Empirical Orthogonal Function Decomposition Research Articles

An EOF-Based Global Plasmaspheric Electron Content Model and Its Potential Role in Vertical-Slant TEC Conversion

Topside total electron content (TEC) data measured by COSMIC/FORMAT-3 during 2008 and 2016 were used to analyze and model the global plasmaspheric electron content (PEC) above 800 km with the help of the empirical orthogonal function (EOF) analysis method, and the potential role of the proposed PEC model in helping Global Navigation Satellite System (GNSS) users derive accurate slant TEC (STEC) from existing high-precision vertical TEC (VTEC) products was validated. A uniform gridded PEC dataset was first obtained using the spherical harmonic regression method, and then, it was decomposed into EOF basis modes. The first four major EOF modes contributed more than 99% of the total variance. They captured the pronounced latitudinal gradient, longitudinal differences, hemispherical differences, diurnal and seasonal variations, and the solar activity dependency of global PEC. A second-layer EOF decomposition was conducted for the spatial pattern and amplitude coefficients of the first-layer EOF modes, and an empirical PEC model was constructed by fitting the second-layer basis functions related to latitude, longitude, local time, season, and solar flux. The PEC model was designed to be driven by whether solar proxy or parameters derived from the Klobuchar model meet the real-time requirements. The validation of the results demonstrated that the proposed PEC model could accurately simulate the major spatiotemporal patterns of global PEC, with a root-mean-square (RMS) error of 1.53 and 2.24 TECU, improvements of 40.70% and 51.74% compared with NeQuick2 model in 2009 and 2014, respectively. Finally, the proposed PEC model was applied to conduct a vertical-slant TEC conversion experiment with high-precision Global Ionospheric Maps (GIMs) and dual-frequency carrier phase observables of more than 400 globally distributed GNSS sites. The results of the differential STEC (dSTEC) analysis demonstrated the effectiveness of the proposed PEC model in aiding precise vertical-slant TEC conversion. It improved by 18.52% in dSTEC RMS on a global scale and performed better in 90.20% of the testing days compared with the commonly used single-layer mapping function.

Attribution analysis and forecast of salinity intrusion in the Modaomen estuary of the Pearl River Delta

Attribution analysis and forecast of salinity intrusion in the Modaomen estuary of the Pearl River Delta

Analysis on characteristics of extreme precipitation indices and atmospheric circulation in Northern Shanxi

This article utilizes daily precipitation data from 28 national meteorological stations in northern Shanxi Province spanning from 1972 to 2020, and the US NCEP/NCAR monthly average reanalysis and ERA5 monthly average reanalysis data. The study employs techniques such as empirical orthogonal function (EOF) decomposition, Mann-Kendall mutation and other methods to investigate the spatiotemporal distribution of extreme precipitation index in northern Shanxi and their correlation with atmospheric circulation. The research results show that: the absolute index, relative index, intensity index and sustained dry period index (CDD) in the continuous index appear from southwest to northeast. The spatial distribution characteristics of the central region decrease, while the continuous wet period (CWD) decreases from the central to the east and west. The three indices Rx1day, Rx5day, and CWD mutated in 1978, 1975, and 1983 respectively, and other extreme precipitation indices all appeared in a sudden change from a low-value period to a high-value period occurred around 2010. In the high-value years of the summer extreme precipitation index, there is a significant negative anomaly in the height field in the mid-high latitude regions of Eurasia. Northern Shanxi is controlled by a broad low-pressure trough in the Lake Baikal area. Water vapor transported via the east, west, and south routes converges in the northern Shanxi region and encounters cold air from the north. There is a strong upward motion anomaly at 500 hPa in the troposphere, and the dynamic conditions of upper-level divergence and lower-level convergence lead to more summer extreme precipitation in the northern Shanxi region. Conversely, in the low-value years of the summer extreme precipitation index, northern Shanxi is affected by a strong high-pressure ridge north of Lake Baikal. There is a downward motion anomaly at 500 hPa, and the northern Shanxi region lacks water vapor. The cold and warm air cannot converge, and both the water vapor conditions and dynamic conditions are poor, which is not conducive to the production of extreme precipitation in northern Shanxi.

