Global Sea Surface Temperature Research Articles

Variability of SST through Koopman Modes

Abstract The majority of dynamical systems arising from applications show a chaotic character. This is especially true for climate and weather applications. We present here an application of Koopman operator theory to tropical and global sea surface temperature (SST) that yields an approximation to the continuous spectrum typical of these situations. We also show that the Koopman modes yield a decomposition of the datasets that can be used to categorize the variability. Most relevant modes emerge naturally, and they can be identified easily. A difference with other analysis methods such as empirical orthogonal function (EOF) or Fourier expansion is that the Koopman modes have a dynamical interpretation, thanks to their connection to the Koopman operator, and they are not constrained in their shape by special requirements such as orthogonality (as it is the case for EOF) or pure periodicity (as in the case of Fourier expansions). The pure periodic modes emerge naturally, and they form a subspace that can be interpreted as the limiting subspace for the variability. The stationary states therefore are the scaffolding around which the dynamics takes place. The modes can also be traced to the Niño variability and in the case of the global SST to the Pacific decadal oscillation (PDO). Significance Statement We compute the Koopman modes for the tropical and global SST, demonstrating that significant dynamical modes can be identified also in complex high-dimensional datasets with continuous spectra. The Koopman modes are then used to identify the stationary subspace, namely, the limit subspace for the invariant evolution of the system.

STA-SST: Spatio-temporal time series prediction of Moroccan Sea surface temperature

Global Sea Surface Temperature (SST) trends have garnered significant attention in several ocean-related domains, including global warming, marine biodiversity, and environmental protection. This involves having an accurate and efficient forecast of future SST to ensure early detection and response in time to these events. Deep learning algorithms have become popular in SST prediction recently, although directly obtaining optimal prediction results from historical observation data is not simple. In this paper, we propose STA-SST, a new deep learning approach for forecasting SST, by combining the temporal dependencies of SST using the Bidirectional Long Short-Term Memory (BiLSTM) model, spatial features extracted from the convolution layer, and relevant information from the attention mechanism. To assess how well the Attention-BiLSTM with convolution layer predicts SST, we conducted a case study in the Moroccan Sea, concentrating on five different areas. The proposed model was compared against alternative forecasting models, including LSTM, XGBoost, Support Vector Regression (SVR), and Random Forest (RF). The experimental results show that STA-STT produces noticeably the best prediction results and is a solid choice for field implementation.

The potential role of insolation in the long-term climate evolution since the early Pleistocene

The distribution of insolation over time and space is a significant driver of climate change on orbital timescales. However, the influence of insolation on long-term climate evolution remains poorly understood due to the absence of a discernible long-term trend regulated by Earth's orbit. In this study, we present a sea surface temperature anomaly (SSTA) stack spanning the past 2 Myr, compiled from 26 millennial-resolved records obtained from the global ocean. The global average sea surface temperature (SST) reveals a 405-kyr cycle, as well as a gradual decrease of 2.34 ± 1.05 °C (1σ) from 2000 ka to 940 ka, followed by a period of relative stability. We introduce an index named the integral of annual mean insolation anomaly (IAMIA), which quantifies the continuous departure of annual mean insolation (AMI) from its “normal” cycle over a specific time interval. We find that the SST leads the variations in AMI and IAMIA at the 405-kyr band, intimating that the cycle evident in global SST does not originate from the changes of eccentricity but rather stems from harmonic or combination tones. Notably, IAMIA exhibits a fundamental shift at 935 ka, coinciding with the “900-ka event” observed in the SST. Modeling results support that the “900-ka event” could be driven by the change of cumulative insolation. Additionally, IAMIA underscores the important role of insolation in Pleistocene climate change on the long-term trend through the cumulative response of ocean heat content (OHC) to successive small step-wise insolation changes. Furthermore, we hypothesize that the fundamental changes of insolation around 935 ka, transitioning from a positive state characterized by substantial amplitude to a negative state typified by diminished amplitude, facilitated the onset and progression of the mid-Pleistocene transition (MPT). This investigation provides invaluable insights into the role of insolation in long-term climate evolution.

