A grid-based spatio-temporal dynamic estimation method for nitrogen and phosphorus runoff loss of crops
Patent Information
- Application Number
- CN202611549885.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-10-08
- Publication Date
- 2026-11-03
AI Technical Summary
Existing technologies are insufficient for the continuous, spatial, and temporal estimation of nitrogen and phosphorus runoff loss from crops over many years. This is especially true in coastal areas with low-lying terrain and dense river networks, where existing methods lack detailed spatial location information and expressions of fertilization time differences, making it difficult to meet the needs of dynamic pollution generation assessment.
By constructing a multi-year continuous crop spatial distribution grid map, combining the differences in fertilization intensity and fertilization dynamic response coefficients across provinces and counties, and combining the time allocation coefficient, a three-dimensional refined estimation of crops, time, and space is achieved, generating gridded spatiotemporal dynamic data of nitrogen and phosphorus runoff loss.
It enables a three-dimensional, refined estimation of nitrogen and phosphorus runoff loss from crops, reflecting the risk and temporal changes of fertilization windows, and supporting risk assessment of agricultural non-point source pollution and optimization of fertilization management.
Smart Images
Figure CN123032185A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of agricultural non-point source pollution estimation and environmental management technology, specifically involving a gridded spatiotemporal dynamic estimation method for nitrogen and phosphorus runoff loss from crops. Background Technology
[0002] Besides being absorbed and utilized by crops, some nitrogen and phosphorus fertilizers applied during agricultural production can enter rivers, lakes, and nearshore waters through surface runoff, posing a significant source of agricultural non-point source pollution. Coastal areas are characterized by low-lying terrain, dense river networks, and numerous channels into the sea, allowing farmland nutrients to quickly enter estuaries and nearshore waters. This loss process is highly sensitive to fertilization timing, rainfall, and runoff conditions; simply calculating annual total amounts is insufficient to meet the needs of dynamic pollution generation assessment.
[0003] Existing methods for calculating crop spatial distribution mainly rely on administrative statistics or remote sensing classification products. While administrative statistics can provide regional planting area, they lack detailed spatial location information. Although remote sensing classification products can provide spatial distribution information, they suffer from issues such as insufficient crop coverage, discontinuous years, and inconsistencies with statistically reported planting areas, making it difficult to directly meet the needs of multi-year, continuous crop-scale nutrient loss accounting.
[0004] Regarding fertilizer input, existing methods typically determine annual fertilizer input based on provincial or county-level statistical data, rarely considering the differences in fertilization intensity between different crops and regions, and also failing to express the differences in the timing of basal fertilizer and topdressing. The timing, frequency, and proportion of basal fertilizer and topdressing application vary significantly among different crops, directly affecting nutrient release and loss risk. Expressing these values solely as annual averages would mask the peak loss during critical fertilization windows.
[0005] In estimating nutrient loss, existing fertilizer loss coefficient methods can estimate nutrient loss at the annual scale, but they struggle to further describe the temporal variations resulting from the combined effects of fertilization timing and runoff processes. While complex hydro-biogeochemical models can simulate nutrient migration processes, they require high-quality data and model parameters, hindering the rapid construction of multi-year, continuous spatial databases of farmland nitrogen and phosphorus loss over large areas. Therefore, there is an urgent need for an estimation method that integrates crop spatial distribution, statistical area, fertilizer input, fertilization timing, and runoff processes, while simultaneously achieving spatial and temporal dynamics of nitrogen and phosphorus loss from different crops. Summary of the Invention
[0006] To address the aforementioned issues, this invention proposes a gridded spatiotemporal dynamic estimation method for nitrogen and phosphorus runoff loss from crops. This method enhances the differences in fertilization intensity through cross-allocation between provinces and counties, characterizes the risk of fertilization windows through fertilization dynamic coefficients, and captures the coupling process between fertilization and runoff through time allocation coefficients. This achieves a three-dimensional, refined estimation of crop, time, and space, which can be used for agricultural non-point source pollution risk assessment and fertilization management optimization.