Climate Variability of Air Temperature and Its Warming Trends in the Arabian Gulf

AbstractOver a 41-year period (1979–2019), the climate of the Arabian Gulf (Gulf) exhibited notable variability. This variation, influenced by mesoscale and globalscale meteorological and climatic determinants, showed temperature fluctuations, predominantly in the Northern Gulf (N Gulf). The N Gulf recorded an average temperature of 25.9 °C, a noticeable 3 °C higher than the broader Gulf average. This period marked extreme temperature records, with a peak at 53.8 °C in July 2016, paralleling the highest known temperatures in Asia. Data reveals a warming trend (0.1–0.6 °C per decade) in the region, with the N Gulf’s rise surpassing global trends. The years 2014–2019 were notably warmer, impacting overall trends, and 1997 was the transitional year shifting from cooler to warmer climates. The N Gulf’s seasonal dynamics are also transforming, with extended summers and shortened winters. In this study, we employed empirical orthogonal function (EOF) decomposition analysis to examine the interannual to multidecadal fluctuations in air temperature within the Gulf from 1979 to 2019, using hourly air temperature data from the Watch Forcing Dataset (WFDE5) generated from the fifth generation (ERA5) European Centre for Medium-Range Weather Fore-casts (ECMWF). Through this method, we discerned three principal EOF modes, which together accounted for 82% of the air temperature variability in the Gulf, highlighting external influences such as North Atlantic Oscillation (NAO), El Nino-Southern Oscillation (ENSO), and Atlantic Multidecadal Oscillation (AMO). The data for these external influences were sourced from the National Oceanic and Atmospheric Administration (NOAA) Physical Sciences Laboratory. These findings stress the pressing need for detailed climate studies and mitigation strategies for a rapidly warming Gulf region, especially the N Gulf.

Characteristics of Dry and Wet Changes and Future Trends in the Tarim River Basin Based on the Standardized Precipitation Evapotranspiration Index

Global changes in drought and wetness and their future trends in arid regions have recently become a major focus of research attention. The Tarim River Basin (TRB) in Xinjiang, China, is among the most climate-sensitive regions in the world. This study uses data from the past 60 years (1962–2021) to analyze the spatial and temporal features of drought and wetness conditions in the TRB, calculating the Standardized Precipitation Evapotranspiration Index (SPEI). Trend detection for SPEI is performed using the BEAST mutation test, identification of drought events using the theory of operations, and spatial and temporal analyses of dry and wet changes using Empirical Orthogonal Function (EOF) decomposition. Additionally, the CMIP6 dataset is used to estimate future changes. The study results indicate the following: (1) From 1962 to 1998, the TRB exhibited a “warm and wet” trend that suddenly shifted from “wet-to-dry” in 1998 and subsequently transitioned to a pronounced “warm and dry” trend. (2) After the “wet-to-dry” shift, the frequency of drought events noticeably increased. The northern section of the basin witnessed more frequent drought events, albeit with lower severity, while the southern part had fewer occurrences but with higher severity. The spatial distribution of drought event frequency and severity is inconsistent. (3) The EOF decomposition results for SPEI-variable fields at 1-, 3-, and 6-month time scales show that the cumulative variance contribution rate of the first three principal spatial modal feature vectors exceeds 70%. The spatial distribution of the modes includes a consistent pattern across the entire basin, a north–south opposite pattern, and an east–west opposite pattern. (4) The future trend of drought in the TRB is expected to intensify, manifesting a spatial pattern characterized by dryness in the middle of the basin and wetness around the periphery. These research findings can provide support for decisions addressing regional drought risks.

Temporal and Spatial Surface Heat Source Variation in the Gurbantunggut Desert from 1950 to 2021

