Pollen discharge and concentration simulation system and method

Abstract

The invention discloses a pollen discharge and concentration simulation system and a pollen discharge and concentration simulation method, which are used for solving the problem of low pollen discharge and concentration simulation accuracy, and receiving a pollen discharge and concentration simulation request sent by a terminal, predicting the total pollen discharge yield of the to-be-observed vegetation at the specified observation station in the pollen discharge period according to the acquired meteorological observation data of the specified observation station in each day in the pollen discharge period of the to-be-observed vegetation and a pollen yield prediction model; determining a pollen discharge potential capacity value of the tth day according to the temperature cumulant of the tth day of the pollen discharge period, the accumulated temperature threshold of the pollen starting period, the accumulated temperature threshold of the pollen ending period, the number of days of the pollen discharge period and a standardized pollen discharge potential model; determining the pollen discharge amount of the tth day according to the total pollen discharge yield and the potential pollen discharge capacity value of the tth day; and determining a target pollen discharge amount of the tth day according to the pollen discharge amount of the tth day and the meteorological factors of the tth day, and simulating the pollen discharge and concentration of each day in the pollen discharge period.

Description

Technical Field

[0001] This application relates to the field of pollen emission concentration analysis technology, and in particular to a pollen emission and concentration simulation system and method. Background Art

[0002] Pollen emission and concentration simulation plays a vital role in environmental science and public health, especially during allergy seasons. Tree pollen is a significant source and trigger of spring allergies, and simulations of tree pollen emissions and concentrations provide important guidance for individual health management and public health policy development. Furthermore, pollen emission and concentration simulations are significant for research on the characteristics of different vegetation types within the field of environmental science.

[0003] Currently, pollen emission and concentration simulation is mainly based on traditional statistical methods, using historical observation data and simple linear regression models to perform regression analysis to simulate tree pollen emission and concentration. However, due to the simple structure of the linear regression model, it is difficult to capture the complex nonlinear relationship between pollen emission and concentration and other influencing factors (such as environmental factors), resulting in low accuracy of pollen emission and concentration simulation results.

[0004] Therefore, improving the accuracy of pollen emission and pollen concentration simulation is one of the technical problems that urgently need to be solved in the existing technology. Summary of the Invention

[0005] To address the issue of low accuracy in simulating pollen emissions and concentrations, embodiments of this application provide a pollen emissions and concentration simulation system and method.

[0006] In a first aspect, embodiments of this application provide a pollen emission and concentration simulation system, including: The communication module is used to receive pollen emission and concentration simulation requests sent by the terminal. The pollen emission and concentration simulation requests carry information on vegetation to be observed, information on designated observation stations, and information on the year to be simulated. The data acquisition module is used to acquire meteorological observation data of a specified observation station for each day during the pollen emission period of the vegetation to be observed, based on the year simulated in the request. The processor is configured to predict the total pollen emission of the vegetation under observation at the designated observation station during the pollen emission period, based on meteorological observation data from the designated observation station on each day of the pollen emission period and a pollen yield prediction model. The pollen yield prediction model is trained using a preset machine learning model based on historical sample meteorological observation data from the sample observation station and sample pollen observation data of the vegetation under observation. The processor also acquires the cumulative temperature on day t of the pollen emission period, and calculates the cumulative temperature based on the cumulative temperature on day t, the accumulated temperature threshold at the beginning of the pollen season, the accumulated temperature threshold at the end of the pollen season, and the pollen emission period... The pollen emission potential capacity value for day t is determined by using the number of days and a standardized pollen emission potential model. The standardized pollen emission potential model is used to calculate the pollen emission potential capacity value. The pollen emission amount for day t is determined based on the total pollen emission and the pollen emission potential capacity value for day t. Meteorological factors for day t are obtained, and the target pollen emission amount for day t is determined based on the pollen emission amount and the meteorological factors for day t. The pollen emission and concentration for each day of the pollen emission period are simulated, and the target pollen emission amount is the pollen emission amount after adding meteorological disturbances. The response module is used to return to the terminal the simulated results of pollen emission and concentration for each day of the pollen emission period.

[0007] Secondly, embodiments of this application provide a method for simulating pollen emission and concentration, including: The receiving terminal sends a pollen emission and concentration simulation request, which carries information about the vegetation to be observed, information about the designated observation station, and information about the year to be simulated. Based on the year simulated in the request, obtain meteorological observation data from a specified observation station for each day of the pollen emission period of the vegetation to be observed. Based on the meteorological observation data of the designated observation station for each day during the pollen emission period of the vegetation to be observed and the pollen yield prediction model, the total pollen emission of the vegetation to be observed at the designated observation station during the pollen emission period is predicted. The pollen yield prediction model is obtained by training a preset machine learning model based on the historical sample meteorological observation data of the sample observation station and the sample pollen observation data of the vegetation to be observed. The cumulative temperature on day t of the pollen emission period is obtained. Based on the cumulative temperature on day t, the accumulated temperature threshold at the beginning of the pollen period, the accumulated temperature threshold at the end of the pollen period, the number of days in the pollen emission period, and the constructed standardized pollen emission potential model, the pollen emission potential value on day t is determined. The standardized pollen emission potential model is used to calculate the pollen emission potential value. The pollen emission amount on day t is determined based on the total pollen emission production and the pollen emission potential capacity value on day t. Obtain the meteorological factors for day t, determine the target pollen emission for day t based on the pollen emission amount and the meteorological factors for day t, and simulate the pollen emission and concentration for each day of the pollen emission period. The target pollen emission amount is the pollen emission amount after adding meteorological disturbances. The simulated pollen emission and concentration results for each day of the pollen emission period are returned to the terminal.

[0008] Thirdly, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the pollen emission and concentration simulation method described in this application.

[0009] Fourthly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the pollen emission and concentration simulation method described in this application.

[0010] The beneficial effects of this application are as follows: The pollen emission and concentration simulation system provided in this application includes a communication module, a data acquisition module, a processor, and a response module. The communication module receives pollen emission and concentration simulation requests sent by a terminal. These requests carry information about the vegetation to be observed, information about designated observation stations, and information about the year to be simulated. The data acquisition module acquires meteorological observation data from designated observation stations for each day of the pollen emission period of the vegetation to be observed, based on the year to be simulated. The processor predicts the total pollen emission of the vegetation to be observed during the pollen emission period based on the meteorological observation data from designated observation stations for each day of the pollen emission period and a pollen yield prediction model. The pollen yield prediction model is trained using a preset machine learning model based on historical sample meteorological observation data from sample observation stations and sample pollen observation data from the vegetation to be observed. The system obtains the cumulative temperature on day t of the pollen emission period. Based on the cumulative temperature on day t, the accumulated temperature thresholds for the pollen initiation and termination periods, the number of days in the pollen emission period, and the constructed standardized pollen emission potential model, it determines the pollen emission potential value on day t. The standardized pollen emission potential model is used to calculate the pollen emission potential value. Based on the total pollen emission and the pollen emission potential value on day t, it determines the pollen emission amount on day t. It obtains the meteorological factors on day t, determines the target pollen emission amount on day t based on the pollen emission amount and the meteorological factors on day t, and simulates the pollen emission and concentration for each day of the pollen emission period. The target pollen emission amount is the pollen emission amount after adding meteorological disturbances. The response module is used to return the simulation results of the pollen emission and concentration for each day of the pollen emission period to the terminal.In this embodiment, when the communication module receives a pollen emission and concentration simulation request from the terminal, it extracts the information of the vegetation to be observed, the information of the designated observation station, and the year information to be simulated carried in the pollen emission and concentration simulation request, and sends it to the acquisition module. The acquisition module obtains the meteorological observation data of the designated observation station for each day of the pollen emission period of the vegetation to be observed in the year to be simulated. The processor uses a pollen yield prediction model pre-trained based on a machine learning model to predict the total pollen emission yield of the vegetation to be observed during the pollen emission period. The machine learning model can capture the complex historical meteorological observation data of the designated observation station. Nonlinear relationships (such as the impact of climate, geographical location, and year on pollen production) are explored to improve the accuracy of total pollen emission prediction, providing a reliable basis for determining daily pollen emissions. After predicting the total pollen emission, the processor combines a set dual-threshold pollen accumulated temperature (i.e., the accumulated temperature threshold for the start and end of pollen emission) with a constructed standardized pollen emission potential model to predict the pollen emission potential of the observed vegetation for each day of its pollen emission period. The dual-threshold pollen accumulated temperature—the start and end accumulated temperature thresholds—serves as temperature constraints, which, compared to a single accumulated temperature threshold, is more dynamically adaptable to the impact of climate change on pollen production. The impact of pollen emission potential capacity is addressed by using a standardized pollen emission potential model constructed based on a dual threshold of pollen accumulated temperature. This model accurately simulates the temporal distribution of pollen emission potential capacity, avoiding errors caused by uniform distribution, better matching the actual pollen emission curve, and improving the prediction accuracy of pollen emission potential capacity on a daily scale. Furthermore, the processor calculates the daily pollen emission amount based on the total pollen emission yield of the observed vegetation during the pollen emission period and the pollen emission potential capacity value for each day. Finally, combining the daily pollen emission amount with the corresponding meteorological factors, the target pollen emission amount for each day is determined, i.e., the pollen emission amount after incorporating meteorological disturbances. This study simulates pollen emissions and concentrations on each day of the pollen emission period. By combining daily pollen emissions with meteorological factors, the simulation considers the impact of meteorological factors on pollen emissions and concentrations (such as wind speed affecting pollen diffusion speed and range, precipitation inhibiting pollen emissions and promoting sedimentation, and humidity affecting pollen emission activity and suspension capacity). By combining daily pollen emissions with environmental diffusion conditions, the study simulates more accurate daily pollen emissions and concentrations, providing a more reliable data foundation for the analysis and research of spring pollen emissions and concentrations, as well as for the research on the impact of pollen emissions and concentrations on health risks (such as pollen allergies).