[0007] To achieve the above objectives, the present invention adopts the following technical solution: A gridded spatiotemporal dynamic estimation method for nitrogen and phosphorus runoff loss from crops includes the following steps: S1. Acquire crop spatial distribution data, remote sensing image data, crop suitability data and agricultural statistics for the target area, and construct a spatial distribution grid map of the target crops and their rotation patterns over many years. S2. Obtain nitrogen and phosphorus fertilizer application rates and recommended fertilization times for different crops and regions through statistical yearbooks and public data at all levels, and construct provincial and county-level databases of target crop fertilization intensity and fertilization time series. S3. Based on the cross-allocation of provincial and county-level fertilization data, calculate the nitrogen and phosphorus fertilization amounts and fertilization intensity per unit area for different crops at the county level; S4. Calculate the annual nitrogen and phosphorus runoff loss of the target crop based on the fertilizer intensity per unit area of crops at the county and city level, the planting area and the corresponding nutrient loss coefficient, and spatially correlate it with the crop spatial distribution grid data. S5. Construct the cumulative response law of nitrogen and phosphorus runoff loss to fertilization events under different crops, fertilization amounts and fertilization times, and generate fertilization dynamic response coefficients. S6. The dynamic response coefficient of integrated fertilization and the dynamic allocation coefficient of nitrogen and phosphorus loss over time are used to allocate the annual nutrient runoff loss to each time unit. S7. Spatially couple the annual nutrient runoff loss and temporal dynamic distribution results with the crop spatial distribution grid data to generate nitrogen and phosphorus runoff spatial distribution grid data with crop attributes, spatial attributes and temporal attributes. S8. Verify the estimation effect based on actual monitoring data at both the crop and time scales.
[0008] Preferably, in step S1, existing crop spatial distribution data is directly used for crops with continuous spatial distribution data; for crops lacking continuous spatial distribution data, a crop spatial distribution probability map is constructed based on one or more of the following: remote sensing spectral features, vegetation index features, phenological features, climate features, topographic features, water resource features, and crop suitability features. Spatial allocation is carried out using the planting area of administrative units as a total constraint. When creating the annual crop distribution map, the existing perennial crop distribution locations are masked to prevent overlapping planting locations. When the same grid is repeatedly allocated by different crops over consecutive years, it is regarded as a crop rotation area and the temporal rotation information of multiple crops is retained.
[0009] Preferably, in step S3, the formula for calculating the county / city-level crop fertilizer allocation ratio is as follows: In the formula, This represents the county-level crop fertilization allocation ratio for the i-th crop type in the j-th county-level administrative unit; i is the crop type number; n is the total number of crop types; j is the county-level administrative unit number; p is the province number; Let be the provincial crop nitrogen or phosphorus fertilizer input intensity per unit area in the p-th province for the i-th crop type; The planting area of the i-th crop type in the j-th county-level administrative unit; The formula for calculating the amount of nitrogen or phosphorus fertilizer input for crops at the county / city level is as follows: In the formula, For the i-th crop type and the j-th county-level administrative unit, the amount of nitrogen or phosphorus fertilizer input for crops at the county or city level. This represents the total nitrogen or phosphorus fertilizer input at the county / city level for the j-th county / city-level administrative unit. The formula for calculating the nitrogen or phosphorus fertilizer application intensity per unit area of crops at the county / city level is as follows: In the formula, For the i-th crop type and j-th county-level administrative unit, the nitrogen or phosphorus fertilizer application intensity per unit area of the crop; Let represent the county-level crop planting area of the i-th crop type in the j-th county-level administrative unit.
[0010] Preferably, in step S4, the formula for calculating the annual nitrogen or phosphorus runoff loss is: In the formula, For the i-th crop type and the j-th county-level administrative unit, the annual nitrogen or phosphorus runoff loss is denoted as ; The nitrogen or phosphorus fertilizer loss coefficient is the nitrogen or phosphorus fertilizer loss coefficient for the i-th crop type. The nitrogen or phosphorus fertilizer loss coefficient is determined according to the crop type and nutrient type through one or more of the following: agricultural pollution survey data, experimental monitoring data, literature statistical analysis results, or existing parameter databases.