Based on data from the Gurbantunggut Desert, the largest fixed/semi-fixed desert in China, and ERA5-Land reanalysis data, the long-term variations and spatial surface heat source (SHS) differences in the Gurbantunggut Desert are discussed herein. The results show the following: (1) The hourly SHS at the Kelameili station during the 2013–2021 period was a weak heat source at night; contrastingly, it was a strong heat source during the day. The duration of the hourly SHS increased gradually from January to July, but it decreased gradually from July to December. The daily SHS showed obvious seasonal variation, reaching the maximum in summer and the minimum in winter. The ERA5-Land reanalysis can reproduce all the variation characteristics of the SHS well. (2) The climatology (i.e., multi-year mean) of the monthly SHS intensity was lower than 50 W/m2 during the January–March and September–December periods in the Gurbantunggut Desert, indicating a weak heat source. On the other hand, the climatology recorded in April–August was higher than 50 W/m2, with a strong heat source. From the perspective of spatial distribution, the eastern and western regions of the Gurbantunggut Desert show strong heat sources, while the central region shows weak heat sources. The spatial distribution of the first and second modes of the empirical orthogonal function (EOF) decomposition reflected the consistent spatial variability and a north–south (or east–west) polarity variation of the monthly SHS in the Gurbantunggut Desert, respectively. (3) The yearly SHS showed negative anomalies during the 1950–1954, 1964–1982 and 2004–2015 periods, and positive anomalies during the 1955–1963, 1983–2003 and 2016–2021 periods in the Gurbantunggut Desert. Additionally, the time series of the SHS anomalies was positively correlated with the Interdecadal Pacific Oscillation (IPO) index. During the negative IPO phase, the yearly SHS showed a negative anomaly in the Gurbantunggut Desert, while the yearly SHS showed a positive anomaly during the positive IPO phase in most regions of the Gurbantunggut Desert.

On the Vertical Structure of Deep-Ocean Subinertial Variability

Abstract The vertical structure of subinertial variability is examined using full-depth horizontal velocity and vertical isopycnal displacement observations derived from the Ocean Observatory Initiative (OOI). Vertical profiles on time scales between 100 h and 1 yr or longer are characterized through empirical orthogonal function decomposition and qualitatively compared with theoretical modal predictions for the cases of flat, sloping, and rough bathymetry. OOI observations were obtained from mooring clusters at four deep-ocean sites: Argentine Basin, Southern Ocean, Station Papa, and Irminger Sea. Because no single OOI mooring in these arrays provides temperature, salinity, and horizontal velocity information over the full water column, sensor observations from two or more moorings are combined. Depths greater than ∼150–300 m were sampled by McLane moored profilers; in three of the four cases, two profilers were utilized on the moorings. Because of instrument failures on the deployments examined here, only ∼2 yr of full-ocean-depth observations are available from three of the four sites and some 3+ yr from the other. Results from the OOI “global” sites are contrasted with a parallel analysis of 3.5 yr of observations about the axis of the Gulf Stream where much of the subinertial variability is associated with stream meandering past the moorings. Looking across the observations, no universal vertical structure is found that characterizes the subinertial variability at the five sites examined; regional bathymetry, stratification, baroclinicity, nonlinearity, and the forcing (both local and remote) likely all play a role in shaping the vertical structure of the subinertial variability in individual ocean regions.

Enhancing the robustness of ocean sound speed profile representation via interpretable deep matrix decomposition.

Developing an effective and robust representation model for ocean sound speed profiles (SSPs) is crucial for numerous ocean acoustic applications. However, the performance of existing sound speed profile (SSP) representation methods, such as empirical orthogonal function and K-singular value decomposition, heavily relies on the number of selected basis functions. This could lead to overfitting of noise, as these methods are unable to distinguish between signals and noise during the basis function learning process. To overcome these limitations and effectively learn a large number of basis functions with strong representation power from potentially noisy SSP data, we propose a novel algorithm called deep matrix decomposition (deep MD). This algorithm utilizes untrained deep neural networks as priors to reject noise within the interpretable matrix decomposition framework. To achieve optimal performance with deep MD, we propose a stopping strategy based on the rank estimate to determine the termination epoch. Experimental results using real-life datasets demonstrate that deep MD is robust against various types of noise and outperforms traditional SSP representation methods in terms of SSP reconstruction and characterizing the transmission loss in underwater acoustics.

Analysis of the relationship between dust aerosol and precipitation in spring over East Asia using EOF and SVD methods

This study utilized the Empirical Orthogonal Function (EOF) and Singular Value Decomposition (SVD) methods to investigate the spatial and temporal patterns and trends of dust aerosol and precipitation, and to identify the coupled modes between them. The research employed MODIS and CALIPSO retrieved dust aerosol optical depth (DAOD) data to represent dust aerosol information and CMORPH data to provide precipitation information. The results indicated that specific dust source regions were associated with the primary modes of spring dust in East Asia, while atmospheric circulation and land-sea monsoon were closely related to the primary modes of spring precipitation. Additionally, the study revealed that the impact of dust aerosol on precipitation varied based on the source region within the coupled modes. The first coupled mode, with dust sources in the Mongolian Gobi and Taklamakan Desert, demonstrated a pattern of increased dust aerosol and reduced precipitation in most regions. The second coupled mode, with dust sources focused on the Mongolian Gobi, exhibited a consistent pattern of increased dust aerosol and a significant increase in precipitation in northern China. This study highlights the significance of considering dust source regions when examining the relationship between dust aerosol and precipitation, providing new insights into the potential impact of dust aerosol on precipitation in East Asia during the spring.