Assessment of relationship between sea surface temperature (SST) changes and precipitation types in Nigeria from 2000 to 2022

This study investigated the circulation patterns associated with rainfall variations across Nigeria’s different climatic zones (2000–2022). Data was acquired from MERRA2, gridded ERSST5 provided by the NOAA, and observation data records obtained from the Climatic Research Unit (CRU). Principal Component Analysis (PCA) was employed to compare different precipitation types, while the Multinomial Logistic Regression Model was employed to examine the influence of SSTa on precipitation presenting regression coefficients with a significance level set at p < 0.05. Five distint standard precipitation index (SPI) classes: very dry, dry, normal, wet and very wet were categorised using prcipitation data. Analysis was done in R-studio and involved data preparation, model training, and evaluation, with emphasis on interpreting the coefficients to discern the impact of SST anomalies on precipitation for each specified level. The results show that TOP, AVP, LSP, and CNP varied: spatially, the northern region received low moisture budget from the Atlantic Ocean while the temporal distribution of precipitation across different climatic zones indicate high variability in precipitation across these zones. The mean SSTa in the WAf region were predominantly positive (0.5 and 1). The lowest (highest) global SST values were prevalent during the DJF (JJA) season(s) whereas, the monthly distribution of SSTa for the WAf region reveal neutral (2000–2016) and El Niño (2016–2022) episodes. The analysis of NIF values indicates a varied but generally stronger relationship between WAf SST anomalies and precipitation types compared to nino3.4 SST versus precipitation types. As a signal for prediction of seasonal and spatial distribution of precipitation across Nigeria’s different climatic zones, this outcome can support planning for food security, water and biodiversity conservation, and climate change adaptation and mitigation.

Depth-dependent warming of the Gulf of Eilat (Aqaba)

The Gulf of Eilat (Gulf of Aqaba) is a semi-enclosed basin situated at the northern end of the Red Sea, renowned for its exceptional marine ecosystem. To evaluate the response of the Gulf to climate variations, we analyzed various factors including temperature down to 700 m, surface air temperature, and heat fluxes. We find that the sea temperature is rising at all depths despite inconclusive trends in local atmospheric variables, including the surface air temperature. The Gulf’s sea surface temperature (SST) warms at a rate of a few hundredths of a degree Celsius per year, which is comparable to the warming of the global SST and the Mediterranean Sea. The increase in sea warming is linked to fewer winter deep mixing events that used to occur more frequently in the past. Based on the analysis of the ocean-atmosphere heat fluxes, we conclude that the lateral advection of heat from the southern part of the Gulf likely leads to an increase in water temperature in the northern part of the Gulf. Our findings suggest that local ocean warming is not necessarily associated with local processes, but rather with the warming of other remote locations.

Effective attention model for global sea surface temperature prediction

Sea Surface Temperature (SST) is an essential parameter of the ocean–atmosphere system, so accurately predicting the future SST is of great significance. To address the low prediction accuracy due to insufficient consideration of spatiotemporal correlations in existing machine learning models, this paper proposes an Effective Attention Model (EAM) to achieve daily global SST prediction (predicting the SST of the next 15 days using those of the past 15 days) and monthly global SST prediction (predicting the SST of the next 12 months using those of the past 12 months). The EAM adopts an attention mechanism to avoid information redundancy and improve prediction accuracy. This is the first work to realize high-accuracy global SST prediction with a high spatial resolution of 0.25° on two time scales (daily and monthly) based on deep learning. Experiments show that (1) the prediction results have a consistent global spatiotemporal distribution with the ground truth, with the Root Mean Squared Difference (RMSD) of daily prediction ranging from 0.212 to 0.932 ℃ and the RMSD of monthly prediction ranging from 0.666 to 0.782 ℃. The EAM is superior to the deep learning models (ConvLSTM, 3D-CNN, and LSTM) and the numerical model (HYCOM); (2) Besides global prediction, the EAM also achieves good performance in a climate change sensitive area (Niño 3.4 sea area), with the lowest RMSD of 0.231 ℃. This study may not only benefit the research on climate events (El Niño and La Niña) but also serve to predict other marine disasters (such as heat waves and storms). In the future, the proposed attention module can also be adapted to other prediction scenarios such as meteorology.

