Regional rainfall inter-year prediction method and device and storage medium

Abstract

The invention discloses a regional rainfall inter-annual prediction method and device and a storage medium, and belongs to the technical field of rainfall prediction, and the method comprises the steps: dividing a prediction region into a plurality of sub-regions according to a rainfall main mode, and generating a regional rainfall index of observation data; calculating an interannual variation rate based on the obtained power mode return data, and generating a regional rainfall index of the power mode return data; environment variable key areas are extracted, and predictable source factors are screened; training and optimizing a gradient boosting regression tree prediction model by using the screened predictable source factor and the regional rainfall index of the dynamic mode return data to obtain an optimized gradient boosting regression tree prediction model; on the basis of the optimized gradient lifting regression tree prediction model, predicting the rainfall index of each sub-region in a future set year, and generating a regional rainfall inter-year prediction result; the rainfall condition of the future set year is predicted in advance, computing resources are saved, and meanwhile the rainfall prediction skill is improved.

Description

Technical Field

[0001] The invention relates to a method, a device and a storage medium for regional precipitation interdecadal prediction, and belongs to the technical field of precipitation prediction. Background Art

[0002] The eastern part of my country is significantly affected by the East Asian summer monsoon, with summer precipitation accounting for 49% of the total annual precipitation. At the same time, summer precipitation in eastern China also shows obvious interdecadal variation characteristics. Effective interdecadal prediction is of great significance for water resources management and energy supply adjustment. At present, dynamical models are the main technical means of prediction, but due to the influence of initial and boundary conditions, the prediction skills for summer precipitation in eastern my country are generally very low and far from meeting actual needs. At present, precipitation prediction models based on statistical or machine learning methods are basically aimed at weather to seasonal scales, and there are few models for interdecadal prediction. Summary of the invention

[0003] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a method, device and storage medium for regional precipitation interdecadal prediction, which can predict the precipitation conditions of future set years in advance, save computing resources, and improve the precipitation prediction skills, which will help relevant departments to prepare for disaster prevention and mitigation, water resources regulation, etc. in advance.

[0004] To achieve the above object, the present invention is implemented by adopting the following technical solutions: In a first aspect, the present invention provides a method for regional precipitation decadal prediction, comprising: The main precipitation mode is calculated based on the observation data of the monthly average grid points in the forecast area, and the forecast area is divided into multiple sub-areas according to the main precipitation mode to generate the regional precipitation index of the observation data; Calculate the interdecadal variability based on the acquired dynamical model return data, and generate the regional precipitation index of the dynamical model return data based on the divided sub-regions; The key areas of environmental variables are extracted from the regional precipitation index of observation data and the regional precipitation index of dynamical model return data, and the predictability source factors are screened according to the extracted key areas of environmental variables; Using the selected predictability source factors and the regional precipitation index of the dynamical model return data, the gradient boosting regression tree prediction model is trained and optimized to obtain the optimized gradient boosting regression tree prediction model; Based on the optimized gradient boosting regression tree prediction model, the precipitation index of each sub-region in the future set years is predicted to generate regional precipitation decadal prediction results.

[0005] Furthermore, the calculation of the main precipitation mode based on the observation data of the monthly average grid points in the forecast area, dividing the forecast area into multiple sub-areas according to the main precipitation mode, and generating the regional precipitation index of the observation data include: Obtain observation data of monthly average grid points in the prediction area; The average precipitation in any season at each grid point is calculated based on the observed data, and a five-year sliding average is performed to generate decadal variability; Anomalies of decadal variability are calculated and the main modes of precipitation are extracted through empirical orthogonal decomposition; According to the latitudinal tripole distribution characteristics of the main precipitation mode, the forecast area is divided into three sub-regions, and the regional average precipitation index of each sub-region is calculated respectively.

[0006] Furthermore, the interdecadal variability is calculated based on the acquired dynamical model report data, and the regional precipitation index of the dynamical model report data is generated based on the divided sub-regions, including: Obtain dynamic model return data for the forecast area; Based on the dynamical model return data, the decadal variability of the annual average of future set years is extracted and spliced ​​into a sequence; Regional precipitation indices are generated from dynamical model return data based on the spliced ​​series of decadal variability.

