Monthly Mean Sea Research Articles

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.

Deconstructing the causes of July sea level variability in the East Sea from 1994 to 2021

In 2021 the East Sea experienced its highest July sea level (65.09 cm), as well as the highest July sea surface and atmospheric temperatures, in the 29 years between 1993 and 2021. We present several methodologies to identify the more important causes of sea level change (SLC) in a semi-enclosed sea and explore the critical fluctuation of ocean mass transport divergence during a period of rapid sea level rise. Based on satellite altimeter data in the East Sea, the SLC, as reflected in the absolute dynamic topography (ADT), which is the ADT difference between the last and first days of a month, was 7.18 cm in July 2021. This may reflect a combination of oceanic and atmospheric factors: ocean heat transport divergence among the Korea, Tsugaru, and Soya Straits was found to contribute 2.32 cm of SLC, atmospheric heat flux at the sea surface contributed 2.87 cm of SLC, and mass divergence (including errors) accounted for 1.98 cm of SLC. The monthly mean sea level (ADT) variation in the East Sea should be examined in terms of the oceanic and atmospheric fluxes, and the SLC sources can additionally be divided into heat and others (mass + errors). The proportional contribution of heat to SLC from 1994 to 2021 was 97.4%. Although the contributions of mass and errors were small, there were substantial temporal fluctuations, as their standard deviation reached up to 96.7% of that of SLC in ADT. In the near future, a more precise analysis of the contributions of mass and errors to SLC is required.

Using EEMD mode decomposition in combination with machine learning models to improve the accuracy of monthly sea level predictions in the coastal area of China

In the context of climate change and human activities, the global sea level is facing a rising trend, which poses serious challenges to the ecological environment of coastal areas. In this study, we selected the monthly mean sea level (MSL) time series of 9 stations in the coastal areas of China as the research object. First, we analyzed the spatiotemporal distribution characteristics of the monthly MSL in the coastal areas of China. Secondly, we analyzed the ability of ensemble empirical mode decomposition (EEMD) to decompose the monthly MSL series. Finally, we choose three machine learning models, namely Back Propagation (BP), K-Nearest Neighbor (KNN), and Long Short-Term Memory (LSTM) neural network models to compare model prediction effect between single machine learning models with machine learning models combined with EEMD. The results show that except for the YANTAI (YT) station, which showed an insignificant downward trend, the monthly MSL of other stations showed an upward trend, indicating that the coastal areas of China are facing the risk of sea level rise. EEMD can effectively reduce the complexity of the original monthly MSL time series, and different intrinsic mode functions (IMFs) reflect changes in monthly MSL at different frequencies. Comparing the single machine learning model and the machine learning model combined with EEMD, it is found that the simulation effect of the machine learning model combined with EEMD is better than that of the single model. The model with the best prediction effect on monthly MSL in the coastal areas of China is LSTM-EEMD, followed by KNN-EEMD. This study provides an important reference for systematically understanding sea level changes and selecting an appropriate monthly MSL prediction model in the coastal areas of China.

The sea level time series of Trieste, Molo Sartorio, Italy (1869–2021)

Abstract. The sea level observations carried out at Trieste, Molo Sartorio, from 1869 to 2021 have been revised and updated. Information on the tide gauges and on the geodetic benchmarks on Molo Sartorio during that period have been collected. Basic quality checks have been applied. The hourly data for the 1917–1938 period, digitized from the original charts, have allowed us to build a time series of hourly sea level heights from 1905 to 2021. Gaps of up to 24 h have been filled by interpolation. The errors affecting the monthly and annual mean sea levels have been estimated. The availability of monthly and annual means prior to 1904 allowed us to build a mean sea level time series spanning 153 years, characterized by linear trends of an observed sea level of 1.36 ± 0.17 mm yr−1 and of an inverse-barometer-corrected sea level of 1.45 ± 0.13 mm yr−1. A significant acceleration of 0.008 ± 0.004 mm yr−2 was estimated from the inverse-barometer-corrected sea level time series. This data set represents the most up-to-date data set of sea level observations and ancillary information relative to the tide-gauge station of Trieste, Molo Sartorio. The data are available through SEANOE (https://doi.org/10.17882/62758, Raicich, 2022).