Constructing a Regional Ionospheric TEC Model in China with Empirical Orthogonal Function and Dense GNSS Observation

Using Global Navigation Satellite Systems (GNSS) observation data for developing a high-precision ionospheric Total Electron Content (TEC) model is one of the essential subjects in ionospheric physics research and the application of satellite navigation correction. In this study, we integrate the Empirical Orthogonal Function (EOF) method with the TEC data provided by the Center for Orbit Determination in Europe (CODE), and observed by the dense GNSS receivers operated by the Crustal Movement Observation Network of China (CMONOC) to construct a regional ionospheric TEC model over China. The EOF analysis of CODE TEC in China from 1998 to 2010 shows that the first-order EOF component accounts for 90.3813% of the total variation of the ionospheric TEC in China. Meanwhile, the average value of CODE TEC is consistent with the spatial and temporal distribution characteristics of the first-order EOF base function, which mainly reflects the latitude and diurnal variations of TEC in China. The first-order coefficient after EOF decomposition shows an obvious 11-year period and semi-annual variations. The maximum amplitude of semi-annual variation mainly appears in March and October, which is closely associated with the variation in geographical longitude, the semi-annual change of the low-latitude electric field, and the ionospheric fountain effect. The second-order coefficient has an evident annual variation, the minimum amplitude mainly occurs in March, August, and September, and the amplitude values in the high solar activity years are more significant than those in the low solar activity years. The third-order coefficient mainly shows the characteristics of annual variation, and the fourth-order coefficient shows the noticeable semi-annual and annual variations. The third and fourth-order coefficients are both modulated by the solar activity index F10.7. The ionospheric TEC model in China, driven by CMONOC real-time GNSS observation data, can better reflect the latitude, local time and seasonal variation characteristics of ionospheric TEC over China. In particular, it can clearly show the spring and autumn asymmetry of ionospheric TEC in the low latitudes. The root mean square error of the absolute error between the model and the actual observation is mainly distributed around 2.45 TECU (1 TECU = 1016 electrons/m2). The values of the TEC model constructed in this study are closer to the actual observed values than those of the CODE TEC in China.

Extended-Range Forecast of Regional Persistent Extreme Cold Events Based on Deep Learning

Regional persistent extreme cold events are meteorological disasters that cause serious harm to people’s lives and production; however, they are very difficult to predict. Low-temperature weather systems and their effects have a significant low-frequency oscillation period (10–20 d and 30–60 d). This paper uses deep learning to analyze the extended-range time scale and predict regional persistent extreme cold events. The dominant low-frequency oscillation components of cold events are obtained via wavelet transform and Butterworth filtering. The low-frequency oscillation component is decomposed via empirical orthogonal function decomposition to extract the main spatial mode and time coefficient. A convolutional neural network is used to establish the correlation between large-scale circulations and the time coefficient of the low-frequency oscillation component of the lowest temperature. The proposed deep learning model exhibits good prediction accuracy for regional persistent extreme cold events with low-frequency oscillations.

Regional forecasting of PM2.5 concentrations: A novel model based on the empirical orthogonal function analysis and Nadaraya–Watson kernel regression estimator

Predicting regional air pollutant concentrations quickly and accurately is still a challenge when meteorological data that affect pollutant concentrations are not available. To avoid modeling single points, which is tedious, it is necessary to explore more efficient and reliable regional methods with fewer input data. In this study, the principle and procedure of a new hybrid model, EOF-NWKRE, based on empirical orthogonal function (EOF) decomposition and the N–W kernel regression estimator (NWKRE) are proposed, then, the PM2.5 concentrations of 13 cities in Jiangsu Province is simulated by the EOF-NWKRE. The results show that the average prediction accuracy of EOF-NWKRE model is 74.38%, with more than 92% of the cumulative variance. The model is completely self-driven by PM2.5 data without covariate data input, which reduces the computational burden, improves the computational efficiency, and results in better prediction accuracy. It is suitable for the prediction of pollutant concentrations in a large area.

