in the south), seasonal changes (such as the high incidence of heavy pollution in autumn and winter), or emergency response status; the methodology framework can also be extended to access new data sources such as soil moisture and cloud water resource assessment to adapt to future business development needs.

[0023] 6. Promote the construction of cross-departmental collaboration and scientific decision-making mechanisms To provide technical tools for establishing a consultation mechanism for artificial rain enhancement needs that is led by meteorology and involves multiple departments, promote the transformation from experience-based judgment to data-driven approaches and from passive response to proactive planning, and enhance the comprehensive protection capabilities of national water resources security, ecological security and public safety. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention 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 embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0025] Figure 1 This is a flowchart of the method for calculating the demand index of artificial rain enhancement operations based on multi-source data fusion provided in this embodiment of the invention. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] This invention discloses a method for calculating the demand index for artificial rain enhancement operations based on multi-source data fusion, referring to... Figure 1 As shown, it includes: S1. Obtain multi-source heterogeneous data of the target area; the multi-source heterogeneous data includes meteorological drought comprehensive index, air quality index, reservoir water status information and forest fire weather level; the target area includes multiple district and county-level administrative regions; S2. Normalize the multi-source heterogeneous data according to the needs of artificial rain enhancement operations to obtain four types of normalized feature values. S3. The four types of normalized feature values ​​are assigned to district / county-level administrative regions through spatial interpolation or attribute aggregation to form a four-dimensional feature vector corresponding to each district / county included in the target area. S4. The four-dimensional feature vectors of each district and county are weighted and fused, and the drought level of the district and county is introduced for constraint adjustment to obtain the artificial rain enhancement operation demand index of each district and county.

[0028] This embodiment uses a province in China as the target area to calculate and visualize the current artificial rain enhancement operation demand index at the county-level administrative unit scale. First, multi-source data access and preprocessing are performed: monitoring and forecast data from meteorological, environmental protection, water conservancy, and forestry departments are accessed in real time and standardized. Second, data normalization processing is performed to meet rain enhancement demand: different normalization functions are designed for each type of data, mapping them uniformly to the contribution range to rain enhancement demand (e.g., [0,1]), where a larger value indicates a more urgent demand for rain enhancement. Spatial scale standardization: all data are unified to the county-level administrative unit scale through spatial interpolation or attribute aggregation methods, forming spatially aligned feature vectors. Weighted fusion and constraint adjustment: the four types of normalized feature values ​​are weighted and summed to form an initial comprehensive index; subsequently, constraints based on drought levels are introduced to forcibly adjust the lower limit of the initial index, ensuring that the index reflects the minimum operational demand under specific drought conditions. Finally, the index is divided into levels and visualized: the final index is divided into multiple levels (such as none, low, average, high) and visualized in the form of a map.

[0029] The implementation steps of this embodiment are described in detail below: Follow step S1 to obtain multi-source heterogeneous data of the target area.

[0030] Obtain the daily meteorological drought composite index (MCI) data and its corresponding drought level released by the meteorological department.

[0031] Obtain the daily Air Quality Index (AQI) data from various air quality monitoring stations released by the environmental protection department.

[0032] Obtain daily water level data for each reservoir released by the water resources department, including at least information such as reservoir water level and water flow status.

[0033] Obtain the forest fire weather risk level forecast product issued by the forestry department and read the forecast fire risk level for each district and county on that day.

[0034] According to step S2, the acquired multi-source heterogeneous data are normalized to meet the needs of artificial rain enhancement operations, resulting in four types of normalized feature values.

[0035] This embodiment normalizes the meteorological drought composite index to obtain drought demand characteristic values. a 1.

[0036] against MCI The index employs a reverse normalization method based on historical extreme values. The formula is as follows:

[0037] in, Max histThis represents the historical maximum drought value. Min hist This represents the historical minimum drought value. In this embodiment... Min hist and Max hist For nearly a year MCI Statistical extrema, this design makes MCI The smaller the value (the more severe the drought), the higher the normalized value. a The higher the value of 1, the greater the demand for rain enhancement.

[0038] This embodiment uses a piecewise function to normalize the air quality index and calculates the air demand characteristic value. a 2. Expressed as a formula:

[0039] when AQI <100, a 2= AQI / 100; when AQI When ≥100, a 2=1; When pollution reaches a certain level, the demand for rain enhancement reaches saturation.

[0040] This embodiment normalizes the reservoir water state information and calculates the characteristic value of water demand. a 3.

