Application method of novel efficient multi-scale coupling assimilation system in iCESM

Abstract

The invention belongs to the technical field of palaeoclimate data assimilation, and discloses an application method of a novel efficient multi-scale coupling assimilation system in iCESM, comprising: acquiring an observation increment at an observation point, and acquiring a projection of the observation increment; solving of observation increment is improved, and updated projection is obtained; an MSHea-EnKF coupling assimilation algorithm is embedded into an earth system mode CESM1.3; and an MSHea-EnKF coupling assimilation algorithm is embedded into an earth system mode iCESM1.3 mode containing isotope change of an f19g16 resolution version. According to the invention, the construction of the novel efficient multi-scale coupling assimilation algorithm MSHea-EnKF in the f19g16-resolution isotope earth system mode iCESM1.3 mode is realized, the multi-scale assimilation function is set to be selectable switch control, the calculation cost can be reduced, and the complex change of the climate system can be accurately captured.

Description

Technical Field

[0001] The present invention belongs to the technical field of paleoclimate data assimilation, and in particular relates to an application method of a novel high-efficiency multi-scale coupled assimilation system in iCESM. Background Art

[0002] Earth system models are important tools for understanding the mechanisms of past climate and environmental evolution and for predicting potential global change scenarios in the future. Proxy data that can record past climate change have the defect of uneven temporal and spatial distribution. Coupled data assimilation can integrate models and proxy data to obtain continuous temporal and spatial evolution characteristics of the atmosphere and ocean that are closer to the true value. Using coupled data assimilation systems, combining the dynamics and physics framework of coupled Earth system models, integrating proxy data information into coupled Earth system models, and producing a balanced and coordinated coupled model reanalysis state are the indispensable core elements for paleoclimate reanalysis. Paleoclimate research can provide important historical similarities for the present, and its importance is increasingly valued. If traditional ensemble assimilation algorithms (such as EnKF, ensemble Kalman filtering) are used, it is necessary to integrate N model members to obtain the background error covariance, which will require N times the computing resources of a single model, with high computing costs and low assimilation efficiency. This is undoubtedly not feasible for long-term paleoclimate simulations, which originally require huge computing resources. Summary of the invention

[0003] The purpose of the present invention is to provide an application method of a novel and efficient multi-scale coupled assimilation system in iCESM to solve the problems existing in the above-mentioned prior art.

[0004] To achieve the above object, the present invention provides a novel and efficient multi-scale coupled assimilation system application method in iCESM, comprising:

[0005] Obtain observations and pattern estimates at observation points;

[0006] calculating an observation increment based on the observation value and the mode estimate;

[0007] Projecting the observation increment onto the surrounding pattern grid points affected by the current observation increment to obtain the observation increment projection;

[0008] The solution of observation increments is improved through the efficient multi-scale coupled assimilation algorithm MSHea-EnKF to obtain updated projections;

[0009] After the projection update, the MSHea-EnKF coupled assimilation algorithm is embedded into the Earth system model CESM1.3 under the f19_g16 resolution version and the Earth system model iCESM1.3 after isotope embedding, realizing the setting of multiple scale assimilation functions as selectable switch control.

[0010] Calculating the observation increment based on the observation value and the pattern estimation value specifically includes:

[0011] A standard deviation between the observed value and the mode estimate is calculated, and an observation increment is calculated based on the observed value, the mode estimate and the standard deviation.

[0012] The observation increment is calculated based on the observed value, the model estimated value and the standard deviation. The specific calculation formula is:

[0013]

[0014] In the formula, is the observation increment at the observation point, σ P is the standard deviation of the pattern estimate at the observation point, σ o is the observed standard deviation, is the average of the mode values ​​of each member at the observation point, is the estimated value of the mode of each member at the observation point.

[0015] Project the observation increment onto the surrounding pattern grid points affected by the current observation increment to obtain the observation increment projection. The specific calculation formula is:

[0016]

[0017] In the formula, is the analysis increment at the projected model grid point, the superscript u represents the update value, cov(Δx,Δy) is the covariance of the model value at the model grid point and the model estimate at the observation point, is the variance of the pattern estimate at the observation point, is the first observed increment obtained.