Modeling of Precipitation Prediction Based on Causal Analysis and Machine Learning

The factors influencing precipitation in western China are quite complex, which increases the difficulty in determining accurate predictors. Hence, this paper models the monthly measured precipitation data from 240 meteorological stations in mainland China and the precipitation data from the European Centre for Medium-Range Weather Forecasts and the National Climate Centre and employs 88 atmospheric circulation indices to develop a precipitation prediction scheme. Specifically, a high-quality grid-point field is created by fusing and revising the precipitation data from multiple sources. This field is combined with the Empirical Orthogonal Function decomposition and the causal information flow. Next, the best predictors are screened through Empirical Orthogonal Function decomposition and causal information flow, and a data-driven precipitation prediction model is established using a Back Propagation Neural Network and a Random Forest algorithm to conduct the 1-month, 3-month, and 6-month precipitation predictions. The results show that: The machine learning-based precipitation prediction model has high accuracy and is generally able to predict the precipitation trend in the western region better. The Random Forest algorithm significantly outperforms the Back Propagation Neural Network algorithm in the prediction of the three starting times, and the prediction ability of both models gradually decreases as the starting time increases. Compared with the 2022 flood season prediction scores of the Institute of Atmospheric Sciences of the Chinese Academy of Sciences, the model improves the prediction of 1-month and 3-month precipitation in the western region and provides a new idea for the short-term climate prediction of precipitation in western China.

Spatiotemporal changes of the aridity index in Xinjiang over the past 60 years

The Aridity Index (AI), calculated from 99 homogeneous meteorological stations from 1961 to 2020, was used to analyze the variation of dryness/wetness climate change in Xinjiang in the past 60 years. The results show that the annual AI in Xinjiang has demonstrated a significant decreasing trend over the past 60 years; that is, the climate in Xinjiang, especially southern Xinjiang, has shown a relatively wetting trend. The interdecadal variations from the 1960s to 2010s explained that the total station ratio of arid and extremely arid areas showed a decreasing trend. In contrast, the semi-arid, humid, and semi-humid areas showed increasing trends, especially since the beginning of the twenty-first century. The interdecadal spatial evolution characteristics show that Xinjiang’s dryness/wetness climate reversed in the 1990s. An abrupt change in the annual AI occurred in 1986, after which the study region was continuously wetting. The first empirical orthogonal function (EOF) decomposition mode is consistent in Xinjiang; that is, the climate in Xinjiang is generally dry or wet, and the intensity of this change varies among different regions. The second mode reflects the opposite spatial distribution characteristics of the dryness/wetness climate in southern and northern Xinjiang with the Tianshan Mountains. Dryness/wetness climate changes in Xinjiang mainly exhibit 2.5-year and 6-year oscillation periods, between which the 6-year period is even more significant.

Seasonal prediction of summer extreme precipitation frequencies over Southwest China based on machine learning

It is challenging to predict extreme precipitation over regions with complex terrain, such as Southwest China (SWC). This study develops a statistical seasonal prediction scheme for summer extreme precipitation frequencies (EPFs) over SWC. In this scheme, the year-to-year increment method is first applied to the SWC summer EPFs and boundary forcings in preceding seasons. The first six leading modes of SWC summer EPFs are extracted by empirical orthogonal function (EOF) decomposition. The relationships between the six modes and preceding boundary forcings are investigated. According to these relationships, physics-based predictors for the principal components (PCs) of the modes are selected. The first two modes are taken as examples to analyze the mechanisms by which the predictors influence SWC summer EPFs. Two machine learning (ML) models, random forest and extreme gradient boosting, are trained using the data during 1980–2011. The ML models that perform best for each PC on the training data are used to be independently tested on data during 2012–2021. The predicted EPF patterns can be reconstructed by predicted PCs and observed EOFs. Prediction results indicate that the optimal ML models can well capture the variations in the six modes and spatial patterns of SWC EPFs and outperform the traditional stepwise regression models. The prediction skills for SWC summer EPF anomaly patterns show year-to-year variations which depend on the prediction results of the PCs. This scheme may help the seasonal predictions of SWC summer EPFs.