[0007] Furthermore, extracting the key areas of environmental variables from the regional precipitation index of the observation data and the regional precipitation index of the dynamic model report data, and screening the predictability source factors according to the extracted key areas of environmental variables, includes: Based on the regional precipitation index of the observed data and the environmental variables of the pre-generated reanalysis data, the key areas of environmental variables in the observations are determined through correlation analysis; Using the regional precipitation index of the dynamical model return data, the key areas of environmental variables in the dynamical model are extracted; Verify the predictive power of key areas of environmental variables in the dynamical model and their consistency with observed physical mechanisms; Screen environmental variables that meet the consistency between predictive ability and physical mechanism as predictability source factors.

[0008] Furthermore, the verification of the predictive capability of the key areas of environmental variables in the dynamic model and the consistency with the observed physical mechanisms includes: Verify that the key areas of environmental variables in the dynamical model are consistent with the key areas of environmental variables in observations, that the confidence level in the correlation distribution map between the dynamical model report data and the reanalysis data exceeds the set threshold, and that the dynamical model can reproduce the physical process by which environmental variables in observations affect precipitation.

[0009] Furthermore, when the regional precipitation index of the screened predictability source factors and dynamic model return data is used to train and optimize the gradient boosting regression tree prediction model, the hyperparameters are optimized by a five-fold cross validation method.

[0010] Furthermore, the optimized gradient boosting regression tree prediction model is used to predict the precipitation index of each sub-region in a future set year, and then a linear regression model and linear trend correction are combined to generate a regional precipitation decadal prediction result.

[0011] In a second aspect, the present invention provides a regional precipitation decadal prediction device, comprising: The first generation module is used to calculate the main precipitation mode based on the observation data of the monthly average grid points in the prediction area, divide the prediction area into multiple sub-areas according to the main precipitation mode, and generate the regional precipitation index of the observation data; The second generation module is used to calculate the interdecadal variability based on the acquired dynamical model return data, and generate a regional precipitation index of the dynamical model return data based on the divided sub-regions; A screening module is used to extract key areas of environmental variables from the regional precipitation index of the observation data and the regional precipitation index of the dynamic model return data, and screen predictability source factors based on the extracted key areas of environmental variables; A training and optimization module is used to train and optimize the gradient boosting regression tree prediction model using the screened predictability source factors and the regional precipitation index of the dynamical model return data to obtain an optimized gradient boosting regression tree prediction model; The prediction module is used to predict the precipitation index of each sub-region in the future set year based on the optimized gradient boosting regression tree prediction model, and generate regional precipitation decadal prediction results.

[0012] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the aforementioned methods.

[0013] In a fourth aspect, the present invention provides a computer device, comprising: Memory, for storing computer programs / instructions; A processor is used to execute the computer program / instructions to implement the steps of any of the aforementioned methods.

[0014] In a fifth aspect, the present invention provides a computer program product, comprising a computer program / instruction, which implements the steps of any of the aforementioned methods when executed by a processor.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a method, device and storage medium for regional precipitation interdecadal prediction. Based on the optimized gradient boosting regression tree prediction model, the precipitation index of each sub-region in a future set year is predicted, and the regional precipitation interdecadal prediction results are generated. The precipitation conditions in the future set year are predicted in advance, which saves computing resources and improves the precipitation prediction skills, which helps relevant departments to make preparations for disaster prevention and mitigation, water resources regulation, etc. in advance. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 The present invention provides a flowchart of a method for regional precipitation decadal prediction. DETAILED DESCRIPTION

[0017] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0018] Embodiment 1: This embodiment introduces a method for regional precipitation decadal prediction, including: The main precipitation mode is calculated based on the observation data of the monthly average grid points in the forecast area, and the forecast area is divided into multiple sub-areas according to the main precipitation mode to generate the regional precipitation index of the observation data; Calculate the interdecadal variability based on the acquired dynamical model return data, and generate the regional precipitation index of the dynamical model return data based on the divided sub-regions; The key areas of environmental variables are extracted from the regional precipitation index of observation data and the regional precipitation index of dynamical model return data, and the predictability source factors are screened according to the extracted key areas of environmental variables; Using the selected predictability source factors and the regional precipitation index of the dynamical model return data, the gradient boosting regression tree prediction model is trained and optimized to obtain the optimized gradient boosting regression tree prediction model; Based on the optimized gradient boosting regression tree prediction model, the precipitation index of each sub-region in the future set years is predicted to generate regional precipitation decadal prediction results.

