Search

Abstract Western Europe experienced an unprecedentedly hot July in 2022, which significantly impacted ecosystems and society. Our observational and numerical modeling study reveals that this event was influenced by anomalous North Atlantic and Eurasian jet streams. The northeastward shift of the North Atlantic jet stream, driven by sea surface temperature gradients, and the curving of the Eurasian jet stream, affected by rainfall anomalies in Pakistan, enhanced atmospheric subsidence over western Europe. This research highlights the crucial role of the synergistic behavior of the North Atlantic and Eurasian jet streams in driving extreme heat over Western Europe. Furthermore, CMIP6 climate model projections suggest that under the SSP585 scenario, similar jet stream configurations could lead to even more intense extreme temperatures (~7.02 ± 0.61 °C) compared to the current climatological mean.

Oceanic mesoscale eddies are prevalent throughout the global ocean, playing a critical role in material and energy transport while significantly influencing climate change. Accurate characterization of their three-dimensional structures and movement is essential for a quantitative analysis of their transport processes. Traditional eddy detection algorithms have lower successful detection rate and with more limitations, so they fall short in the complex and dynamic ocean environment. The rising trend of applying artificial intelligence (AI) algorithms, due to their efficiency, precision, and automation, addresses this challenge. This study employs the 3D-U-Res-Net algorithm to identify the three-dimensional structures of mesoscale eddies in the Southern Ocean using GLORYS12V1 data from 2011 to 2020. A vector geometry-based eddy detection algorithm (VG) initially identified 1587292 eddy snapshots in the Southern Ocean (2011–2019), which were used for training the 3D-U-Res-Net algorithm. Data from 2020 served as the ground truth and validation set. The successful detection rate of 3D-U-Res-Net algorithm is 100%, which means that it identified all 135734 eddy snapshots from the VG dataset in 2020. For eddy tracking, the VG algorithm counted 18168 eddy tracks, whereas the 3D-U-Res-Net counted 18559, reflecting a 2.15% bias. To reduce uncertainty, eddies with lifespans shorter than two weeks were excluded. The average lifespans and traveling distances for eddies detected by the 3D-U-Res-Net (VG) algorithm were 29.35 (29.61) days and 77.78 (37.60) km, respectively, with the 3D-U-Res-Net identifying eddies with longer traveling distances. The mean radius of eddies detected by the VG algorithm was 43.16 km, while the 3D-U-Res-Net detected eddies with a mean radius of 43.74 km, a 0.58 km increase. We categorized eddies into four three-dimensional structures: bowl-shaped, cone-shaped, lens-shaped, and cylindrical. The VG algorithm identified these structures in proportions of 32%, 31%, 25%, and 12%, respectively, whereas the 3D-U-Res-Net algorithm found 19.48%, 19.58%, 0.04%, and 60.9%, respectively. The 3D-U-Res-Net identified more cylindrical eddies and was approximately ten times faster than the VG algorithm. Overall, this algorithm has good performance and higher efficiency. It is an attempt of using AI for oceanic research, and more works can be carried out in the future.

The Lofoten Vortex (LV), which is identified as a quasi-permanent anticyclonic eddy, strengthens through continuous merging with external anticyclonic eddies. Our investigation used the Lagrangian method to monitor the LV on a daily basis. Utilizing satellite altimeter data, we conducted multi-year tracking and statistical analysis of merging events involving the LV. The results indicate a characteristic radius of approximately 42.72 km and a mean vorticity at the eddy center of approximately −2.23 × 10−5 s−1. The eddy exhibits oscillatory motion within the sea basin depression, centered at 70°N, 3°E, characterized by counterclockwise trajectories between 0.5°E and 6°E and between 69°N and 70.5°N. There are two types of merging events: fusion events (55%), in which eddies of similar strengths interact within a closed flow line and then merge to form a new eddy; and absorption events (45%), in which the stronger LV absorbs the weaker anticyclonic eddies without destroying the structure of the LV itself. The nodes where strong vorticity advection occurs correspond to the nodes where merging occurs, suggesting that their effect on merging can be well characterized by the vorticity advection time series. We also observe occasional fluctuations and substitution events involving the LV and external anticyclonic eddies, suggesting a dynamic succession rather than a single vortex entity.