An in-depth investigation of global sea surface temperature behavior utilizing chaotic modeling.

Sea surface temperature (SST), with its complex and dynamic behavior, is a major driver of ocean-atmosphere interactions. The purpose of this study is to investigate the behavior of SST and its prediction using a chaotic approach. Average mutual information (AMI) and Cao methods were used to reconstruct the phase space. The Lyapunov exponent and correlation dimension were used to investigate chaos. The Lyapunov exponent index was used to predict SST with a 5-year average prediction horizon using the local prediction method between 2023 and 2027. The results showed a 3-month delay time for the Pacific and Antarctic Oceans, and a 2-month delay time for the Atlantic, Indian, and Arctic Oceans. The optimal embedding dimension for all oceans is between 6 and 7. Our analysis reveals that the dynamics of SST in all oceans exhibit varying degrees of chaos, as indicated by the correlation dimension. The local prediction method achieves relatively accurate short-term SST predictions due to the clustering of SST points around specific attractors in the phase space. However, in the long term, the accuracy of this method decreases as the points in the phase space of SST can spread randomly. The model performance ranking with a Percent Mean Relative Absolute Error shows that the Indian Ocean has the best performance compared to other oceans, while the Atlantic, Pacific, and Antarctic and Arctic Oceans are in the next ranks. This study contributes to understanding the dynamics of SST and has practical value for use in the development of climate models.

Exploring the contribution of low‐frequency internal variability modes to global mean sea surface temperature variability based on large ensembles

AbstractBased on three large ensemble (LE) historical simulations, this study explored the contribution of low‐frequency internal variability (IV) modes to global mean sea surface temperature anomaly (GmSST) during 1920–2010. The dominant contributions of low‐frequency IV modes to GmSST in the three large ensembles are quite different. The low‐frequency IV modes in the Indo‐Pacific Ocean (i.e., IPO and IOBW) dominate the low‐frequency GmSST signal for the CESM2 and MPI LEs (the ensemble median contribution value reaches 70%–80%), while the low‐frequency IV modes in the North Atlantic dominate for the FGOALS‐g3 LE (contribute more than 50%). CESM2 LE and MPI LE can reasonably reproduce the impact of low‐frequency modes in the Indo‐Pacific ocean as observed, but FGOALS‐g3 LE underestimates this impact and overestimates the extratropical impact (on GmSST) of the North Atlantic. After removing the external forcing, it still retains the excessive signs in low‐frequency IV modes in FGOALS‐g3 LE over the North Atlantic. Furthermore, we used a pattern adjustment approach to revise the surplus effect in the North Atlantic. After revision, the contribution of decadal IVs in the North Atlantic to GmSST is reduced by 30%. Meanwhile, the tropical contribution in the Indo‐Pacific Ocean is increased and that is closer to the observed one. This approach can be employed to revise the weak or strong IV signals in other regions or LEs, which is meaningful for reducing the uncertainty of IV signals.

TEOS-10 Equations for Determining the Lifted Condensation Level (LCL) and Climatic Feedback of Marine Clouds

At an energy flux imbalance of about 1 W m−2, the ocean stores 90% of the heat accumulating by global warming. However, neither the causes of this nor the responsible geophysical processes are sufficiently well understood. More detailed investigations of the different phenomena contributing to the oceanic energy balance are warranted. Here, the role of low-level marine clouds in the air–sea interaction is analysed. TEOS-10, the International Thermodynamic Equation of State of Seawater—2010, is exploited for a rigorous thermodynamic description of the climatic trends in the lifted condensation level (LCL) of the marine troposphere. Rising sea surface temperature (SST) at a constant relative humidity (RH) is elevating marine clouds, cooling the cloud base, and reducing downward thermal radiation. This LCL feedback effect is negative and counteracts ocean warming. At the current global mean SST of about 292 K, the net radiative heat flux from the ocean surface to the LCL cloud base is estimated to be 24 W m−2. Per degree of SST increase, this net flux is expected to be enhanced by almost 0.5 W m−2. The climatic LCL feedback effect is relevant for the ocean’s energy balance and may be rigorously thermodynamically modelled in terms of TEOS-10 equations. LCL height may serve as a remotely measured, sensitive estimate for the sea surface’s relative fugacity, or conventional relative humidity.