[0011] Preferably, in step S5, the formula for calculating the fertilization dynamic response coefficient is: In the formula, is the fertilization dynamic response coefficient of the i-th crop type in time unit t; k is the fertilization event number; Let be the fertilization amount influence coefficient for the i-th crop type in the k-th fertilization event; This represents the time unit in which the k-th fertilization event occurs. For the i-th crop type, after fertilization interval The time-related influence coefficient; the fertilizer application rate influence coefficient is positively correlated with the single application rate; the time-related influence coefficient decreases as the interval after fertilization increases, and its parameters are determined through historical monitoring data, field trial data, or literature meta-analysis results.
[0012] Preferably, in step S6, the formula for calculating the time dynamic allocation coefficient is: In the formula, The time dynamic allocation coefficient for the i-th crop type in time unit t; The runoff is the runoff in time unit t; the time dynamic allocation coefficient is normalized so that the sum of the time dynamic allocation coefficients of the same crop in each time unit within the year is equal to 1.
[0013] Preferably, the spatial allocation in step S3 satisfies the annual total constraint at the administrative unit level, so that the sum of crop fertilizer application in each county-level administrative unit is equal to the total crop fertilizer application in the corresponding provincial-level administrative unit; the time allocation in step S6 satisfies the conservation of total annual nutrient loss, so that the sum of nutrient loss after allocation in each time unit is equal to the total annual nutrient loss.
[0014] Preferably, when a crop rotation system exists in the target area, crop spatial distribution data and fertilization time series data are established according to the crop planting status of different years, and the nitrogen and phosphorus nutrient runoff loss of the corresponding year is calculated respectively; the time unit is one or more of annual, quarterly, monthly or fertilization stage.
[0015] Preferably, the spatial distribution raster data generated in step S7 can be statistically and spatially analyzed according to crop type, administrative region, nutrient type, year and time unit, and output the nitrogen and phosphorus nutrient runoff loss and its spatiotemporal distribution results for different crops and different regions.
[0016] Preferably, in step S8, the actual monitoring data includes statistical data from the Ministry of Agriculture and Rural Affairs, field trial data, and measured data from literature. The estimation results are verified at both the crop type scale and the time unit scale, and the impact of changes in key parameters on the estimation results is evaluated through sensitivity analysis.
[0017] By adopting the above technical solution, the present invention has the following beneficial effects: 1. This invention integrates crop spatial distribution, county and city-level fertilization intensity, and fertilization time-series response into a unified spatialized loss estimation framework, which can output a refined three-dimensional nitrogen and phosphorus runoff loss distribution in terms of crop type, time, and space.
[0018] 2. This invention enhances the ability to express spatial differences in fertilization intensity by cross-allocating statistics at the provincial and county / city levels, and realizes the crop and regional differentiation of fertilization time information. Furthermore, it characterizes the risk of fertilization window through the dynamic response coefficient of fertilization, reflects the process of enhanced short-term loss after fertilization, and makes up for the shortcomings of annualized estimation.
[0019] 3. The time dynamic allocation coefficient of this invention couples fertilization response with runoff conditions, which can capture the coupling relationship between concentrated runoff during the flood season and rapid nutrient loss after fertilization. The final output results can be directly used for agricultural non-point source pollution risk assessment, fertilization measure optimization and responsibility area identification, and can adapt to regions with different data conditions based on multi-source data fusion. Attached Figure Description
[0020] Figure 1 This is a flowchart of the present invention. Detailed Implementation
[0021] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.