Seasonal Prediction of Tropical Cyclones and Storms over the Southwestern Indian Ocean Region Using the Generalized Linear Models

Tropical cyclones (TCs) and storms (TSs) are among the devastating events in the world and southwestern Indian Ocean (SWIO) in particular. The seasonal forecasting TCs and TSs for December to March (DJFM) and November to May (NM) over SWIO were conducted. Dynamic parameters including vertical wind shear, mean zonal steering wind and vorticity at 850 mb were derived from NOAA (NCEP-NCAR) reanalysis 1 wind fields. Thermodynamic parameters including monthly and daily mean Sea Surface Temperature (SST), Outgoing Longwave Radiation (OLR) and equatorial Standard Oscillation Index (SOI) were used. Three types of Poison regression models (i.e. dynamic, thermodynamic and combined models) were developed and validated using the Leave One Out Cross Validation (LOOCV). Moreover, 2 × 2 square matrix contingency tables for model verification were used. The results revealed that, the observed and cross validated DJFM and NM TCs and TSs strongly correlated with each other (p ≤ 0.02) for all model types, with correlations (r) ranging from 0.62 - 0.86 for TCs and 0.52 - 0.87 for TSs, indicating great association between these variables. Assessment of the model skill for all model types of DJFM and NM TCs and TSs frequency revealed high skill scores ranging from 38% - 70% for TCs and 26% - 72% for TSs frequency, respectively. Moreover, results indicated that the dynamic and combined models had higher skill scores than the thermodynamic models. The DJFM and NM selected predictors explained the TCs and TSs variability by the range of 0.45 - 0.65 and 0.37 - 0.66, respectively. However, verification analysis revealed that all models were adequate for predicting the seasonal TCs and TSs, with high bias values ranging from 0.85 - 0.94. Conclusively, the study calls for more studies in TCs and TSs frequency and strengths for enhancing the performance of the March to May (MAM) and December to October (OND) seasonal rainfalls in the East African (EA) and Tanzania in particular.

A deep learning model for forecasting global monthly mean sea surface temperature anomalies

Sea surface temperature (SST) variability plays a key role in the global weather and climate system, with phenomena such as El Niño-Southern Oscillation (ENSO) regarded as a major source of interannual climate variability at the global scale. The ability to make long-range forecasts of SST variations and extreme marine heatwave events have potentially significant economic and societal benefits, especially in a warming climate. We have developed a deep learning time series prediction model (Unet-LSTM), based on more than 70 years (1950–2021) of ECMWF ERA5 monthly mean SST and 2-m air temperature data, to predict global 2-dimensional SSTs up to a 24-month lead. Model prediction skills are high in the equatorial and subtropical Pacific. We have assessed the ability of the model to predict SST anomalies in the Niño3.4 region, an ENSO index in the equatorial Pacific, and the Blob marine heatwave events in the northeast Pacific in detail. An assessment of the predictions of the 2019–2020 El Niño and the 2016–2017 and 2017–2018 La Niña show that the model has skill up to 18 months in advance. The prediction of the 2015–2016 extreme El Niño is less satisfactory, which suggests that subsurface ocean information may be crucial for the evolution of this event. Note that the model makes predictions of the 2-d monthly SST field and Nino 3.4 is just one region embedded in the global field. The model also shows long lead prediction skills for the northeast Pacific marine heatwave, the Blob. However, the prediction of the marine heatwaves in the southeast Indian Ocean, the Ningaloo Niño, shows a short lead prediction. These results indicate the significant potential of data-driven methods to yield long-range predictions of SST anomalies.

