Multi-mode north pole climate change prediction and regulation mechanism analysis method

Abstract

The invention relates to the field of meteorological prediction, and provides a multi-mode north pole climate change prediction and regulation and control mechanism analysis method. The purpose of the invention is to predict the climate change of the arctic region through the relationship among NPO, AMOC and arctic near-surface air temperature. According to the method, by constructing and verifying a large sample data set containing AMOC substitute data and north pole SAT historical observation data, the performance and deviation of the CMIP6 mode in simulation of AMOC and north pole SAT changes are evaluated. Then, a physical mechanism of the influence of the AMOC on the north pole SAT is disclosed, and the regulation and control effect of the NPO internal change rate on the relationship between the AMOC and the north pole SAT is explored; based on this, the future climate change scene is simulated, the mode parameters are adjusted to accurately reflect the influence of AMOC and NPO on the north pole SAT, and the change trend of the future north pole SAT and AMOC is predicted to evaluate the reliability of mode prediction. The method is used for predicting the climate change of the arctic region, and the prediction accuracy of the future change of the arctic climate is improved by deeply understanding the interaction between different climate modes.

Description

Technical Field

[0001] The present invention relates to the field of meteorological forecasting and provides a method for multi-modal Arctic climate change forecasting and its regulation mechanism analysis. Background Art

[0002] In recent years, the impact of climate change on the Arctic has attracted widespread attention. In particular, from 1950 to 2014, the Arctic surface air temperature (SAT) and the Atlantic Meridional Overturning Circulation (AMOC) showed significant in-phase changes on the interdecadal time scale. This phenomenon reveals two key factors: one is the rapid warming of the Arctic, and the other is the impact of changes in the Atlantic Meridional Overturning Circulation on the Arctic climate.

[0003] The rapid warming in the Arctic, known as poleward intensification, is associated with a variety of physical processes, including local radiative feedbacks and poleward ocean and atmospheric heat transport. These processes mainly affect Arctic SAT on interannual timescales, while changes in Arctic SAT are more positively correlated with AMOC on decadal timescales. The strengthening of AMOC is accompanied by enhanced ocean heat transport to the mid- and high-latitudes of the Atlantic, leading to basin-scale warming of North Atlantic SST and melting of Arctic sea ice.

[0004] In addition, changes in Arctic SAT are also affected by the variability of Pacific SST, especially the relationship with the Pacific Decadal Oscillation (PDO). Observations and simulations show that the in-phase changes of PDO and AMO can cause warming or cooling in the Arctic. The local atmospheric circulation in the Arctic and the atmospheric circulation in other regions can also adjust the decadal changes of Arctic SAT and sea ice.

[0005] However, current research still does not fully understand how low-frequency climate modes affect Arctic warming. In order to gain a deeper understanding of these relationships, researchers used a variety of proxy data and climate model simulation data to explore the relationship between AMOC and Arctic SAT and its possible physical mechanisms.

[0006] On the decadal time scale, the North Pacific Oscillation (NPO) affects the relationship between AMOC and Arctic SAT through poleward water vapor transport and cross-basin atmospheric circulation processes. Specifically, the synchronous phase changes of NPO and AMOC can explain 14%-41% of the AMOC-Arctic SAT covariance.

[0007] In general, the change of Arctic SAT is a complex process, which is affected by many factors, including changes in AMOC, Pacific SST variability, and the North Pacific Oscillation. The interaction of these factors has an important impact on the future changes of the Arctic climate.

[0008] In summary, the change of Arctic SAT is a complex process, which is affected by many factors, including changes in AMOC, Pacific SST variability, and the North Pacific Oscillation. The interaction of these factors has an important impact on the future changes of the Arctic climate. Summary of the invention

[0009] The purpose of the present invention is to predict climate change in the Arctic region through the relationship between the North Pacific Oscillation (NPO), the Atlantic Meridional Overturning Circulation (AMOC) and the Arctic Surface Air Temperature (SAT).

[0010] In order to achieve the above purpose, the present invention adopts the following technical solutions:

[0011] A method for multi-modal Arctic climate change prediction and its regulation mechanism analysis, characterized by comprising the following steps:

[0012] Step 1: Dataset construction and verification

[0013] 1.1 Collect and organize observational data, including proxy data of AMOC and historical observational data of Arctic SAT;

[0014] 1.2 Build a large sample dataset for CMIP6 models to ensure data quality and integrity;

[0015] 1.3 Compare observational data with model output to verify the reliability of the dataset;

[0016] 1.4 Evaluate the consistency and accuracy of the dataset through statistical tests (e.g., correlation coefficient, mean square error).

[0017] Step 2: Comparison of model simulations with observational data

[0018] 2.1 Use the constructed dataset to simulate the CMIP6 model and obtain the simulation results of AMOC and Arctic SAT;

[0019] 2.2 Compare the simulation results with observational data to assess the model's ability to simulate changes in AMOC and Arctic SAT;

[0020] 2.3 Through comparative analysis, identify the deviations and uncertainties of the model during the simulation process.