[0022] like Figure 1 As shown. A gridded spatiotemporal dynamic estimation method for crop nitrogen and phosphorus runoff loss includes the following steps: S1. Acquire crop spatial distribution data, remote sensing image data, crop suitability data and agricultural statistics for the target area, and construct a spatial distribution grid map of the target crops and their rotation patterns over many years. In step S1, existing crop spatial distribution data is directly used for crops with continuous spatial distribution data; for crops lacking continuous spatial distribution data, a crop spatial distribution probability map is constructed based on one or more of the following: remote sensing spectral features, vegetation index features, phenological features, climate features, topographic features, water resource features, and crop suitability features. Spatial allocation is carried out using the planting area of administrative units as a total constraint. When creating the annual crop distribution map, the existing perennial crop distribution locations are masked to prevent overlapping planting locations. When the same grid is repeatedly allocated by different crops in consecutive years, it is regarded as a rotation area and the temporal rotation information of multiple crops is retained. In one specific embodiment, the target crop is a crop type participating in the nitrogen and phosphorus nutrient runoff accounting, including but not limited to rice, corn, wheat, sugarcane, tea gardens, orchards, and vegetables, which can be determined based on the planting structure, fertilizer input, and contribution to nutrient loss in the study area. Existing crop spatial distribution data, remote sensing image data, crop suitability data, land use data, and county / city-level crop statistical planting area data for the target area are acquired. Crop distribution mapping employs a three-layer progressive masking process: The first layer directly uses existing distribution maps for crops with high-resolution continuous spatial distribution data; the second layer, for perennial crops lacking continuous spatial distribution data, constructs a spatial distribution probability map using a random forest model based on features such as remote sensing spectrum, vegetation index, phenology, climate, topography, water resources, and crop suitability. After excluding occupied grid cells using the first layer's distribution results as a mask, the area is allocated according to probability weights using the county / city-level statistical planting area as the total constraint; the third layer, for annual crops lacking continuous spatial distribution data, uses the second layer's overlay results as a mask and allocates only to the remaining available grid cells. The second and third layers are both constrained by the total planting area statistically recorded at the county and city levels. After allocation, each layer is checked to ensure that the sum of the allocated areas for each crop equals the statistical area. For annual crops, if the same grid is repeatedly allocated by different crops in consecutive years of mapping, it is considered a crop rotation area, and its multi-crop time-series rotation information is retained without deduplication. During area verification, it is checked separately for each crop and each year. The mapping accuracy is verified using area data from a more detailed statistical yearbook.
[0023] S2. Obtain nitrogen and phosphorus fertilizer application rates and recommended fertilization times for different crops and regions through statistical yearbooks and public data at all levels, and construct provincial and county-level databases of target crop fertilization intensity and fertilization time series. In one specific embodiment, the fertilization time database is established based on crop planting systems, methods recommended by regional agricultural technology extension departments, and fertilization records in statistical yearbooks, including the number of basal fertilizer applications, topdressing applications, and corresponding months.
[0024] S3. Based on the cross-allocation of provincial and county-level fertilization data, calculate the nitrogen and phosphorus fertilization amounts and fertilization intensity per unit area for different crops at the county level; In step S3, the formula for calculating the county-level crop fertilizer allocation ratio is as follows: In the formula, This represents the county-level crop fertilization allocation ratio for the i-th crop type in the j-th county-level administrative unit; i is the crop type number; n is the total number of crop types; j is the county-level administrative unit number; p is the province number; Let be the provincial crop nitrogen or phosphorus fertilizer input intensity per unit area in the p-th province for the i-th crop type; The planting area of the i-th crop type in the j-th county-level administrative unit; The formula for calculating the amount of nitrogen or phosphorus fertilizer input for crops at the county / city level is as follows: In the formula, For the i-th crop type and the j-th county-level administrative unit, the amount of nitrogen or phosphorus fertilizer input for crops at the county or city level. This represents the total nitrogen or phosphorus fertilizer input at the county / city level for the j-th county / city-level administrative unit. The formula for calculating the nitrogen or phosphorus fertilizer application intensity per unit area of crops at the county / city level is as follows: In the formula, For the i-th crop type and j-th county-level administrative unit, the nitrogen or phosphorus fertilizer application intensity per unit area of the crop; The planting area of the i-th crop type in the j-th county-level administrative unit; The spatial allocation in step S3 satisfies the annual total constraint at the administrative unit level, so that the sum of crop fertilizer application in each county-level administrative unit is equal to the total crop fertilizer application in the corresponding provincial-level administrative unit. By calculating the fertilizer allocation ratio, nitrogen or phosphorus fertilizer input, and nitrogen or phosphorus fertilizer intensity per unit area of crops at the county and city levels, the spatial redistribution of provincial fertilizer input to the county and city levels is achieved. This approach preserves differences at the county and city scale while ensuring that the aggregated results at each county and city level are consistent with the provincial total. Furthermore, by considering different fertilization regimes for different crops, the proportion of fertilizer application in different time units is determined.