First Analyses of the TIMELINE AVHRR SST Product: Long-Term Trends of Sea Surface Temperature at 1 km Resolution across European Coastal Zones

Coastal areas are among the most productive areas in the world, ecologically as well as economically. Sea Surface Temperature (SST) has evolved as the major essential climate variable (ECV) and ocean variable (EOV) to monitor land–ocean interactions and oceanic warming trends. SST monitoring can be achieved by means of remote sensing. The current relatively coarse spatial resolution of established SST products limits their potential in small-scale, coastal zones. This study presents the first analysis of the TIMELINE 1 km SST product from AVHRR in four key European regions: The Northern and Baltic Sea, the Adriatic Sea, the Aegean Sea, and the Balearic Sea. The analysis of monthly anomaly trends showed high positive SST trends in all study areas, exceeding the global average SST warming. Seasonal variations reveal peak warming during the spring, early summer, and early autumn, suggesting a potential seasonal shift. The spatial analysis of the monthly anomaly trends revealed significantly higher trends at near-coast areas, which were especially distinct in the Mediterranean study areas. The clearest pattern was visible in the Adriatic Sea in March and May, where the SST trends at the coast were twice as high as that observed at a 40 km distance to the coast. To validate our findings, we compared the TIMELINE monthly anomaly time series with monthly anomalies derived from the Level 4 CCI SST anomaly product. The comparison showed an overall good accordance with correlation coefficients of R &gt; 0.82 for the Mediterranean study areas and R = 0.77 for the North and Baltic Seas. This study highlights the potential of AVHRR Local Area Coverage (LAC) data with 1 km spatial resolution for mapping long-term SST trends in areas with high spatial SST variability, such as coastal regions.

Spatiotemporal Expansion of Algal Blooms in Coastal China Seas.

The coastal seas of China are increasingly threatened by algal blooms, yet their comprehensive spatiotemporal mapping and understanding of underlying drivers remain challenging due to high turbidity and heterogeneous water conditions. We developed a singular value decomposition-based algorithm to map these blooms using two decades of MODIS-Aqua satellite data, spanning from 2003 to 2022. Our findings indicate significant algal activity along the Chinese coastline, impacting an average annual area of approximately 1.8 × 105 km2. The blooms exhibit peak intensity in August, while the maximum affected area occurs in September, featuring multifrequency outbreaks in spring, and pronounced large-scale events in summer and autumn. Notably, our analysis demonstrates a robust 67% increase in bloom occurrences over the study period. This expansion is primarily attributed to increased nutrient inflow from terrestrial sources linked to human activity and precipitation, compounded by rising global sea surface temperatures. These spatiotemporal insights are critical for devising effective management strategies and policies to mitigate the impacts of algal blooms.