[0018] The solution of observation increment is improved by using the efficient multi-scale coupled assimilation algorithm MSHea-EnKF. The specific calculation formula is:

[0019]

[0020] In the formula, Δy o,(τ) is the observation increment at the observation point, σ (τ) The standard deviation of the statistical data set of the three time scale filtering schemes, σ (o) is the observation standard deviation, τ=1, Γ represents Γ time scales, is the estimated value of the pattern at the observation point.

[0021] The calculation process for obtaining the observation increment projection is:

[0022]

[0023] In the formula, Δx u is the analytical increment at the projected mode grid point, is the observation increment at the first observation point, α (τ) is the adjustment weight coefficient in the three time scale filtering schemes, cov(Δx (τ) ,Δy (τ) ) represents the covariance relationship between the model values ​​at the model grid points of each time scale filtering scheme and the model estimation values ​​at the observation grid points.

[0024] The MSHea-EnKF coupled assimilation algorithm is embedded into the Earth System Model CESM1.3 at the f19_g16 resolution version and the Earth System Model iCESM1.3 including isotopes, including:

[0025] The MSHea-EnKF atmospheric assimilation algorithm is embedded into the Earth System Model CESM1.3 with the resolution of f19_g16.

[0026] The MSHea-EnKF ocean assimilation algorithm is embedded into the Earth System Model CESM1.3 with the resolution of f19_g16.

[0027] The MSHea-EnKF atmosphere and ocean assimilation algorithms are coupled and embedded into the Earth System Model CESM1.3 with a resolution of f19_g16.

[0028] The MSHea-EnKF coupled assimilation algorithm is embedded into the isotope Earth system model iCESM1.3 with the resolution of f19_g16.

[0029] The MSHea-EnKF atmospheric assimilation algorithm is embedded in the CESM1.3 model with the resolution of f19_g16, including:

[0030] Configure the f19_g16 resolution version of the Earth System Model CESM1.3 model example;

[0031] Load observation data and prepare f19_g16 resolution mode startup file; interpolate the loaded observation data, and construct background data and observation data of different frequencies for the atmospheric assimilation system based on the interpolated observation data;

[0032] Modify the parameters of the atmospheric assimilation system to control the number of resolution grid points, compile and run cross-validation, debug and modify the bugs of the assimilation code, and set the atmospheric assimilation function at different scales to be selectable switch control;

[0033] The MSHea-EnKF atmospheric assimilation algorithm is embedded into the f19_g16 resolution version of the CESM1.3 model to realize the atmospheric assimilation of surface air pressure, and the atmospheric assimilation system is tested and verified under the framework of the perfect twin experiment.

[0034] The MSHea-EnKF ocean assimilation algorithm is embedded in the CESM1.3 model with the resolution of f19_g16, including:

[0035] Configure the f19_g16 resolution version of the Earth System Model CESM1.3 model example;

[0036] Load observation data and prepare f19_g16 resolution mode startup file; interpolate the loaded observation data, and construct background data and observation data of different frequencies for the ocean assimilation system based on the interpolated observation data;

[0037] Modify the parameters of the ocean assimilation system to control the number of resolution grid points, compile and run cross-validation, debug and modify the bugs of the assimilation code, and set the ocean assimilation function at different scales to be selectable switch control;

[0038] The MSHea-EnKF ocean assimilation algorithm is embedded into the f19_g16 resolution version of the CESM1.3 model to realize ocean SST assimilation, and the ocean assimilation system is tested and verified under the framework of the perfect twin experiment.

[0039] The MSHea-EnKF coupled assimilation algorithm is embedded into the isotope Earth system model iCESM1.3 with the resolution of f19_g16, including:

[0040] The MSHea-EnKF atmosphere and ocean coupled assimilation algorithm was embedded into the CESM1.3 model with a resolution of f19_g16 to achieve the assimilation of atmospheric surface pressure and ocean SST. The coupled assimilation system was tested and verified under the framework of the perfect twin experiment.