[0021] Step 3: Uncovering the physical mechanism

[0022] 3.1 Comparative analysis of observational and simulated data to diagnose the physical process of the AMOC's influence on the Arctic SAT, based on the calculation formula of the meridional overturning stream function (MSF):

[0023]

[0024] where lev B is the depth of the ocean floor, lon w is the western boundary of the ocean e is the eastern boundary of the ocean, v(time, lev, lat, lon) represents the northward flow velocity of the ocean, lev represents the vertical level, indicating the different depth layers in the ocean, lat represents latitude, lon represents longitude, dlon represents the differential element of longitude, which is used for integration operations, and dlev represents the differential element of vertical level, which is used for integration operations;

[0025] The AMOC index is calculated as the maximum MSF value below 500 m in the Atlantic region north of 20°N, according to the definition of the Arctic Gradient Index AGI:

[0026] AGI=SAT 60N-90N -SAT 0-90N

[0027] Among them, SAT 60N-90N and SAT 0-90N It is the area-weighted average, where 60N-90N represents the area from 60 degrees north latitude to 90 degrees north latitude, and 0-90N represents the area from the equator to 90 degrees north latitude (the North Pole);

[0028] 3.2 Through thermodynamic and kinetic analysis, the air-sea interaction and the heat transport mechanism of the atmosphere-ocean towards the poles are revealed. The calculation formula of atmospheric heat transport (AHT) is:

[0029]

[0030] where φ is the latitude, φ s is the latitude of Antarctica, Flx s and Flx TOA are the net downward heat flux from the surface and the net incident radiation at the top of the atmosphere, R E represents the radius of the Earth, λ represents the radial longitude, and ocean heat transport is the model diagnostic output;

[0031] 3.3 Establishing the quantitative relationship between physical processes and the changes in AMOC and Arctic SAT, the energy budget of SAT is calculated as:

[0032]

[0033] On the left side of the equation, T air represents SAT, Δ represents annual anomaly, positive anomaly corresponds to downward direction, S u / S d Represents upward / downward shortwave radiation, F latent / F sensible represents latent heat / sensible heat flux, l dis the long-wave radiation, R is the residual term, T s is the surface temperature, σ is the Stefan-Boltzmann constant, and ε is the surface emissivity.

[0034] Step 4: The regulatory role of internal variability

[0035] 4.1 The impact of internal variability such as NPO on the relationship between AMOC and Arctic SAT is analyzed. The NPO index is defined as the normalized sea level anomaly in the mid-latitudes of the North Pacific (45°-65°N, 170°E-140°W) minus the normalized sea level pressure anomaly in the subtropics (20°-40°N, 170°E-140°W). The calculation formula of the NPO index is as follows:

[0036]

[0037] SLP mid : is the normalized sea level pressure anomaly in the mid-latitude North Pacific (45°-65°N, 170°E-140°W);

[0038] SLP sub : is the normalized sea level pressure anomaly in the subtropical region (20°-40°N, 170°E-140°W);

[0039] It represents the average of the difference between the sea level pressure anomalies in two regions;

[0040] σSLP mid and σSLP sub are the standard deviations of the sea level pressure anomalies in the two regions.

[0041] 4.2 Determine the correlation between NPO and AMOC and Arctic SAT through statistical analysis:

[0042]

[0043] r NPO，AMOC : represents the correlation coefficient between NPO and AMOC; r NPO，SAT : represents the correlation coefficient between NPO and SAT; NPO i 、AMOC i , SAT i are the NPO index, AMOC index and Arctic SAT value at the i-th time point, respectively; n: is the total number of time points;

[0044] 4.3 Explore how NPO regulates atmospheric and oceanic circulation and thus affects Arctic SAT.

[0045] 4.3.1. Evaluation of the impact of NPO on meridional atmospheric heat transport (AHT):

[0046] ΔAHT=CΔv

[0047] where ΔAHT represents the change in meridional atmospheric heat transport, C is a proportionality constant that represents the sensitivity of meridional atmospheric heat transport to changes in ocean northward flow velocity, and Δv represents the change in northward flow velocity between the positive and negative phases of the NPO:

[0048]

[0049] and Represent the ocean northward flow speed during the positive and negative phases of NPO respectively.

[0050] 4.3.2 How does NPO affect the Atlantic Meridional Overturning Circulation (AMOC) through air-sea interaction, and thus affect the Arctic SAT? Here, the change of Δv may lead to the change of AMOC, which is expressed as:

[0051]

[0052] ΔAMOC reflects the change in AMOC intensity caused by the phase change of NPO. The change of Δv will affect ΔAMOC through the following channels:

[0053] If Δv>0, it indicates that the northward flow velocity increases, which will lead to more heat transport to the North Atlantic and enhance AMOC, that is, ΔAMOC>0;

[0054] If Δv < 0, it indicates that the northward flow velocity decreases, which will lead to a decrease in heat transport and weaken AMOC, that is, ΔAMOC < 0;

[0055] The NPO affects Arctic SAT by modulating atmospheric and oceanic circulations, through influencing the northward flow velocity Δv and the AMOC strength ΔAMOC.

[0056] Step 5: Predictions of future climate change

[0057] 5.1. Simulate future climate change scenarios based on the revealed physical mechanisms and the regulation of internal variability using the compared models;

[0058] 5.1.1. In simulating future climate change, incorporate the AMOC impact mechanism on Arctic SAT revealed in step 3 into the model, and correctly represent the changes in heat transport, salinity and ocean productivity in the model;

[0059] 5.1.2. Integrate the regulatory role of the NPO internal variability analyzed in step 4 into the simulation, including setting up parameterization schemes in the model that can simulate NPO changes, which will feed back to AMOC and Arctic SAT;

[0060] 5.2 Adjust model parameters to better reflect the effects of observed physical processes and internal variability;

[0061] 5.2.1. Based on the results of steps 3 and 4, adjust model parameters to ensure that the model can better simulate the influence of AMOC and NPO on Arctic SAT, including adjusting the air-sea coupling parameters to reflect the transport of heat and salinity;

[0062] 5.2.2 Adjust the internal variability parameters in the models, including the representation of the NPO, to ensure that the models can simulate the changes in internal variability and its impact on the AMOC and Arctic SAT;

[0063] 5.3 Input future RCP or SSP emission scenarios and simulate the changes of AMOC and Arctic SAT in the next few decades to centuries.