[0019] The application process of the regional precipitation interdecadal prediction method provided in this embodiment specifically involves the following steps: Step 1: Calculate the interdecadal variability of summer precipitation in eastern China: 1) Select the monthly average CN05.1 grid observation data with a horizontal resolution of 0.25°×0.25°, and calculate the summer average (June-August) of each grid point; 2) Calculate the five-year sliding average of the summer average of each grid point to obtain the interdecadal variability, and mark it as the year in the middle of the five years. For example, the average value of 1961-1965 is the decadal variability marked as 1963, and use the value of that year to subtract the climate state (the average of the summer average of 1981-2010) to obtain the anomaly; 3) For the three years from 1963 to the year of advance prediction ( The empirical orthogonal decomposition (EOF) analysis of precipitation anomalies during the period from 2010 to 2011 is performed to obtain the main mode of precipitation anomalies. For example, the predicted year is 2013, but only data up to 2012 are available. The interdecadal variability calculated by the five-year sliding average can only reach 2010, so the prediction is three years in advance for the current year. The spatial distribution of the main mode presents a latitudinal tripole distribution (the spatial distribution does not differ much when the year is changed). Therefore, the eastern region of China is divided into three sub-regions, namely South China, the Yangtze-Huaihe River Basin and North China, and the regional average value of each sub-region is calculated as the precipitation index.

[0020] Step 2: Process the dynamical model data set and reanalysis data set: 1) Use the 9 dynamical model report data of DCPPA in CMIP6, which reports predictions every year. The average value of the next 1-5 years predicted each year is used as the interdecadal variability, marked as the year in the middle of the five years, and the interdecadal variability predicted in each year is spliced ​​and combined into a sequence. The variables include: precipitation (PRE), sea temperature (SST), sea ice (SIC), geopotential high speed (HGT) and latitudinal wind field (UV); 2) Calculate the precipitation index of the three sub-regions predicted by the dynamic model; 3) Select ERA5 monthly reanalysis data with a horizontal resolution of 1°×1°, and obtain the interdecadal variability through a five-year sliding average, marked as the year in the middle of the five years, and the variables include: sea temperature (SST), sea ice (SIC), geopotential high speed (HGT) and latitudinal wind field (UV).

[0021] Step 3: Find the sources of precipitation predictability and identify the physical mechanisms: 1) Use the bilinear interpolation method to interpolate all dynamical model data, observations and reanalysis data into 1°×1° for easy evaluation and comparison, and remove linear trends; 2) Find the key sea temperature and sea ice areas in the same period, previous winter and spring that are related to the precipitation index of each sub-region in the observation (areas with a confidence level of more than 90% in the correlation distribution map of observed precipitation and reanalysis sea temperature or sea ice data), and use ERA5 reanalysis data to explore the physical processes that affect precipitation; 3) Find the key sea temperature and sea ice areas in the same period, previous winter and spring that are related to the precipitation index of each sub-region in the dynamical model data (in the correlation map of the dynamical model precipitation ensemble mean and the dynamical model sea temperature or sea ice ensemble mean Areas with a confidence level of more than 90% in the distribution map); 4) For the key SST and sea ice found in the previous step, evaluate the ability of the ensemble mean of all dynamical models (the average of all initial fields of all models) to predict their interdecadal predictions, and use the temporal correlation coefficient (CC) as an indicator; 5) If the SST or sea ice key area is consistent with the observation, the dynamical model has predictive skills for the SST or sea ice in this area (the confidence level in the correlation distribution map between the dynamical model data and the reanalysis data is more than 90%), and the dynamical model can reproduce the physical process of the factor affecting precipitation in the observation, then these key variables are considered to be sources of predictability, and the regional average is calculated as the factor index; 6) Repeat processes (2) to (5) for the three sub-areas to select all sources of predictability in each sub-area.

[0022] Step 4: Construct a machine learning model for decadal prediction of summer precipitation in eastern China and conduct independent test reports: 1) For each sub-region, the predictability source factor index sequence of the selected model prediction (9 model average sequence, one factor has 9 sequences) and the regional average sequence of precipitation in the corresponding sub-region predicted by the model prediction (9 model average sequence, a total of 9 sequences) are used to build a prediction model in combination with the gradient boosting regression tree (GBRT) model. The three years from 1963 to the year of advance prediction are selected as the training period, and the optimal hyperparameters are determined by a five-fold cross-validation. The sub-region average of summer precipitation in the next five years is used to predict the interdecadal prediction of summer precipitation in eastern China. The average index is used to predict the precipitation index of the three sub-regions. For example, in 2013, the interdecadal variability from 2011 to 2015 was predicted using the interdecadal data from 1963 to 2010. 2) For each sub-region, a linear regression model is constructed using the relationship between each grid point and the regional average index in the three years from 1963 to the year of advance prediction. Combined with the predicted precipitation index of the region in the next five years, the predicted values ​​of each grid point in the eastern China region are obtained, and the corresponding linear trends of each grid point are added to obtain the final predicted spatial distribution of summer precipitation in eastern China in the next five years.