Variability and trends of temperature and rainfall over three agro-ecological zones in North Shewa, Central Ethiopia

This study was conducted to analyze rainfall and temperature variability and trends over three Agro-Ecological Zones (AEZs) in Central Ethiopia. Global weather data for Soil and Water Assessment Tool (SWAT) and global mean monthly sea surface temperature (SST) data series were used for analysis. Mann–Kendall (MK) test was utilized to analyze rainfall and temperature trends. Sen’s slope estimator was employed to find out the rate of change. The study detected an annual increase of 0.07 °C (p < 0.001) in mean maximum temperature at Kolla AEZ. It also showed a 0.06 °C rise per annum (p < 0.001) for both Dega and Woina Dega AEZs. The average annual minimum temperature has increased by 0.03 °C per year at Kolla (p < 0.001), Woina Dega (p < 0.05), and Dega (p < 0.01), which means an increase of 1.05 °C between 1979 and 2013. Results from precipitation concentration index (PCI) revealed the highest percentage (97.1%) of irregular distributions in annual rainfall pattern at Kolla AEZ, followed by Woina Dega (82.9%). Standardized rainfall anomalies (SRAs) calculated in the study also showed higher percentage (28.6%) of drought in Kolla AEZ, which experienced drought once in every 3 or 4 years. The study revealed negative annual rainfall anomalies for 18 years in Kolla and 16 years in both Dega and Woina Dega AEZs. Rising temperatures and irregular rainfall patterns may induce some adverse impacts on the livelihoods of rural populations in general, their subsistence agricultural production processes and food availability in particular. Policymakers and stakeholders should give priority in designing and introducing pro-poor plus geographically differentiated adaptive strategies.

Application of Nonlinear Autoregressive Neural Network Model to Forecast Local Mean Sea Level

The primary purpose of this study was to apply the nonlinear autoregressive neural network to model the long-term records of the monthly mean sea level from January 1978 to October 2020 at Grand Isle, Louisiana, as extracted from the National Oceanic and Atmospheric Administration Tides and Currents database. In this study, the empirical results revealed that the Bayesian Regularization algorithm was the best-suit training algorithm for its high regression R-value and low mean square error compared to the Levenberg-Marquardt and Scaled Conjugate Gradient algorithms for the nonlinear autoregressive neural network. Understanding past sea levels is important for the analysis of current and future sea level changes. In order to sustain these observations, research programs utilizing the existing data should be able to improve our understanding and significantly narrow our projections of ffuture sea-level changes.

Simulation and Analysis of Sea-Level Change from Tide Gauge Station by Using Artificial Neural Network Models

Sea level change is one of the most certain results of global warming. Sea level change would increase erosion in coastal areas, result in intrusion into water supplies, inundate coastal marshes and other important habitats, and make the coastal property more vulnerable to erosion and flooding. This situation coincides with the massive socio-economic development of the coastal city areas. The coastal areas of the East Coast of Peninsular Malaysia are vulnerable to sea-level change, flooding, and extreme erosion events. The monthly Mean Sea Level (MSL) change was simulated by using two Artificial Neural Network (ANN) models, Feed Forward- Neural Network (FF-NN) and Nonlinear Autoregressive Exogenous- Neural Network (NARX-NN) models. Both models did well in recreating sea levels and their fluctuating patterns, according to the data. The NARX-NN model with architecture (5-6-1) and four lag options, on the other hand, got the greatest results. The findings of the model's mean sea level rise simulation show that Kuala Terengganu would have a growing and upward trend of roughly 25.34 mm/year. This paper shows that the eastern coast of Malaysia is highly vulnerable to sea-level rise and therefore, requires sustainable adaptation policies and plans to manage the potential impacts. It recommends that various policies, which enable areas to be occupied for longer before the eventual retreat, could be adapted to accommodate vulnerable settlements on the eastern coast of Malaysia.