[0064] 5.4 Perform statistical analysis on the simulation results to extract the predicted trends of future Arctic SAT and AMOC, extract the future trends of AMOC and Arctic SAT, and compare them with the findings in steps 3 and 4 to assess the reliability of the model predictions.

[0065] Because the present invention adopts the above technical means, it has the following beneficial effects:

[0066] By constructing and verifying a large sample dataset containing AMOC proxy data and Arctic SAT historical observation data, the problem of dataset construction and verification in Arctic climate change prediction was solved, achieving the effect of improving data reliability and accuracy, and considering multi-source data more comprehensively than existing technologies.

[0067] By simulating and comparing the CMIP6 model output with observational data, the performance and deviations of the model in simulating AMOC and Arctic SAT changes were evaluated and identified, which solved the problem of matching model simulation with actual observations, thereby improving the accuracy of model simulation. Compared with existing technologies, it is better able to identify and adjust deviations in the simulation process.

[0068] The physical mechanism of the AMOC's impact on the Arctic SAT was revealed, and quantitative relationships were established through diagnostic analysis, thermodynamics and kinetic studies, and related physical parameters were calculated. This solved the problem of unclear physical mechanisms in Arctic climate change predictions, achieved the effect of improving the scientific nature of the predictions, and provided a deeper understanding of the internal interactions of the climate system than existing technologies.

[0069] The authors explored the regulatory effect of NPO internal variability on the relationship between AMOC and Arctic SAT, determined their correlation through statistical analysis, and evaluated how NPO regulates Arctic SAT by influencing atmospheric and oceanic circulation. This solved the unclear role of internal variability in Arctic climate change predictions, achieved the effect of improving prediction precision, and considered the complexity of the climate system more comprehensively than existing technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 , shows the temporal evolution of the AMOC and AGI indices observed during 1995-2014.

[0071] Figure 2 , quantify the regulation of the decadal NPO on the correlation between AMOC and AGI. Scatter plots of the AMOC index and AGI index (unit: ℃) when the decadal NPO and AMOC indices are in the same phase and opposite phase from 1950 to 2014. (a, b) Multiple sets of AMOC proxy indices; (c, d) FGOALS-g3 large sample ensemble simulation; (e, f) piControl large ensemble simulated by FGOALS-g3; "In phase" ("opposite phase") means that the NPO and AMOC indices are in the same (opposite) phase on the decadal scale, and the absolute values ​​of their indices are greater than 0.5 times the standard deviation. The red and blue lines are regression lines, and the black line is the 95% confidence interval. The linear correlation coefficient (r), p value and number of sample points of the horizontal and vertical axes are shown.

[0072] Figure 3 Composite analysis of near-surface air temperature (SAT, unit: °C, shaded) and sea level pressure (SLP) anomalies (unit: Pa, contour lines, interval: 5 Pa) simulated by the FGOALS-g3 large sample ensemble under different NPO and AMOC phases. (a) NPO- and AMOC+ phases; (b) NPO+ and AMOC+ phases; (c) NPO- and AMOC- phases; (d) NPO+ and AMOC- phases. The standardized NPO and AMOC indices are both greater than ±0.5 standard deviations, and the NPO index leads the AMOC index by three years. (c) is (a) minus (b), and (f) is (d) minus (e). The dotted line indicates negative SLP anomalies, the solid line indicates positive anomalies, and the thick black line is the zero line. The black / red lines in the sub-figures are the latitudinal averages of the SAT / SLP anomalies. The dotted areas pass the signal-to-noise ratio test (signal is greater than noise). DETAILED DESCRIPTION

[0073] The following is a detailed description of the embodiments of the present invention. Although the present invention will be described and illustrated in conjunction with some specific embodiments, it should be noted that the present invention is not limited to these embodiments. On the contrary, modifications or equivalent substitutions made to the present invention should all be included in the scope of the claims of the present invention.

[0074] In addition, in order to better illustrate the present invention, numerous specific details are given in the following specific embodiments. It will be understood by those skilled in the art that the present invention can also be implemented without these specific details.

[0075] The present invention provides a method for predicting multimodal Arctic climate change and analyzing its regulation mechanism. In the embodiment, HadCRUT5 (Morice et al. 2021) is used to observe SAT, and conclusions similar to HadCRUT5 are obtained from the GISTEMP (England et al. 2021; Hansen et al. 2010) data set. The observed sea surface heat flux and radiation data refer to JRA55 (Kobayashi et al. 2015) and NCEP-NCAR reanalysis data (Kalnay et al. 1996). ERSSTv5 is used to observe sea surface temperature, and Had1ey Centre Sea Level Pressure (A1lanand Ansell 2006) is used to observe sea level pressure. Due to the lack of early (before 1950) ocean observations and susceptibility to the uncertainty of observation instruments (Chanet al. 2019; Kennedy 2014), only observations and simulation results after 1950 are analyzed (such as Zhao et al. 2022). Unless otherwise specified, all observational data in this paper have been linearly detrended. The climate state is selected from 1961 to 1990, and climate anomalies are defined as deviations from the climate state.