[0023] like Figure 1The figure shows the prediction process of the entire model. First, the interdecadal variability of observed precipitation is calculated, and then the main mode obtained by EOF analysis is used to partition. After removing the linear trend, the key sea temperature and sea ice factors of each region are found, and the corresponding physical mechanism process is clarified. Similarly, the precipitation index of each region is calculated using the data predicted by the dynamic model, and the key sea temperature and sea ice factors are found in the dynamic model data without linear trend. If the factor satisfies the three conditions of being consistent with the key factor area in the observation, being able to reproduce the process of the factor affecting precipitation in the observation in the dynamic model, and the dynamic model having prediction skills for the factor, then the factor is identified as the predictability source factor of precipitation in the sub-region, and combined with the precipitation index predicted by the model, the GBRT model is used to predict the precipitation index. Finally, the relationship between each point in the sub-region and the corresponding precipitation index is used to construct a linear model to predict the value of each point, and the corresponding linear trend of each point is added to obtain the final predicted spatial distribution of summer precipitation in eastern China.

[0024] The contents involved in the above embodiment are described below in conjunction with a preferred embodiment.

[0025] The specific implementation plan is illustrated by taking the prediction of the interdecadal variability of summer precipitation in eastern China from 2015 (average from 2013 to 2017) to 2019 (average from 2017 to 2021) as an example.

[0026] Step 1: Select the monthly average CN05.1 grid observation data from 1961 to 2016, calculate the summer average (June to August) of each grid point, and perform a five-year sliding average to obtain the decadal variability, mark it as the year in the middle of the five years, and use the value of the current year to subtract the climate state (the average of the summer average from 1981 to 2010) to obtain the anomaly. After processing, the anomaly of the decadal variability of summer precipitation from 1963 to 2014 is obtained, and the empirical orthogonal decomposition (EOF) analysis is performed to obtain the main mode of the precipitation anomaly. The spatial distribution of the main mode shows a latitudinal tripole distribution. Therefore, the eastern region of China is divided into three sub-regions, namely South China, the Yangtze-Huaihe River Basin and North China, and the regional average of each sub-region is calculated as the precipitation index.

[0027] Step 2: Processing models and reanalyzing data. The nine dynamical model return data of DCPPA in CMIP6, which were reported annually from 1960 to 2016, were selected. The average value of the next 1-5 years predicted each year was used as the interdecadal variability, marked as the year in the middle of the five years, and the interdecadal variability predicted in each year was spliced ​​and combined into a sequence, and the years became 1963-2019. The variables included: precipitation (PRE), sea temperature (SST), sea ice (SIC), geopotential high speed (HGT) and latitudinal and longitudinal wind field (UV), and the precipitation index of the three sub-regions predicted by the dynamic model from 1963 to 2019 was calculated; the monthly reanalysis data of ERA5 from 1961 to 2016 were selected, and the interdecadal variability was obtained through a five-year sliding average, marked as the year in the middle of the five years, and the time range was 1963-2014. The variables included: sea temperature (SST), sea ice (SIC), geopotential high speed (HGT) and latitudinal and longitudinal wind field (UV).

[0028] Step 3: Find all sources of predictability in each sub-region. 1) First, use the bilinear interpolation method to interpolate all dynamical model data, observations and reanalysis data into 1°×1°, and remove linear trends. Then find the key sea temperature and sea ice areas in the same period, previous winter and spring that are related to the precipitation index of each sub-region in the observations during the period 1963-2014 (areas with a confidence level of more than 90% in the correlation distribution map between observed precipitation and reanalysis sea temperature or sea ice data), and use ERA5 reanalysis data to explore the physical processes that affect precipitation; 2) Find the key sea temperature and sea ice areas in the same period, previous winter and spring that are related to the precipitation index of each sub-region in the dynamical model data during the period 1963-2014 (areas with a confidence level of more than 90% in the correlation distribution map between the dynamical model precipitation ensemble mean and the dynamical model sea temperature or sea ice ensemble mean); 3 ) For the key SST and sea ice found in the previous step, evaluate the ability of the ensemble mean of all dynamical models (the average of all initial fields of all models) to make decadal predictions, and use the temporal correlation coefficient (CC) as an indicator; 4) If the following three conditions are met: a. The SST or sea ice in the key area is consistent with the observation; b. The dynamical model has prediction skills for the SST or sea ice in the area (the confidence level in the correlation distribution map between the dynamical model data and the reanalysis data exceeds 90%); c. The dynamical model can reproduce the physical process of the factor affecting precipitation in observations; then the key variable is identified as a source of predictability, and the regional average is calculated as the factor index. The index series time range is 1963-2019; 5) Repeat processes (2)-(4) for the three sub-areas until all the predictability source factor indices for each sub-area are selected.