[0041] Configure the iCESM1.3 model example with f19_g16 resolution version;

[0042] Obtain proxy data, convert units of proxy data, and produce proxy data required to build the f19_g16 resolution model assimilation system;

[0043] Load observation data and prepare the f19_g16 resolution model startup file in the iCESM version mode; produce background data of different frequencies required to build the f19_g16 resolution model assimilation system;

[0044] The coupled MSHea-EnKF atmosphere and ocean coupled assimilation algorithm is embedded into the f19_g16 resolution version of the iCESM1.3 model to realize the atmospheric assimilation of surface pressure and the ocean assimilation of SST. The coupled assimilation system is then tested and verified under the framework of the perfect twin experiment.

[0045] The technical effects of the present invention are:

[0046] The present invention discloses the application of a novel and efficient multi-scale coupled assimilation system in iCESM, integrates the MSHea-EnKF atmosphere and ocean coupled assimilation algorithm into the iCESM1.3 mode with a resolution of f19_g16, and realizes the setting of the atmospheric and ocean multi-scale assimilation functions as selectable switch control; reduces the demand for computing resources for assimilation; provides an important simulation test site for paleoclimate assimilation simulation research; promotes research in related fields such as paleoclimate, global change and climate prediction, and the development of paleoclimate reanalysis data; and contributes scientific and technological strength to disaster prevention and mitigation, and safeguards social and economic development and the well-being of the people. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0049] Figure 1 It is a schematic diagram of the application process of the new efficient multi-scale coupled assimilation system in iCESM in an embodiment of the present invention. DETAILED DESCRIPTION

[0050] Various exemplary embodiments of the present invention will now be described in detail. This detailed description should not be considered as limiting the present invention, but should be understood as a more detailed description of certain aspects, features, and embodiments of the present invention.

[0051] It should be understood that the terms described in the present invention are only for describing special embodiments and are not intended to limit the present invention. In addition, for the numerical range in the present invention, it should be understood that each intermediate value between the upper and lower limits of the scope is also specifically disclosed. Each smaller range between the intermediate value in any stated value or stated range and any other stated value or intermediate value in the described range is also included in the present invention. The upper and lower limits of these smaller ranges can be independently included or excluded in the scope.

[0052] Unless otherwise indicated, all technical and scientific terms used herein have the same meanings as those of ordinary skill in the art generally understood by the present invention. Although the present invention describes only preferred methods, any method similar or equivalent to that described herein may also be used in the implementation or testing of the present invention. All documents mentioned in this specification are incorporated by reference to disclose and describe the methods related to the documents. In the event of a conflict with any incorporated document, the content of this specification shall prevail.

[0053] It will be apparent to those skilled in the art that various modifications and variations may be made to the specific embodiments of the present invention description without departing from the scope or spirit of the present invention. Other embodiments derived from the present invention description will be apparent to those skilled in the art. The present application description and examples are exemplary only.

[0054] The words “include,” “including,” “have,” “contain,” etc. used in this article are open-ended terms, meaning including but not limited to.

[0055] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0056] Embodiment 1

[0057] like Figure 1 As shown, this embodiment provides an application method of a novel efficient multi-scale coupled assimilation system in iCESM, including: obtaining observation values ​​and model estimation values ​​at observation points; calculating observation increments based on the observation values ​​and the model estimation values; projecting the observation increments onto surrounding model grid points affected by the current observation increments to obtain observation increment projections; improving the solution of the observation increments by using the efficient multi-scale coupled assimilation algorithm MSHea-EnKF to obtain updated projections; after the projection is updated, embedding the MSHea-EnKF coupled assimilation algorithm into the earth system model CESM1.3 under the f19_g16 resolution version and into the earth system model iCESM1.3 containing isotopes, so as to realize the setting of multiple scale assimilation functions as selectable switch control.