[0076] FGOALS-g3 and CESM2 large sample collection

[0077] FGOALS-g3 is a climate model (Li et al. 2020), whose large sample set of historical simulations (1850-2014) and future climate projections (2015-2099) includes 110 samples (Lin et al. 2022). The horizontal resolution of the atmospheric / ocean latitude and longitude grid of FGOALS-g3 is approximately 2°×2.25° / 1°×0.76°. The initial value of each sample is 1850, and the difference between each sample is only the initial value. The historical external forcing uses the external forcing field of the CMIP6 standard (Eyring et al. 2016). The sample initial values ​​are selected from 110 different years from 900 to 2000 simulated by the pre-industrial control experiment. The original evaluation showed that the FGOALS-g3 large ensemble can reasonably simulate SAT changes similar to observations (Lin et al. 2022).

[0078] In order to verify the results of FGOALS-g3, the CESM2 historical large sample set (100 samples) was used (Rodgers et al. 2021). CESM2 is an Earth system model with an atmospheric / ocean latitude and longitude grid horizontal resolution of approximately 1.25°×0.9° / 1.125°×0.44°, and the historical forcing is the historical forcing of the CMIP6 standard. The CESM2 analysis results are similar to those of FGOALS-g3. Therefore, unless otherwise specified, the main text shows the FGOALS-g3 simulation results.

[0079] In order to verify the historical simulation results, the piContro1 large sample set was constructed based on the pre-industrial control experiment (piControl) of FGOALS-g3 and CESM2. Referring to previous studies (Maher et al. 201 8; Zhao et al. 2022), each sample in the piContro1 large sample set is truncated from piContro1 with a time length of 65 years. The piContro1 large sample set of FGOALS-g3 / CESM2 contains 43 / 89 samples. The appendix file provides a more detailed introduction to the model simulation. Unless otherwise emphasized, all analyzed model variables are deducted from the ensemble mean.

[0080] AMOC, AGI, NPO and PDO indices

[0081] The calculation of the meridional overturning stream function (MSF) is:

[0082]

[0083] where lev B is the depth of the ocean floor, lon w / lon eis the western / eastern boundary of the ocean, and v(time, lev, lat, lon) represents the northward flow velocity of the ocean. The model AMOC index is calculated as the maximum MSF value below 500 m depth in the Atlantic region north of 20°N ((Zhang and Wang 2013).

[0084] Many studies have quantified the Arctic poleward intensification from different perspectives (Bekryaev et al. 2010; Davy et al. 2018; Ono et al. 2022). The Arctic poleward intensification index used is based on Davy et al. (2018), which is defined as the area-weighted average of SAT in the Arctic (60°N-90°N) minus the area-weighted average of SAT in the Northern Hemisphere. This index represents the change in Arctic SAT relative to low-latitude SAT, so it is called the Arctic Gradient Index (AGI):

[0085] AGI=SAT 60N-90N -SAT 0-90N #(2)

[0086] SAT 60N-90N and SAT 0-90N is an area-weighted average. Other definitions of AA have phase transitions similar to the AGI index on the interdecadal time scale (see Davy et al. 2018).

[0087] Referring to the work of Furtado et al. (2012), the NPO index is defined as the normalized sea level anomaly in the mid-latitudes of the North Pacific (45°-65°N, 170°E-140°W) minus the normalized sea level pressure anomaly in the subtropics (20°-40°N, 170°E-140°W). Other NPO indices (Yeh et al. 2018) have obtained similar results. The FGOALS-g3 and CESM2 models can simulate the observed NPO mode well.

[0088] AMOC and NPO in phase (or out of phase) is defined as the standardized AMOC and NPO indices being in phase and their absolute values ​​being greater than 0.5 standard deviations. Standard deviation thresholds of different strengths yielded similar conclusions.

[0089] Referring to the work of Newman et al. (2016), the calculation of the PDO index has the following steps: first, the global mean sea temperature anomaly is deducted to eliminate the impact of global warming, and then the first EOF of the residual sea temperature anomaly in the North Pacific (20°-70°N, 110°E-110°W) is calculated as the PDO mode, and its corresponding standardized principal component time series is defined as the PDO index.

[0090] Meridional Atmospheric Heat Transport (AHT)

[0091] Referring to He et al. (2019), the calculation formula of the meridian AHT is:

[0092]

[0093] where φ is the latitude, φ s is the latitude of Antarctica, Flx s and Flx TOA They represent the net downward heat flux from the surface and the net incident radiation at the top of the atmosphere, respectively. E represents the radius of the Earth and λ represents the radial longitude. Ocean heat transport is a model diagnostic output. The climatological states of poleward ocean / atmosphere heat transport simulated by FGOALS-g3 and CESM2 are similar (not shown).

[0094] Energy budget of SAT

[0095] Since Arctic SAT is temporally correlated with surface temperature, the energy budget of SAT anomalies

[0096] (Lee et al. 2017) can be calculated as:

[0097]

[0098] On the left side of the equation, T air represents SAT, Δ represents annual anomaly, S u / S d Represents upward / downward shortwave radiation, F latent / F sensible represents the latent heat / sensible heat flux (positive anomaly corresponds to the downward direction), l d is the downward longwave radiation, R is the residual term, T is the surface temperature, σ is the Stefan-Boltzmann constant, and ε is the surface emissivity.