[0011] Other features and advantages of this application will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings. Attached Figure Description

[0012] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a schematic diagram of the pollen emission and concentration simulation system provided in the embodiments of this application; Figure 2 This is a schematic flowchart of the pollen emission and concentration simulation method provided in the embodiments of this application; Figure 3(a) is a scatter plot of the predicted and observed total pollen production of Cupressaceae and Salicaceae plants at a certain observation site in the first stage of spring, simulated by the pollen production prediction model corresponding to Cupressaceae and Salicaceae plants provided in the embodiments of this application. Figure 3(b) is a scatter plot of the predicted and observed total pollen production of a certain observation site in the second stage of spring, simulated by the pollen production prediction model corresponding to pine vegetation provided in the embodiments of this application. Figure 4 A schematic diagram of the process for determining the pollen emission potential value on day t, provided in an embodiment of this application; Figure 5 A flowchart illustrating the process of determining the cumulative pollen fraction emitted from the start time of accumulated temperature to day t of the pollen emission period and the flowering probability on day t, provided for embodiments of this application; Figure 6 A time-series comparison chart of station average observed values ​​and simulated values ​​of spring pollen emissions in a certain city, provided as an embodiment of this application; Figure 7(a) is a schematic diagram of the spatial distribution characteristics of pollen emissions in a certain city during the first stage of spring, provided in an embodiment of this application. Figure 7(b) is a schematic diagram of the spatial distribution characteristics of pollen emissions in a certain city during the second stage of spring, provided in an embodiment of this application. Figure 8 A schematic diagram of the architecture of the pollen simulation model provided in the embodiments of this application; Figure 9 A geopotential height distribution map of a simulated region provided in the embodiments of this application; Figure 10 The following is a time series comparison chart of the station average observed values ​​and simulated values ​​of spring pollen concentration in a certain city based on the WRF-Chem-Pollen model, provided for the embodiments of this application; Figure 11 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. DETAILED DESCRIPTION

[0013] To address the issue of low accuracy in simulating pollen emissions and concentrations, embodiments of this application provide a pollen emissions and concentration simulation system and method.

[0014] The preferred embodiments of this application are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit this application. Furthermore, the embodiments and features in the embodiments of this application can be combined with each other without conflict.

[0015] like Figure 1 As shown, this is a schematic diagram of the structure of the pollen emission and concentration simulation system provided in an embodiment of this application. The pollen emission and concentration simulation system 11 includes: Communication module 111 is used to receive pollen emission and concentration simulation requests sent by terminal 10. The pollen emission and concentration simulation requests carry information on vegetation to be observed, information on designated observation stations, and information on the year to be simulated. The data acquisition module 112 is used to acquire meteorological observation data of a specified observation station for each day during the pollen emission period of the vegetation to be observed, according to the year to be simulated. Processor 113 is used to predict the total pollen emission of the vegetation under observation at a designated observation station during the pollen emission period, based on meteorological observation data from the designated observation station and a pollen emission prediction model for each day of the pollen emission period of the vegetation under observation. The pollen emission prediction model is trained using a preset machine learning model based on historical sample meteorological observation data from the sample observation station and sample pollen observation data from the vegetation under observation. Processor 113 is also used to: acquire the cumulative temperature on day t of the pollen emission period; determine the pollen emission potential value on day t based on the cumulative temperature on day t, the accumulated temperature threshold at the beginning of the pollen period, the accumulated temperature threshold at the end of the pollen period, the number of days in the pollen emission period, and a standardized pollen emission potential model; calculate the pollen emission potential value using the standardized pollen emission potential model; determine the pollen emission amount on day t based on the total pollen emission and the pollen emission potential value on day t; acquire the meteorological factors on day t; determine the target pollen emission amount on day t based on the pollen emission amount and the meteorological factors on day t; and simulate the pollen emission and concentration for each day of the pollen emission period. The response module 114 is used to return the simulation results of pollen emission and concentration for each day of the pollen emission period to the terminal.

[0016] In one implementation, the processor 113 is specifically configured to determine the cumulative pollen fraction of the vegetation to be observed from the start time of the accumulated temperature to the day t and the flowering probability on the day t based on the accumulated temperature on the day t, the accumulated temperature threshold at the beginning of the pollen period, the accumulated temperature threshold at the end of the pollen period, and the number of days in the pollen emission period. The pollen emission potential value on day t is determined based on the cumulative pollen fraction emitted from the start of accumulated temperature to day t, the flowering probability on day t, and the standardized pollen emission potential model.

[0017] In one implementation, the vegetation to be observed is divided into at least two categories according to different pollen emission start times, with each category of vegetation corresponding to a different pollen emission period. The processor 113 is specifically used to determine the cumulative pollen fraction of the vegetation to be observed from the start time of the accumulated temperature to the day t, based on the accumulated temperature on day t of the pollen emission period corresponding to the vegetation to be observed, the accumulated temperature threshold at the beginning of the pollen period, and the accumulated temperature threshold at the end of the pollen period, for each type of vegetation to be observed. Based on the accumulated temperature on day t, the accumulated temperature threshold for the start of pollen emission, the accumulated temperature threshold for the end of pollen emission, and the number of days in the pollen emission period corresponding to the vegetation to be observed, determine the probability of flowering at the start and the probability of flowering at the end of the pollen emission period on day t of the corresponding pollen emission period for the vegetation to be observed. The flowering probability of the observed vegetation on day t of the corresponding pollen emission period is determined by the probability of flowering at the beginning and the probability of flowering at the end of the pollen emission period on day t of the corresponding pollen emission period.

[0018] In one implementation, processor 113 is specifically configured to calculate the cumulative pollen fraction emitted by the observed vegetation from the start time of accumulated temperature to day t using the following formula:

[0019] in, Indicates the time from the start of accumulated temperature for the vegetation to be observed. The cumulative pollen fraction emitted up to day t of the pollen emission period corresponding to the vegetation to be observed; This represents the cumulative temperature on day t of the pollen emission period corresponding to the vegetation being observed. , This represents the temperature on day t of the pollen emission period corresponding to the vegetation being observed. This indicates the preset daily average temperature threshold; This indicates the accumulated temperature threshold at the beginning of the pollen season; This indicates the accumulated temperature threshold at the end of the pollen season.

[0020] In one implementation, the processor 113 is specifically configured to calculate the probability of flowering of the observed vegetation at the start of the pollen emission period on day t of the corresponding pollen emission period using the following formula:

[0021] in, This represents the probability that the observed vegetation will begin flowering on day t of the corresponding pollen emission period; This represents the cumulative temperature on day t of the pollen emission period corresponding to the vegetation being observed. , This represents the temperature on day t of the pollen emission period corresponding to the vegetation being observed. This indicates the preset daily average temperature threshold. Indicates the start time of accumulated temperature; This indicates the accumulated temperature threshold at the beginning of the pollen season; This represents the tolerance factor for relative temperature accumulation, used to construct the transition range for temperature accumulation. This indicates the number of days in the pollen emission period corresponding to the vegetation being observed. ,in, This indicates the end time of pollen emission of the vegetation to be observed. Indicates the start time of pollen emission from the vegetation being observed; and The probability of flowering at the end of the pollen emission period on day t of the observed vegetation period can be calculated using the following formula:

[0022] in, This represents the probability that the pollen emission period of the observed vegetation ends on day t of the corresponding pollen emission period; Indicates the accumulated temperature threshold at the end of the pollen season; and The flowering probability of the observed vegetation on day t of the corresponding pollen emission period can be calculated using the following formula:

[0023] in, This represents the probability of the observed vegetation flowering on day t of the corresponding pollen emission period.

[0024] In one implementation, processor 113 is specifically configured to calculate the pollen emission potential of the observed vegetation on day t of the corresponding pollen emission period using the following formula:

[0025] in, This represents the potential pollen emission capacity of the observed vegetation on day t of the corresponding pollen emission period. Indicates the time from the start of accumulated temperature for the vegetation to be observed. The cumulative pollen fraction emitted up to day t of the pollen emission period corresponding to the vegetation to be observed; This represents the probability of the observed vegetation flowering on day t of the corresponding pollen emission period; This indicates the land cover of the vegetation to be observed; This represents the preset scaling factor, which characterizes the pollen emission enhancement effect parameter caused by temperature changes.

[0026] In one embodiment, the processor 113 is specifically configured to calculate the pollen emission of the observed vegetation on day t of the corresponding pollen emission period using the following formula:

[0027] in, This represents the pollen emission of the observed vegetation on day t of the corresponding pollen emission period. This represents the total pollen production of the observed vegetation during the corresponding pollen emission period.

[0028] In one implementation, meteorological factors characterize the degree of influence of meteorology on pollen emission, including wind factors, precipitation factors, and relative humidity factors. Processor 113 is specifically used to input the pollen emission amount on day t into the pollen simulation model. The pollen simulation model runs on processor 113 and is an improved version of the WRF-Chem model. The pollen simulation model includes a WRF module, a Chem module, and a pollen emission regulation module. The WRF module calculates the meteorological factors on day t and sends them to the pollen emission regulation module. The pollen emission regulation module calculates the target pollen emission amount of the observed vegetation on day t based on the pollen emission amount and the meteorological factors on day t, and sends the target pollen emission amount on day t to the Chem module. The Chem module simulates pollen concentration based on the target pollen emission amount of the observed vegetation on each day of the corresponding pollen emission period. The pollen emission regulation module calculates the target pollen emission of the observed vegetation on day t of the corresponding pollen emission period using the following formula:

[0029] in, This represents the target pollen emission of the vegetation to be observed on day t of the corresponding pollen emission period; This represents the pollen emission on day t of the pollen emission period corresponding to the vegetation to be observed. The wind factor represents the day t of the pollen emission period corresponding to the vegetation to be observed. The precipitation factor represents the day t of the pollen emission period corresponding to the vegetation to be observed. The relative humidity factor represents the day t of the pollen emission period corresponding to the vegetation to be observed.