[0058] This embodiment uses the observed value and the model estimated value at the observation point and the standard deviation of the observed value and the model estimated value to obtain the observation increment; the observation increment is projected onto the surrounding model grid points affected by it to obtain the projection of the observation increment; the solution of the observation increment is improved by the efficient multi-scale coupled assimilation algorithm MSHea-EnKF to obtain the updated projection; the MSHea-EnKF coupled assimilation algorithm is embedded in the f19_g16 resolution version of the Earth system model CESM1.3; the MSHea-EnKF coupled assimilation algorithm is embedded in the f19_g16 resolution version of the embedded isotope Earth system model iCESM1.3 model. The present invention realizes the construction of the new efficient multi-scale coupled assimilation algorithm MSHea-EnKF in the f19_g16 resolution isotope Earth system model iCESM1.3 model, and at the same time realizes the setting of multiple scale assimilation functions as selectable switch control. It can effectively reduce the computing cost, more efficiently integrate the substitute observation information, and more accurately capture the complex changes of the climate system.

[0059] This embodiment discloses the application of a new and efficient multi-scale coupled assimilation system in iCESM, integrates the MSHea-EnKF atmosphere and ocean coupled assimilation algorithm into the iCESM1.3 model with a resolution of f19_g16, and realizes the setting of the atmosphere and ocean multi-scale assimilation function as an optional switch control, which can reduce the demand for computing resources for assimilation and provide an important simulation test field for paleoclimate assimilation simulation research. The present invention can promote research in related fields such as paleoclimate, global change and climate prediction, as well as the development of paleoclimate reanalysis data, and contribute scientific and technological strength to disaster prevention and mitigation, and to ensuring social and economic development and people's health and well-being.

[0060] The specific implementation process of this embodiment includes:

[0061] The observation increment is obtained by using the observed value and the model estimated value at the observation point and the standard deviation of the observed value and the model estimated value;

[0062] Projecting the observation increment onto the surrounding pattern grid points affected by the observation increment to obtain the projection of the observation increment;

[0063] Improve the solution of observation increments through a new and efficient multi-scale coupled assimilation system to obtain updated projections;

[0064] The MSHea-EnKF atmospheric assimilation algorithm is embedded into the CESM1.3 model with the resolution of f19_g16.

[0065] The MSHea-EnKF ocean assimilation algorithm is embedded into the CESM1.3 model with the resolution of f19_g16.

[0066] Improvement of efficient multi-scale assimilation algorithm

[0067] EnKF can be summarized as a two-step method to obtain the observed increments of the pattern.

[0068] The first step is to calculate the observation increment using the observed values ​​and model estimates at the observation points and their standard deviations. The calculation formula is:

[0069]

[0070] In the formula, is the observation increment at the observation point, σ P is the standard deviation of the pattern estimate at the observation point, σ o is the observed standard deviation. is the average of the mode values ​​of each member at the observation point, is the estimated value of the pattern of each member at the observation point,

[0071] The first item on the right is the adjusted ensemble mean, which is the optimal estimate of the analysis value obtained based on the Bayesian theorem using the observed value and the pattern estimate at the observation point. When the prior ensemble variance is greater than the observed variance (σ P >σ o ), the ensemble mean adjustment will be closer to the observed value, otherwise it will be closer to the model estimate; the second item is the ensemble spread adjustment, which represents the nonlinear part of the observed value and the model estimate. When the prior ensemble variance is greater than the observed variance (σ P >σ o ), the degree of set dispersion will be greatly reduced by the observation, otherwise it will be closer to the degree of set dispersion of the prior estimate; the first two items are added together to obtain the analysis value at the observation point, minus the mode estimate of each member at the third observation point Get the observed increment

[0072] The second step is to project the observation increment to the surrounding pattern grid points affected by it after obtaining the observation increment at the observation point. The calculation formula is:

[0073]

[0074] In the formula, is the analysis increment at the projected model grid point, the superscript u represents the update value (update), cov(Δx,Δy) is the covariance of the model value at the model grid point and the model estimate at the observation point, is the variance of the pattern estimate at the observation point, is the observation increment obtained in step 1.