[0099] NPO regulation of the relationship between AMOC and Arctic SAT

[0100] To explore the potential impact of NPO. Figure 2 Figures a and b show the scatter plots of AMOC and AGI when the phases of AMOC and NPO are the same or opposite, respectively. As shown in the figure, when the phases of AMOC and NPO are opposite, the correlation coefficient between AMOC and AGI is larger (r = 0.77), while when the phases of AMOC and NPO are the same, the correlation coefficient between the two is smaller (r = 0.42). This indicates that the negative phase of NPO contributes more to the connection between AMOC and AGI. Based on historical simulations and the piControl ensemble, the same calculation method was used to re-examine this relationship ( Figure 2cf). The results also show that when AMOC and NPO are in opposite phases, the correlation coefficient between AMOC and AGI (r=0.61 in historical simulation and r=0.48 in piControl) is greater than the correlation coefficient when the two are in the same phase (r=0.35 in historical simulation and r=-0.02 in piControl). Both the observational and FGOALS-g3 large sample ensemble results show that the difference in the correlation coefficient between AMOC and AGI is significant when NPO and AMOC are in different phases and in the same phase (exceeding the 95% confidence level, Table 2). In addition, the simulation results of the CESM2 historical large sample ensemble and the corresponding piControl ensemble are similar to the FGOALS-g3 results (Table 2). In order to quantify the impact of NPO phase change on the association between AMOC and Arctic SAT on the interdecadal time scale, based on observations and simulations, we calculated the explained covariance (the square of the correlation coefficient, i.e., r) caused by NPO phase change. 2 ) and their changes (Table 3). The results show that the contribution of the phase changes of NPO and AMOC to the explained covariance between AMOC and AGI is 14%-41%, indicating that NPO plays an important role.

[0101] Table 2. Figure 2 Correlation coefficients (outside brackets) and their Fisher's r-to-z test values ​​(inside brackets). The values ​​in brackets are confidence intervals (95% significance level) estimated based on Fisher's r-to-z test. HIST(PI) stands for Historical (piControl) Large Sample Set.

[0102]

[0103]

[0104] Table 3. Square values ​​of correlation coefficients in Table 2 (r 2 ). HIST(PI) represents the historical (piControl) large sample set.

[0105]

[0106] The regulatory role of NPO on the impact of AMOC on SAT was further investigated. Figure 3 The spatial composite field of SAT and SLP anomalies under different combinations of AMOC and NP0, where NPO leads AMOC by three years. Figure 3 In the negative phase of AMOC ( Figure 3The above regions have significant cold SAT anomalies. This indicates that AMOC dominates the SAT changes on the interdecadal time scale in the Arctic. When the AMOC is in a certain phase, the NPO- phase is more conducive to warming in northern North America and the Arctic Ocean than the NPO+ phase ( Figure 3 In addition, when AMOC remains in the negative phase, the SAT amplitude over North America and the Arctic Ocean is in the NPO-phase ( Figure 3 f) is better than NPO+phase ( Figure 3 c) is larger (about 0.1℃). In the NPO-phase, the zonal averages of SAT and SLP north of 60°N are also larger than those in the NPO+phase. In the NPO-phase, the increase in TCWV is greater than that in the NPO+phase. This verifies that in different phases of AMOC, the NPO-phase promotes the warming of Arctic SAT through the meridional water vapor advection process. In addition, the impact of NPO is greater in the negative phase of AMOC.

[0107] Figure 1 Schematic diagram showing the physical processes of the combination of different NPO phases and AMOC types affecting Arctic SAT. (Left) NPO negative phase (NPO-) and cross-basin AMOC, (Right) NPO positive phase (NPO+) and Atlantic internal circulation AMOC.

[0108] Example 1

[0109] A method for multi-modal Arctic climate change prediction and its regulation mechanism analysis includes the following steps:

[0110] Step 1: Construct and validate a large sample dataset containing AMOC proxy data and Arctic SAT historical observation data, and ensure its reliability and accuracy through comparison and statistical tests;

[0111] Step 1.1 Collect and organize observational data, including proxy data of AMOC and historical observational data of Arctic SAT;

[0112] Step 1.2 Build a large sample dataset of CMIP6 models to ensure data quality and integrity;

[0113] Step 1.3: Compare observational data with model output to verify the reliability of the dataset;

[0114] Step 1.4 evaluates the consistency and accuracy of the dataset through statistical tests.

[0115] Step 2: Assess and identify the performance and bias of the models in simulating AMOC and Arctic SAT changes by simulating and comparing CMIP6 model outputs with observations;

[0116] Step 2.1 Use the constructed dataset to simulate the CMIP6 model and obtain the simulation results of AMOC and Arctic SAT;

[0117] Step 2.2 compares the simulation results with observational data to assess the model's ability to simulate AMOC and Arctic SAT changes;

[0118] Step 2.3 identifies the bias and uncertainty of the model during the simulation process through comparative analysis.