Atmospheric forcings are significant physical factors that influence the variation of sea surface temperature (SST) and are often used as essential input variables for ocean numerical models. However, their contribution to the prediction of SST based on machine-learning methods still needs to be tested. This study presents a prediction model for SST in the East China Sea (ECS) using two machine-learning methods: Random Forest and SA-ConvLSTM algorithms. According to the Random Forest feature importance scores and correlation coefficients R, 2 m air temperature and longwave radiation were selected as the two most important key atmospheric factors that can affect the SST prediction performance of machine-learning methods. Four datasets were constructed as input to SA-ConvLSTM: SST-only, SST-T2m, SST-LWR, and SST-T2m-LWR. Using the SST-T2m and SST-LWR, the prediction skill of the model can be improved by about 9.9% and 9.43% for the RMSE and by about 8.97% and 8.21% for the MAE, respectively. Using the SST-T2m-LWR dataset, the model’s prediction skill can be improved by 10.75% for RMSE and 9.06% for MAE. The SA-ConvLSTM can represent the SST in ECS well, but with the highest RMSE and AE in summer. The findings of the presented study requires much more exploration in future studies.

The marine environment of the Arctic Ocean has changed rapidly in recent decades. We used reanalysis data and observational data to explore the variations in the upper ocean heat content (OHC) of the Canadian Basin (CB) and the variations in the temperature profiles of the Southern Canadian Basin (SCB). Both the reanalysis data and observational data show increasing trends for the OHC of the CB from 1993 to 2023. Compared to the World Ocean Atlas data (WOA 18/23), the reanalysis data (ORAS5 or GLORYS12V1) significantly underestimated the values of the upper OHC of the Canadian Basin. To explain the OHC differences, the Ice-Tethered Profiler (ITP) observational data were used to analyze the variations in the vertical temperature profiles. We found that the reanalysis data remarkably underestimated the maximum temperatures of the subsurface Pacific warm water and its increasing trend. Based on the short-term prediction results from the Bi-LSTM neural network, we forecasted that the upper OHC will continue to increase in the SCB, mainly due to the warming of the intermediate Atlantic warm water. The research results provide a valuable reference for assessing and improving climate-coupled models.

Measurement(s) mesoscale eddy • chlorophyll • thermohaline profile Technology Type(s) remote sensing • argo float

The East Greenland Polar Front (EGPF) is an important front with strong salinity and temperature gradients in the Nordic Seas. It is formed by the interaction between Arctic-origin and Atlantic-origin water. The variations of EGPF are closely linked with sea ice melting and heat content transport associated with North Atlantic water recirculation. For a three-dimensional (3D) daily analysis, we use the global ocean eddy resolution reanalysis product (GLORYS12V1) from 1993 to 2018 to calculate the salinity and temperature horizontal gradient in the upper ocean and obtain the spatiotemporal distribution and intensity characteristics of EGPF. After assessment, the thresholds of the salinity and temperature fronts are set to 0.04 psu/km and 0.09°C/km, respectively. Compared with satellite observations of sea ice concentration, a significant spatial relationship is observed between the main position of EGPF and the ice edge before the sea ice shrinks to the continental shelf sea area. Affected by the freshening of the Arctic-origin water due to the melting of the sea ice, the intensity and area of EGPF show significant seasonal variations. Against the background of global warming, the sea ice area presents an obvious decreasing trend in the Greenland Sea. The melting of sea ice increased annually every summer from 1998 to 2018. The heat content transport of the Atlantic-origin water has also increased in recent years. The 3D characteristics (intensity and volume) of EGPF as salinity and temperature fronts exhibit increasing trends.

The Drake Passage is known for its abundant mesoscale eddies, but little is known about their three-dimensional characteristics, which hinders our understanding of their impact on eddy-induced transport and deep-sea circulation. A 10-year study was conducted using GLORYS12 Mercator Ocean reanalysis data from 2009 to 2018. The study analyzed the statistical characteristics of eddies in the Drake Passage, spanning from the surface down to a depth of 2000 m in three dimensions. The findings indicate that the mean radius of the eddies is 35.5 km, with a mean lifespan of 12.3 weeks and mean vorticity of 2.2 × 10−5 s−1. The eddies are most active and energetic near the three main fronts and propagate north-eastward at an average distance of 97.8 km. The eddy parameters vary with water depth, with more anticyclones detected from the surface to 400 m, displaying a larger radius and longer propagation distance. Cyclones have longer lifespans and greater vorticity. However, beyond 400 m, there is not much difference between anticyclones and cyclones. Approximately 23.3% of the eddies reach a depth of 2000 m, with larger eddies tending to penetrate deeper. The eddies come in three different shapes, bowl-shaped (52.7%), lens-shaped (27.1%) and cone-shaped (20.2%). They exhibit annual and monthly distribution patterns. Due to its high latitude location, the Drake Passage has strong rotation and weak stratification, resulting in the generation of small and deep-reaching eddies. These eddies contribute to the formation of Antarctic intermediate water and lead to modulation of turbulent dissipation.