[0075] In the traditional algorithm, the ensemble statistics and background error covariance are obtained from N dynamic sets mi=1,N, and the observation increments corresponding to the N mode integrals are obtained after convolution with the observation information; in the efficient multi-scale algorithm, the prepared values ​​of the single mode integrals of N years, N months, N days and N steps are used as N statistical data to construct the "ensemble members" of multi-scale filtering assimilation at the annual scale, monthly scale, daily scale and hourly scale, respectively, and the observation increments are obtained after the N statistical data at each scale are Gaussian convolved with the observation information.

[0076] The new efficient multi-scale coupled assimilation algorithm improves the “two-step method” formula for calculating observation increments in the traditional EnKF algorithm. Specifically:

[0077] The first step is to update the formula for the observation increment at the observation point to:

[0078]

[0079] In the formula, Δy o,(τ) is the observation increment at the observation point, σ (τ) The standard deviation of the statistical data set constructed for the three time scale filtering schemes (steady, low frequency slow variation and high frequency fast variation), σ (o) is the observed standard deviation, τ=1, Γ represent Γ time scales respectively, in the present invention, Γ=3, is the estimated value of the pattern at the observation point.

[0080] The first item on the right is the analysis value at the observation point, and the second item is is the estimated value of the pattern at the observation point, obtained using 8-point interpolation.

[0081] In the second step, the observation increments obtained in the first step are projected onto the model grid. The update equation for the projection is:

[0082]

[0083] In the formula, Δx u is the analytical increment at the projected mode grid point, represents the observation increment at the observation point obtained in the first step, α (τ) represents the adjustment weight coefficient in the three time scale filtering schemes, cov(Δx (τ) ,Δy (τ) ) represents the covariance relationship between the model values ​​at the model grid points of each time scale filtering scheme and the model estimation values ​​at the observation grid points.

[0084] The MSHea-EnKF assimilation algorithm includes three scales of sequential assimilation filtering processes, namely, the monthly average scale of constant filter assimilation (filter_mon), the daily average scale of low-frequency filter assimilation (filter_day) and the hourly average scale of high-frequency filter assimilation (filter_hour). The assimilation of the three scales does not affect each other, and the assimilation increments of the three scales are added to the model variables in a certain proportion. The statistics of the ensemble members of each scale are generated by the model integration. For example, the 20 statistics of filter_day are the average of the results of the previous 20 days of the current state of the model, and the statistics of filter_mon are the average of the results of the previous 20 months of the current state of the model, and so on.

[0085] In addition to the steady-state filter assimilation at the monthly scale, the statistical data at other scales will be updated during the model operation. Since the statistical data that are continuously updated with the model operation are used to replace the original dynamic set members, the demand for computing resources is greatly reduced. For example, originally only 6 members could be run using 32×6 cores, but now only 32×1 core is needed to assimilate 20 statistical members.

[0086] 1) The MSHea-EnKF atmospheric assimilation algorithm is embedded into the CESM1.3 model with a resolution of f19_g16 to realize the atmospheric assimilation of surface pressure, and the assimilation system is tested and verified under the framework of the perfect twin experiment.

[0087] 2) Embed the MSHea-EnKF ocean assimilation algorithm into the f19_g16 resolution version of the CESM1.3 model to realize ocean SST assimilation, and test and verify the assimilation system under the framework of the perfect twin experiment.

[0088] 3) The MSHea-EnKF atmosphere and ocean assimilation algorithms are coupled to realize the assimilation of surface pressure in the atmosphere and SST in the ocean in the f19_g16 resolution version of the CESM1.3 model, and the assimilation system is tested and verified under the framework of the perfect twin experiment.

[0089] 4) The MSHea-EnKF atmosphere and ocean assimilation coupling algorithm is embedded in the f19_g16 resolution version of the iCESM1.3 model to achieve the assimilation of surface pressure into the atmosphere and SST into the ocean, and the assimilation system is tested and verified under the framework of the perfect twin experiment.