[0029] Step 4: Construct a machine learning model for decadal prediction of summer precipitation in eastern China and conduct independent trial reports. 1) The predictability source factor index of the selected model predictions from 1963 to 2019 (9 model average sequences, one factor has 9 sequences) and the regional average sequence of precipitation in the corresponding area predicted by the model from 1963 to 2019 (9 model average sequences, a total of 9 sequences) were used to build a prediction model in combination with the GBRT model. 1963-2014 was selected as the training period, and the optimal hyperparameters were determined using a five-fold cross-validation. The three precipitation indices for the next five years (2015-2019) were predicted to obtain the precipitation indices predicted for the three regions; 2) For each sub-region, a linear regression model was constructed using the relationship between each grid point and the regional average index in the sub-region during the period 1963-2014. Combined with the regional predicted precipitation index from 2015 to 2019, the predicted values ​​of each grid point in the eastern China region were obtained. Together with the corresponding linear trends of each grid point, the final predicted spatial distribution of summer precipitation in eastern China from 2015 to 2019 was obtained.

[0030] Embodiment 2: This embodiment provides a regional precipitation decadal prediction device, comprising: The first generation module is used to calculate the main precipitation mode based on the observation data of the monthly average grid points in the prediction area, divide the prediction area into multiple sub-areas according to the main precipitation mode, and generate the regional precipitation index of the observation data; The second generation module is used to calculate the interdecadal variability based on the acquired dynamical model return data, and generate a regional precipitation index of the dynamical model return data based on the divided sub-regions; A screening module is used to extract key areas of environmental variables from the regional precipitation index of the observation data and the regional precipitation index of the dynamic model return data, and screen predictability source factors based on the extracted key areas of environmental variables; A training and optimization module is used to train and optimize the gradient boosting regression tree prediction model using the screened predictability source factors and the regional precipitation index of the dynamical model return data to obtain an optimized gradient boosting regression tree prediction model; The prediction module is used to predict the precipitation index of each sub-region in the future set year based on the optimized gradient boosting regression tree prediction model, and generate regional precipitation decadal prediction results.

[0031] The specific functional implementation of each of the above modules can be found in the relevant contents of the method in Example 1 and will not be elaborated here.

[0032] Embodiment 3, this embodiment provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps of any one of the methods described in Embodiment 1 are implemented.

[0033] Embodiment 4: This embodiment provides a computer device, including: Memory, for storing computer programs / instructions; A processor, configured to execute the computer program / instructions to implement the steps of any one of the methods described in Example 1.

[0034] Embodiment 5: This embodiment provides a computer program product, including a computer program / instruction, which implements the steps of any method described in Embodiment 1 when executed by a processor.

[0035] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

[0036] It should be understood by those skilled in the art that the embodiments of the present disclosure may be provided as methods, systems or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware aspects. Moreover, the present disclosure may take the form of a computer program product implemented 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.

[0037] The present disclosure is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present disclosure. It should be understood that each process and / or block in the flowchart and / or block diagram and the combination of processes and / or blocks in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0038] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate 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 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0039] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0040] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure rather than to limit its protection scope. Although the present disclosure has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that after reading the present disclosure, those skilled in the art can still make various changes, modifications or equivalent substitutions to the specific implementation methods of the invention, but these changes, modifications or equivalent substitutions are all within the protection scope of the disclosed claims to be approved.

Claims

1. A method for regional precipitation decadal prediction, characterized in that: include: The main precipitation mode is calculated based on the observation data of the monthly average grid points in the forecast area, and the forecast area is divided into multiple sub-areas according to the main precipitation mode to generate the regional precipitation index of the observation data; Calculate the interdecadal variability based on the acquired dynamical model return data, and generate the regional precipitation index of the dynamical model return data based on the divided sub-regions; The key areas of environmental variables are extracted from the regional precipitation index of observation data and the regional precipitation index of dynamical model return data, and the predictability source factors are screened according to the extracted key areas of environmental variables; Using the selected predictability source factors and the regional precipitation index of the dynamical model return data, the gradient boosting regression tree prediction model is trained and optimized to obtain the optimized gradient boosting regression tree prediction model; Based on the optimized gradient boosting regression tree prediction model, the precipitation index of each sub-region in the future set years is predicted to generate regional precipitation decadal prediction results.