[0030] In one implementation, different types of vegetation to be observed correspond to different accumulated temperature thresholds for the pollen initiation and pollen termination periods; and

[0031] The processor 113 is further configured to optimize the accumulated temperature thresholds for the pollen initiation period, the accumulated temperature threshold for the pollen initiation period, the accumulated temperature threshold for the pollen end period, and the preset daily average temperature threshold for the vegetation to be observed before determining the accumulated pollen fraction emitted from the accumulated temperature start time to the pollen end period and the flowering probability on the pollen day based on the accumulated temperature on day t, the accumulated temperature threshold for the pollen initiation period, the accumulated temperature start time and the preset daily average temperature threshold for the vegetation to be observed.

[0032] The pollen emission and concentration simulation system in this application embodiment can be set on a server, or on any other electronic device with computing capabilities, such as a terminal device. This application embodiment does not limit this.

[0033] like Figure 2 As shown, this is a schematic diagram of the implementation process of the pollen emission and concentration simulation method provided in this application embodiment. The pollen emission and concentration simulation method can be applied to, for example... Figure 1 The pollen emission and concentration simulation system shown may specifically include the following steps: S21. Receive the pollen emission and concentration simulation request sent by the terminal. The pollen emission and concentration simulation request carries the vegetation information to be observed, the information of the designated observation station, and the year information to be simulated.

[0034] In practice, the communication module in the pollen emission and concentration simulation system receives the pollen emission and concentration simulation request sent by the terminal, extracts the vegetation information to be observed, the designated observation station information, and the year information to be simulated from the pollen emission and concentration simulation request, and sends the vegetation information to be observed, the designated observation station information, and the year information to be simulated to the acquisition module in the pollen emission and concentration simulation system.

[0035] S22. Obtain meteorological observation data from designated observation stations for each day of the pollen emission period of the vegetation to be observed, based on the year requested for simulation.

[0036] In practice, designated observation stations can be set according to actual needs. This application uses six designated observation stations in a certain city (located in six districts of the city: District 1, District 2, District 3, District 4, District 5, and District 6) as an example. Pollen emission in this city typically begins in March. The dominant allergenic pollen species in spring mainly include Cupressaceae, Salicaceae, and Pinaceae. The pollen emitted by these three families accounts for a significant portion of the total spring pollen concentration. Therefore, this application uses vegetation from these three families as the vegetation to be observed. The vegetation to be observed is divided into at least two categories according to different pollen emission start times (SOS), with each category corresponding to a different pollen emission period. For the pollen emission end time (EOS), when the accumulated pollen concentration reaches a certain percentage of the total pollen emission concentration in spring (i.e., the observation season)... or At that time, pollen emission can be used as the end time ( Based on historical observations (2012-2019) of pollen emission start and end times (EOS) for Cupressaceae, Salicaceae, and Pinaceae at designated observation stations, the pollen emission start times for Cupressaceae and Salicaceae are relatively consistent, and their pollen emission end times also show strong consistency. The pollen emission start and end times for Pinaceae lag significantly behind those of Cupressaceae and Salicaceae. Therefore, to more accurately simulate the temporal variation trend of spring pollen emission concentration in a certain city, this application's embodiments divide the spring pollen emission period into... The pollen release period is divided into two phases: the first phase is the pollen release period for Cupressaceae and Salicaceae, and the second phase is the pollen release period for Pinaceae. That is, Cupressaceae and Salicaceae in the observed vegetation have the same start and end times for pollen release. The start time of pollen release for Pinaceae is later than that for Cupressaceae and Salicaceae, and similarly, the end time of pollen release for Pinaceae is also later than that for Cupressaceae and Salicaceae. The number of days between the end date and the start date of pollen release in the observed vegetation is the pollen release period of the observed vegetation. Although there is some overlap between the two phases, this division method better reflects the temporal characteristics of actual pollen release. According to observations, the peak pollen emission period for Cupressaceae and Salicaceae plants in a certain city is generally from mid-March to early April, while the peak pollen emission period for Pinaceae plants is generally from late April to early May. In practice, the pollen emission start time for Cupressaceae and Salicaceae in a certain city can be set as early March and the pollen emission end time as mid-to-late April, resulting in a pollen emission period for Cupressaceae and Salicaceae from early March to mid-to-late April. Similarly, the pollen emission start time for Pinaceae can be set as late March to early April and the pollen emission end time as mid-to-late May, resulting in a pollen emission period for Pinaceae from late March to early April to mid-to-late May. There is some overlap in the pollen emission periods between the two phases.

[0037] In this step, after the acquisition module in the pollen emission and concentration simulation system receives the information of the vegetation to be observed, the information of the designated observation stations, and the information of the year to be simulated sent by the communication module, it can obtain the meteorological observation data of each designated observation station for each day of the pollen emission period of each type of vegetation to be observed (Cypressaceae and Salicaceae are one type, and Pinaceae are another type) from the daily value dataset of surface climate data provided by the National Meteorological Science Data Center, and send it to the processor in the pollen emission and concentration simulation system. The daily meteorological observation data may include, but is not limited to, at least one or more combinations of the following meteorological data: (daily) average temperature (TEM_Avg), maximum temperature (TEM_Max), minimum temperature (TEM_Min), sunshine duration (SSH), station altitude (Alti) (i.e., the altitude of the meteorological station), average air pressure (PRS_Avg), maximum air pressure (PRS_Max), minimum air pressure (PRS_Min), maximum wind speed (WIN_S_Max), maximum wind speed (WIN_S_Inst_Max), average 2-minute wind speed (WIN_S_2mi_Avg), and ground temperature (GST_Avg_Xcm, X=5, 10,15, 20, 40, 80, 160). The data include, but are not limited to, the average surface temperature (GST_Avg), the minimum surface temperature (GST_Min), the maximum surface temperature (GST_Max), the average relative humidity (RHU_Avg), the minimum relative humidity (RHU_Min), the average water vapor pressure (VAP_Avg), the precipitation from 20:00 to 20:00 (PRE_Time_2020), and the precipitation from 08:00 to 08:00 (PRE_Time_0808), etc. The embodiments of this application do not limit these.

[0038] S23. Based on the meteorological observation data of the designated observation station and the pollen production prediction model for each day during the pollen emission period of the vegetation to be observed, predict the total pollen emission of the vegetation to be observed at the designated observation station during the pollen emission period.

[0039] The pollen yield prediction model is trained by the processor using historical meteorological observation data from sample observation stations and pollen observation data from the vegetation to be observed, according to a preset machine learning model. This preset machine learning model can be, but is not limited to, the XGBoost model; this embodiment does not limit this. Since different types of vegetation have different pollen emission periods, separate pollen yield prediction models are trained for each type of vegetation during the training process.

[0040] For each type of vegetation, historical meteorological observation data from all sample observation stations were collected every day during the pollen emission period of that type of vegetation in historical years (e.g., 2006-2020) using the data acquisition module. The total pollen emission of that type of vegetation at each sample observation station during the pollen emission period was also collected. The sample observation stations could be any of the six designated observation stations in a specific city, randomly selected from those stations. The historical meteorological observation data of the selected sample observation stations and the total pollen emission yield of each selected sample observation station during the pollen emission period were used as the training set. The remaining data... The historical meteorological observation data of the sample observation stations and the total pollen emission of each sample observation station during the pollen emission period were used as the test set to verify the simulation accuracy of the pollen production prediction model for the total pollen emission during the pollen observation period.

[0041] When training the pollen yield prediction model, the processor uses historical meteorological observation data from each sample observation station for each day of the pollen emission period of this type of vegetation as input to the XGBoost model. It performs iterative machine training based on the error between the predicted total pollen emission of this type of vegetation at that sample observation station during the pollen emission period and the sample total pollen emission of this type of vegetation at that sample observation station during the pollen emission period (i.e., the actual total pollen emission). The parameters of the XGBoot model are adjusted until the model converges, resulting in the trained pollen yield prediction model for that vegetation type. In implementation, to improve the prediction performance of the XGBoot model, the GridSearchCV (enumerated grid search) method can be used for hyperparameter optimization during the training phase, combined with five-fold cross-validation to prevent overfitting and enhance the model's robustness and generalization ability. It should be noted that the total pollen emission of the sampled plants during the first spring stage, i.e., the pollen emission period of Cupressaceae and Salicaceae plants, fluctuated significantly around 2013. This change may be related to adjustments in vegetation planting structure or changes in climate conditions. Using all-time data for model training may introduce large errors. Therefore, in this embodiment of the application, for modeling the total pollen emission of Cupressaceae and Salicaceae plants during the first spring stage, the observed data can be divided into two subsets according to time: before 2013 and after 2013. The XGBoost model can be trained independently for each subset to improve simulation accuracy. The total pollen emission of Pinaceae plants during the second spring stage, i.e., the pollen emission period, was relatively stable between 2006 and 2020, with a gentle trend. Therefore, the model training for this stage can use all-time data for unified modeling, thereby improving the model's sensitivity and generalization ability to the spatiotemporal variation characteristics of total pollen emission in different stages, and providing more robust basic data support for subsequent regional-scale total pollen emission estimation and numerical simulation. Figure 3(a) shows the predicted total pollen production of Cupressaceae and Salicaceae plants at a certain observation site during the first stage of spring (i.e., the pollen emission period of Cupressaceae and Salicaceae plants), simulated by the trained pollen production prediction model for Cupressaceae and Salicaceae plants. Simulated values ​​and observed values ​​(i.e.: The scatter plot between the observed and predicted pollen production values ​​of a certain observation site during the second spring stage (i.e., the pollen emission period of pine and willow trees) is shown in Figure 3(b). This is a scatter plot of the predicted and observed total pollen production values ​​of pine trees at a certain observation site, simulated by a trained pollen production prediction model for pine vegetation. As can be seen from Figures 3(a) and 3(b), the predicted and observed values ​​exhibit a strong linear correlation in both stages, with the correlation coefficient R between the predicted and observed values ​​being greater than or equal to [a certain value]. And all of them passed. The significance test results show that the simulation method of total pollen emission based on the XGBoost algorithm has high accuracy and reliability in different vegetation types and time stages, providing a solid data foundation for subsequent high-resolution pollen spatial simulation.