The Lofoten Vortex (LV) is a quasi-permanent anticyclonic eddy with the characteristic of periodic regeneration in the Lofoten Basin (LB), which is one of the major areas of deep vertical mixing in the Nordic Sea. Our analysis of the LV contributes to our understanding of the variations in convective mixing in the LB. Based on drifter data and satellite altimeter data, the climatological results show that the LV has the sea surface characteristics of relative stability in terms of its spatial position and significant seasonal variations in its physical characteristics. Combined with the temperature and salinity data of Argo profiles, the vertical structures of the LV are presented here in terms of their spatial distribution and monthly variations. The wavelet analysis of the satellite sea surface temperature (SST) data shows that the period of SST anomaly (SSTA) in the LV sea area is 8–16 years. In the stage marked by a decreasing (increasing) trend of SSTA, the vertical mixing is strengthened (weakened). Current vertical mixing is clearly revealed by the Argo profiles, and the SSTA shows a significant impact of cooling. However, against a background of warming and freshening, this vertical mixing will be greatly weakened in the next increasing trending stage of the SSTA.

Cross-slope eddies play an important role in the exchange of water, salt, heat, nutrients, chlorophyll, phytoplankton and other biogeochemical elements between basin and shelf in the South China Sea. The cross-slope transport process by a mesoscale anticyclonic eddy is studied by ROMS model and satellite data. The 1000 m isobath was considered as a proxy for the slope. The anticyclone shows different features at different places on the slope: (a) the volume transport at the northeast of the slope was off-slope, while at the southwest was on-slope; (b) both on and off-slope transports were greatly enhanced during the cross-slope process, and gradually weakened after crossing the slope. The total cross-slope water transport was 5.97 Sv, which was higher than the along-slope component with −0.58 Sv. The Eulerian results also showed that enhanced cross-slope transport was related to the distance between the eddy and slope, the eddy radius, and the eccentricity of the eddy. The offline passive tracer experiment showed that particles were floating up during and after the crossing process, mainly due to the strong Ekman pumping.

Deep learning methods have excellent prospects for application in wave forecasting research. This study employed the convolutional LSTM (ConvLSTM) algorithm to predict the South China Sea (SCS) significant wave height (SWH). Three prediction models were established to investigate the influences of setting different parameters and using multiple training data on the forecasting effects. Compared with the SWH data from the China–France Ocean Satellite (CFOSAT), the SWH of WAVEWATCH III (WWIII) from the pacific islands ocean observing system are accurate enough to be used as training data for the ConvLSTM-based SWH prediction model. Model A was preliminarily established by only using the SWH from WWIII as the training data, and 20 sensitivity experiments were carried out to investigate the influences of different parameter settings on the forecasting effect of Model A. The experimental results showed that Model A has the best forecasting effect when using three years of training data and three hourly input data. With the same parameter settings as the best prediction performance Model A, Model B and C were also established by using more different training data. Model B used the wind shear velocity and SWH as training and input data. When making a 24-h SWH forecast, compared with Model A, the root mean square error (RMSE) of Model B is decreased by 17.6%, the correlation coefficient (CC) is increased by 2.90%, and the mean absolute percentage error (MAPE) is reduced by 12.2%. Model C used the SWH, wind shear velocity, wind and wave direction as training and input data. When making a 24-h SWH forecast, compared with Model A, the RMSE of Model C decreased by 19.0%, the CC increased by 2.65%, and the MAPE decreased by 14.8%. As the performance of the ConvLSTM-based prediction model mainly rely on the SWH training data. All the ConvLSTM-based prediction models show a greater RMSE in the nearshore area than that in the deep area of SCS and also show a greater RMSE during the period of typhoon transit than that without typhoon. Considering the wind shear velocity, wind, and wave direction also used as training data will improve the performance of SWH prediction.

The ocean chlorophyll-a (Chl-a) concentration is an important variable in the marine environment, the abnormal distribution of which is closely related to the hazards of red tides. Thus, the accurate prediction of its concentration in the East China Sea (ECS) is greatly important for preventing water eutrophication and protecting the coastal ecological environment. Processed by two different pre-processing methods, 10-year (2011–2020) satellite-observed chlorophyll-a data and logarithmic data were used as the long short-term memory (LSTM) neural network training datasets in this study. The 2021 data were used for comparison to prediction results. The past 15 days’ data were used to predict the concentration of chlorophyll-a for the five following days. Results showed that the predictions obtained by both pre-processing methods could simulate the seasonal distribution of the Chl-a concentration in the ECS effectively. Moreover, the prediction performance of the model driven by the original values was better in the medium- and low-concentration regions. However, in the high-concentration region, the prediction of extreme concentrations by the two data-driven LSTM models showed underestimation, considering that the prediction performance of the model driven by the original values was better. Results of sensitivity experiments showed that the prediction accuracy of the model decreased considerably when the backward prediction time step increased. In this study, the neural network was driven only by chlorophyll-a, whose concentration in the ECS was forecasted, and the effect of other relevant marine elements on Chl-a was not considered, which is the current weakness of this study.