Analysis of the Temporal and Spatial Changes of Sea Level in Korean Coastal Zone

In this study, the temporal and spatial changes of sea level in Korean coastal zone were analyzed. First, the error value of 1 hour tide data provided by the Korea Hydrographic and Oceanographic Agency was excluded through preprocessing, the analysis of sea level change was performed by considering long period components through low-pass filtering. Using low-pass filtering data, the monthly mean sea level was calculated and the temporal changes of sea level were verified by data fitting using linear fit and multidimensional smoothing spline. The results of linear fit show that the tide tends to rise at most stations. And the results of multidimensional smoothing spline fit show that the increase and decrease of sea level is repeated every year. Although the amounts of change vary every year, sea levels have been repeated seasonally, with sea levels rising in summer and falling in winter. Also, Hovmöller diagram was prepared for the change of sea level in each east, west, and south coast, and the temporal and spatial changes of sea level were analyzed. Compared to the past, recent sea level changes increased in all points. Especially, in the west coast, the sea level increased more in the south (Mokpo) area than in the north (Yeongjong Bridge), and the south (Mokpo) area of the west coast could be a vulnerable area due to the rise in sea level.

Time Series Modeling and Forecasting of Monthly Mean Sea Level (1978 – 2020): SARIMA and Multilayer Perceptron Neural Network

The primary purpose of this study was to demonstrate the role of time series model in predicting process and to pursue analysis of time series data using long-term records of monthly mean sea level from January 1978 to October 2020 at Grand Isle, Louisiana. Following the Box–Jenkins methodology, the ARIMA(1,1,1)(2,0,0)[12] with drift model was selected to be the best fitting model for the time series, according to the lowest AIC value in this study. Empirically, the results revealed that the MLP neural network model performed better compared to the ARIMA(1,1,1)(2,0,0)[12] with drift model at its smaller MSE value. Hence, the MLP neural network model not only can provided information which are important in decision making process related to the future sea level change impacts, but also can be employed in forecasting the future performance for local mean sea level change outcomes. Understanding past sea level is important for the analysis of current and future sea level changes. In order to sustain these observations, research programs utilizing the resulting data should be able to improve significantly our understanding and narrow projections of future sea level rise and variability.

Extreme Level Fluctuations in the Seas of the Russian Arctic in the Context of Global Climate Change

The presented article is devoted to the results of detailed studies of the features of extreme sea level fluctuations in the Russian Arctic under the conditions of global climate change. Estimates of the rate of mean sea level rise in the Arctic are obtained from the monthly mean sea level data at coastal tide gauges and satellite altimetry. It was found that for the Arctic Ocean coast from the Kara to the Chukchi Seas, the highest tides take place in the Khatan- ga Bay (Preobrazheniya Island), 131 cm. In the White, Kara, Laptev and Chukchi Seas, the seasonal variability of harmonic tidal constants was studied; it occurs due to changes in hydrometeorological conditions during the year. The ADCIRC hydrodynamic model was adapted for the conditions of the Laptev Sea. Numerical modeling of sea level oscillations from 1980 to 2000 was performed. According to the simulation results, the maximum ranges of residual sea level fluctuations are observed both for the total level and at the mouth of the Khatanga (489 cm) and Oleneksky Bay (444 cm). Based on the data of coastal observations and numerical modeling, the authors obtained estimates of extreme level fluctuations with different recurrence periods.