[0042] In this step, the processor inputs the meteorological observation data of the designated observation station for each day of the pollen emission period of the vegetation to be observed into the pollen yield prediction model corresponding to the vegetation to be observed, and obtains the predicted total pollen emission of the vegetation to be observed at the designated observation station during its pollen emission period.

[0043] S24. Obtain the cumulative temperature on day t of the pollen emission period. Based on the cumulative temperature on day t, the accumulated temperature threshold at the beginning of the pollen period, the accumulated temperature threshold at the end of the pollen period, the number of days in the pollen emission period, and the constructed standardized pollen emission potential model, determine the pollen emission potential capacity value on day t.

[0044] In specific implementation, it can be done according to the following: Figure 4 The procedure shown determines the potential pollen emission capacity on day t, including the following steps: S31. Based on the accumulated temperature on day t, the accumulated temperature threshold at the beginning of pollen season, the accumulated temperature threshold at the end of pollen season, and the number of days in the pollen emission period, determine the cumulative pollen fraction of the vegetation to be observed from the start of the accumulated temperature to day t and the flowering probability on day t.

[0045] Since the pollen emission start and end times differ for different types of vegetation under observation, and spring temperatures gradually increase over time, this application, in order to improve the accuracy of predicting the pollen emission potential of different types of vegetation under observation, pre-sets different accumulated temperature thresholds for the pollen initiation period for different types of vegetation under observation. SumT s and pollen end-of-period accumulated temperature threshold SumT e The start time of accumulated temperature for different types of vegetation to be observed Similarly, the calculation of the cumulative temperature for each day during the pollen emission period is based on the same preset cumulative temperature start time. It can be set to January 1st, with preset daily average temperature thresholds corresponding to different types of vegetation to be observed. The settings can be based on experience values, and this application does not impose any restrictions on them.

[0046] When implementing, it can be done according to the following: Figure 5 The procedure shown determines the cumulative pollen fraction emitted from the start of accumulated temperature to day t of the pollen emission period and the flowering probability on day t, including the following steps: S41. For each type of vegetation to be observed, the cumulative pollen fraction of the vegetation to be observed from the start time of the accumulated temperature to the day t is determined based on the accumulated temperature on day t of the pollen emission period corresponding to the vegetation to be observed, the accumulated temperature threshold at the beginning of the pollen period, and the accumulated temperature threshold at the end of the pollen period.

[0047] In practice, the processor can calculate the cumulative pollen fraction emitted by the vegetation under observation from the start time of accumulated temperature to day t using the following formula:

[0048] in, Indicates the time from the start of accumulated temperature for the vegetation to be observed. The cumulative pollen score up to day t of the pollen emission period corresponding to the vegetation to be observed; where the cumulative pollen score represents the cumulative score of pollen emission capacity, and the cumulative temperature starting time of the vegetation to be observed. The cumulative pollen fraction emitted up to day t of the pollen emission period corresponding to the vegetation to be observed, that is: the cumulative temperature fraction of the vegetation to be observed from the start time of the accumulated temperature. The cumulative score of pollen emission capacity up to day t of the pollen emission period corresponding to the vegetation to be observed. The higher the cumulative score, the stronger the pollen emission capacity. This represents the cumulative temperature on day t of the pollen emission period corresponding to the vegetation being observed. , This represents the temperature on day t of the pollen emission period corresponding to the vegetation to be observed (i.e., the daily average temperature on day t). Indicates the daily average temperature threshold; This indicates the accumulated temperature threshold for the pollen initiation period, which is the cumulative temperature threshold required for the observed vegetation to begin releasing pollen. This represents the accumulated temperature threshold at the end of the pollen season, which is the cumulative temperature threshold required for the observed vegetation to stop releasing pollen.

[0049] In this way, the time from the start of accumulated temperature for the vegetation under observation can be calculated. The cumulative pollen fraction emitted each day during the pollen emission period corresponding to the vegetation to be observed.

[0050] Different types of vegetation to be observed correspond to different accumulated temperature thresholds for the pollen initiation and pollen termination periods, while different types of vegetation to be observed correspond to the same accumulated temperature start time and daily average temperature threshold. In one embodiment, the accumulated temperature threshold for the pollen initiation period for the vegetation to be observed is... Pollen end-of-season accumulated temperature threshold Accumulated temperature start time and daily average temperature threshold It can be preset based on experience values.

[0051] In one implementation, to improve the accuracy of assessing pollen emission potential, the time from the start of accumulated temperature for the vegetation under observation is determined. Before the cumulative pollen fraction emitted on day t of the pollen emission period corresponding to the vegetation to be observed and the flowering probability on day t, the simulated annealing algorithm can be used to optimize the accumulated temperature thresholds for the pollen initiation period, the pollen end period, the accumulated temperature start time, and the preset daily average temperature thresholds for the vegetation to be observed, so as to obtain the optimized accumulated temperature thresholds for the pollen initiation period, the pollen end period, the accumulated temperature start time, and the daily average temperature thresholds for the vegetation to be observed.

[0052] Simulated annealing is a probabilistic heuristic global optimization algorithm inspired by the metal annealing process. It avoids getting trapped in local optima by accepting a certain probability of "dominant solutions" and can be used for complex nonlinear optimization problems. The specific process of optimizing the accumulated temperature thresholds for the pollen initiation and end stages, as well as the accumulated temperature start time and daily average temperature thresholds, based on the simulated annealing algorithm in this application is as follows: First, initialize the parameter space and set the value range of each parameter based on empirical values. For example, the accumulated temperature start time can be set. The value range is 1 to 90 days (e.g., starting from January), setting the daily average temperature threshold. The value range is 0~15℃, and the accumulated temperature threshold for the pollen initiation period is set. The value range is 50~300℃, the accumulated temperature threshold for the end of pollen season. The value range is 500~1500℃, but this application does not limit it in the embodiments.

[0053] Further define the objective function, which measures the difference between simulated and observed values, and it can be set as follows:

[0054] in, Let be the objective function. This represents the simulated phenological days in year i. Let N represent the phenological day observed in the i-th year, where the phenological day can be the pollen emission day of the vegetation to be observed, and N is the number of years.

[0055] Furthermore, initialize the annealing parameters as shown in Table 1: Table 1

[0056] Then, the annealing process is simulated, and the specific implementation process is as follows: Step 1: Generate initial solution: Randomly generate a set of initial parameters Calculate its objective function value .

[0057] Step 2: Circulating cooling: Current temperature T = T0 (initial temperature), when At the minimum temperature, the current solution X is perturbed to generate a new solution X_new. Each parameter is randomly perturbed with a set step size (e.g., ...). ), calculate the objective function of the new solution ,if If so, accept the new solution X_new; otherwise, use probability. Accept the inferior solution, update the current optimal solution, and cool it down to: .

[0058] Step 3: Output the optimal solution: Final output: , The optimized daily average temperature threshold, The optimized accumulated temperature start time, The optimized accumulated temperature threshold for the pollen initiation period, This is the optimized accumulated temperature threshold for the end of pollen season.

[0059] Table 2 shows examples of optimized parameters for Cupressaceae, Salicaceae, and Pinaceae vegetation. Table 2

[0060] Among them, the start time of accumulated temperature The unit is days, and can be an integer, for example, In this case, the start time for accumulated temperature can be set to January 23.

[0061] S42. Based on the accumulated temperature on day t, the accumulated temperature threshold for the start of pollen emission, the accumulated temperature threshold for the end of pollen emission, and the number of days in the pollen emission period corresponding to the vegetation to be observed, determine the probability of flowering at the start and the probability of flowering at the end of the pollen emission period on day t of the corresponding pollen emission period for the vegetation to be observed.

[0062] Probability of flowering at the start of pollen release period The probability of flowering at the end of the pollen emission period It is also piecewise linear, affected by the accumulated temperature threshold at the beginning of pollen season. and pollen end-of-period accumulated temperature threshold And the effect of accumulated temperature. Normally, flowering doesn't begin suddenly, but rather gradually begins around the time the accumulated temperature reaches a threshold, until full bloom. In this application, the threshold is set to [a certain value]. It begins to bloom when the accumulated temperature threshold is reached. When the flower is in full bloom, the probability of flowering increases linearly from the start of flowering to the end of flowering.