[0090] The construction process of the new efficient multi-scale coupled assimilation system includes:

[0091] (1) Example configuration:

[0092] Configure the f19_g16 resolution version of the CESM1.3 model example.

[0093] Configure the iCESM1.3 model example with the resolution version f19_g16.

[0094] (2) Prepare the data required for the construction of the f19_g16 resolution assimilation system:

[0095] Download observation (substitute) data and prepare the resolution model startup file. Interpolate the observation data to create background data and observation (substitute) data of three frequencies required to build the resolution model assimilation system.

[0096] (3) Realize the construction of f19_g16 resolution assimilation system in iCESM1.3 model:

[0097] Modify the resolution-related grid point parameter settings in the assimilation system; embed the MSHea-EnKF atmosphere and ocean coupled assimilation algorithm into the f19_g16 resolution version of the iCESM1.3 model. Realize the assimilation of surface pressure in the atmosphere and SST in the ocean. Test and verify the assimilation system under the framework of the perfect twin experiment.

[0098] The above is only a preferred specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A new and efficient multi-scale coupled assimilation system application method in iCESM, characterized by: include: Obtain observations and pattern estimates at observation points; calculating an observation increment based on the observation value and the mode estimate; Projecting the observation increment onto the surrounding pattern grid points affected by the current observation increment to obtain the observation increment projection; The solution of observation increments is improved through the efficient multi-scale coupled assimilation algorithm MSHea-EnKF to obtain updated projections; After the projection update, the MSHea-EnKF coupled assimilation algorithm is embedded into the Earth system model CESM1.3 under the f19_g16 resolution version and the Earth system model iCESM1.3 including isotopes, realizing the setting of multiple scale assimilation functions as selectable switch control.

2. The application method of a novel and efficient multi-scale coupled assimilation system in iCESM according to claim 1 is characterized in that: Calculating the observation increment based on the observation value and the pattern estimation value specifically includes: A standard deviation between the observed value and the mode estimate is calculated, and an observation increment is calculated based on the observed value, the mode estimate and the standard deviation.

3. The application method of a novel and efficient multi-scale coupled assimilation system in iCESM according to claim 2 is characterized in that: The observation increment is calculated based on the observed value, the model estimated value and the standard deviation. The specific calculation formula is: In the formula, is the observation increment at the observation point, σ P is the standard deviation of the pattern estimate at the observation point, σ o is the observed standard deviation, is the average of the mode values ​​of each member at the observation point, is the estimated value of the mode of each member at the observation point.

4. The application method of a novel and efficient multi-scale coupled assimilation system in iCESM according to claim 1 is characterized in that: Project the observation increment onto the surrounding pattern grid points affected by the current observation increment to obtain the observation increment projection. The specific calculation formula is: In the formula, is the analysis increment at the projected model grid point, the superscript u represents the update value, cov(Δx,Δy) is the covariance of the model value at the model grid point and the model estimate at the observation point, is the variance of the pattern estimate at the observation point, is the first observed increment obtained.

5. The application method of a novel and efficient multi-scale coupled assimilation system in iCESM according to claim 1 is characterized in that: The solution of observation increment is improved by using the efficient multi-scale coupled assimilation algorithm MSHea-EnKF. The specific calculation formula is: In the formula, Δy o,(τ) is the observation increment at the observation point, σ (τ) The standard deviation of the statistical data set of the three time scale filtering schemes, σ (o) is the observation standard deviation, τ=1, Γ represents Γ time scales respectively, is the estimated value of the pattern at the observation point.

6. The application method of a novel and efficient multi-scale coupled assimilation system in iCESM according to claim 1 is characterized in that: The calculation process for obtaining the observation increment projection is: In the formula, Δx u is the analytical increment at the projected mode grid point, is the observation increment at the first observation point, α (τ) is the adjustment weight coefficient in the three time scale filtering schemes, cov(Δx (τ) ,Δy (τ) ) represents the covariance relationship between the model values ​​at the model grid points of each time scale filtering scheme and the model estimation values ​​at the observation grid points.