[0025] S4. Calculate the annual nitrogen and phosphorus runoff loss of the target crop based on the fertilizer intensity per unit area of crops at the county and city level, the planting area and the corresponding nutrient loss coefficient, and spatially correlate it with the crop spatial distribution grid data. In step S4, the formula for calculating the annual nitrogen or phosphorus runoff loss is: In the formula, For the i-th crop type and the j-th county-level administrative unit, the annual nitrogen or phosphorus runoff loss is denoted as ; The nitrogen or phosphorus fertilizer loss coefficient for the i-th crop type is determined according to crop type and nutrient type, through one or more of the following: agricultural pollution survey data, experimental monitoring data, literature statistical analysis results, or existing parameter databases. In one specific embodiment, for crops lacking directly corresponding parameters, parameters of similar crops can be used to determine the parameters, and the uncertainty of the parameters can be evaluated through sensitivity analysis. After calculating the annual nitrogen or phosphorus runoff loss, it is spatially correlated with the corresponding crop spatial distribution grid data. Using the crop distribution grid as a link, the annual loss calculated at the administrative unit scale is allocated to each spatial unit according to the crop distribution weight within the grid. S5. Construct the cumulative response law of nitrogen and phosphorus runoff loss to fertilization events under different crops, fertilization amounts and fertilization times, and generate fertilization dynamic response coefficients. In step S5, the formula for calculating the fertilization dynamic response coefficient is: In the formula, is the fertilization dynamic response coefficient of the i-th crop type in time unit t; k is the fertilization event number; Let be the fertilization amount influence coefficient for the i-th crop type in the k-th fertilization event; This represents the time unit in which the k-th fertilization event occurs. For the i-th crop type, after fertilization interval The time-related influence coefficient; the fertilizer application rate influence coefficient is positively correlated with the single application rate; the time-related influence coefficient decreases as the interval after fertilization increases, and its parameters are determined through historical monitoring data, field trial data or literature meta-analysis results; S6. The dynamic response coefficient of integrated fertilization and the dynamic allocation coefficient of nitrogen and phosphorus loss over time are used to allocate the annual nutrient runoff loss to each time unit. In step S6, the formula for calculating the dynamic time allocation coefficient is: In the formula, The time dynamic allocation coefficient for the i-th crop type in time unit t; The runoff in time unit t can be obtained through regional hydrological model simulation, precipitation-runoff relationship calculation or actual measurement data from hydrological stations; the time dynamic allocation coefficient is normalized so that the sum of the time dynamic allocation coefficients of each time unit of the same crop in the year is equal to 1, ensuring that the total annual nutrient loss does not change due to time allocation. The time allocation in step S6 satisfies the conservation of the total annual nutrient loss, so that the sum of the nutrient loss after the allocation of each time unit is equal to the total annual nutrient loss. When a crop rotation system exists in the target area, crop spatial distribution data and fertilization time series data are established according to the crop planting status of different years, and the nitrogen and phosphorus nutrient runoff loss of the corresponding year is calculated respectively; the time unit is one or more of annual, quarterly, monthly or fertilization stage; S7. Spatially couple the annual nutrient runoff loss and temporal dynamic distribution results with the crop spatial distribution grid data to generate nitrogen and phosphorus runoff spatial distribution grid data with crop attributes, spatial attributes and temporal attributes. In one specific embodiment, crop type, spatial location, nutrient type, year, time unit, and corresponding nutrient (nitrogen or phosphorus) runoff loss are written into spatial grid cells to form a spatial distribution grid database of farmland nitrogen and phosphorus runoff loss.