A Statistical Prediction Model for Summer Precipitation in China Based on TSD Method and EOF Modes’ Time Coefficients

It is a challenge to improve the skill of seasonal precipitation prediction, because there are many factors affecting summer precipitation in China, which are found on different time scales and have complex interactions with each other. For these reasons, we establish a prediction model with the time-scale decomposition (TSD) method to investigate whether the TSD has an improving effect on the prediction skill of summer precipitation in China. Using this statistical model, the predictors and predictands will be separated into interannual and interdecadal time scales, after which Empirical Orthogonal Function (EOF) decomposition is performed on these two components, and their time coefficients are predicted, respectively. The hindcast cross-validation results show that the model without TSD has prediction skills only in some regions of East China and South China. Compared with the model without TSD, surprisingly, the model with TSD can significantly improve the prediction performance in more regions in China, such as Xinjiang Province and Northeast China. The anomaly correlation coefficients (ACC) between hindcast precipitation with TSD and observation are higher in most years than that without TSD. The results of the independent sample test show that the forecast model with TSD has a stable and gratifying prediction skill, and the averaged ACC is increased by more than 0.1.

Intraseasonal variations in winter surface air temperature over China and its prediction skill in ECMWF System5

AbstractBased on the daily average temperature data of 2374 stations in China from 1993 to 2022 and the monthly average temperature data of European Centre for Medium‐Range Weather Forecasts (ECMWF) System5 in winter, this article analyses the intraseasonal variations of winter surface air temperature (SAT) over China by using seasonal empirical orthogonal function decomposition (S‐EOF) and compare the prediction skill differences of the first three dominant modes in ECMWF System5. The first mode (S‐EOF1) is characterized by out‐of‐phase changes in SAT anomalies (SATA) between December and January. Both the Siberian high (SH) and the central Pacific El Niño‐Southern Oscillation (CP ENSO) can impact S‐EOF1. The second mode (S‐EOF2) is characterized by consistency in warm anomalies over the Northeast and Qinghai‐Tibet Plateau's variation, while the other areas are characterized by the out‐of‐phase change in SATA. S‐EOF2 is closely related to Arctic oscillation (AO). The third mode is characterized by alternating changes over 3 months in winter, which is related to the Pacific meridional mode (PMM) and the North Atlantic SST tripole (NAT). Comparing the ECMWF System5 prediction skill for the first three leading modes for the intraseasonal variations in winter SAT, the prediction skill is not high. The model did not predict the extent of warm anomalies and the intraseasonal fluctuation in S‐EOF1, which may be due to the poor prediction skills for the range and intensity of the CP ENSO and the SH. The second mode has the highest prediction skill among the three modes. For the prediction skill of AO is good. The model can capture the intraseasonal reverse SATA between December and January but did not capture the reverse SATAs between January and February. Because the model does not predict obvious PMM and NAT well.

The role of synoptic eddies on remote response of Eurasian atmosphere to the tropical forcing associated with 2021/2022 winter

AbstractIn this paper, the role of eddy feedback in the formation and excitation of stream function anomaly pattern over Eurasian atmosphere (SFA) during 2021/2022 December–February (DJF) season is examined in terms of a low‐order linear model with a synoptic eddy and low frequency (SELF) feedback empirical closure onto the empirical orthogonal function decomposition bases (EOF‐bases). The results show that the SFA can be produced roughly and the SELF feedback plays important role in shaping the SFA during 2021/2022 DJF over Eurasian atmosphere. The SFA can be produced roughly owing to the internal action of the linear dynamical system. However, the importance of the external forcing on the formation of SFA depends on the role of the eddy feedback to a different degree. For the North Atlantic oscillation (NAO)‐like pattern with the wave train at high latitudes, the external forcing in the tropical Atlantic contributes greater with eddy feedback. While for the wave train at lower latitudes, the forcing will have little influence with the increase of the SELF feedback. A set of experiments with isolated forcing show that the forcing over tropical Atlantic Ocean can induce the wave train patterns alike the SFA, but the forcing to the east over South Asian to western Indian Ocean also contributes for shaping the SFA, especially the anomalous cyclone over Tibetan Plateau (TP). The combined effects of the response to both the tropical Atlantic Ocean and the eastern forcing over South Asian to western Indian Ocean contributes much to the formation of the two wave trains over Eurasia, thereby a negative stream function centre over TP, which might be the main cause for the excessive DJF rainfall over South China during 2021/2022. The diagnosis of the role of synoptic eddy on the atmospheric response to the external forcing for specific events can be helpful for understanding the causes for the formation of extreme climate events deeply.