7. The application method of a novel and efficient multi-scale coupled assimilation system in iCESM according to claim 1 is characterized in that: The MSHea-EnKF coupled assimilation algorithm is embedded into the Earth System Model CESM1.3 at the f19_g16 resolution version and the Earth System Model iCESM1.3 including isotopes, including: The MSHea-EnKF atmospheric assimilation algorithm is embedded into the Earth System Model CESM1.3 with the resolution of f19_g16. The MSHea-EnKF ocean assimilation algorithm is embedded into the Earth System Model CESM1.3 with the resolution of f19_g16. The MSHea-EnKF atmosphere and ocean assimilation algorithms are coupled and embedded into the Earth System Model CESM1.3 with a resolution of f19_g16. The MSHea-EnKF coupled assimilation algorithm is embedded into the isotope Earth system model iCESM1.3 with the resolution of f19_g16.

8. The application method of a novel and efficient multi-scale coupled assimilation system in iCESM according to claim 7 is characterized in that: The MSHea-EnKF atmospheric assimilation algorithm is embedded in the CESM1.3 model with the resolution of f19_g16, including: Configure the f19_g16 resolution version of the Earth System Model CESM1.3 model example; Load observation data and prepare f19_g16 resolution mode startup file; interpolate the loaded observation data, and construct background data and observation data of different frequencies for the atmospheric assimilation system based on the interpolated observation data; Modify the parameters of the atmospheric assimilation system to control the number of resolution grid points, compile and run cross-validation, debug and modify the bugs of the assimilation code, and set the atmospheric assimilation function at different scales to be selectable switch control; The MSHea-EnKF atmospheric assimilation algorithm is embedded into the f19_g16 resolution version of the CESM1.3 model to realize the atmospheric assimilation of surface air pressure, and the atmospheric assimilation system is tested and verified under the framework of the perfect twin experiment.

9. The application method of a novel and efficient multi-scale coupled assimilation system in iCESM according to claim 7 is characterized in that: The MSHea-EnKF ocean assimilation algorithm is embedded in the f19_g16 resolution version of the Earth System Model CESM1.3, including: Configure the f19_g16 resolution version of the Earth System Model CESM1.3 model example; Load observation data and prepare f19_g16 resolution mode startup file; interpolate the loaded observation data, and construct background data and observation data of different frequencies for the ocean assimilation system based on the interpolated observation data; Modify the parameters of the ocean assimilation system to control the number of resolution grid points, compile and run cross-validation, debug and modify the bugs of the assimilation code, and set the ocean assimilation function at different scales to be selectable switch control; The MSHea-EnKF ocean assimilation algorithm is embedded into the f19_g16 resolution version of the CESM1.3 model to realize ocean SST assimilation, and the ocean assimilation system is tested and verified under the framework of the perfect twin experiment.

10. The application method of a novel and efficient multi-scale coupled assimilation system in iCESM according to claim 7 is characterized in that: The MSHea-EnKF coupled assimilation algorithm is embedded into the isotope Earth system model iCESM1.3 with the resolution of f19_g16, including: The MSHea-EnKF atmosphere and ocean coupled assimilation algorithm was embedded into the CESM1.3 model with a resolution of f19_g16 to achieve the assimilation of atmospheric surface pressure and ocean SST. The coupled assimilation system was tested and verified under the framework of the perfect twin experiment. Configure the iCESM1.3 model example with f19_g16 resolution version; Obtain proxy data, convert units of proxy data, and produce proxy data required to build the f19_g16 resolution model assimilation system; Load observation data and prepare the f19_g16 resolution model startup file in the iCESM version mode; produce background data of different frequencies required to build the f19_g16 resolution model assimilation system; The coupled MSHea-EnKF atmosphere and ocean coupled assimilation algorithm is embedded into the f19_g16 resolution version of the iCESM1.3 model to realize the atmospheric assimilation of surface pressure and the ocean assimilation of SST. The coupled assimilation system is then tested and verified under the framework of the perfect twin experiment.

Images