[0026] The spatial distribution raster data generated in step S7 can be statistically and spatially analyzed according to crop type, administrative region, nutrient type, year and time unit, and output the nitrogen and phosphorus nutrient runoff loss and its spatiotemporal distribution results for different crops and different regions; S8. Verify the estimation effect from crop-scale and time-scale based on actual monitoring data; In step S8, the actual monitoring data includes statistical data from the Ministry of Agriculture and Rural Affairs, field trial data, and measured data from literature. The estimation results are verified at both the crop type scale and the time unit scale, and the impact of changes in key parameters on the estimation results is evaluated through sensitivity analysis.
[0027] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.
Claims
1. A gridded spatiotemporal dynamic estimation method for nitrogen and phosphorus runoff loss from crops, characterized in that, Includes the following steps: S1. Acquire crop spatial distribution data, remote sensing image data, crop suitability data and agricultural statistics for the target area, and construct a spatial distribution grid map of the target crops and their rotation patterns over many years. S2. Obtain nitrogen and phosphorus fertilizer application rates and recommended fertilization times for different crops and regions through statistical yearbooks and public data at all levels, and construct provincial and county-level databases of target crop fertilization intensity and fertilization time series. S3. Based on the cross-allocation of provincial and county-level fertilization data, calculate the nitrogen and phosphorus fertilization amounts and fertilization intensity per unit area for different crops at the county level; S4. Calculate the annual nitrogen and phosphorus runoff loss of the target crop based on the fertilizer intensity per unit area of crops at the county and city level, the planting area and the corresponding nutrient loss coefficient, and spatially correlate it with the crop spatial distribution grid data. S5. Construct the cumulative response law of nitrogen and phosphorus runoff loss to fertilization events under different crops, fertilization amounts and fertilization times, and generate fertilization dynamic response coefficients. S6. The dynamic response coefficient of integrated fertilization and the dynamic allocation coefficient of nitrogen and phosphorus loss over time are used to allocate the annual nutrient runoff loss to each time unit. S7. Spatially couple the annual nutrient runoff loss and temporal dynamic distribution results with crop spatial distribution grid data to generate nitrogen and phosphorus runoff spatial distribution grid data with crop attributes, spatial attributes and temporal attributes. S8. Verify the estimation effect based on actual monitoring data at both the crop and time scales.
2. The gridded spatiotemporal dynamic estimation method for crop nitrogen and phosphorus runoff loss as described in claim 1, characterized in that: In step S1, existing crop spatial distribution data is directly used for crops with continuous spatial distribution data; for crops lacking continuous spatial distribution data, a crop spatial distribution probability map is constructed based on one or more of the following: remote sensing spectral features, vegetation index features, phenological features, climate features, topographic features, water resource features, and crop suitability features. Spatial allocation is carried out using the planting area of administrative units as a total constraint. When creating the annual crop distribution map, the existing perennial crop distribution locations are masked to prevent overlapping planting locations. When the same grid is repeatedly allocated by different crops over consecutive years, it is regarded as a crop rotation area and the temporal rotation information of multiple crops is retained.
3. The gridded spatiotemporal dynamic estimation method for crop nitrogen and phosphorus runoff loss as described in claim 1, characterized in that, In step S3, the formula for calculating the county-level crop fertilizer allocation ratio is as follows: In the formula, The fertilizer allocation ratio for the i-th crop type in the j-th county-level administrative unit; i represents the crop type code; n represents the total number of crop types; j represents the county / city-level administrative unit code; p represents the province code; Let be the provincial crop nitrogen or phosphorus fertilizer input intensity per unit area in the p-th province for the i-th crop type; The planting area of the i-th crop type in the j-th county-level administrative unit; The formula for calculating the amount of nitrogen or phosphorus fertilizer input for crops at the county / city level is as follows: In the formula, For the i-th crop type and the j-th county-level administrative unit, the amount of nitrogen or phosphorus fertilizer input for crops at the county or city level. This represents the total nitrogen or phosphorus fertilizer input at the county / city level for the j-th county / city-level administrative unit. The formula for calculating the nitrogen or phosphorus fertilizer application intensity per unit area of crops at the county / city level is as follows: In the formula, For the i-th crop type and j-th county-level administrative unit, the nitrogen or phosphorus fertilizer application intensity per unit area of the crop; Let represent the county-level crop planting area of the i-th crop type in the j-th county-level administrative unit.