[0041] Map reservoir water level status codes to numerical values. For example, "declining" is mapped to 1 (high demand), "neutral" to 0 (moderate demand), and "rising" to -1 (low demand). Iterate through all reservoirs in each district / county, and obtain the final value for each district / county. a 3” takes the maximum value of this value among all reservoirs in the district / county, reflecting the most urgent water conservancy demand; that is, as long as there is a reservoir “falling”, there is a water conservancy demand.

[0042] This embodiment normalizes the forest fire weather risk level and calculates the characteristic value of fire risk demand. a 4.

[0043] The forest fire risk level is converted into a binary variable, expressed by the formula:

[0044] Map fire hazard levels (1-4) to numerical values. Level 1 is 0, and Level 2 and above are 1. Record this value as... a 4.

[0045] The four types of normalized eigenvalues ​​obtained in this embodiment are a 1. a 2. a 3. a 4.

[0046] According to step S3, the four types of normalized feature values ​​are assigned to district and county-level administrative regions through spatial interpolation or attribute aggregation to form a four-dimensional feature vector for each district and county in the target area.

[0047] This embodiment aggregates the above four types of data into a unified district / county-level administrative division: For site-type data MCI and AQI Continuous raster surfaces are generated using ordinary kriging interpolation, and then the average values ​​for each district / county are calculated using zonal statistics, which are then used as... a 1j and a 2j ; a 1j Indicates the first j Drought demand characteristics of each district and county, a 2j Indicates the first j Air demand characteristics of each district and county.

[0048] For point-based reservoir data: Based on administrative region codes, the maximum value is directly taken as the classification. a 3j ; a 3j Indicates the first j Water demand characteristics of each district and county.

[0049] For area fire hazard rating data: if the fire hazard of any grid point is ≥ Level 2, then the district / county a 4j =1. a 4j Indicates the first j Fire risk demand characteristics of each district and county.

[0050] Ultimately, each district and county will be formed. j Four-dimensional feature vectors: V j =[ a 1j ,a 2j ,a 3j ,a 4j ].

[0051] Following step S4, the four-dimensional feature vectors of each district and county are weighted and fused, and drought level constraints are introduced for adjustment to obtain the artificial rain enhancement operation demand index for each district and county.

[0052] First, in this embodiment, the four-dimensional feature vectors of each district / county are weighted and fused, as expressed by the formula:

[0053]

[0054] Among them, I j,init represents the initial value of the artificial rainfall enhancement operation demand index. w 1. w 2. w 3 and w 4 respectively represent the weights corresponding to drought demand, air demand, water conservancy demand, and fire risk demand.

[0055] The optimal weight configuration in this embodiment is: w 1 = 0.6, w 2 = 0.1, w 3 = 0.1, w 4 = 0.2.

[0056] Secondly, constraint adjustment is performed based on the drought level. [[ID=3...]]

[0057] In this embodiment, the drought level is introduced as a mandatory constraint condition to adjust the lower limit of the initial value of the artificial rainfall enhancement operation demand index. The adjustment rules are defined as follows: If the drought level of the district or county j is light drought, that is, MCI ∈(0, 1.0], then the final index I j = max( I j,init , T1); <00003...]] If the drought level of the district or county j is moderate drought, that is, MCI ∈(1.0, 1.5], then I j = max( I j,init , T2); If the drought level of the district or county j is severe drought or extreme drought, that is, MCI > 1.5, then I j = max( I j,init , T3).

[0058] Among them, T1, T2, and T3 respectively represent the thresholds of the artificial rainfall enhancement operation demand index for different drought levels defined in advance. And 0 < T1 < T2 < T3 ≤ 1. For example, T1 = 0.25, T2 = 0.5, T3 = 0.75. This step ensures that when a drought occurs, the comprehensive index can reflect its basic urgency. This mechanism ensures that even if other factors are weak, severely drought-stricken areas are still identified as high-demand areas.

[0059] Finally, this embodiment will calculate the rainfall enhancement demand index for each district / county. I j It is limited to the interval [0,1].

[0060] The index levels are divided according to preset level thresholds, such as: None: [0, 0.25), Low: [0.25, 0.5), Normal: [0.5, 0.75), High: [0.75, 1.0].

[0061] Generate data including district / county codes, names, and normalized values ​​for each component. a 1- a 4) Final Index ( I j (and graded forms).

[0062] By combining a geographic information system, this embodiment uses different colors to render each district and county, thereby realizing a visual map display of the spatial distribution of the demand index.

[0063] This invention proposes a weighted fusion model integrating four types of data: drought, environmental protection, water conservancy, and forest fire risk. It features clear physical meaning, standardized calculation procedures, and strong interpretability of results, providing reliable decision support for the scientific scheduling and precise implementation of weather modification operations.