Highly stratified mid-Pliocene Southern Ocean in PlioMIP2

Abstract. During the mid-Pliocene warm period (mPWP; 3.264–3.025 Ma), atmospheric CO2 concentrations were approximately 400 ppm, and the Antarctic Ice Sheet was substantially reduced compared to today. Antarctica is surrounded by the Southern Ocean, which plays a crucial role in the global oceanic circulation and climate regulation. Using results from the Pliocene Model Intercomparison Project (PlioMIP2), we investigate Southern Ocean conditions during the mPWP with respect to the pre-industrial period. We find that the mean sea surface temperature (SST) warming in the Southern Ocean is 2.8 °C, while global mean SST warming is 2.4 °C. The enhanced warming is strongly tied to a dramatic decrease in sea ice cover over the mPWP Southern Ocean. We also see a freshening of the ocean (sub)surface, driven by an increase in precipitation over the Southern Ocean and Antarctica. The warmer and fresher surface leads to a highly stratified Southern Ocean that can be related to weakening of the deep abyssal overturning circulation. Sensitivity simulations show that the decrease in sea ice cover and enhanced warming is largely a consequence of the reduction in the Antarctic Ice Sheet. In addition, the mPWP geographic boundary conditions are responsible for approximately half of the increase in mPWP SST warming, sea ice loss, precipitation, and stratification increase over the Southern Ocean. From these results, we conclude that a strongly reduced Antarctic Ice Sheet during the mPWP has a substantial influence on the state of the Southern Ocean and exacerbates the changes that are induced by a higher CO2 concentration alone. This is relevant for the long-term future of the Southern Ocean, as we expect melting of the western Antarctic Ice Sheet in the future, an effect that is not currently taken into account in future projections by Coupled Model Intercomparison Project (CMIP) ensembles.

Performance of high-resolution MUR satellite sea surface temperature data as a proxy for near-surface in situ temperatures on neotropical reefs

As ocean temperatures increase, so does the need for accurate, regular, and widespread temperature data. This is especially important for researchers studying corals, which can be pushed over their bleaching thresholds by small, sustained changes in water temperature. Satellite data products offer longitudinal estimates of sea surface temperature (SST) taken at regular intervals without the ongoing costs and geospatial limits associated with underwater temperature loggers. However, previous work indicates satellite data are sometimes characterized by significant temperature biases, particularly in stratified coastal waters. Here, we compared the foundation sea surface temperature (fSST) estimated from a highly processed, gridded satellite data set (GHRSST MUR L4 Global SST) to in situ temperature (IST) data collected from coral reef habitats in the top 20 m of the water column at 19 sites in Panama. Comparisons between Pacific and Caribbean coasts and between upwelling and non-upwelling seasons provide a robust test of the utility of this product to estimate subsurface temperatures. The mean differences between fSST and IST show that MUR fSST accurately reflects temperatures at shallow (&lt;9 m) depths on the Caribbean coast (mean bias ≤0.55°C ± standard error (SE) 0.01°C) and during the non-upwelling season on the Pacific coast (mean bias ≤0.26°C ± SE 0.03°C). MUR fSST poorly reflects IST during the Pacific upwelling season, especially at sites 12 m and deeper. In such cases, 66% of satellite estimates deviated from ISTs by &gt;1°C. Our findings underscore how highly processed SST data products should be ground-truthed before being used as a substitute for IST, especially in upwelling areas.

Dynamic pathway linking Pakistan flooding to East Asian heatwaves.

In July to August 2022, Pakistan suffered historic flooding while record-breaking heatwaves swept southern China, causing severe socioeconomic impacts. Similar extreme events have frequently coincided between two regions during the past 44 years, but the underlying mechanisms remain unclear. Using observations and a suite of model experiments, here, we show that the upper-tropospheric divergent wind induced by convective heating over Pakistan excites a barotropic anomalous anticyclone over eastern China, which further leads to persistent heatwaves. Atmospheric model ensemble simulation indicates that this dynamic pathway linking Pakistan flooding and East Asian heatwaves is intrinsic to the climate system, largely independent of global sea surface temperature forcing. This dynamic connection is most active during July to August when convective variability is large over Pakistan and the associated divergent flow excites barotropic Rossby waves that propagate eastward along the upper troposphere westerly waveguide. This robust waveguide and the time delay offer hopes for improved subseasonal prediction of extreme events in East Asia.