Sea surface temperature (SST) is an important physical factor in the interaction between the ocean and the atmosphere. Accurate monitoring and prediction of the temporal and spatial distribution of SST are of great significance in dealing with climate change, disaster prevention, disaster reduction, and marine ecological protection. This study establishes a prediction model of sea surface temperature for the next five days in the East China Sea using long-term and short-term memory neural networks (LSTM). It investigates the influence of different parameters on prediction accuracy. The sensitivity experiment results show that, based on the same training data, the length of the input data of the LSTM model can improve the model’s prediction performance to a certain extent. However, no obvious positive correlation is observed between the increase in the input data length and the improvement of the model’s prediction accuracy. On the contrary, the LSTM model’s performance decreases with the prediction length increase. Furthermore, the single-point prediction results of the LSTM model for the estuary of the Yangtze River, Kuroshio, and the Pacific Ocean are accurate. In particular, the prediction results of the point in the Pacific Ocean are the most accurate at the selected four points, with an RMSE of 0.0698 °C and an R2 of 99.95%. At the same time, the model in the Pacific region is migrated to the East China Sea. The model was found to have good mobility and can well represent the long-term and seasonal trends of SST in the East China Sea.

This paper utilizes a fine scale parameterization scheme based on strain spectra and characterizes the temporal and spatial distributions of diapycnal turbulent mixing in the West Pacific Warm Pool. We use Array for Real‐Time Geostrophic Oceanography (Argo) profiles from 2011 to 2022 to generate the turbulent dissipation rate (ε) as a representative of internal‐wave‐induced mixing. We then discuss the influence of El Niño‐Southern Oscillation (ENSO) on turbulent mixing by considering seafloor topography, thermocline depth, and ENSO phases. Results indicate that ENSO can influence the vertical stratification in the ocean by changing the depth of the thermocline, causing the change of interaction between internal tides and seafloor topography, thereby affecting the intensity and vertical distribution of diapycnal turbulent mixing. During La Niña years, the value of ε is larger compared to that of El Niño years. Compared to rough regions, the thermocline variations induced by ENSO have a more pronounced impact in the smooth regions. The variability in turbulent mixing influenced by ENSO is crucial for heat and substance exchange in the equatorial Pacific. [ABSTRACT FROM AUTHOR]

Using multi-source observational data and GLORYS12V1 reanalysis data, we conducted a comparative analysis of different responses of two warm eddies, AE1 and AE2 in the northern South China Sea to Typhoon Kalmaegi during September 2014. The findings of our research are as follows: (1) For horizontal distribution, the area and the sea surface temperature (SST) of AE1 and AE2 decreased by about 31% (36%) and 0.4 °C (0.6 °C).The amplitude, Rossby number (Ro) and eddy kinetic energy (EKE) of AE1 increased by 1.3 cm, 1.4x10-2 and 107.2 cm² s-2 after the typhoon, respectively, while AE2 weakened and the amplitude, vorticity and EKE decreased by 3.1 cm, 1.6x10-2 and 38.5 cm² s-2, respectively. (2) In vertical direction, AE1 demonstrated enhanced convergence, leading to an increase in temperature and a decrease in salinity above 150 m. The response below the mixing layer depth (MLD) was particularly prominent (1.3 °C). In contrast, AE2 experienced cooling and a decrease in salinity above the MLD. Below the MLD, it exhibited a subsurface temperature drop and salinity increase due to the upwelling of cold water induced by the suction effect of the typhoon. (3) The disparity in the responses of the two warm eddies can be attributed to their different positions relative to Typhoon Kalmaegi. Warm eddy AE1, with its center located on the left side of the typhoon's path, experienced a positive work effect as the typhoon passed by. This induced a strong negative wind stress curl and triggered a negative Ekman pumping velocity (EPV), further enhanced by the converging sinking of the upper warm water, thereby strengthening AE1. On the other hand, warm eddy AE2, situated closer to the center of the typhoon, weakened due to the cold suction caused by the strong positive wind stress curl in the typhoon's center. These findings underscore the importance of relative positions of eddies in their interactions with typhoons. [ABSTRACT FROM AUTHOR]