Опыт разработки методики долгосрочного прогноза сроков вскрытия ото льда рек бассейна р. Вятки

An attempt is made to develop a method for long-term forecasting of the ice breakup time for the Vyatka River basin, to identify the impact of the distribution of sea surface temperature and geopotential height in the informative regions at the levels H100 and H500 over the Northern Hemisphere on the river ice breakup. The location and boundaries of the informative regions in the fields of H100 and H500 were revealed by the discriminant analysis, the EOF expansion coefficients of the fields of anomalies of monthly mean values of H100 and H500 for January and February and the anomalies of monthly mean sea surface temperature in the North Atlantic and Northwest Pacific were used as potential predictors. The stepwise regression analysis allowed deriving good and satisfactory (S/σ = 0.45–0.73) complex prognostic equations for forecasting the ice breakup time for the Vyatka River basin. The essential influence of H100 and H500 geopotential height fields and the spatial distribution of sea surface temperature anomalies in the North Atlantic and Northwest Pacific in January and February on the river ice breakup time is revealed. It is proposed to improve the method by considering the impact of air temperature, maximum ice thickness per winter, and other indirect characteristics on the processes of river ice breakup in the Vyatka River basin. Keywords: ice regime, long-range forecast, river ice breakup, expansion coefficients, geopotential height fields, spring ice phenomena, energy-active zones of the oceans, complex prognostic equation

Опыт разработки методики долгосрочного прогноза сроков вскрытия ото льда рек бассейна р. Вятки

An attempt is made to develop a method for long-term forecasting of the ice breakup time for the Vyatka River basin, to identify the impact of the distribution of sea surface temperature and geopotential height in the informative regions at the levels H100 and H500 over the Northern Hemisphere on the river ice breakup. The location and boundaries of the informative regions in the fields of H100 and H500 were revealed by the discriminant analysis, the EOF expansion coefficients of the fields of anomalies of monthly mean values of H100 and H500 for January and February and the anomalies of monthly mean sea surface temperature in the North Atlantic and Northwest Pacific were used as potential predictors. The stepwise regression analysis allowed deriving good and satisfactory (S/σ = 0.45–0.73) complex prognostic equations for forecasting the ice breakup time for the Vyatka River basin. The essential influence of H100 and H500 geopotential height fields and the spatial distribution of sea surface temperature anomalies in the North Atlantic and Northwest Pacific in January and February on the river ice breakup time is revealed. It is proposed to improve the method by considering the impact of air temperature, maximum ice thickness per winter, and other indirect characteristics on the processes of river ice breakup in the Vyatka River basin. Keywords: ice regime, long-range forecast, river ice breakup, expansion coefficients, geopotential height fields, spring ice phenomena, energy-active zones of the oceans, complex prognostic equation

Sea level simulation with signal decomposition and machine learning

In recent years, under the background of global warming, influenced by changing climate conditions and human activities, the sea level time series have shown great non-stationarity and randomness, making it difficult to predict sea levels scientifically and accurately. Sea level simulations and predictions under climate change are hot topics in global environment and climate studies. In this paper, three prediction models based on signal decomposition (time-varying filtering based empirical mode decomposition (TVF-EMD), wavelet transform (WT), and complementary ensemble empirical mode decomposition (CEEMD) coupled with the Elman neural network (ENN) are established, and the prediction performances of the monthly mean sea level (MMSL, time series length between 16 and 61 years) at 12 stations and daily mean sea level (DMSL, time series length is 790 days) at six stations are evaluated. The results show that the MMSL has multiple time scales. Among the TVF-EMD-ENN, WT-ENN and CEEMD-ENN models, the TVF-EMD-ENN model is the optimal model for predicting the MMSL and DMSL, in MMSL and DMSL prediction, the Nash-Sutcliffe efficiency coefficient (NSE) range is 0.98–1.0 and 0.974–0.995, respectively, NSE is greater than 0.97 and performing better than the ENN and long short-term memory (LSTM) model. The difference in the prediction results between the TVF-EMD-ENN, WT-ENN, and CEEMD-ENN are mainly caused by the first subcomponent (Sc-1). The performance of the TVF-EMD-ENN model is optimal because the contribution of Sc-1 to the original sequence is small, and the complexity of the Sc-1 model is low. The predicted future MMSL exhibits a period of alternating sea level rise and decline, and the rate of increase is relatively low. The proposed approach has great significance for sea-level prediction and environmental management.