[0119] Step 3: Reveal the physical mechanism of the AMOC’s impact on Arctic SAT, establish quantitative relationships through diagnostic analysis, thermodynamic and kinetic studies, and calculate relevant physical parameters;

[0120] Step 3.1 Compare and analyze the observed and simulated data to diagnose the physical process of the AMOC's impact on the Arctic SAT, according to the calculation formula of the meridional overturning stream function (MSF):

[0121]

[0122] where lev B is the depth of the ocean floor, lon w is the western boundary of the ocean e is the eastern boundary of the ocean, v(time, lev, lat, lon) represents the northward flow velocity of the ocean, lev represents the vertical level, indicating the different depth layers in the ocean, lat represents latitude, lon represents longitude, dlon represents the differential element of longitude, which is used for integration operations, and dlev represents the differential element of vertical level, which is used for integration operations;

[0123] The AMOC index is calculated as the maximum MSF value below 500 m in the Atlantic region north of 20°N, according to the definition of the Arctic Gradient Index AGI:

[0124] AGI=SAT 60N-90N -SAT 0-90N

[0125] Among them, SAT 60N-90N and SAT 0-90N It is the area-weighted average, where 60N-90N represents the area from 60 degrees north latitude to 90 degrees north latitude, and 0-90N represents the area from the equator to 90 degrees north latitude (the North Pole);

[0126] Step 3.2: Through thermodynamic and kinetic analysis, the air-sea interaction and the heat transport mechanism of the atmosphere-ocean towards the poles are revealed. The calculation formula of atmospheric heat transport (AHT) is:

[0127]

[0128] where φ is the latitude, φ s is the latitude of Antarctica, Flx s and Flx TOA are the net downward heat flux from the surface and the net incident radiation at the top of the atmosphere, R E represents the radius of the Earth, λ represents the radial longitude, and ocean heat transport is the model diagnostic output;

[0129] Step 3.3 Establish a quantitative relationship between the physical process and the changes in AMOC and Arctic SAT. The energy budget of SAT is calculated as:

[0130]

[0131] On the left side of the equation, T air represents SAT, Δ represents annual anomaly, positive anomaly corresponds to downward direction, S v / S d Represents upward / downward shortwave radiation, F latent / Fs ensible represents latent heat / sensible heat flux, l d is the long-wave radiation, R is the residual term, T s is the surface temperature, σ is the Stefan-Boltzmann constant, and ε is the surface emissivity.

[0132] Step 4: Explore the regulatory role of NPO internal variability on the relationship between AMOC and Arctic SAT, determine its correlation through statistical analysis, and evaluate how NPO regulates Arctic SAT by affecting atmospheric and oceanic circulation;

[0133] Step 4.1 Analyze the impact of internal variability such as NPO on the relationship between AMOC and Arctic SAT. The NPO index is defined as the standardized sea level anomaly in the mid-latitude North Pacific (45°-65°N, 170°E-140°W) minus the standardized sea level pressure anomaly in the subtropics (20°-40°N, 170°E-140°W). The calculation formula of the NPO index is as follows:

[0134]

[0135] SLP mid : is the normalized sea level pressure anomaly in the mid-latitude North Pacific (45°-65°N, 170°E-140°W);

[0136] SLP sub : is the normalized sea level pressure anomaly in the subtropical region (20°-40°N, 170°E-140°W);

[0137] It represents the average of the difference between the sea level pressure anomalies in two regions;

[0138] σSLP mid and σSLP sub are the standard deviations of the sea level pressure anomalies in the two regions.

[0139] Step 4.2 Determine the correlation between NPO and AMOC and Arctic SAT through statistical analysis:

[0140]

[0141] r NPO，AMOC : represents the correlation coefficient between NPO and AMOC; r NPO，SAT : represents the correlation coefficient between NPO and SAT; NPO i 、AMOC i , SAT i are the NPO index, AMOC index and Arctic SAT value at the i-th time point, respectively; n: is the total number of time points;

[0142] Step 4.3 explores how the NPO modulates atmospheric and oceanic circulation and thus affects Arctic SAT.

[0143] Step 4.3.1. Evaluate the impact of NPO on meridional atmospheric heat transport (AHT):

[0144] ΔAHT=CΔv

[0145] where ΔAHT represents the change in meridional atmospheric heat transport, C is a proportionality constant that represents the sensitivity of meridional atmospheric heat transport to changes in ocean northward flow velocity, and Δv represents the change in northward flow velocity between the positive and negative phases of the NPO:

[0146]

[0147] and Represent the ocean northward flow speed during the positive and negative phases of NPO respectively.

[0148] 4.3.2 How does NPO affect the Atlantic Meridional Overturning Circulation (AMOC) through air-sea interaction, and thus affect the Arctic SAT? Here, the change of Δv may lead to the change of AMOC, which is expressed as:

[0149]

[0150] ΔAMOC reflects the change in AMOC intensity caused by the phase change of NPO. The change of Δv will affect ΔAMOC through the following channels:

[0151] If Δv>0, it indicates that the northward flow velocity increases, which will lead to more heat transport to the North Atlantic and enhance AMOC, that is, ΔAMOC>0;

[0152] If Δv < 0, it indicates that the northward flow velocity decreases, which will lead to a decrease in heat transport and weaken AMOC, that is, ΔAMOC < 0;

[0153] The NPO affects Arctic SAT by modulating atmospheric and oceanic circulations, through influencing the northward flow velocity Δv and the AMOC strength ΔAMOC.