Predicting Summer Precipitation Anomalies in the Yunnan–Guizhou Plateau Using Spring Sea-Surface Temperature Anomalies

By constructing a correlation network between global sea surface temperature anomalies (SSTAs) and summer precipitation anomalies in the Yunnan–Guizhou Plateau, key SST regions influencing summer precipitation anomalies in the Yunnan–Guizhou Plateau were selected. It was found that spring SSTAs in the Bay of Bengal, southwestern Atlantic, and eastern Pacific are crucial for influencing summer precipitation anomalies in the Yunnan–Guizhou Plateau. Setting SSTAs from these three regions as predictor variables 3 months in advance, we constructed multiple linear regression (MLR), ridge regression (RR), and lasso regression (LR) models to predict summer precipitation anomalies over the Yunnan–Guizhou region. The training phase involved data spanning from 1961 to 2005, which aimed to predict precipitation anomalies in the Yunnan–Guizhou Plateau for the period extending from 2006 to 2022. Based on MLR, RR, and LR models, the correlations between predicted values and observed summer precipitation anomalies in Yunnan–Guizhou were 0.48, 0.46, and 0.46, respectively. These values were all higher than the correlation coefficients of the NCC_CSM model’s predicted and observed values. Additionally, its performance in predicting summer precipitation anomalies over the Yunnan–Guizhou region, based on key SST regions, was assessed using performance metrics such as anomaly correlation coefficient (ACC), anomaly sign consistency rate (PC), and trend anomaly comprehensive score (PS score). The average ACC of MLR, RR, and LR models was higher than that of the NCC_CSM model’s predictions. For MLR, RR, LR, and NCC_CSM models, the PCs exceeding 50% of the year were 14, 14, 11, and 10, respectively. Furthermore, the average PS score for predicting summer precipitation anomalies over the Yunnan–Guizhou region using MLR, RR, and LR was approximately 73 points; 8 higher than the average PS score of the NCC_CSM model. Therefore, predicting summer precipitation anomalies over the Yunnan–Guizhou region based on key SST regions is of great significance for improving the prediction skills of precipitation anomalies in this region.

Satellite-based time-series of sea-surface temperature since 1980 for climate applications

A 42-year climate data record of global sea surface temperature (SST) covering 1980 to 2021 has been produced from satellite observations, with a high degree of independence from in situ measurements. Observations from twenty infrared and two microwave radiometers are used, and are adjusted for their differing times of day of measurement to avoid aliasing and ensure observational stability. A total of 1.5 × 1013 locations are processed, yielding 1.4 × 1012 SST observations deemed to be suitable for climate applications. The corresponding observation density varies from less than 1 km−2 yr−1 in 1980 to over 100 km−2 yr−1 after 2007. Data are provided at their native resolution, averaged on a global 0.05° latitude-longitude grid (single-sensor with gaps), and as a daily, merged, gap-free, SST analysis at 0.05°. The data include the satellite-based SSTs, the corresponding time-and-depth standardised estimates, their standard uncertainty and quality flags. Accuracy, spatial coverage and length of record are all improved relative to a previous version, and the timeseries is routinely extended in time using consistent methods.

Covariability of decadal surface air temperature variability over Myanmar with sea surface temperature based on singular value decomposition analysis

The Myanmar’ Southeast Asian country is currently experiencing environmental changes, with temperature change being one the major contributing factors. Although many studies have shown the contribution of anthropogenic activities, the factors sustaining the observed increase in air temperature (TEMP) are not fully understood. We examined the interdependence of the surface TEMP with the sea surface temperature (SST) from 1971 to 2020 on a decadal timescale to predict changes in TEMP over a longer time period. Our analysis showed a pronounced interdecadal change in TEMP, with the highest intensity observed in the 2010s. The results show that the dominant modes of the global SST significantly influence the TEMP variation in the region at the decadal time scale. Indian Ocean (IO) SST-singular value decomposition (SVD)1 presents significant positive (negative) correlations in the southeast, and central (southwest) related to warming (cooling) TEMP in the east, northern tip, and northwest (south-Yangon and Taninthayi, and some parts of the north). The second and third modes of SST-SVD are cooling (warming) SST over the east of the Bay of Bengal (BoB), and along the Myanmar coast, which are associated with cooling(warming) TEMP patterns in the region. Significant negative correlations of decadal TEMP at the annual scale with Pacific Decadal Oscillation (PDO) and North Pacific (NP), the MAMJ with Atlantic Multidecadal Oscillation (AMO) and November–December–January–February (NDJF) with PDO were evidenced. Meanwhile, significant positive correlations were obtained between TEMP and AMO (NP) at the annual scale (NDJF), respectively. These findings provide valuable insights into decadal-scale TEMP patterns and their relationships with SSTs, contributing to a better understanding of TEMP variability in Myanmar, which may be helpful in climate prediction. Predicting surface TEMP on a decadal timescale is helpful in environmental management.