[0063] In this embodiment of the application, the flowering probability at the start of the pollen emission period is defined. The probability of flowering at the end of the pollen emission period Respectively compared with the accumulated temperature threshold at the start of pollen season Pollen end-of-season accumulated temperature threshold It is related to, rather than to, the total pollen production of the observed vegetation during its pollen emission period. This indicates that the pollen emission period does not end with the cessation of pollen emission, but rather depends on the accumulated temperature threshold, thus avoiding the decline due to the total pollen emission yield. Simulation errors can cause pollen release periods to end abnormally early or late.

[0064] In practice, the processor can calculate the probability of flowering of the observed vegetation on day t of the corresponding pollen emission period using the following formula:

[0065] in, This represents the probability that the observed vegetation will begin flowering on day t of the corresponding pollen emission period; This represents the cumulative temperature on day t of the pollen emission period corresponding to the vegetation being observed. , This represents the temperature on day t of the pollen emission period corresponding to the vegetation to be observed (i.e., the daily average temperature on day t). This indicates the preset daily average temperature threshold. Indicates the start time of accumulated temperature; This indicates the accumulated temperature threshold at the beginning of the pollen season; This represents the tolerance factor for relative temperature accumulation, used to construct the transition range for temperature accumulation. This indicates the number of days in the pollen emission period corresponding to the vegetation being observed. ,in, This indicates the end time of pollen emission of the vegetation to be observed. This indicates the start time of pollen emission from the vegetation to be observed.

[0066] In this way, the probability of flowering at the start of pollen emission on each day of the corresponding pollen emission period of the vegetation under observation can be calculated.

[0067] The processor can calculate the flowering probability of the observed vegetation at the end of the pollen emission period on day t of the corresponding pollen emission period using the following formula:

[0068] in, This represents the probability that the pollen emission period of the observed vegetation ends on day t of the corresponding pollen emission period; This represents the accumulated temperature threshold for the pollen end period.

[0069] In this way, the flowering probability of the observed vegetation at the end of the pollen emission period on day t of the corresponding pollen emission period can be calculated.

[0070] In the embodiments of this application, Based on experience, it can be set as, but is not limited to, as follows: Controlling the gradual increase and decrease of pollen shedding, specifically when the accumulated temperature reaches the threshold for the initial pollen stage. The loose powder won't be stopped immediately afterward, but will be gradually reduced until it's finished.

[0071] S43. Based on the flowering probability at the beginning and the flowering probability at the end of the pollen emission period on day t of the corresponding pollen emission period, determine the flowering probability of the observed vegetation on day t of the corresponding pollen emission period.

[0072] In practice, the processor can calculate the flowering probability of the observed vegetation on day t of the corresponding pollen emission period using the following formula:

[0073] in, This represents the probability of the observed vegetation flowering on day t of the corresponding pollen emission period; This represents the probability that the observed vegetation will begin flowering on day t of the corresponding pollen emission period; This represents the probability that the pollen emission period of the observed vegetation ends on day t of the corresponding pollen emission period.

[0074] In this way, the flowering probability of the observed vegetation on each day during the corresponding pollen emission period can be calculated.

[0075] S32. Based on the cumulative pollen fraction emitted from the start time of accumulated temperature to day t, the flowering probability on day t, and the standardized pollen emission potential model, determine the pollen emission potential value on day t.

[0076] Among them, pollen emission potential characterizes the potential pollen emission capacity of the vegetation to be observed, and the standardized pollen emission potential model is used to calculate the pollen emission potential value.

[0077] In practice, the processor can calculate the pollen emission potential of the observed vegetation on day t of the corresponding pollen emission period using the following standardized pollen emission potential model:

[0078] in, This represents the potential pollen emission capacity of the observed vegetation on day t of the corresponding pollen emission period. Indicates the time from the start of accumulated temperature for the vegetation to be observed. The cumulative pollen fraction emitted up to day t of the pollen emission period corresponding to the vegetation to be observed; Accumulated pollen score The derivative with respect to time t; ;in, It is the Heaviside function. when hour, ,when hour, =1; This represents the probability of the observed vegetation flowering on day t of the corresponding pollen emission period; This indicates the land cover of the vegetation to be observed; This represents the preset scaling factor, which characterizes the pollen emission enhancement effect parameter caused by temperature changes.

[0079] In this embodiment, the land cover of the vegetation to be observed refers to the proportion of that type of vegetation per unit area. For the land use dataset, this application uses Community Land Model 4 (CLM4). This dataset contains 25 plant functional types, including coniferous forests, broad-leaved forests, shrublands, herbs, and crops, with a spatial resolution of [missing information]. Since this application primarily simulates total pollen concentration in spring (tree pollen), the plant functionality of spring tree pollen shedding is calculated using the sum of land cover of all coniferous and broadleaf forests in the CLM4 data.

[0080] This application takes into account the explosive increase in pollen emission concentration in a certain city during the early spring as temperature rises. To more accurately depict the rapid changes in pollen levels during the early spring, a scaling factor is introduced. This is used to characterize the enhanced pollen emission effect caused by temperature changes. The scaling factor varies with time t, and can be specifically set as follows: after a specified date each year (e.g., March 1st), if the temperature shows the first periodic change process of "rising first and then falling", the pollen emission potential is amplified during this period. times, It can be set to 2; at other times, the default is 2. .

[0081] In this way, the potential pollen emission capacity of the vegetation under observation can be calculated for each day during the corresponding pollen emission period.

[0082] S25. Determine the pollen emission amount on day t based on the total pollen emission production and the pollen emission potential capacity value on day t.

[0083] In practice, the processor can calculate the pollen emission of the observed vegetation on day t of the corresponding pollen emission period using the following formula:

[0084] in, This represents the pollen emission of the observed vegetation on day t of the corresponding pollen emission period. This represents the total pollen production of the observed vegetation during the corresponding pollen emission period.

[0085] In this way, the pollen emission of the vegetation under observation can be calculated for each day during the corresponding pollen emission period.

[0086] S26. Obtain the meteorological factors for day t, determine the target pollen emission for day t based on the pollen emission amount and the meteorological factors for day t, and simulate the pollen emission and concentration for each day of the pollen emission period.

[0087] Meteorological factors characterize the degree of influence of meteorological conditions on pollen emissions. These factors can include at least: wind factors, precipitation factors, and relative humidity factors. The wind factor represents the degree of influence of wind on pollen emissions, the precipitation factor represents the degree of influence of precipitation on pollen emissions, and the relative humidity factor represents the degree of influence of relative humidity on pollen emissions. The target pollen emission amount is the pollen emission amount after incorporating meteorological disturbances.

[0088] In practice, the processor can calculate the wind factor on day t of the pollen emission period corresponding to the vegetation to be observed using the following formula. :

[0089] in, The wind factor represents the day t of the pollen emission period corresponding to the vegetation to be observed. The wind speed at 10m near the ground on day t of the pollen emission period corresponding to the vegetation to be observed; The vertical turbulent wind speed represents the pollen emission period corresponding to the vegetation to be observed on day t.

[0090] wind factor wind speed at near ground level 10m Vertical turbulent wind speed It is related to the index.

[0091] The processor can calculate the precipitation factor on day t of the pollen emission period corresponding to the observed vegetation using the following formula. :

[0092] in, The precipitation factor represents the day t of the pollen emission period corresponding to the vegetation to be observed. This represents the precipitation on day t during the pollen emission period of the vegetation to be observed. Indicates a low precipitation threshold. This indicates the high precipitation threshold.

[0093] When precipitation is below the low precipitation threshold When the precipitation factor is set to 1, and the precipitation amount is higher than the high precipitation threshold, the precipitation factor is set to 1. When the precipitation factor is 0, and the precipitation amount is between the low precipitation threshold, the precipitation factor is set to 0. and high precipitation threshold In the interval, the precipitation factor is the quotient of the difference between the high precipitation threshold and the precipitation amount, and the threshold difference. High precipitation threshold With low precipitation threshold It can be set based on experience values, such as... , However, the embodiments in this application do not limit this.

[0094] The processor can calculate the relative humidity factor on day t of the pollen emission period corresponding to the vegetation under observation using the following formula. :

[0095] in, The relative humidity factor represents the day t of the pollen emission period corresponding to the vegetation to be observed. The relative humidity represents the day t of the pollen emission period corresponding to the vegetation to be observed. This indicates the high relative humidity threshold. This indicates the low relative humidity threshold.

[0096] During implementation, the high relative humidity threshold With low relative humidity threshold It can be set based on experience values, such as... , However, the embodiments in this application do not limit this.

[0097] Furthermore, after determining the values ​​of each meteorological factor, the processor can calculate the target pollen emission of the observed vegetation on day t of the corresponding pollen emission period using the following formula:

[0098] in, This represents the target pollen emission of the vegetation to be observed on day t of the corresponding pollen emission period; This represents the pollen emission on day t of the pollen emission period corresponding to the vegetation to be observed. The wind factor represents the day t of the pollen emission period corresponding to the vegetation to be observed. The precipitation factor represents the day t of the pollen emission period corresponding to the vegetation to be observed. The relative humidity factor represents the day t of the pollen emission period corresponding to the vegetation to be observed.

[0099] In this way, the target pollen emission of the vegetation to be observed at a specified observation site can be calculated for each day during its pollen emission period, thereby simulating the pollen emission.