[0154] Step 5: Based on the understanding of the physical mechanisms and the role of internal variability, simulate future climate change scenarios, adjust model parameters to accurately reflect the impact of AMOC and NPO on Arctic SAT, and input emission scenarios to predict the future trend of Arctic SAT and AMOC to evaluate the reliability of model predictions;

[0155] Step 5.1: Simulate future climate change scenarios by combining the revealed physical mechanisms and the regulation of internal variability with the compared models;

[0156] Step 5.1.1. When simulating future climate change, incorporate the AMOC impact mechanism on Arctic SAT revealed in Step 3 into the model, and correctly represent the changes in heat transport, salinity and ocean productivity in the model;

[0157] Step 5.1.2: Integrate the regulation of the internal variability of the NPO analyzed in step 4 into the simulation, including setting up parameterization schemes in the model that can simulate the changes in the NPO, which will feed back to the AMOC and the Arctic SAT;

[0158] Step 5.2: Adjust model parameters to better reflect the effects of observed physical processes and internal variability;

[0159] Step 5.2.1. Based on the results of Steps 3 and 4, adjust the model parameters to ensure that the model can better simulate the influence of AMOC and NPO on Arctic SAT, including adjusting the sea-air coupling parameters to reflect the transport of heat and salinity.

[0160] Step 5.2.2: Adjust the internal variability parameters in the model, including the representation of the NPO, to ensure that the model can simulate the changes in internal variability and its impact on the AMOC and Arctic SAT;

[0161] Step 5.3 inputs the future RCP or SSP emission scenario and simulates the changes in AMOC and Arctic SAT over the next few decades to centuries.

[0162] Step 5.4 performs statistical analysis on the simulation results to extract the predicted trends of future Arctic SAT and AMOC, extract the future trends of AMOC and Arctic SAT, and compare them with the findings in steps 3 and 4 to assess the reliability of the model predictions.

Claims

1. A multi-modal Arctic climate change prediction and regulation mechanism analysis method, characterized in that: The following steps are involved: Step 1: Construct and validate a large sample dataset containing AMOC proxy data and Arctic SAT historical observation data, and ensure its reliability and accuracy through comparison and statistical tests; Step 2: Assess and identify the performance and bias of the models in simulating AMOC and Arctic SAT changes by simulating and comparing CMIP6 model outputs with observations; Step 3: Reveal the physical mechanism of the AMOC’s impact on Arctic SAT, establish quantitative relationships through diagnostic analysis, thermodynamic and kinetic studies, and calculate relevant physical parameters; Step 4: Explore the regulatory role of NPO internal variability on the relationship between AMOC and Arctic SAT, determine its correlation through statistical analysis, and evaluate how NPO regulates Arctic SAT by affecting atmospheric and oceanic circulation; Step 5: Based on the understanding of the physical mechanisms and the regulating role of internal variability, simulate future climate change scenarios, adjust model parameters to accurately reflect the impact of AMOC and NPO on Arctic SAT, and input emission scenarios to predict future trends in Arctic SAT and AMOC to evaluate the reliability of model predictions.

2. A multi-modal Arctic climate change prediction and regulation mechanism analysis method according to claim 1, characterized in that: Step 1 includes the following steps: Step 1.1: Collect and organize observational data, including proxy data of AMOC and historical observational data of Arctic SAT; Step 1.2: Build a large sample dataset of CMIP6 models to ensure data quality and integrity; Step 1.3: Compare observational data with model output to verify the reliability of the dataset; Step 1.4: Evaluate the consistency and accuracy of the dataset through statistical tests.

3. A multi-modal Arctic climate change prediction and regulation mechanism analysis method according to claim 1, characterized in that: Step 2 includes the following steps: Step 2.1: Use the constructed dataset to simulate the CMIP6 model and obtain the simulation results of AMOC and Arctic SAT; Step 2.2: Compare the simulation results with observational data to assess the model's ability to simulate AMOC and Arctic SAT changes; Step 2.3: Identify the deviations and uncertainties of the model during the simulation process through comparative analysis.

4. A multi-modal Arctic climate change prediction and regulation mechanism analysis method according to claim 1, characterized in that: Step 3 includes the following steps: Step 3.1: Compare and analyze the observational and simulated data to diagnose the physical process of the AMOC's influence on the Arctic SAT, according to the calculation formula of the meridional overturning stream function (MSF): where lev B is the depth of the ocean floor, lon w is the western boundary of the ocean e is the eastern boundary of the ocean, v(time, lev, lat, lon) represents the northward flow velocity of the ocean, lev represents the vertical level, indicating the different depth layers in the ocean, 1at represents latitude, 1on represents longitude, and d1o n The microelement representing the longitude is used for integral calculation, and d1ev represents the microelement at the vertical level, which is used for integral calculation; The AMOC index is calculated as the maximum MSF value below 500 m in the Atlantic region north of 20°N, according to the definition of the Arctic Gradient Index AGI: AGI=SAT 60N-90N -VILLAGE 0-90N Among them, SAT 60N-90N and SAT 0-90N It is the area-weighted average, where 60N-90N represents the area from 60 degrees north latitude to 90 degrees north latitude, and 0-90N represents the area from the equator to 90 degrees north latitude (the North Pole); Step 3.2: Through thermodynamic and kinetic analysis, the air-sea interaction and the heat transport mechanism of the atmosphere-ocean towards the poles are revealed. The calculation formula of atmospheric heat transport (AHT) is: where φ is the latitude, φ s is the latitude of Antarctica, Flx s and Flx TOA are the net downward heat flux from the surface and the net incident radiation at the top of the atmosphere, R E represents the radius of the Earth, λ represents the radial longitude, and ocean heat transport is the model diagnostic output; Step 3.3: Establish a quantitative relationship between physical processes and changes in AMOC and Arctic SAT. The energy budget of SAT is calculated as: On the left side of the equation, T air represents SAT, Δ represents annual anomaly, positive anomaly corresponds to downward direction, S u / S d Represents upward / downward shortwave radiation, F latent / F sensible represents latent heat / sensible heat flux, l d is the long-wave radiation, R is the residual term, T s is the surface temperature, σ is the Stefan-Boltzmann constant, and ε is the surface emissivity.