Study of Time-Frequency Domain Characteristics of the Total Column Ozone in China Based on Wavelet Analysis

Ozone is a very important trace gas in the atmosphere, it is like a “double-edged sword”. Because the ozone in the stratosphere can effectively help the earth’s organisms to avoid the sun’s ultraviolet radiation damage, the ozone near the ground causes pollution. Therefore, it is essential to explore the time-frequency domain variation characteristics of total column ozone and have a better understanding of its cyclic variation. In this paper, based on the monthly scale dataset of total column ozone (TCO) (September 2002 to February 2023) from Atmospheric Infrared Sounder (AIRS) carried by NASA’s Aqua satellite, linear regression, coefficient of variation, Mann-Kendall (M-K) mutation tests, wavelet analysis, and empirical orthogonal function decomposition (EOF) analysis were used to analyze the variation characteristics of the TCO in China from the perspectives of time domain, frequency domain, and spatial characteristics. Finally, this study predicted the future of TCO data based on the seasonal autoregressive integrated moving average (SARIMA) model in the time series algorithm. The results showed the following: (1) From 2003 to 2022, the TCO in China showed a slight downward trend, with an average annual change rate of −0.29 DU/a; the coefficient of variation analysis found that TCO had the smallest intra-year fluctuations in 2008 and the largest intra-year fluctuations in 2005. (2) Using the M-K mutation test, it was found that there was a mutation point in the total amount of column ozone in 2016. (3) Using wavelet analysis to analyze the frequency domain characteristics of the TCO, it was observed that TCO variation in China had a combination of 14-year, 6-year, and 4-year main cycles, where 14 years is the first main cycle with a 10-year cycle and 6 years is the second main cycle with a 4-year cycle. (4) The spatial distribution characteristics of the TCO in China were significantly different in each region, showing a distribution characteristic of being high in the northeast and low in the southwest. (5) Based on the EOF analysis of the TCO in China, it was found that the variance contribution rate of the first mode was as high as 52.85%, and its spatial distribution of eigenvectors showed a “-” distribution. Combined with the trend analysis of the time coefficient, this showed that the TCO in China has declined in the past 20 years. (6) The SARIMA model with the best parameters of (1, 1, 2) × (0, 1, 2, 12) based on the training on the TCO data was used for prediction, and the final model error rate was calculated as 1.34% using the mean absolute percentage error (MAPE) index, indicating a good model fit.

Time–Frequency Characteristics and SARIMA Forecasting of Atmospheric Water Vapor in East Asia

Given the increasing impact of extreme rainfall and flooding on human life, studying and predicting changes in atmospheric water vapor (AWV) becomes particularly important. This paper analyzes the moderate-resolution imaging spectroradiometer (MODIS) data of the East Asian region from January 2003 to February 2023. The AWV data are examined in the time and frequency domain using methods such as empirical orthogonal function (EOF), Mann–Kendall (MK) analysis, and others. Additionally, four prediction models are applied to forecast the monthly average AWV data for the next year. The accuracy of these models is evaluated using metrics such as mean square error (MSE), mean absolute error (MAE), root mean square error (RMSE), and coefficient of determination (R2). The findings reveal several key insights: (1) The East Asian region exhibits highly variable seasonal variability in AWV, with identified mutation points after the MK test. (2) Spatial analysis shows high AWV data in the southern coastal areas of China, Thailand, Myanmar, Nansha Islands, and other regions during winter, while the Qinghai-Tibet Plateau region experiences low AWV during summer. (3) The first mode obtained through EOF decomposition contributes over 60% of the variance. Analysis of this mode reveals an increasing trend in AWV data for regions such as the Indian peninsula, Mongolia, and central and northeastern China over the past nine years. Conversely, the Bay of Bengal, Spratly Islands, eastern coast, and certain areas display a decreasing trend. (4) Employing the ensemble empirical mode decomposition (EEMD), the study identifies AWV data as a non-stationary series with an overall decreasing trend from 2003 to 2022. The filtered AWV series undergoes fast Fourier transform (FFT), uncovering periodicities of 2.6 years, 5 years, and 19 years. (5) Among the four forecasting models compared, the seasonal autoregressive integrated moving average model (SARIMA) demonstrates superior performance with the smallest MSE of 0.00782, MAE of 0.06977, RMSE of 0.08843, and the largest R2 value of 0.98454. These results clearly indicate that the SARIMA model provides the best fit. Therefore, the SARIMA forecasting model can be effectively utilized for forecasting AWV data, offering valuable insights for studying weather variability.