[0064] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0065] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for calculating the demand index for artificial rain enhancement operations based on multi-source data fusion, characterized in that, include: S1. Obtain multi-source heterogeneous data of the target area; the multi-source heterogeneous data includes meteorological drought comprehensive index, air quality index, reservoir water status information and forest fire weather level; the target area includes multiple district and county-level administrative regions; S2. Normalize the multi-source heterogeneous data according to the needs of artificial rain enhancement operations to obtain four types of normalized feature values. S3. The four types of normalized feature values ​​are assigned to district / county-level administrative regions through spatial interpolation or attribute aggregation to form a four-dimensional feature vector corresponding to each district / county included in the target area. S4. The four-dimensional feature vectors of each district and county are weighted and fused, and the drought level of the district and county is introduced for constraint adjustment to obtain the artificial rain enhancement operation demand index of each district and county.

2. The method as described in claim 1, characterized in that, Step S2 specifically includes: ( a Based on the extreme values ​​of the historical meteorological drought composite index, the meteorological drought composite index is back-normalized to obtain the drought demand characteristic value. a 1; ( b Air Quality Index AQI Normalization is performed using a piecewise function to calculate the characteristic value of air demand. a 2; when 0≤ AQI <100, a 2= AQI / 100; when AQI When ≥100, a 2 = 1; ( c The reservoir water potential information is mapped to numerical values, and the maximum value of the mapped values ​​of all reservoirs within each district / county-level administrative region is taken as the water demand characteristic value of the current district / county. a 3; ( d The forest fire weather risk level is mapped to a binary variable; when the fire risk level is Level 1, the fire risk demand characteristic value is... a 4=0; When the fire hazard level is level two or above, the characteristic value of fire hazard demand is... a 4 = 1.

3. The method as described in claim 2, characterized in that, The meteorological drought composite index is inversely normalized based on the extreme values ​​of historical meteorological drought composite index to obtain drought demand characteristic values. a 1; expressed by the formula: in, MCI This represents the comprehensive meteorological drought index. Max hist This represents the historical maximum drought value. Min hist This represents the lowest historical drought value.

4. The method as described in claim 2, characterized in that, The numerical mapping rule for the reservoir water potential state information is as follows: When the reservoir water level status information is "falling", it is mapped to 1, indicating a high demand for artificial rain enhancement operations; When the reservoir water level status information is "flat", the mapping is 0, indicating a general demand for artificial rain enhancement operations; When the reservoir water level is "rising", it is mapped to -1, indicating a low demand for artificial rain enhancement operations.

5. The method as described in claim 2, characterized in that, Step S3 specifically includes: Meteorological drought composite index MCI and air quality index AQI, As for site-type data, the corresponding feature values ​​are interpolated into continuous raster surfaces using the Kriging interpolation method, and then the average value of the corresponding feature values ​​in each district and county is obtained through regional statistics. The reservoir water level status information, as point data, is first located in the administrative region to which it belongs, and then the water conservancy demand characteristic values ​​are aggregated by district and county, and the maximum value is taken. Forest fire weather risk levels, as areal data, are directly correlated with fire risk demand characteristic values ​​according to administrative divisions.

6. The method as described in claim 1, characterized in that, In step S4, the four-dimensional feature vectors of each district / county are weighted and fused, as expressed by the formula: in, I j,init This represents the initial value of the demand index for artificial rain enhancement operations. a 1j Indicates the first j Drought demand characteristic values ​​for each district and county, a 2j Indicates the first j Air demand characteristics of each district and county, a 3j Indicates the first j Water demand characteristics of each district and county a 4j Indicates the first j Fire risk demand characteristics of each district and county, w 1. w 2. w 3 and w 4 represents the weights of drought demand, air demand, water demand, and fire risk demand, respectively.

7. The method as described in claim 6, characterized in that, In step S4, drought level constraints are adjusted, specifically including: A drought level is introduced as a mandatory constraint to adjust the lower limit of the initial value of the artificial rain enhancement operation demand index: When the drought level of the corresponding district / county is mild drought, the final demand index for artificial rain enhancement operations is: I j =max( I j,init ,T1); When the drought level of the corresponding district / county is moderate drought, the final demand index for artificial rain enhancement operations is: I j =max( I j,init ,T2); When the drought level of the corresponding district / county is severe drought or extreme drought, the final demand index for artificial rain enhancement operations is: I j =max( I j,init ,T3); Where T1, T2 and T3 represent the predefined threshold values ​​for the demand index of artificial rain enhancement operations at different drought levels.

Images