Sea Level Seasonal, Interannual and Decadal Variability in the Tropical Pacific Ocean

The satellite altimeter data, temperature and salinity data, and 1.5-layer reduced gravity model are used to quantitatively evaluate the contributions of the steric effect and the dynamic process to sea level variations in the Tropical Pacific Ocean (TPO) on different time scales. Concurrently, it also analyses the influence of wind forcing over the different regions of the Pacific Ocean on the sea level variations in the TPO. Seasonal sea level variations in the TPO were the most important in the middle and eastern regions of the 5°–15°N latitude zone, explaining 40–60% of the monthly mean sea level variations. Both the steric effect and dynamic process jointly affected the seasonal sea level variations. Among them, the steric effect was dominant, contributing over 70% in most regions of the TPO, while the dynamic process primarily acted near the equator and southwest regions, contributing approximately 55–85%. At the same time, the seasonal dynamic sea level variations were caused by the combined actions of primarily local wind forcing, alongside subtropical north Pacific wind forcing. On the interannual to decadal time scale, the sea level interannual variations were significant in the northwestern, southwestern, and middle eastern regions of the TPO and explained 45–60% of the monthly mean sea level variations. The decadal sea level variations were the most intense in the eastern Philippine Sea, contributing 25–45% to the monthly mean sea level variations. The steric effect and the dynamic process can explain 100% of the interannual to decadal sea level variations. The contribution of the steric effect was generally high, accounting for more than 85% in the regions near the equator. The impact of the dynamic process was mainly concentrated in the northwest, northeast, and southern regions of the TPO, contributing approximately 55–80%. Local wind forcing is the leading role of interannual to decadal sea level variations. The combined actions of El Niño–Southern Oscillation (ENSO) and the Pacific Decadal Oscillation (PDO) can explain 90% of the interannual to decadal sea level variations in the northwestern and eastern of the TPO.

Distribution patterns of intertidal oribatid mites (Acari, Oribatida) from South African shores and their relationship to temperature

A faunistic study of the intertidal oribatid mite fauna of South Africa’s coastline revealed the presence of four species from three families, showing specific biogeographic patterns. Their occurrences show a clear east–west divide, with a small gap near East London between the Podacaridae and the other two families, the Selenoribatidae and Fortuyniidae, clearly coinciding with suggested marine biogeographic ecoregions. The podacarid Halozetes capensis is confined to the cooler warm-temperate Agulhas Ecoregion; the fortuyniid Fortuynia elamellata micromorpha and the selenoribatid Schusteria ugraseni to the warmer subtropical Natal Ecoregion; and the selenoribatid Selenoribates divergens to the tropical Delagoa Ecoregion. These distributions are an indication that the oceanic climate may be the primary factor shaping their biogeography, particularly the seawater temperatures along the coast, which are in turn affected by the Agulhas and the Benguela currents. A mean monthly sea surface temperature of approximately 22 °C apparently represents the climatic border of the distributions, with the podacarid persisting only below this temperature, and the fortuyniid and selenoribatids only above it. Global warming will certainly change these geographic ranges, and in South Africa the warm-adapted fortuyniid and selenoribatids can be expected to expand their distributions southwards, while the occurrence of the cold-adapted podacarid might be reduced to a few southwestern coastal areas in the next few decades.

Increasing Trends in Air and Sea Surface Temperature in the Central Adriatic Sea (Croatia)

The Adriatic Sea and its coastal region have experienced significant environmental changes in recent decades, aggravated by climate change. The most prominent effects of climate change (namely, an increase in sea surface and air temperature together with changes in the precipitation regime) could have an adverse effect on social and environmental processes. In this study, we analyzed the time series of sea surface temperature and air temperature measured at three meteorological stations in the Croatian part of the Adriatic Sea. To assess the trends and variations in the time series of sea surface and air temperature, different statistical methods were employed, i.e., linear and quadratic regressions, Mann–Kendall test, Rescaled Adjusted Partial Sums method, and autocorrelation. The results evidenced increasing trends in the mean annual sea surface temperature and air temperature; furthermore, sudden variations in values were observed in 1998 and 1992, respectively. Increasing trends in the mean monthly sea surface temperature and air temperature occurred in the warmer parts of the year (from March to August). The results of this study could provide a foundation for stakeholders, decision–makers, and other scientists for developing effective measures to mitigate the negative effects of climate change in the scattered environment of the Adriatic islands and coastal region.