Multimodal fusion-based spatiotemporal incremental learning for ocean environment perception under sparse observation

Accurate ocean environment perception is crucial for weather and climate prediction. Environmental limitations and deployment costs constrain satellite and buoy real-time observation, leading to sparse data availability. This paper proposes a novel approach, multimodal fusion-based spatiotemporal incremental learning, enhancing the ocean environment perception under sparse observations. This method uses sparse real-time observations to comprehend, reconstruct, and predict the full environment. First, spatiotemporal disentanglement decouples intrinsic features by integrating physical principles and data learning. Subsequently, incremental extension captures the dynamic environment through stable representation updating and dynamic behavior learning. Then, multimodal information fusion synergizes multisource intrinsic features, enabling the full perception of the ocean environment. Finally, the methodology is supported by convergence analysis and error boundary evaluation. Validation with global sea surface temperature and western Pacific Ocean high-dimensional temperature datasets demonstrates its potential for advancing ocean research and applications using sparse real-time observation.

Harnessing the Power of Graph Representation in Climate Forecasting: Predicting Global Monthly Mean Sea Surface Temperatures and Anomalies

AbstractThe variability of sea surface temperatures (SSTs) is crucial in climate dynamics, influencing marine ecosystems and human activities. This study leverages graph neural networks (GNNs), specifically a GraphSAGE model, to forecast SSTs and their anomalies (SSTAs), focusing on the global scale structure of climatological data. We introduce an improved graph construction technique for SST teleconnection representation. Our results highlight the GraphSAGE model's capability in 1‐month‐ahead global SST and SSTA forecasting, with SST predictions spanning up to 2 years with a recursive approach. Notably, regions with persistent currents exhibited enhanced SSTA predictability, contrasting with equatorial and Antarctic areas. Our GNN outperformed both the persistence model and traditional methods. Additionally, we offer an SST and SSTA graph data set based on ERA5 and a graph generation tool. This GNN case study has shown how the GraphSAGE can be used in SST and SSTA forecasting, and will provide a foundation for further refinements such as adjusting graph construction, optimizing imbalanced regression techniques for extreme SSTAs, and integrating GNNs with other temporal pattern learning methods to improve long‐term predictions.

Global Sea Surface Temperature Analysis Based on Domestic Ocean Satellite Data. Part II: Comparative Analysis of Multiple Sets of Global Sea Surface Temperature Products in 2022

With more and more global gap-free fusion products of the sea surface temperature, understanding the consistency and discrepancy of the different SST fusion products will not only help data providers to improve their algorithms, but also help them to select the one that may better suit their applications. In this article, we have compared and analysed 10 sets of global gap-free fusion products of sea surface temperature in 2022, with different fusion techniques and related configurations. It is found that each SST analysis product has the same spatial distribution, with the minimum NMEFC mean value (20.11) and the maximum MGDSST mean value (20.31). Compared with ARGO in-situ data, the RMSE ranges are from 0.3233 (OSTIA) to 0.5180 (MGDSST). The RMSE between NMEFC fusion products and Argo in-situ data is 0.3861, ranked fifth out of 10 sets of fusion products. Compared with GMPE analysis among 9 sets of fusion products, the RMSE ranges are from 0.1579 (CMC) to 0.3199 (K10), and the NMEFC fusion product has a RMSE of 0.3040, which is at the intermediate level, ranked sixth out of 9 fusion products.