[0100] Assuming the terminal requests simulations of six designated observation sites in a city, covering the years from 2006 to 2020, the above process can calculate the target pollen emission amount for each type of vegetation at each designated observation site during its corresponding pollen emission period, as well as the time series of pollen emission amounts for each type of vegetation at each designated observation site during its corresponding pollen emission period. Furthermore, the average total pollen emission amount for each type of vegetation at each designated observation site can be calculated for each day during the two pollen emission periods (including overlapping dates). The average total pollen emission amount for a given day is calculated as follows: first, the sum of the pollen emissions for each type of vegetation at each designated observation site on that day is calculated as the total pollen emission amount for that site on that day; then, the average of the total pollen emission amounts for all designated observation sites on that day is taken. Figure 6 As shown, this is a time series comparison of the station average observed values ​​and simulated values ​​of pollen emissions in a certain city during spring, including data from 2006 to 2020. Figure 6 The blue dots represent the daily measured average total pollen emissions (average total pollen emissions at each designated observation site), while the red solid line represents the corresponding daily simulated average total pollen emissions. To assess the consistency between simulated and observed values, we used the correlation coefficient R and root mean square error (RMSE) as evaluation metrics. Figure 6 As can be seen, the simulation results are highly consistent with the observed values ​​in most years. In the years when the correlation coefficient between the two exceeded Especially in 2008, 2009, 2016, 2017, and 2018, the R-value exceeded [a certain value]. This indicates that the pollen emission potential model possesses strong pollen emission calculation capabilities. Meanwhile, spring pollen emissions typically exhibit two peaks: the first peak is mainly caused by early spring tree species such as cypress and willow, while the second peak corresponds to the late spring stage dominated by pines. Simulation results show that the emission characteristics of these two stages are well reproduced in most years, especially in 2008, 2013, 2017, 2018, and 2019, where the simulated values ​​highly match the observed values, demonstrating the model's excellent performance in these years. Notably, the simulated early spring pollen start stage in 2020 highly matches the observed values, reflecting the explosive growth of early spring pollen, while the simulation of late spring pollen emissions also shows good results. Therefore, the pollen emission potential model constructed in this application can accurately simulate the start and end times of spring pollen in a certain city, as well as the time series changes throughout the pollen season, demonstrating good simulation capabilities and application potential.

[0101] Figures 7(a) and 7(b) show the spatial distribution characteristics of pollen emissions in a certain city during the first and second spring phases (Cypressaceae and Salicaceae) from 2006 to 2020. Overall, cypress and poplar pollen emissions exhibit significant spatiotemporal differences and marked interannual fluctuations, while pine pollen emissions show a relatively stable interannual trend. High pollen-producing areas are mainly concentrated in the northern and western parts of the city, reflecting high planting density of related tree species or good vegetation growth in these areas. In terms of interannual variation, the overall concentration of pollen emissions during the first spring phase was relatively low from 2006 to 2012, but increased significantly starting in 2013, indicating that the vegetation structure may have changed during this period, or that human planting activities (such as urban greening) have affected the population size of cypress and poplar. After 2017, pollen emissions eased somewhat but remained at a high level. For pine pollen, the spatial distribution of pollen emissions remained relatively stable throughout the 15 years. Cypress and poplar pollen constitute a significant portion of the total spring pollen count in a certain city, and their emission characteristics are influenced by multiple factors, including vegetation distribution, meteorological conditions, and human intervention. In-depth research on the emission of these allergenic pollens is crucial for accurately identifying high-risk exposure areas, developing individualized protection strategies, and selecting appropriate urban greening plants.

[0102] To further simulate pollen emission concentration, the following methods can be used: Figure 8 The pollen simulation model shown simulates pollen emission concentration.

[0103] In specific implementation, a pollen simulation model can be set on the processor. The pollen simulation model is an improvement on the WRF-Chem (Weather Research and Forecasting model with Chemistry) model. The pollen simulation model includes WRF module 501, Chem module 502 and pollen emission regulation module 503. The processor inputs the pollen emission amount on day t into the pollen emission regulation module 503 of the pollen simulation model. The WRF module 501 calculates the meteorological factors on day t based on the acquired Final Reanalysis Data (FNL) and sends these meteorological factors to the pollen emission regulation module 503. The pollen emission regulation module then calculates the target pollen emission amount for the observed vegetation on day t within the corresponding pollen emission period, based on the pollen emission amount and meteorological factors on day t. This target pollen emission amount is then sent to the Chem module 502. The Chem module 502 simulates pollen concentration based on the target pollen emission amount for each day within the corresponding pollen emission period. The calculation methods for meteorological factors by the WRF module 501 and for the target pollen emission amount by the pollen emission regulation module 503 are the same as those used by the processor in calculating meteorological factors and target pollen emissions, and will not be elaborated here.

[0104] like Figure 8 As shown, this is a schematic diagram of the pollen simulation model provided in this application embodiment. This pollen simulation model can be referred to as the WRF-Chem-Pollen model, an improvement upon the WRF-Chem model. The WRF-Chem model is a new generation atmospheric simulation and forecasting model developed by the National Oceanic and Atmospheric Administration (NOAA). It achieves true spatiotemporal resolution coupling based on meteorological and chemical transport models, realizing online biochemical transport. This model not only has good simulation and forecasting effects on pollutant gas and aerosol emissions and chemical transport processes, but can also be used to analyze air quality, cloud-chemical interactions, etc. Furthermore, because this model has comprehensive physical parameterization processes, a high degree of coupling with chemical models, convenient operation, and high computational efficiency, it is widely used in simulating meteorological processes and atmospheric pollution processes. This application was filed in WRF-Chem ( Based on the existing model, pollen is considered as a coarse-mode particle and incorporated into the WRF-Chem model to construct the WRF-Chem-Polen model, which simulates the physical processes of pollen transport, diffusion, and deposition in the atmosphere. In addition to WRF module 501 and Chem module 502, the WRF-Chem-Polen model introduces a pollen emission regulation module 503. WRF module 501 provides meteorological simulation functions to calculate daily meteorological factors based on global reanalysis data. Pollen emission regulation module 503 is used to determine the daily target pollen emission amount based on the daily pollen emission amount of the observed vegetation during its pollen emission period and meteorological factors, reflecting the influence of meteorology on pollen emission. Chem module 502 is used to simulate the pollen emission process in the atmosphere and output the simulated pollen concentration. The proposed Secondary Organic Aerosol Model (SORGAM) mechanism was incorporated into... The proposed ModalAerosol Dynamics Model for Europe (MADE) mechanism comprises the MADE / SORGAM aerosol mechanism. The MADE / SORGAM mechanism employs a modal approach to describe three log-normally distributed aerosol particle size modes, including the Aitken Mode. Accumulation Mode ) and Coarse Mode (coarse particle mode) In addition to the three modalities mentioned above, this application further considers pollen as a separate modality, and considers the pollen density to be approximately... The diameter is approximately Other parameters, such as Standard Deviation and Hygroscopicity, can be set according to the parameters of Coarse Mode. In this embodiment, the daily pollen emission of the vegetation under observation during its pollen emission period can be calculated offline. The daily pollen emission of the vegetation under observation during its pollen emission period and the global reanalysis data are used as pollen emission sources and input into the pollen emission regulation module 503 and WRF module 501 of the constructed WRF-Chem-Pollen model, respectively. The WRF module 501 determines the meteorological data of the designated observation station each day based on the global reanalysis data, calculates the meteorological factors for each day based on the meteorological data of the designated observation station, and transmits the calculated meteorological factors for each day to the pollen emission regulation module 503. The pollen emission regulation module 503 calculates the target pollen emission for each day based on the daily pollen emission of the vegetation under observation during its pollen emission period and the daily meteorological factors, and then transmits the target pollen emission for each day to the Chem module 502 to simulate the transport and evolution of pollen in the atmosphere. The pollen emission regulation module 503 in the WRF-Chem-Pollen model constructed in this application can also be set as a sub-module of the Chem module 502, and this application embodiment does not limit this.

[0105] In this embodiment, the WRF-Chem-Pollen model is used to simulate the influence of daily meteorological factors such as temperature, precipitation, humidity, and wind speed on pollen concentration at six observation stations in a city. It also describes physical processes such as advection diffusion, convective transport, and wet and dry deposition, providing an analysis of the atmospheric dissemination and deposition behavior of pollen in spring. Figure 9 The figure shown is a geopotential height distribution map of a simulated area provided in this embodiment of the application. The triangles mark six observation stations, and the circles mark meteorological monitoring stations. In the simulation, the Lambert projection method is used, and the horizontal grid is set to... The resolution is 3000m, and the integration time step can be set to 15s. When calculating meteorological factors, Global Reanalysis (FNL) data is used as the initial and boundary fields input to the model's WRF module, with a spatial accuracy of [missing information]. With a time resolution of 6 hours, the WRF module calculates the daily meteorological factors for each observation station during the pollen emission period of the observed vegetation based on global reanalysis data. Each observation station is located within a grid-structured region. For each station, the meteorological factors are calculated using the average meteorological data from the grids covering that station. For example, to calculate the wind factor on day t of the pollen emission period for a given observation station, the average near-surface 10m wind speed and the average vertical turbulent wind speed are calculated across all grids covered by the station on day t. The average near-surface 10m wind speed is used as the near-surface 10m wind speed for that station on day t, and the average vertical turbulent wind speed is used as the vertical turbulent wind speed for that station on day t. The WRF module then calculates the wind factors based on these parameters. The calculation formula is used to obtain the wind factor of the observation station on day t. The calculation methods for precipitation factor and relative humidity factor are similar and will not be elaborated here. The Noah East Asian regional land surface process scheme, Lin cloud microphysics scheme, RRTM longwave radiation scheme and Goddard shortwave radiation scheme are adopted. The chemical field adopts the output data every 6 hours of the MOZART (Model for Ozone and Related chemical Tracers) global atmospheric chemistry model input. Anthropogenic emissions adopt the Tsinghua University MEIC (Multi-resolution Emission Inventory for China) inventory. Meteorological emissions adopt the RADM2 (Regional Acid Deposition Model Version 2) chemical mechanism. Aerosols adopt the MADE / SORGAM mechanism. Biological emissions adopt the online MEGAN (Model of Emissions of Gases and Aerosols from Nature) inventory. The simulation period is from March 1 to June 30 of each year from 2006 to 2020. The specific configuration of the WRF-Chem-Pollen model is shown in Table 3. Table 3

[0106] To prevent data anomalies, outliers in pollen emission concentration are removed, and pollen emission concentrations are adjusted according to a preset ratio (e.g., can be set to...). Remove the highest pollen concentration, that is, remove the pollen emission concentration. Pollen concentration data, retain the remaining This application uses pollen concentration data for simulation. Simultaneously, a 5-day moving average is used to smooth the pollen concentration data, which not only eliminates the influence of noise on the pollen concentration data but also reduces the impact of daily meteorological changes and advection diffusion on daily pollen emissions, allowing for a clearer exploration of the daily variation trend of pollen emission concentration. This application employs an improved WRF-Chem-Pollen model for simulation, which can deeply analyze the influence of key factors such as meteorological elements and advection diffusion on daily pollen concentration. This model can accurately reflect the influence of daily meteorological elements such as temperature, precipitation, humidity, and wind speed on pollen emissions and concentration, while describing key physical processes such as advection diffusion, convective transport, and dry and wet deposition, thus providing a comprehensive analysis of pollen propagation and deposition behavior in the atmosphere.