The relationship between the eddy-driven jet stream and northern European sea level variability

Wintertime sea level variability over the northern European continental shelf is largely wind-driven. Using daily gridded sea level anomaly from altimetry, we examine both the spatial and the temporal relationship between northern European sea level variability and large-scale atmospheric circulation patterns as represented by the jet cluster paradigm. The jet clusters represent different configurations of the eddy-driven jet stream and, therefore, provide a physical description of the atmospheric variability in the North Atlantic. We find that each of the four jet clusters is associated with a distinct northern European sea level anomaly pattern whose magnitudes are comparable to those of typical sea level variations on the shelf. In certain locations, such as the German Bight and the east coast of England, sea level anomalies are mainly associated with one single jet cluster. In other locations, such as the interior and the northern part of the North Sea, sea level anomalies are found to be sensitive to at least two jet configurations. Based on these regional sea level variations, we map out the locations on the shelf where each jet cluster or combination of clusters is most active before discussing the role of Ekman transport in inducing the resulting patterns. Through a multiple linear regression model, we also find that the jet clusters reconstruct up to 50% of the monthly mean sea level anomaly variance over the northern European continental shelf. The model best performs in the interior and the western part of the North Sea, suggesting that wind direction rather than wind speed plays a more prominent role over these regions. We conclude that the jet cluster approach gives valuable new insights compared to linear regression techniques for characterising wind-driven sea level variability over the northern European continental shelf.

Spatial and temporal variability of temperature, salinity and chlorophyll-a in the Magdalena River mouth, Caribbean Sea

Variations in the physico-chemical characteristics of estuaries, such as surface water salinity and temperature, lead to the establishment of gradients that are closely related with the distribution of nutrients and suspended sediment. This affects light penetration, which in turn influences Chlorophyll-a (Chl-a) concentration and primary productivity. We used MODIS imagery to identify spatio-temporal patterns of sea surface salinity, temperature, and Chl-a concentrations from 2003 to 2017, to explore relationships between these variables and oceanographic factors, such as streamflows, winds and currents in the mouth of Colombia's Magdalena River, which discharges to the Caribbean Sea. Sea surface salinity (SSS) in the study zone varied in time and space from estuarine to marine. Mean SSS was 10.8 ± 3.4 at Bocas de Ceniza, and 28.4 ± 0.4 in the Caribbean Sea, with the horizontal salinity gradient providing evidence for the existence of a salinity plume. Mean monthly sea surface temperatures (SST) averaged across all years were 27.6 ± 1.5 °C at Bocas de Ceniza, and 27.6 ± 1.3 °C in the Caribbean Sea. A significant, increasing trend in temperature was observed throughout the years of the study period. Average Chl-a values were 3.3 ± 1.4 mg m−3 at Bocas de Ceniza, and 1.5 ± 1.2 mg m−3 in the Caribbean Sea and the calculated average Trophic State Index (TSI) for Bocas de Ceniza indicated that the estuary trophic state varied between oligo-mesotrophic (30 < TSI≤40) and mesotrophic (40 < TSI≤50). The highest concentrations of Chl-a were found in intermediate salinities in the estuarine zone. Outside the saline plume, there is a considerable decrease in Chl-a concentrations (<0.5 mg m−3). Winds played the most important role in influencing spatio-temporal distribution of chemical and physical variables in the study zone. Our results emphasize the importance of physical processes on biological dynamics in the Magdalena River mouth.