5. The method for multi-modal Arctic climate change prediction and its regulation mechanism analysis according to claim 1, characterized in that: Step 4 includes the following steps: Step 4.1: Analyze the impact of internal variability such as NPO on the relationship between AMOC and Arctic SAT. The NPO index is defined as the normalized sea level anomaly in the mid-latitudes of the North Pacific (45°-65°N, 170°E-140°W) minus the normalized sea level pressure anomaly in the subtropics (20°-40°N, 170°E-140°W). The calculation formula of the NPO index is as follows: SLP mid : is the normalized sea level pressure anomaly in the mid-latitude North Pacific (45°-65°N, 170°E-140°W); SLP sub : is the normalized sea level pressure anomaly in the subtropical region (20°-40°N, 170°E-140°W); It represents the average of the difference between the sea level pressure anomalies in two regions; σSLP mid and σSLP sub are the standard deviations of the sea level pressure anomalies in the two regions; Step 4.2: Determine the correlation between NPO and AMOC and Arctic SAT through statistical analysis: r NPO,AMOC : represents the correlation coefficient between NPO and AMOC; r NPO,SAT : represents the correlation coefficient between NPO and SAT; NPO i 、AMOC i , SAT i are the NPO index, AMOC index and Arctic SAT value at the i-th time point, respectively; n: is the total number of time points; Step 4.3: Explore how NPO regulates atmospheric and oceanic circulation and thus affects Arctic SAT.

6. A multi-modal Arctic climate change prediction and regulation mechanism analysis method according to claim 5, characterized in that: Step 4.3 includes the following steps: Step 4.3.

1. Evaluate the impact of NPO on meridional atmospheric heat transport (AHT): ΔAHT=CΔ v where ΔAHT represents the change in meridional atmospheric heat transport, C is a proportionality constant that represents the sensitivity of meridional atmospheric heat transport to changes in ocean northward flow velocity, and Δv represents the change in northward flow velocity between the positive and negative phases of the NPO: and They represent the ocean northward flow speed during the positive and negative phases of NPO, respectively; Step 4.3.2: How does NPO affect the Atlantic Meridional Overturning Circulation (AMOC) through air-sea interaction, and thus affect the Arctic SAT? Here, the change of Δv may lead to the change of AMOC, which is expressed as: ΔAMOC reflects the change in AMOC intensity caused by the phase change of NPO. The change of Δv will affect ΔAMOC through the following channels: If Δv>0, it indicates that the northward flow velocity increases, which will lead to more heat transport to the North Atlantic and enhance AMOC, that is, ΔAMOC>0; If Δv < 0, it indicates that the northward flow velocity decreases, which will lead to a decrease in heat transport and weaken AMOC, that is, ΔAMOC < 0; The NPO affects Arctic SAT by modulating atmospheric and oceanic circulations, through influencing the northward flow velocity Δv and the AMOC strength ΔAMOC.

7. The method for multi-modal Arctic climate change prediction and its regulation mechanism analysis according to claim 1, characterized in that: Step 5 includes the following steps: Step 5.1: Simulate future climate change scenarios by combining the revealed physical mechanisms and the regulation of internal variability with the compared models; Step 5.2: Adjust model parameters to better reflect the effects of observed physical processes and internal variability; Step 5.3: Input the future RCP or SSP emission scenario and simulate the changes of AMOC and Arctic SAT in the next few decades to hundreds of years; Step 5.4: Perform statistical analysis on the simulation results to extract the predicted trends of future Arctic SAT and AMOC, extract the future trends of AMOC and Arctic SAT, and compare them with the findings in Steps 3 and 4 to assess the reliability of the model predictions.

8. A multi-modal Arctic climate change prediction and regulation mechanism analysis method according to claim 7, characterized in that: Step 5.1 includes the following steps: Step 5.1.

1. When simulating future climate change, incorporate the AMOC impact mechanism on Arctic SAT revealed in Step 3 into the model, and correctly represent the changes in heat transport, salinity and ocean productivity in the model; Step 5.1.2: Integrate the regulatory role of the NPO internal variability analyzed in step 4 into the simulation, including setting up a parameterization scheme in the model that can simulate NPO changes, which will feed back to AMOC and Arctic SAT.

9. A multi-modal Arctic climate change prediction and regulation mechanism analysis method according to claim 7, characterized in that: Step 5.2 includes the following steps: Step 5.2.

1. Based on the results of Steps 3 and 4, adjust the model parameters to ensure that the model can better simulate the influence of AMOC and NPO on Arctic SAT, including adjusting the sea-air coupling parameters to reflect the transport of heat and salinity. Step 5.2.2: Adjust the internal variability parameters in the model, including the representation of the NPO, to ensure that the model can simulate the changes in internal variability and its impact on the AMOC and Arctic SAT.

Images