[0107] like Figure 10 As shown in the figure, this is a time series comparison of the average observed and simulated values ​​of pollen concentration at various stations in a certain city during spring, based on the WRF-Chem-Pollen model. The red solid line represents the simulated average total pollen concentration at six observation stations in the city, while the blue dots represent the observed average total pollen concentration at the same six stations. The red and blue shading represents the daily average absolute deviation, respectively. The figure shows that the WRF-Chem-Pollen model can capture the temporal trend of plant pollen concentration during the spring pollen season relatively well, generally showing a trend of first increasing and then decreasing. However, influenced by meteorological factors such as temperature, precipitation, and relative humidity, as well as physical processes such as advection diffusion, convective transport, and wet and dry deposition, the daily pollen emission concentration exhibits significant fluctuations. While the pollen concentration simulated by the WRF-Chem-Pollen model generally follows the same trend as the observed pollen concentration, certain differences still exist.

[0108] Influenced by various factors such as meteorological conditions, the annual start date (SOS) of pollen emission and the total pollen emission during the pollen emission period in a certain city both exhibit certain interannual variations. Figure 10 As can be seen, the peak average total pollen concentration of grasses (including Cupressaceae, Salicaceae, and Pinaceae vegetation) in spring from 2006 to 2019 was mainly distributed in [location missing]. Between, the peak average total pollen concentration in 2020 exceeded The concentration was significantly higher than in previous years. It was also observed that before 2013 (excluding 2013), the peak average total pollen concentration in this region was generally lower than... However, after 2013 (including 2013), it basically remained at... The above, especially in 2018 and 2020, exceeded 700 and 2020 respectively. This significant upward trend may be related to changes in the composition of plant functional types around 2013, or it may be related to adjustments in pollen observation methods; the specific reasons are still unclear. Figure 10 As can be seen from the shaded areas, there are significant differences in pollen concentration among different designated observation sites, reflecting the significant impact of spatial heterogeneity in meteorological conditions and vegetation distribution on pollen concentration. Overall, the WRF-Chem-Pollen model accurately simulates pollen concentration during the spring pollen season (i.e., the two-stage pollen emission period) and can reproduce the temporal variation characteristics of the observed data well. Regarding the simulation of daily concentration variations, the model successfully captures the "bimodal structure" of pollen concentration commonly found in spring each year. The first peak typically occurs from mid-March to early April, while the second peak is concentrated between late April and early May, corresponding to the main pollen emission periods of Cupressaceae / Saussureaceae and Pinaceae vegetation, respectively. It is worth noting that, due to meteorological conditions, the peak times vary in different years, occurring earlier or later.

[0109] Through comprehensive evaluation, the WRF-Chem-Pollen model showed good consistency between its simulations and observations of spring pollen concentrations in most years, with the correlation coefficient R between predicted and observed values ​​generally higher than that of observed values. The R-values ​​reached [values] in 2009, 2014, 2015, 2017, and 2018. That's all. (Approximately) The year R value exceeds RMSE in The fact that the pollen simulation model constructed in this application has high accuracy and reliability in simulating the spring pollen concentration change trend of grasses in a certain city (excluding 2020) not only verifies the effectiveness of the pollen emission potential model constructed based on machine learning algorithms in the early stage, but also provides strong data support for the analysis and research of pollen-related health risks.

[0110] S27. Return the simulation results of pollen emission and concentration for each day of the pollen emission period to the terminal.

[0111] In practice, the response module in the pollen emission and concentration simulation system returns to the terminal the simulation results of pollen emission and pollen concentration of the vegetation to be observed at the designated observation site for each day of the pollen emission period, so that they can be displayed on the terminal.

[0112] The pollen emission and concentration simulation method provided in this application pre-trains a pollen yield prediction model based on a preset machine learning model using historical meteorological observation data from sample observation stations and sample pollen observation data from the vegetation to be observed. This model predicts the total pollen emission yield of the vegetation to be observed during the pollen emission period. The machine learning model can capture complex nonlinear relationships (such as the influence of climate, geographical location, and year on pollen yield) in the historical meteorological observation data of a specified observation station, improving the accuracy of the total pollen emission yield prediction and providing a basis for subsequently determining the pollen emission concentration for each day. This provides a reliable foundation. After predicting the total pollen emission yield, it combines the established dual pollen accumulated temperature thresholds (i.e., the accumulated temperature thresholds for the start and end of pollen emission) with a standardized pollen emission potential model to predict the pollen emission potential of the observed vegetation for each day of its pollen emission period. The dual pollen accumulated temperature thresholds—the start and end accumulated temperature thresholds—serve as temperature constraints, which, compared to a single accumulated temperature threshold, are more dynamically adaptable to the impact of climate change on pollen emission potential. The standardized pollen emission potential model built based on the dual pollen accumulated temperature thresholds can accurately simulate… This study aims to understand the temporal distribution pattern of pollen emission potential, avoid errors caused by uniform distribution, better match actual pollen emission curves, and improve the prediction accuracy of daily pollen emission potential. Furthermore, it calculates daily pollen emission based on the total pollen emission of the observed vegetation during the pollen emission period and the daily pollen emission potential value. Finally, it combines the daily pollen emission amount with corresponding meteorological factors to determine the target daily pollen emission amount, i.e., the pollen emission amount after incorporating meteorological disturbances. This analysis then analyzes the daily pollen emission and concentration during the pollen emission period. This simulation, combining daily pollen emissions with meteorological factors, simulates pollen emissions and concentrations. It considers the impact of meteorological factors on pollen emissions and concentrations (e.g., wind speed affects pollen dispersion speed and range, precipitation inhibits pollen emissions and promotes sedimentation, humidity affects pollen emission activity and suspension capacity, etc.). By combining daily pollen emissions with environmental diffusion conditions, it simulates more accurate daily pollen emissions and concentrations, providing a more reliable data foundation for the analysis and research of spring pollen emissions and concentrations, as well as the research on the impact of pollen emissions and concentrations on health risks (such as pollen allergies). This application employs multi-model fusion and staged modeling to reduce model complexity, making errors at each step traceable. The pollen yield prediction model and pollen emission potential model can be independently optimized, improving the accuracy of subsequent pollen emission and concentration simulation results.

[0113] Based on the same technical concept, this application also provides an electronic device 600, referring to... Figure 11As shown, the electronic device 600 is used to implement the pollen emission and concentration simulation method described in the above-described method embodiments. The electronic device 600 in this embodiment may include: a memory 601, a processor 602, and a computer program stored in the memory and executable on the processor, such as a pollen emission and concentration simulation program. When the processor executes the computer program, it implements the steps in the various pollen emission and concentration simulation method embodiments described above.

[0114] This application embodiment does not limit the specific connection medium between the memory 601 and the processor 602. This application embodiment... Figure 11 The memory 601 and the processor 602 are connected via a bus 603, and the bus 603 is in Figure 11 The connections between other components are shown in bold lines only and are not intended to be limiting. The bus 603 can be divided into address bus, data bus, control bus, etc. For ease of illustration, Figure 11 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0115] Memory 601 may be volatile memory, such as random-access memory (RAM); memory 601 may also be non-volatile memory, such as read-only memory, flash memory, hard disk drive (HDD), or solid-state drive (SSD); or memory 601 may be any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto. Memory 601 may be a combination of the above-described memories.

[0116] The processor 602 is used to implement the pollen emission and concentration simulation method provided in the embodiments of this application.

[0117] This application also provides a computer-readable storage medium storing computer-executable instructions required to execute the processor, including a program required to execute the processor.

[0118] In some possible implementations, various aspects of the pollen emission and concentration simulation method provided in this application can also be implemented in the form of a program product, which includes program code that, when the program product is run on an electronic device, causes the electronic device to perform the steps in the pollen emission and concentration simulation method according to various exemplary embodiments of this application described above.