4. The gridded spatiotemporal dynamic estimation method for crop nitrogen and phosphorus runoff loss as described in claim 3, characterized in that, In step S4, the formula for calculating the annual nitrogen or phosphorus runoff loss is: In the formula, For the i-th crop type and the j-th county-level administrative unit, the annual nitrogen or phosphorus runoff loss is denoted as ; The nitrogen or phosphorus fertilizer loss coefficient is the nitrogen or phosphorus fertilizer loss coefficient for the i-th crop type. The nitrogen or phosphorus fertilizer loss coefficient is determined according to the crop type and nutrient type through one or more of the following: agricultural pollution survey data, experimental monitoring data, literature statistical analysis results, or existing parameter databases.
5. The gridded spatiotemporal dynamic estimation method for crop nitrogen and phosphorus runoff loss as described in claim 4, characterized in that, In step S5, the formula for calculating the fertilization dynamic response coefficient is: In the formula, is the fertilization dynamic response coefficient of the i-th crop type in time unit t; k is the fertilization event number; Let be the fertilization amount influence coefficient for the i-th crop type in the k-th fertilization event; This represents the time unit in which the k-th fertilization event occurs. For the i-th crop type, after fertilization interval The time-related influence coefficient; the fertilizer application rate influence coefficient is positively correlated with the single application rate; the time-related influence coefficient decreases as the interval after fertilization increases, and its parameters are determined through historical monitoring data, field trial data, or literature meta-analysis results.
6. The gridded spatiotemporal dynamic estimation method for crop nitrogen and phosphorus runoff loss as described in claim 5, characterized in that, In step S6, the formula for calculating the dynamic time allocation coefficient is: In the formula, The time dynamic allocation coefficient for the i-th crop type in time unit t; The runoff is the runoff in time unit t; the time dynamic allocation coefficient is normalized so that the sum of the time dynamic allocation coefficients of the same crop in each time unit within the year is equal to 1.
7. The gridded spatiotemporal dynamic estimation method for crop nitrogen and phosphorus runoff loss as described in claim 1, characterized in that: The spatial allocation in step S3 satisfies the annual total constraint at the administrative unit level, so that the sum of crop fertilizer application in each county-level administrative unit is equal to the total crop fertilizer application in the corresponding provincial-level administrative unit; the temporal allocation in step S6 satisfies the conservation of total annual nutrient loss, so that the sum of nutrient loss after allocation in each time unit is equal to the total annual nutrient loss.
8. The gridded spatiotemporal dynamic estimation method for crop nitrogen and phosphorus runoff loss as described in claim 1, characterized in that: When a crop rotation system exists in the target area, crop spatial distribution data and fertilization time series data are established according to the crop planting status of different years, and the nitrogen and phosphorus nutrient runoff loss of the corresponding year is calculated respectively; the time unit is one or more of annual, quarterly, monthly or fertilization stage.
9. The gridded spatiotemporal dynamic estimation method for crop nitrogen and phosphorus runoff loss as described in claim 1, characterized in that: The spatial distribution raster data generated in step S7 can be statistically and spatially analyzed according to crop type, administrative region, nutrient type, year and time unit, and output the nitrogen and phosphorus nutrient runoff loss and its spatiotemporal distribution results for different crops and different regions.
10. The gridded spatiotemporal dynamic estimation method for crop nitrogen and phosphorus runoff loss as described in claim 1, characterized in that: In step S8, the actual monitoring data includes statistical data from the Ministry of Agriculture and Rural Affairs, field trial data, and measured data from literature. The estimation results are verified at both the crop type scale and the time unit scale, and the impact of changes in key parameters on the estimation results is evaluated through sensitivity analysis.