2. The method for regional precipitation decadal prediction according to claim 1, characterized in that: The method of calculating the main precipitation mode based on the observation data of the monthly average grid points in the prediction area, dividing the prediction area into multiple sub-areas according to the main precipitation mode, and generating the regional precipitation index of the observation data includes: Obtain observation data of monthly average grid points in the prediction area; The average precipitation in any season at each grid point is calculated based on the observed data, and a five-year sliding average is performed to generate decadal variability; Anomalies of decadal variability are calculated and the main modes of precipitation are extracted through empirical orthogonal decomposition; According to the latitudinal tripole distribution characteristics of the main precipitation mode, the forecast area is divided into three sub-regions, and the regional average precipitation index of each sub-region is calculated respectively.

3. The method for regional precipitation decadal prediction according to claim 1, characterized in that: The method of calculating the interdecadal variability based on the acquired dynamical model report data and generating a regional precipitation index of the dynamical model report data based on the divided sub-regions includes: Obtain dynamic model return data for the forecast area; Based on the dynamical model return data, the decadal variability of the annual average of future set years is extracted and spliced ​​into a sequence; Regional precipitation indices are generated from dynamical model return data based on the spliced ​​series of decadal variability.

4. The method for regional precipitation decadal prediction according to claim 1, characterized in that: The key areas of environmental variables are extracted from the regional precipitation index of the observation data and the regional precipitation index of the dynamic model report data, and predictability source factors are screened according to the extracted key areas of environmental variables, including: Based on the regional precipitation index of the observed data and the environmental variables of the pre-generated reanalysis data, the key areas of environmental variables in the observations are determined through correlation analysis; Using the regional precipitation index of the dynamical model return data, the key areas of environmental variables in the dynamical model are extracted; Verify the predictive power of key areas of environmental variables in the dynamical model and their consistency with observed physical mechanisms; Screen environmental variables that meet the consistency between predictive ability and physical mechanism as predictability source factors.

5. The method for regional precipitation decadal prediction according to claim 4, characterized in that: The verification of the predictive ability of key areas of environmental variables in the dynamic model and their consistency with observed physical mechanisms includes: Verify that the key areas of environmental variables in the dynamical model are consistent with the key areas of environmental variables in observations, that the confidence level in the correlation distribution map between the dynamical model report data and the reanalysis data exceeds the set threshold, and that the dynamical model can reproduce the physical process by which environmental variables in observations affect precipitation.

6. The method for regional precipitation decadal prediction according to claim 1, characterized in that: When training and optimizing the gradient boosting regression tree prediction model using the screened predictability source factors and the regional precipitation index of the dynamical model return data, the hyperparameters are optimized by the five-fold cross validation method.

7. The method for regional precipitation decadal prediction according to claim 1, characterized in that: The optimized gradient boosting regression tree prediction model is used to predict the precipitation index of each sub-region in the future set year, and then a linear regression model and linear trend correction are combined to generate a regional precipitation decadal prediction result.

8. A regional precipitation decadal prediction device, characterized in that: include: The first generation module is used to calculate the main precipitation mode based on the observation data of the monthly average grid points in the prediction area, divide the prediction area into multiple sub-areas according to the main precipitation mode, and generate the regional precipitation index of the observation data; The second generation module is used to calculate the interdecadal variability based on the acquired dynamical model return data, and generate a regional precipitation index of the dynamical model return data based on the divided sub-regions; A screening module is used to extract key areas of environmental variables from the regional precipitation index of the observation data and the regional precipitation index of the dynamic model return data, and screen predictability source factors based on the extracted key areas of environmental variables; A training and optimization module is used to train and optimize the gradient boosting regression tree prediction model using the screened predictability source factors and the regional precipitation index of the dynamical model return data to obtain an optimized gradient boosting regression tree prediction model; The prediction module is used to predict the precipitation index of each sub-region in the future set years based on the optimized gradient boosting regression tree prediction model, and generate regional precipitation decadal prediction results.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method described in any one of claims 1 to 7 are implemented.

10. A computer device, characterized in that: include: Memory, for storing computer programs / instructions; A processor, configured to execute the computer program / instructions to implement the steps of the method according to any one of claims 1 to 7.

Images