[0119] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0120] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (devices), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0121] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0122] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0123] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0124] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A pollen emission and concentration simulation system, characterized in that, include: The communication module is used to receive pollen emission and concentration simulation requests sent by the terminal. The pollen emission and concentration simulation requests carry information on vegetation to be observed, information on designated observation stations, and information on the year to be simulated. The data acquisition module is used to acquire meteorological observation data of a specified observation station for each day during the pollen emission period of the vegetation to be observed, based on the year simulated in the request. The processor is configured to predict the total pollen emission of the vegetation under observation at the designated observation station during the pollen emission period, based on meteorological observation data from the designated observation station on each day of the pollen emission period and a pollen yield prediction model. The pollen yield prediction model is trained using a preset machine learning model based on historical sample meteorological observation data from the sample observation station and sample pollen observation data of the vegetation under observation. The processor also acquires the cumulative temperature on day t of the pollen emission period, and calculates the cumulative temperature based on the cumulative temperature on day t, the accumulated temperature threshold at the beginning of the pollen season, the accumulated temperature threshold at the end of the pollen season, and the pollen emission period... The pollen emission potential capacity value for day t is determined by using the number of days and a standardized pollen emission potential model. The standardized pollen emission potential model is used to calculate the pollen emission potential capacity value. The pollen emission amount for day t is determined based on the total pollen emission and the pollen emission potential capacity value for day t. Meteorological factors for day t are obtained, and the target pollen emission amount for day t is determined based on the pollen emission amount and the meteorological factors for day t. The pollen emission and concentration for each day of the pollen emission period are simulated, and the target pollen emission amount is the pollen emission amount after adding meteorological disturbances. The response module is used to return to the terminal the simulated results of pollen emission and concentration for each day of the pollen emission period.

2. The system as described in claim 1, characterized in that, The processor is specifically used to determine the cumulative pollen fraction emitted by the vegetation under observation from the start time of accumulated temperature to the day t and the flowering probability on the day t based on the accumulated temperature on the day t, the accumulated temperature threshold at the beginning of pollen, the accumulated temperature threshold at the end of pollen, and the number of days in the pollen emission period. The pollen emission potential value on day t is determined based on the cumulative pollen fraction emitted from the start time of accumulated temperature to day t, the flowering probability on day t, and the standardized pollen emission potential model.

3. The system as described in claim 2, characterized in that, The vegetation to be observed is divided into at least two categories according to different pollen emission start times, and each category of vegetation corresponds to a different pollen emission period. The processor is specifically used to determine the cumulative pollen fraction of the vegetation to be observed from the start time of the accumulated temperature to the day t, based on the accumulated temperature on day t of the pollen emission period corresponding to the vegetation to be observed, the accumulated temperature threshold at the beginning of the pollen period, and the accumulated temperature threshold at the end of the pollen period, for each type of vegetation to be observed. Based on the accumulated temperature on day t, the accumulated temperature threshold for the pollen initiation period, the accumulated temperature threshold for the pollen termination period, and the number of days in the pollen emission period corresponding to the vegetation to be observed, the probability of flowering at the beginning and the probability of flowering at the end of the pollen emission period on day t of the corresponding pollen emission period of the vegetation to be observed are determined. The flowering probability of the observed vegetation on day t of the corresponding pollen emission period is determined based on the flowering probability at the beginning and end of the pollen emission period on day t of the observed pollen emission period.

4. The system as described in claim 3, characterized in that, The processor is specifically used to calculate the cumulative pollen fraction emitted by the observed vegetation from the start time of accumulated temperature to day t using the following formula: in, Indicates the time from the start of accumulated temperature for the vegetation to be observed. The cumulative pollen fraction emitted up to day t of the pollen emission period corresponding to the vegetation to be observed; This represents the cumulative temperature on day t of the pollen emission period corresponding to the vegetation being observed. , This represents the temperature on day t of the pollen emission period corresponding to the vegetation being observed. This indicates the preset daily average temperature threshold; This indicates the accumulated temperature threshold for the pollen initiation period; This represents the accumulated temperature threshold for the pollen end period.

5. The system as described in claim 3, characterized in that, The processor is specifically used to calculate the probability of flowering of the observed vegetation on day t of the corresponding pollen emission period using the following formula: in, This represents the probability that the observed vegetation will begin flowering on day t of the corresponding pollen emission period; This represents the cumulative temperature on day t of the pollen emission period corresponding to the vegetation being observed. , This represents the temperature on day t of the pollen emission period corresponding to the vegetation being observed. This indicates the preset daily average temperature threshold. Indicates the start time of accumulated temperature; This indicates the accumulated temperature threshold for the pollen initiation period; This represents the tolerance factor for relative temperature accumulation, used to construct the transition range for temperature accumulation. This indicates the number of days in the pollen emission period corresponding to the vegetation being observed. ,in, This indicates the end time of pollen emission from the observed vegetation. Indicates the start time of pollen emission from the observed vegetation; and The probability of flowering at the end of the pollen emission period on day t of the corresponding pollen emission period of the observed vegetation is calculated using the following formula: in, This represents the probability that the pollen emission period of the observed vegetation ends on day t of the corresponding pollen emission period; Indicates the accumulated temperature threshold of the pollen end period; and The flowering probability of the observed vegetation on day t of the corresponding pollen emission period is calculated using the following formula: in, This represents the probability of the observed vegetation flowering on day t of the corresponding pollen emission period.

6. The system according to any one of claims 3 to 5, characterized in that, The processor is specifically used to calculate the pollen emission potential of the observed vegetation on day t of the corresponding pollen emission period using the following formula: in, This represents the potential pollen emission capacity of the observed vegetation on day t of the corresponding pollen emission period; Indicates the time from the start of accumulated temperature for the vegetation to be observed. The cumulative pollen fraction emitted up to day t of the pollen emission period corresponding to the vegetation to be observed; This represents the probability of the observed vegetation flowering on day t of the corresponding pollen emission period; This indicates the land cover of the vegetation to be observed; This represents a preset scaling factor, which characterizes the pollen emission enhancement effect parameter caused by temperature changes.

7. The system as described in claim 6, characterized in that, The processor is specifically used to calculate the pollen emission of the observed vegetation on day t of the corresponding pollen emission period using the following formula: in, This represents the pollen emission of the observed vegetation on day t of the corresponding pollen emission period; This represents the total pollen production of the observed vegetation during the corresponding pollen emission period.

8. The system as described in claim 3, characterized in that, The meteorological factors characterize the degree of influence of meteorology on pollen emission, and the meteorological factors include wind factors, precipitation factors and relative humidity factors; The processor is specifically used to input the pollen emission on day t into the pollen simulation model, which is an improved version of the WRF-Chem model. The pollen simulation model includes a WRF module, a Chem module, and a pollen emission regulation module. The WRF module calculates the meteorological factors for day t and sends the meteorological factors for day t to the pollen emission regulation module. The pollen emission regulation module calculates the target pollen emission of the vegetation under observation on day t based on the pollen emission on day t and the meteorological factors on day t, and sends the target pollen emission on day t to the Chem module. The Chem module is used to simulate pollen concentration based on the target pollen emission amount of the observed vegetation on each day during the corresponding pollen emission period. The pollen emission regulation module calculates the target pollen emission amount of the observed vegetation on day t of the corresponding pollen emission period according to the following formula: in, This represents the target pollen emission amount of the vegetation under observation on day t of the corresponding pollen emission period; This represents the pollen emission on day t of the pollen emission period corresponding to the vegetation to be observed; The wind factor represents the day t of the pollen emission period corresponding to the vegetation to be observed. This represents the precipitation factor on day t of the pollen emission period corresponding to the vegetation to be observed; The relative humidity factor represents the day t of the pollen emission period corresponding to the vegetation to be observed.

9. The system as described in claim 3 or 5, characterized in that, Different types of vegetation to be observed correspond to different accumulated temperature thresholds for the pollen initiation and pollen termination stages; and The processor is further configured to, before determining the cumulative pollen fraction emitted from the start time of accumulated temperature to the day t and the flowering probability on the day t, use a simulated annealing algorithm to optimize the accumulated temperature thresholds for the start of pollen, the end of pollen, the start time of accumulated temperature, and the preset daily average temperature threshold for the vegetation to be observed, based on the accumulated temperature on the day t, the accumulated temperature threshold for the pollen start period, the accumulated temperature threshold for the pollen end period, and the number of days in the pollen emission period, thereby obtaining the optimized accumulated temperature thresholds for the start of pollen, the accumulated temperature threshold for the pollen end period, the start time of accumulated temperature, and the daily average temperature threshold for the vegetation to be observed.

10. A method for simulating pollen emission and concentration, characterized in that, include: The receiving terminal sends a pollen emission and concentration simulation request, which carries information about the vegetation to be observed, information about the designated observation station, and information about the year to be simulated. Based on the year simulated in the request, obtain meteorological observation data from a specified observation station for each day of the pollen emission period of the vegetation to be observed. Based on the meteorological observation data of the designated observation station for each day during the pollen emission period of the vegetation to be observed and the pollen yield prediction model, the total pollen emission of the vegetation to be observed at the designated observation station during the pollen emission period is predicted. The pollen yield prediction model is obtained by training a preset machine learning model based on the historical sample meteorological observation data of the sample observation station and the sample pollen observation data of the vegetation to be observed. The cumulative temperature on day t of the pollen emission period is obtained. Based on the cumulative temperature on day t, the accumulated temperature threshold at the beginning of the pollen period, the accumulated temperature threshold at the end of the pollen period, the number of days in the pollen emission period, and the constructed standardized pollen emission potential model, the pollen emission potential value on day t is determined. The standardized pollen emission potential model is used to calculate the pollen emission potential value. The pollen emission amount on day t is determined based on the total pollen emission production and the pollen emission potential capacity value on day t. Obtain the meteorological factors for day t, determine the target pollen emission for day t based on the pollen emission amount and the meteorological factors for day t, and simulate the pollen emission and concentration for each day of the pollen emission period. The target pollen emission amount is the pollen emission amount after adding meteorological disturbances. The simulated pollen emission and concentration results for each day of the pollen emission period are returned to the terminal.

Images